Pathogenesis of Insulin Resistance and Atherogenic Dyslipidemia in Nonalcoholic Fatty Liver Disease Daud H

Home , Philip Augustine, Ribero

JCTH Associate Editors Dr. Masahiro Arai Prof. Yen-Hsuan Ni OWNED BY THE SECOND AFFILIATED HOSPITAL Nerima Hikarigaoka Hospital National Taiwan University OF CHONGQING MEDICAL UNIVERSITY Tokyo, Japan Taipei, Taiwan Dr. Roopjeet Bath Prof. Junqi Niu University of Connecticut Jilin University Farmington, USA Changchun, China Dr. Timothy Billiar Dr. Kittichai Promrat Editors-in-Chief University of Pittsburgh School of Medicine Alpert Medical School of Brown University Prof. Hong Ren Pittsburgh, USA Providence, USA General Editor-in-Chief Dr. John Birk Dr. Farzin Roohvand The Second Affiliated Hospital of Chongqing University of Connecticut Pasteur Institute of Iran Medical University, China Farmington, USA Tehran, Iran Dr. Harry Hua-Xiang Xia Dr. Wendy Cao Dr. Arielle Rosenberg Editor-in-Chief New York University Langone Health University Paris Descartes Novartis Pharmaceuticals Corporation, USA New York, USA Paris, France Prof. George Y. Wu Prof. Limin Chen Dr. Michael Schilsky Comprehensive Editor-in-Chief Peking Union Medical College Yale University University of Connecticut Heath Center, USA Chengdu, China New Haven, USA Prof. Aziz A. Chentoufi Dr. Ashwani Singal Mohammed VI University of Health University of South Dakota Managing Editors Sciences Sioux Falls, USA Casablanca, Morocco Huaidong Hu Dr. Qing-Feng Sun Chongqing, China Prof. Mohamed A Daw The Third Affiliated Hospital to Wenzhou University of Tripoli Medical College Sandeep Kumar Karn Tripoli, Libya Wenzhou, China Chongqing, China Prof. Xiaoguang Dou Dr. Tawesak Tanwandee Shengjing Hospital of China Medical Siriraj Hospital, Mahidol University University Bangkok, Thailand Executive Editor Shenyang, China Dr. Fusheng Wang Hua He Prof. Marko Duvnjak 302 Miltitary Hospital of China Houston, USA Clinical Hospital Centre “Sestre milosrdnice” Beijing, China Zagreb, Croatia Prof. Lai Wei Technical Editor Dr. Yu-Chen Fan Tsinghua University Qilu Hospital of Shandong University Beijing, China Wanchen Hu Jinan, China Wuhan, China Prof. Catherine Wu Prof. Jinlin Hou University of Connecticut Nanfang Hospital of Southern Medical Farmington, USA University Guangzhou, China Dr. Xuefeng Xia Third Affiliated Hospital of Guangzhou Prof. Keqin Hu Medical University University of California Guangzhou, China Orange, USA Prof. Jidong Jia Prof. Manfung Yuen Capital Medical University The University of Hong Kong Queen Mary Contact information Beijing, China Hospital Hong Kong, China Telephone +86 23 63727251 Prof. Douglas LaBrecque Fax +86 23 63701383 University of Iowa Prof. Dazhi Zhang Lowa, USA The Second Affiliated Hospital of Chongqing E-mail [email protected] Medical University Dr. Joseph Lim Chongqing, China Address 74 Linjiang Road, Yale University Yuzhong District, New Haven, USA Prof. Lanjing Zhang Chongqing, P. R. China, Prof. Chao-Hong Liu University Medical Center of Princeton Plainsboro, USA 400010 Huazhong University of Science and Technology Dr. Minghua Zheng Wuhan, China The First Affiliated Hospital of Wenzhou Prof. Lungen Lu Medical University Wenzhou, China Publisher Shanghai Jiaotong University School of Xia & He Publishing Inc. Medicine Shanghai, China 14090 Southwest Freeway, Suite 300, Sugar Land, Texas 77478, USA JOURNAL OF CLINICAL AND TRANSLATIONAL HEPATOLOGY

Beginning with 2018, JCTH has been indexed and abstracted in Emerging Sources Citation Index (ESCI).

CONTENTS 2019 7(4):285–393

Editorial

Can Weather Influence the Prevalence of Acute-on-chronic Liver Failure? Matthew McMillin ...... 285

Original Articles

Intrahepatic Cholangiocarcinoma and Thermal Ablation: Long-term Results of An Italian Retrospective Multicenter Study Antonio Giorgio, Pietro Gatti, Luca Montesarchio, Bruno Santoro, Andrea Dell’Olio, Nicola Crucinio, Carmine Coppola, Ferdinando Scarano, Fabio De Biase, Emanuela Ciracì, Stefano Semeraro and Valentina Giorgio ...... 287

Alanine Aminotransferase as the First Test Parameter for Wilson’s Disease Hisao Hayashi, Kazumasa Watanabe, Ayano Inui, Ayako Kato, Yasuaki Tatsumi, Akihiko Okumura, Tomoo Fujisawa and Koichi Kato ...... 293

Association of GCKR Gene Polymorphisms with the Risk of Nonalcoholic Fatty Liver Disease and Coronary Artery Disease in a Chinese Northern Han Population Hui Gao, Shousheng Liu, Zhenzhen Zhao, Xinjuan Yu, Qun Liu, Yongning Xin and Shiying Xuan ...... 297

Minimal Hepatic Encephalopathy in Indians: Psychometric Hepatic Encephalopathy Score and Inhibitory Control Test for Diagnosis and Rifaximin or Lactulose for Its Reversal Vinay B. Pawar, Ravindra G. Surude, Nikhil Sonthalia, Vinay Zanwar, Samit Jain, Qais Contractor and Pravin M. Rathi ...... 304

Liver Stiffness Measurement Can Reflect the Active Liver Necroinflammation in Population with Chronic Liver Disease: A Real-world Evidence Study LeijieWang, Mingyu Zhu, Lihua Cao, Mingjie Yao, Yiwei Lu, XiajieWen, Ying Zhang, Jing Ning, Huiling Long, Yueyong Zhu, Guoxin Hu, Shuangsuo Dang, Qingchun Fu, Liang Chen, Xinxin Zhang, Jingmin Zhao, Zhiliang Gao, Yuemin Nan and Fengmin Lu ...... 313 Baseline Characteristics and Treatment Patterns of the Patients Recruited to the China Registry of Hepatitis B Shan Shan, Hong You, Junqi Niu, Jia Shang, Wen Xie, Yuexin Zhang, Xun Li, Hong Ren, Hong Tang, Huiguo Ding, Xihong Wang, Yuemin Nan, Xiaoguang Dou, Tao Han, Lingyi Zhang, Xiaoqing Liu, Cunliang Deng, Jilin Cheng, XiaozhongWang, Qing Xie, Shumei Lin, Yan Huang, Youqing Xu, Yong Xiong, Wu Li, Xuebing Yan, Hongxin Piao, Wenxiang Huang, Qinghua Lu, Weijin Gong, Shiping Li, Xiaoxuan Hu, Xiaolan Zhang, Shourong Liu, Yufang Li, Dongliang Yang, Hai Li, Caixia Yang, Mingliang Cheng, Liaoyun Zhang, Huanwei Zheng, Xinhua Luo, Feng Lin, Lei Wang, Guanghua Xu, Xiaoyuan Xu, Lai Wei, Jinlin Hou, Zhongping Duan, Hui Zhuang, Xizhong Yang, Yuanyuan Kong and Jidong Jia for the CR-HepB study group, Beijing, China...... 322

Outcomes and Toxicology of Herbal Drugs in Alcoholic Hepatitis – A Single Center Experience from India Cyriac Abby Philips, Sasidharan Rajesh, Tom George, Rizwan Ahamed, Sandeep Kumbar and Philip Augustine . . . 329

The Association between Meteorological Factors and the Prevalence of Acute-on-chronic Liver Failure: A Population-based Study, 2007–2016 Su Lin, Lifen Han, Dongliang Li, Ting Wang, Zimu Wu, Haoyang Zhang, Zhansong Xiao, Yinlian Wu, JiaofengHuang,MingfangWangandYueyongZhu...... 341

Review Articles

Highlights for Dental Care as a Hepatitis C Risk Factor: A Review of Literature Leon D. Averbukh and George Y. Wu ...... 346

The Direct Contribution of Astrocytes and Microglia to the Pathogenesis of Hepatic Encephalopathy Victoria Jaeger, Sharon DeMorrow and Matthew McMillin...... 352

Pathogenesis of Insulin Resistance and Atherogenic Dyslipidemia in Nonalcoholic Fatty Liver Disease Daud H. Akhtar, Umair Iqbal, Luis Miguel Vazquez-Montesino, Brittany B. Dennis and Aijaz Ahmed ...... 362

Role of Granulocyte Colony-stimulating Factor Therapy in Cirrhosis, ‘Inside Any Deep Asking Is the Answering’ Cyriac Abby Philips, Philip Augustine, Rizwan Ahamed, Sasidharan Rajesh, Tom George, Gopakumar C. Valiathan and Solomon K. John ...... 371

Intensive Care Management of Acute Liver Failure: Considerations While Awaiting Liver Transplantation Anil Seetharam...... 384

Reviewer Acknowledgement

2019 Reviewer Acknowledgement ...... 392 Editorial

Can Weather Influence the Prevalence of Acute-on-chronic Liver Failure?

Matthew McMillin*1,2

1Central Texas Veterans Health Care System, Temple, TX, USA; 2Department of Internal Medicine, University of Texas at Austin Dell Medical School, Austin, TX, USA

Citation of this article: McMillin M. Can weather influence the that meteorological factors could have an influence on ACLF. prevalence of acute-on-chronic liver failure? J Clin Transl In regard to this, it is important to understand whether the Hepatol 2019;7(4):285-286. doi: 10.14218/JCTH.2019.00058. direct effects of meteorological changes are inducing ACLF pathology or if these changes are influencing a secondary Acute-on-chronic liver failure (ACLF) results from an acute factor that increases ACLF prevalence. insult that occurs in patients with chronic liver disease, In regard to the direct influences of meteorological ultimately leading to liver failure. The underlying chronic changes on ACLF, hyperthermia has been associated with liver dysfunction can be characterized into three categories: hepatocyte dysfunction and can lead to a disruption in chronic liver disease with no cirrhosis; compensated liver mitochondria function and reduced oxidative phosphoryla- 6 cirrhosis; decompensated cirrhosis. The underlying chronic tion. In addition, during both non-exertional and exertional liver diseases are numerous and include alcoholic liver hyperthermia, acute liver failure has been observed, though disease, viral hepatitis (types B and C), nonalcoholic fatty this occurrence is relatively rare.7 In regards to lower tem- liver disease, and other underlying etiologies.1 Acute liver peratures, therapeutic hypothermia has been used as a injury can arise from viral hepatitis infections (types A, B, strategy primarily to reduce intracranial hypertension in and E), acetaminophen toxicity, drug-induced liver injury, patients with acute liver failure. However, a recent multicen- toxin exposure, and numerous other hepatic and extrahepatic ter clinical study investigating the use of therapeutic hypo- causes including heat stroke.2 Due to the numerous aspects thermia observed no benefit in reducing brain swelling or of pathology that can generate both the chronic and acute increasing the survival rate of acute liver failure patients.8 components of ACLF, these patients can have significantly Therefore, as the study by Lin et al. found that lower temper- different pathology profiles. This makes ACLF difficult to clas- atures were associated with increased prevalence of ACLF, it sify, and this is evident by the different ACLF definitions being was likely not due to any direct effect of hyperthermia or used by the Asian Pacific Association for the Study of the Liver, hypothermia.5 the European Association for the Study of the Liver – Chronic While the influence of seasonal meteorological factors on Liver Failure Consortium, and the North American Consortium liver diseases is an area that has not been extensively for the Study of End-Stage Liver Disease with regards to mor- studied, there is some evidence of seasonality towards the tality, presence of organ failure, inclusion/exclusion criteria, viral hepatitis infections of hepatitis A, B, C, and E as they are and other disease characteristics.3,4 In addition, these differ- more prominent in the spring and summer.9 In addition, ent definitions of ACLF, depending on which definition is used, studies outside the liver have shown bacterial infections can could potentially influence the outcomes of studies depending be influenced by weather as gastroenteritis hospitalizations in upon the metric investigated. Spain were studied and both hot and cold temperatures In the current issue, Lin et al. describe that meteorological increased risk, while precipitation reduced the incidence of patterns can have an influence on the prevalence of ACLF, hospitalization.10 The authors found that certain pathogens which has previously not been investigated.5 In the 3,510 were found in different temperatures, such as rotavirus cases that the authors studied from three hospitals in being associated with cold temperatures, and salmonella Fuzhou City, China, there was a positive correlation of and other foodborne infections agents associated with high average humidity and a negative correlation of temperature temperatures.10 Bacterial infections and bacterial transloca- with the prevalence of ACLF.5 In addition, the highest preva- tion from the gut have been associated with ACLF and bacte- lence of ACLF was observed in January, followed by July.5 rial infections are more common in patients with cirrhosis Rainfall and change of temperature did not affect the inci- compared to patients without cirrhosis.11 Therefore, dence of ACLF.5 The authors conclude that low temperature changes in meteorological factors could increase the rate of and high humidity contribute to the seasonal pattern of viral and bacterial infections in patients with chronic liver ACLF observed in their patient population. This supports disease causing them to progress to ACLF. That being said, more research is needed in this area to identify if this is a causal component of ACLF. Abbreviations: ACLF, acute-on-chronic liver failure. In conclusion, the study by Lin et al. identifies low temper- Received: 12 December 2019; Accepted: 13 December 2019 ature and high humidity as risk factors for ACLF.5 Due to the *Correspondence to: Matthew McMillin, University of Texas at Austin Dell Medical School, 1601 Trinity Street, Building B, Austin, TX 78701, USA. Tel: +1-512-495-5037, lack of studies in this area, it is difficult to know the exact Fax: +1-512-495-5389, E-mail: [email protected] mechanisms of these meteorological factors on liver disease

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 285–286 285

286 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 285–286 Original Article

Intrahepatic Cholangiocarcinoma and Thermal Ablation: Long-term Results of An Italian Retrospective Multicenter Study

Antonio Giorgio*1, Pietro Gatti2, Luca Montesarchio3, Bruno Santoro1, Andrea Dell’Olio4, Nicola Crucinio5, Carmine Coppola6, Ferdinando Scarano6, Fabio De Biase5, Emanuela Ciracì7, Stefano Semeraro7 and Valentina Giorgio8

1Liver Unit and Interventional Ultrasound Unit, Athena Clinical Institute, Piedimonte Matese (CE), Italy; 2Internal Medicine Unit, Brindisi General Hospital, Brindisi, Italy; 3General Surgery Unit, Polla General Hospital, Polla (SA), Italy; 4Department of Radiology, Bisceglie Hospital - ASL BAT, Bisceglie, Italy; 5Gastroenterology Unit, Foggia General Hospital, Foggia, Italy; 6Department of Internal Medicine, Hepatology Interventional Unit, Gragnano Hospital, Gragnano (NA), Italy; 7Internal Medicine Unit, Ostuni Hospital, Ostuni (BR), Italy; 8Fondazione Policlinico A. Gemelli IRCCS, Department of Woman and Child Health and Public Health, Largo A Gemelli, Roma, Italy

Abstract Citation of this article: Giorgio A, Gatti P, Montesarchio L, Santoro B, Dell’Olio A, Crucinio N, et al. Intrahepatic cholan- Background and Aims: Despite resection being considered giocarcinoma and thermal ablation: long-term results of an the treatment of choice for intrahepatic cholangiocarcinoma Italian retrospective multicenter study. J Clin Transl Hepatol (ICC), percutaneous thermal ablation can be an alter- 2019;7(4):287–292. doi: 10.14218/JCTH.2019.00036. native treatment for patients unfit for surgery. Our aim was to compare long-term results of percutaneous sonographi- cally-guided radiofrequency ablation (RFA) with high- Introduction powered microwave ablation (MWSA) in treatment of ICC. Methods: Results of 71 ICC patients with 98 nodules treated Intrahepatic cholangiocarcinoma (ICC) is the second most with RFA (36 patients) or MWSA (35 patients) between common primary malignant tumor of the liver.1,2 ICC January 2008 and June 2018 in 5 Interventional Ultrasound accounts for up to 8–10% of all cholangiocarcinomas and centers of Southern Italy were retrospectively reviewed. Cu- 10–20% of all primary liver tumors.1 Mortality from ICC has mulative overall survival curves were calculated with the Ka- risen globally,2 and recent data have indicated that the occur- plan-Meyer method and differences with the log-rank test. rence of ICC on cirrhosis is increasing.1,2 ICC is considered Eleven possible factors affecting survival were analyzed. more aggressive than hepatocellular carcinoma. Resection is Results: Overall survival of the entire series was 88%, the treatment of choice because it increases survival.3 65%, 45% and 34% at 12, 36, 60 and 80 months, respec- Nevertheless, some researchers have recently reported on tively. Patients treated with MWSA survived longer than pa- ICC patients unfit for surgery who were treated with percuta- tients treated with RFA (p < 0.005). The MWSA group with neous thermal ablation, initially radiofrequency and later # – ICC nodules 3 cm or nodules up to 4 cm survived longer microwaves.4 10 than the RFA group (p < 0.0005). In patients with nodules According to the literature, survival of patients with small >4 cm, no significant difference was found. Disease-free sur- (<3 cm) tumors was better than that of patients with larger vival and progression-free survival were better in the MWSA tumors. In 2017, Shindoh11 reviewed published studies on group compared to the RFA group (p < 0.005). Diameter of ablative therapies of ICC, which at that time were limited in nodules and MWSA were independent factors predicting a number and with small numbers of treated patients. On the better survival. No major complications were observed. basis of literature data and upon the experience of his Hospi- Conclusions: MWSA is superior to RFA in treating ICC unfit tal Institution, this author proposed an algorithm that consid- for surgery, achieving better long-term survival in small ered ablation as an alternative treatment option to surgery in # ( 3 cm) ICC nodules as well as nodules up to 4 cm of neo- unresectable ICC patients. In this algorithm, radiofrequency plastic tumors and should replace RFA. ablation (RFA) appeared to be the technique of choice.11 We also commented12 on the proposal of Shindoh and agreed Keywords: Intrahepatic cholangiocarcinoma; Thermal ablation; Microwaves; that “only limited evidence from a small number of studies Radiofrequency; Percutaneous treatment. has been reported because of the rarity of this tumor”, and Abbreviations: CECT, contrast-enhanced computed tomography; CEMRI, con- that thermal “ablation therapies could be the treatment of trast-enhanced magnetic resonance imaging; CEUS, contrast-enhanced ultra- ” sound; ICC, cholangiocarcinoma; MWSA, microwaves ablation; OS, overall choice for selected cases of ICC . survival; RFA, radiofrequency ablation; US, ultrasound. Nevertheless, we concluded that “most likely RFA should Received: 14 August 2019; Revised: 15 October 2019; Accepted: 6 November be replaced by microwave ablation (MWSA) that is able to 2019 ensure a complete ablation of tumor nodules <3–3.5 cm, *Correspondence to: Antonio Giorgio, Liver Unit and Interventional Ultrasound Unit, Athena Clinical Institute, Piedimonte Matese (CE) (81016), Italy. Tel: +39- determining a sufficient ablative margin of the tumor just for 12 823784666, E-mail: [email protected] the properties inherent in the technique”. In fact, compared

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 287–292 287

Table 1. Baseline characteristics and laboratory findings of patients treated with MWSA and patients treated with RFA

Patients’ characteristics MWSA group, n = 35 RFA group, n =36

Age in years, mean 6 SD 72 6 10 75 6 8 Male (%) 18 (51.4%) 19 (52.8%) Comorbidities (%) None 6 (17.1%) 8 (22.2%) Obesity 13 (37.1%) 11 (30.5%) HCV infection 13 (37.1%) 15 (41.7%) HBV infection 3 (8.6%) 2 (5.6%) Child A cirrhosis 7 (20%) 8 (22.2%) Child B7 cirrhosis 1 (2.8%) 1 (2.8%) Treatment history (%) Primary treatment 30 (85.7%) 32 (88.9%) Post-surgical recurrence 5 (14.3%) 4 (11.1%) Laboratory findings (range) Platelet count in cells/dL, median 97500 (52000–178000) 98500 (56000–176500) INR, median 1.1 (0.9–1.68) 1.2 (0.9–1.70) Ca 19.9 in IU/mL, median 53 (16–250) 57 (15–245) AFP in mg/dL, median 21 (8–54) 23 (10–56)

Abbreviations: AFP, alpha-fetoprotein; HCV, hepatitis B virus; HCV, hepatitis C virus; INR, international normalized ratio; MWSA, microwave ablation; RFA, radiofrequency ablation; SD, standard deviation.

288 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 287–292 Giorgio A. et al: Ablation of intrahepatic cholangiocarcinoma

Table 2. The enhancement pattern of the intrahepatic cholangiocarcinoma treated with MWSA during the arterial phase on CEUS and CECT/CEMR, and according to lesion size

CEUS (%)

CECT or CEMRI Type I Type II Type III Type IV Total

Type I 29 (58) 1 (2) 0 (0) 0 (0) 30 (60) Type II 0 (0) 14 (28) 0 (0) 2 (4) 16 (32) Type III 2 (4) 0 (0) 2 (4) 0 (0) 4 (8) Type IV 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) Total 31 (62) 15 (30) 2 (4) 2 (4) 50 Size, median (range): 3.6 cm (2.2–7.2 cm) #3.0 cm 5 (10) 0 (0) 2 (4) 0 (0) 7 (14) 3.1–5.0 cm 19 (38) 9 (18) 0 (0) 0 (0) 28 (56) >5 cm 7 (14) 6 (12) 0 (0) 2 (4) 15 (30) Total 31 (62) 15 (30) 2 (4) 2 (4) 50

Type I (peripheral irregular rim-like hyperenhancement); Type II (diffuse heterogeneous hyperenhancement); Type III (diffuse homogeneous hyperenhancement); Type IV (diffuse heterogeneous hypoenhancement). Abbreviations: CEMRI, contrast-enhanced magnetic resonance imaging; CETC, contrast-enhanced computed tomography; CEUS, contrast-enhanced ultrasound; MWSA, microwave ablation.

Statistical analysis value, number of nodules, diameter of nodules, surgical recurrence, primary treatment, and type of treatment). Var- The quantitative data were expressed as mean±standard iables with a p value #0.10 on univariate analysis were deviation. Continuous variables and categorical variables included in multivariate regression analysis by using Cox’s were compared using student’s t test and chi-square test, stepwise regression. Statistical significance was defined by respectively. Cumulative overall survival (OS) rate was esti- a p value <0.05 in a two-tailed test. Statistical analysis was mated in months using the Kaplan-Meier method, from the performed using the SPSS 22.0 statistical software (SPSS beginning of the ablation to death or the last visit, and the Inc., Chicago, IL, USA). differences between subgroups of patients in univariate analysis were determined using the log-rank test. Eleven possible Results factors affecting survival were analyzed after their transformation in dichotomous variables (i.e. age, sex, presence of cir- The clinical, laboratory and radiological characteristics were rhosis, total bilirubin level, Ca 19-9 value, alpha-fetoprotein comparable between the two groups (p > 0.05). The follow-up

Table 3. The enhancement pattern of the intrahepatic cholangiocarcinoma treated with RFA during the arterial phase on CEUS and CECT/CEMRI, and according to lesion size

CEUS (%)

CECT or CEMRI Type I Type II Type III Type IV Total

Type I 27 (56.2) 4 (8.3) 0 (0) 0 (0) 31 (64.5) Type II 0 (0) 9 (18.7) 0 (0) 2 (4.2) 11 (22.9) Type III 2 (4.2) 0 (0) 4 (8.3) 0 (0) 6 (12.5) Type IV 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) Total 29 (60.4) 13 (27.0) 4 (8.3) 2 (4.2) 48 Size, median (range): 3.1 cm (2–8 cm) #3.0 cm 4 (8.3) 0 (0) 4 (8.3) 0 (0) 8 (16.7) 3.1–5.0 cm 22 (45.8) 10 (20,8) 0 (0) 0 (0) 32 (66.7) >5 cm 3 (6.2) 3 (6.2) 0 (0) 2 (4.2) 8 (16.7) Total 29 (60.4) 13 (27.0) 4 (8.3) 2 (4.2) 48

Type I (peripheral irregular rim-like hyperenhancement);Type II (diffuse heterogeneous hyperenhancement); Type III, diffuse homogeneous hyperenhancement); Type IV (diffuse heterogeneous hypoenhancement). Abbreviations: CEMRI, contrast-enhanced magnetic resonance imaging; CETC, contrast-enhanced computed tomography; CEUS, contrast-enhanced ultrasound; RFA, radiofrequency ablation.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 287–292 289 Giorgio A. et al: Ablation of intrahepatic cholangiocarcinoma

Fig. 3. Overall survival of patients with nodules ≤3 cm treated with Fig. 1. Overall Survival of all patients. MWSA vs. patients treated with RFA. Abbreviations: MWSA, microwave ablation; RFA, radiofrequency ablation. of the study ranged between 8 and 86 months (median: 48). OS of all patients at 12, 36, 60 and 80 months was 88%, 65%, 45% and 34%, respectively (Fig. 1). Patients treated with and 58% at 12, 36, 60 and 80 months, respectively, vs. MWSA survived longer than patients treated with RFA (OS 86%, 29%, 14 and 0%, p < 0.005) (Fig. 3). In the subgroup was 95%, 75%, 68% and 68% at 12, 36, 60 and 80 of patients with ICC nodules #4 cm, OS of patients treated with months, respectively, in the MWSA group and 86%, 53%, MWSA was better than those treated with RFA (p <0.0005) 26% and 13%, respectively, in the RFA group (p < 0.005) (Fig. 4). This statistically significant difference was no longer (Fig. 2). OS of patients with nodules #3 cm was better in evident when diameter of nodules was >4 cm (p = 0.25). the MWSA group than in the RFA group (93%, 67%, 60% Disease-free survival was better in the MWSA-treated group, with respect to the RFA-treated group (p < 0.005)

Fig. 2. Overall survival of patients treated with MWSA vs. patients treated Fig. 4. Overall survival of patients with nodules ≤4 cm treated with with RFA. MWSA vs. patients treated with RFA. Abbreviations: MWSA, microwave ablation; RFA, radiofrequency Abbreviations: MWSA, microwave ablation; RFA, radiofrequency ablation. ablation.

290 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 287–292 Giorgio A. et al: Ablation of intrahepatic cholangiocarcinoma

Table 4. Significant variables for overall survival in the univariate and multivariate analysis

Univariate analysis, Multivariate analysis, Significant variables OR (95% CI) p OR (95% CI) p

Age in years: #60 vs. >60 1.122 (0.738–1.727) Male sex: yes/no 0.929 (0.532–1.1791) Cirrhosis: yes/no 0.989 (0.841–2.788) 0.002 Total bilirubin level in mg/dL: #2.5 vs. >2.5 3.020 (1.631–5.579) Ca 19.9 in IU/mL: #40 vs. >40 1.149 (0.748–1.729) AFP in mg/dL: #20 vs. >20 1.151 (0.751–1.1734) Size of nodules in cm: #3 vs. >3 1.161 (0.750–1.359) 0.003 1.527 (1.868–2.077) 0.002 Number of nodules 1 vs. >1 1.37 (0.791–1.733) 0.003 Primary treatment vs. surgical recurrence 0.989 (0.754–1.311) 0.986 MWSA vs. RFA 2.079 (1.191–2.730) 0.002 1.899 (1.227–2.888) 0.001

Abbreviations: AFP, alpha-fetoprotein; CI, confidence interval; MWSA, microwave ablation; OR, odds ratio; RFA, radiofrequency ablation.

(Supplementary Fig. 1). Progression-free survival was 79%, groups (RFA and MWSA) was clearly unbalanced in favor of 59%, 55% and 55% at 12, 36, 60 and 80 months, respec- the RFA (88% of nodules). tively, in the MWSA group and 69%, 51%, 8.5% and 8.5 % at Zhang et al.18 reported the clinical and survival outcomes 12, 36, 60 and 80 months, respectively, in the RFA group, of 101 ICC patients treated with MWSA between 2009 and with a significant statistical difference (p < 0.005) (Supple- 2016. According to their results, Child-Pugh class of cirrhosis mentary Fig. 2). On univariate analysis, the presence of and the number of nodules were independent factors affect- cirrhosis, diameter of nodules, size of nodules, and type of ing survival. Xu et al.19 reported a retrospective comparison treatment were independent factors affecting survival. On between MWSA and surgical resection in patients with recur- multivariate analysis, only diameter of nodules and type of rent ICC. They concluded that percutaneous MWSA can be treatment were independent factors predicting survival considered comparable with surgical resection and should (Table 4). No major complications were observed after proce- be a valid alternative to surgery in treating recurrent ICC dures and the hospital stay ranged between 1 and 4 days after hepatectomy. Our results, although retrospective, (mean 1.7). seem to show that MWSA could be a valid alternative to surgery not only in patients with nodules #3 cm but also in the subgroup of ICC patients with nodules between 3 and Discussion 4 cm in diameter. Our study has many limitations. First, it is a retrospective Our retrospective study showed that MWSA is superior to RFA study, with all the implications that this has in the analysis of in treatment of patients with ICC not amenable to surgical the results. In fact, it is well known that retrospective studies resection. Furthermore, results of our series show that MWSA give less reliable results compared to prospective studies. achieves a better survival in patients with nodules #3cmas Moreover, in our study the indication of procedure (i.e. RFA or well as those up to 4 cm, as compared to RFA. In our opinion, MWSA) was not chosen in a randomized manner but on a this can be explained considering the intrinsic characteristics case-by-case basis depending on the patient and the oper- of MWSA with respect to RFA. In fact, compared to RFA, ator. Second, the type of ablation devices used in our series MWSA induces a larger volume of necrosis in a faster time and was quite different among the five different Interventional determinates a spherical shape of the ablated area.13,14 Units and this could have affected the results. It should also Therefore, from an interventional point of view, it is conceiv- be considered that, at the beginning of the study, the skills of able that MWSA, inducing a better local tumor control, allows the physicians involved in the ablation procedures could have an increased survival, both considering the results in terms of been not exactly the same. Finally, only a single electrode OS and considering the disease-free survival and progression- needle (in the case of RFA) and only a single antenna (in the free survival rates of our series. case of MWSA) were used in all procedures. Probably, the use To our knowledge, there are no studies comparing the two of cluster RF electrode needles or more than one MWS thermal techniques in treating ICC not amenable for surgical antenna could have induced a better local tumor control and resection. Generally, previous studies have reported only a better survival in cases of larger nodules. On the other hand, results using one of the two techniques in treating ICC it is conceivable that when using multiple applicators, the risk patients. Recently, however, Takahashi et al.17 retrospectively of bleeding would be doubled or tripled while, in our series, no reviewed the outcome of 20 patients with 50 ICC nodules bleeding event was observed. (mean size: 1.8 ± 1.3 cm) treated mainly with RFA (88% of In conclusion, in our experience, MWSA seems to be nodules) or MWSA between 2006 and 2015. In their experi- superior to RFA in treatment of ICC nodules in patients ence, there was no difference between the two techniques unfit for surgical resection, in nodules #3 cm as well as in for local tumor progression. We must consider that in the nodules up to 4 cm, inducing better OS and better local tumor study of Takahashi and colleagues,17 the ICC-treated control. Because our results were obtained from a real-life nodules were too small and the difference between the two retrospective study and from a limited number of centers,

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 287–292 291 Giorgio A. et al: Ablation of intrahepatic cholangiocarcinoma further prospective controlled randomized trials with a larger [6] Fu Y, Yang W, Wu W, Yan K, Xing BC, Chen MH. Radiofrequency ablation in the number of patients are needed to confirm our results. management of unresectable intrahepatic cholangiocarcinoma. J Vasc Interv Radiol 2012;23:642–649. doi: 10.1016/j.jvir.2012.01.081. [7] Giorgio A, Calisti G, DE Stefano G, Farella N, DI Sarno A, Amendola F, et al. Acknowledgments Radiofrequency ablation for intrahepatic cholangiocarcinoma: retrospective analysis of a single centre experience. Anticancer Res 2011;31: 4575–4580. The authors thank Dr. Massimo De Luca, M.D., Liver Unit at [8] Kim JH, Won HJ, Shin YM, Kim KA, Kim PN. Radiofrequency ablation for the Cardarelli Hospital, Naples, Italy, for his assistance in the treatment of primary intrahepatic cholangiocarcinoma. AJR Am J Roentgenol preparation of the manuscript. 2011;196:W205–W209. doi: 10.2214/AJR.10.4937. [9] Xu HX, Wang Y, Lu MD, Liu LN. Percutaneous ultrasound-guided thermal ablation for intrahepatic cholangiocarcinoma. Br J Radiol 2012;85:1078– Conflict of interest 1084. doi: 10.1259/bjr/24563774. [10] Mizrahi JD, Abdel-Wahab R, Javle M. Clinical trials and novel/emerging The authors have no conflict of interests related to this treatment. In: Pawlik T, Cloyd J, Dillhoff , editors. Intrahepatic Cholangiocar- cinoma. Springer, Cham; 2019:183–208. doi: 10.1007/978-3-030-22258- publication. 1_13. [11] Shindoh J. Ablative therapies for intrahepatic cholangiocarcinoma. Hepato- Author contributions biliary Surg Nutr 2017;6:2–6. doi: 10.21037/hbsn.2016.09.07. [12] Giorgio A, Gatti P, Matteucci P, Giorgio V. Ablative therapies for intrahepatic cholangiocarcinoma. Hepatobiliary Surg Nutr 2018;7:192–194. doi: 10. Study concept and design (AG), acquisition of data (AG, PG, 21037/hbsn.2018.02.06. LM), analysis and interpretation of data (AG, PG, LM), drafting [13] Giorgio A, Gatti P, Montesarchio L, Merola MG, Amendola F, Calvanese A, of the manuscript (AG, VG), critical revision of the manuscript et al. Microwave ablation in intermediate hepatocellular carcinoma in cirrho- for important intellectual content, and administrative, tech- sis: An Italian multicenter prospective study. J Clin Transl Hepatol 2018;6: 251–257. doi: 10.14218/JCTH.2018.00013. nical, or material support (AG, PG, LM, BS, AD, NC, CC, FS, [14] Poggi G, Montagna B, DI Cesare P, Riva G, Bernardo G, Mazzucco M, et al. FDB, EC, SS, VG), study supervision (AG, PG,VG). Microwave ablation of hepatocellular carcinoma using a new percutaneous device: preliminary results. Anticancer Res 2013;33:1221–1227. [15] Chen LD, Xu HX, Xie XY, Lu MD, Xu ZF, Liu GJ, et al. Enhancement patterns of References intrahepatic cholangiocarcinoma: comparison between contrast-enhanced ultrasound and contrast-enhanced CT. Br J Radiol 2008;81:881–889. doi: [1] Gupta A, Dixon E. Epidemiology and risk factors: intrahepatic cholangiocar- 10.1259/bjr/22318475. cinoma. Hepatobiliary Surg Nutr 2017;6:101–104. doi: 10.21037/hbsn. [16] Giorgio A, Merola MG, Montesarchio L, Merola F, Gatti P, Coppola C, et al. 2017.01.02. Percutaneous radiofrequency ablation of hepatocellular carcinoma in cirrhosis: [2] Bertuccio P, Malvezzi M, Carioli G, Hashim D, Boffetta P, El-Serag HB, et al. analysis of complications in a single centre over 20 years. Br J Radiol 2017;90: Global trends in mortality from intrahepatic and extrahepatic cholangiocarci- 20160804. doi: 10.1259/bjr.20160804. noma. J Hepatol 2019;71:104–114. doi: 10.1016/j.jhep.2019.03.013. [17] Takahashi EA, Kinsman KA, Schmit GD, Atwell TD, Schmitz JJ, Welch BT, et al. [3] Weber SM, Ribero D, O’Reilly EM, Kokudo N, Miyazaki M, Pawlik TM. Intra- Thermal ablation of intrahepatic cholangiocarcinoma: Safety, efficacy, and hepatic cholangiocarcinoma: expert consensus statement. HPB (Oxford) factors affecting local tumor progression. Abdom Radiol (NY) 2018;43: 2015;17:669–680. doi: 10.1111/hpb.12441. 3487–3492. doi: 10.1007/s00261-018-1656-3. [4] Carrafiello G, Laganà D, Cotta E, Mangini M, Fontana F, Bandiera F, et al. [18] Zhang K, Yu J, Yu X, Han Z, Cheng Z, Liu F, et al. Clinical and survival out- Radiofrequency ablation of intrahepatic cholangiocarcinoma: preliminary comes of percutaneous microwave ablation for intrahepatic cholangiocarci- experience. Cardiovasc Intervent Radiol 2010;33:835–839. doi: 10. noma. Int J Hyperthermia 2018;34:292–297. doi: 10.1080/02656736. 1007/s00270-010-9849-3. 2017.1327678. [5] Chiou YY, Hwang JI, Chou YH, Wang HK, Chiang JH, Chang CY. Percutaneous [19] Xu C, Li L, Xu W, Du C, Yang L, Tong J, et al. Ultrasound-guided percutaneous ultrasound-guided radiofrequency ablation of intrahepatic cholangiocarci- microwave ablation versus surgical resection for recurrent intrahepatic chol- noma. Kaohsiung J Med Sci 2005;21:304–309. doi: 10.1016/S1607-551X angiocarcinoma: intermediate-term results. Int J Hyperthermia 2019;36: (09)70125-1. 351–358. doi: 10.1080/02656736.2019.1571247.

292 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 287–292 Original Article

Alanine Aminotransferase as the First Test Parameter for Wilson’s Disease

Hisao Hayashi*1, Kazumasa Watanabe2, Ayano Inui3, Ayako Kato1, Yasuaki Tatsumi1, Akihiko Okumura2, Tomoo Fujisawa3 and Koichi Kato1

1Department of Medicine, Aichi Gakuin University School of Pharmacy, Chikusa-ku, Nagoya, Japan; 2Department of Gastroenterology, Kainan Hospital, Yatomi, Japan; 3Department of Pediatric Gastroenterology and Hepatology, Saiseikai Yokohama Toub Hospital, Yokohama, Kanagawa, Japan

Abstract Wilson’s disease (WD) and a simple approach to diagnose asymptomatic liver damage in the early stage of WD. The Background and Aims: The liver is the first organ affected classical form is a primary copper-induced liver disease, by toxic copper in the classical and severe hepatic forms of which may be complicated by lenticular degeneration at a Wilson’s disease (WD). Because their associated chronic later stage, and which was initially described by SAK Wilson liver damage is mostly asymptomatic, an intervention using in 1912.1–3 In addition to increased hepatic copper of 250 mg/g a special test including serum alanine aminotransferase dry liver or more, two mutant alleles in the ATP7B gene (ALT) activity is needed for detecting WD. Methods: Using and reduced serum levels of ceruloplasmin (CP) are found in the modified international criteria for the diagnosis of WD, typical cases.1,3 Other subtype tests include detection of 45 patients were selected from the collective databases Kayser Fleischer rings, neuropsychiatric signs, and Coombs- of our institutions, and 7 infants were reviewed from the negative hemolytic anemia.1,3 The severe hepatic form is literature. Two patients had the severe hepatic form, with normoceruloplasminemia and no mutations in ATP7B.TherapidALT another copper-induced liver disease, the former name of change during hemolytic anemia was adjusted for a baseline. which was idiopathic copper toxicosis, proposed by Scheinberg 1,4–6 The diagnostic potential of the ALT test was assessed from and Sternlieb in 1994. There are currently no tests for the the age-dependent natural course of the liver damage of WD. severe hepatic form, other than detection of increased hepatic 1,6 Results: The natural course had three stages. ALTs were still copper and urinary copper secretion, and the new subtype low in some infants younger than 4 years-old. They were high tests proposed by Tatsumi et al.1 are combined results of in all children between the ages of 4 and 8 years-old; then, they normal CP levels and no mutations in ATP7B. reduced to low levels in some patients over 9 years of age. The Another recommendation that Tatsumi et al.1 propose is a high ALT stage represents chronic active hepatitis, and the sub- treatment for the copper toxicosis of WD. Every patient sequent low ALT stage is due to silent cirrhosis. The hepatic cop- with WD should receive a short-term treatment trial involving per content is a reliable but invasive test, while urinary copper anti-copper regimens, followed by life-long maintenance secretion is an alternative, non-invasive test for copper toxicosis treatment. For the former, a poor response to anti-copper of WD. The serum ceruloplasmin and ATP7B analyses are sub- regimens indicates non-WD. Some patients with WD, espe- type tests of WD. The response to anti-copper regimens is the cially the severe hepatic form, suddenly present with hepatic final test result. Conclusions: ALTcould be the first parameter to failure and require a liver transplantation without the liver test to detect WD in children between the ages of 4 and 8 years. biopsy essential for diagnosing WD. In such cases, the postop- Citation of this article: Hayashi H, Watanabe K, Inui A, erative diagnosis of WD should be confirmed by the increased Kato A, Tatsumi Y, Okumura A, et al. Alanine aminotransferase copper content of the removed liver. as the first test parameter for Wilson’s disease. J Clin Transl Hep- It is now an urgent task to establish liver tests for patients atol 2019;7(4):293–296. doi: 10.14218/JCTH.2019.00042. with asymptomatic liver damage of WD. Aspartate aminotransferase (AST) and alanine aminotransferase (ALT) are favorable liver tests, but they are hampered by the following Introduction problems in WD patients: (1) rapid falls of AST and ALT during Coombs-negative hemolytic anemia;7 (2) natural courses of New terminology proposed by Tatsumi et al.1 may facilitate a the first increasing and second decreasing stages of AST and better understanding of the primary copper toxicosis of ALT;8 and (3) AST having an erythrocyte fraction in WD, where it is significantly affected by copper-induced hemoly- Keywords: Alanine aminotransferase; ATP7B; Ceruloplasmin; Chronic active sis.7,9 When the serum levels of ASTand ALTare introduced as ’ hepatitis; Wilson s disease. tests for the primary liver damage of WD, all problems should Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; CP, ceruloplasmin; WD, Wilson’s disease. be overcome to minimize misdiagnosis. Received: 5 September 2019; Revised: 25 October 2019; Accepted: 30 October In this study, we evaluated the potential of serum levels of 2019 ALT as representative of age-dependent liver disease activity *Correspondence to: Hisao Hayashi, Department of Medicine, Aichi Gakuin Uni- versity School of Pharmacy, 1–100 Kusumoto-cho, Chikusa-ku, Nagoya 464- of WD and determined the optimal age for the ALT test in WD 8650, Japan. Tel: +81-52-757-6779, E-mail: [email protected] patients.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 293–296 293

Table 1. Fifty-two patients with WD and their subtypes using ATP7B analysis and serum CP levels

Serum CP Patient Major Group of patients ATP7B (>20 mg/dL) number subtype

Literature, n = 7 Two mutant alleles Low 7 Classical Normal 0 Classical Patient databases, n = 45 Two mutant alleles Low 38 Classical Normal 1 Classical One mutant allele Low 4 Classical Normal 0 ? No mutant allele Low 0 ? Normal 2 Severe hepatic

Fifty-two patients with WD were subtyped using ATP7B analysis and serum levels of CP. The presence of mutant alleles in ATP7B and hypoceruloplasminemia less than 20 mg/dL strongly suggests the classical form of WD, while their absence suggests the severe hepatic form of WD. A total of 45 patients (7 from the literature review and 38 from the patient databases) had typical features of the classical form, while 2 patients had the severe hepatic form. Therefore, the tests, either incidental or intentional, should be adaptive for the primary liver diseases of WD consisting of the two subtypes of classical and severe hepatic forms. Abbreviations: CP, ceruloplasmin; WD, Wilson’s disease.

Methods of serum CP. Five patients had atypical classical forms, with one mutant allele of ATP7B. Patients Methods International proposals were used for the diagnosis of WD, with minor modifications.3 The original scoring system for the diag- The study protocol was approved by the ethics committees of nosis consists of Kayser-Fleischer rings (2 points), neuropsychi- 3 institutes (Aichi Gakuin University School of Pharmacy, atric symptoms (2), Coombs-negative hemolytic anemia (1), urinary copper (max. increase: 2), hepatic copper quantitative (max. increase: 2), serum CP level (max. decrease: 2),10 and ATP7B mutation (2 alleles: 4, one allele: 1). Diagnostic WD is assigned with a score of 4 or more points. Probable WD is assigned with score of 2–3 points. Unlikely WD is assigned with score of 0–1 points. In the patients saved by liver transplantation, the copper content of the removed liver should receive a maximum of 2 points (>250 mg/g dry liver weight). In patients with 2–3 points (probable WD), the diagnosis was retrospectively confirmed by giving an additional 2 points for a good response to anti-copper regimens. A removed liver with copper less than 250 mg/g dry liver and a poor response to anti-copper regimens were set as exclusion criteria. WD cases complicated with either chronic cholestasis or other chronic liver disease, including chronic viral hepatitis, nonalcoholic steatohepatitis, etc., were also excluded from the study. A total of 52 patients with primary copper toxicosis of WD were enrolled in the study. Forty-five Japanese patients were selected from the databases of three institutes (Aichi Gakuin Fig. 1. Natural course of WD represented by serum ALT levels adjusted for the baseline liver disease and onset ages of patients. Rapid ALT University School of Pharmacy, Kainan Hospital, and Saiseikai changes during Coombs-negative hemolytic anemia of WD were adjusted to a Yokohama Toub Hospital), and 7 infant patients were cited baseline at either the prehemolytic anemia or postrecovery stage. In the patients based on a literature review of PubMed to identify relevant whose WD was found after acute diseases, such as fever episodes and gastro- papers written in English since 1993.11–16 Clinical data of two intestinal problems, the baseline ALT was obtained at the recovery stage. Three ALT stages of WD associated with the ages of patients are shown. The first stage Japanese patients were analyzed twice because they were includes 7 infants with development of WD before 4 years old; ALT levels are low followed for years without anti-copper regimens. in 4 and high in 3 infants. The second stage consists of 10 patients aged between Based on ATP7B analysis and serum levels of CP, the 4 and 8 years whose ALT levels are all high, over 150 IU/L. The third stage includes 35 patients with a wide range of ALT levels; fifteen patients have patho- primary copper toxicosis of WD in the studied patients was logically high ALT levels, while 20 patients have near-normal levels. Especially, all subtyped into two forms — classical and severe hepatic WD 7 patients aged 35 years or older show ALT levels below 50 IU/L. The ALT test is (Table 1).1 Forty-five patients had typical features of the clas- reliable for detecting WD at the high ALT stage of children aged between 4 and 8 ’ ● sical form, with two mutant alleles in ATP7B and low levels of years. ALT, alanine aminotransferase; WD, Wilson sdisease. : Classical form of Japanese patients; :: Severe hepatic form (non-ATP7B) of Japanese patients; serum CP (<20 mg/dL); two other patients had the severe B: Infants with the ATP7B classical form cited from the international liter- hepatic form, with no mutation in ATP7B and normal levels ature.11–16

294 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 293–296 Hayashi H. et al: ALT test for Wilson’s disease

Table 2. Recommended diagnostic approach to the primary liver disorders of WD

Classical Form Severe Hepatic Form

1 Liver test Serum ALT test: > 150 IU/L in children aged 4 – 8 yrs 2 Copper tests Hepatic copper more than 250 mg/g dry liver Increased urinary copper secretion (basic and post-penicillamine) 3 Subtype tests Hypoceruloplasminemia Normoceruloplasminemia and no ATP7B mutation Mutations in ATP7B 4 Anti-copper tests Good responses on ALT reduction

Copper-induced chronic active hepatitis of WD, either classical or severe hepatic forms, should be initially detected by an ALT test and confirmed by tests for copper toxicosis. Serum levels of ceruloplasmin and ATP7B are useful for subtyping. Patients with both the classical and severe hepatic forms should show good responses to anti-copper regimens. Abbreviation: ALT, alanine aminotransferase; WD, Wilson’s disease.

Kainan Hospital, and Saiseikai Yokohama Toub Hospital). the central nervous system is also associated with silent cir- Serum levels of ALT were determined by one point at entry in rhosis formation in classical WD.8 As a result, ALT tests all patients with WD, except for those with Coombs-negative partially lose the potential to detect copper-induced liver dis- hemolytic anemia and complicated acute diseases. The base- eases in patients over 9 years-old. line ALT levels of hemolysis patients were obtained at the Based on the natural disease course presented in this stage of pre-hemolytic anemia or recovery from anemia.7 In study, we can postulate that most patients aged over 9 years the patients whose WD was found after acute diseases, such had asymptomatically experienced the high ALT stage when as fever episodes and gastrointestinal problems, the baseline between the ages of 4 and 8 years-old. Therefore, children ALTwas obtained at the recovery stage. Using the serum levels aged between 4 and 8 years may be suitable candidates of ALT and onset ages of 52 patients aged between 0.7 and for special intervention to detect asymptomatic but biochemi- 47 years, a virtual image was made for the natural course of cally active liver damage of WD, as summarized in Table 2. WD. The diagnostic potential of the age-dependent ALT levels Increased hepatic copper and urinary copper secretion are was assessed for the primary liver damage of WD. reliable tests for the copper toxicosis of WD favored by the international proposal.3 Hypoceruloplasminemia1,10 and Results ATP7B mutations1,3 strongly support the classical form. In contrast, their respective absence does not exclude WD; the Virtual images of the natural course of WD using serum ALT negative combination of the two major subtyping tests for the levels representative of primary liver disease activities and classical form may be transient subtype tests for the severe entry ages of patients are presented in Fig. 1. There are hepatic form that has no definitive subtype tests, including three ALT stages associated with the onset ages of patients. gene analysis.1,6 In the first stage, some infants younger than 4 years-old A final diagnosis of WD should be confirmed in all showed low ALT levels. In the second stage, all of the 10 children with WD, by anti-copper regimens. A recent study children aged between 4 and 8 years showed high ALT on young patients under 10 years of age revealed that zinc levels (>150 IU/L). In the third stage, patients over the monotherapy reduced serum ALT levels ranging from 243 ± age of 9 years showed a wide range of ALT levels; fifteen of 151 to 70 ± 29 IU/L in 1 month.23 It is also important that thirty-five patients showed pathologically high ALT levels, copper-overload conditions secondary to chronic cholesta- while 20 patients had levels that were near-normal. Espe- sis24 or chronic liver diseases other than WD25,26 could be cially, all seven patients aged 35 years or older showed ALT differentiated by the short-term treatment test. levels below 50 IU/L. Because neuropsychiatric symptoms of classical WD appear at a late cirrhotic stage associated with low ALT Discussion levels,8 an intentional ALT test for children at the high ALT stage is prophylactic for neuropsychiatric complications. Our observations strongly suggest that a standard liver test — A smaller number of children with WD may have been affected ALT — has the potential to detect most WD patients between by Coombs-negative hemolytic anemia before such a sched- the ages of 4 and 8 years-old. Serum ALT levels higher than uled ALT test.8 A similar episode in a 7-year-old girl among 150 IU/L are representative of active chronic hepatitis of the 10 patients in the high ALT stage was self-limiting, and she 17,18 WD. Underlying mechanisms of the low ALT levels are recovered from anemia without any sequelae. ALT interven- different in the first and third stages. The former means that tion specialized to detect asymptomatic WD should be con- there is no active liver damage in some patients as yet, due to firmed in a large number of patients. a small amount of toxic copper in the liver,8 while the latter involves silent cirrhosis in some patients. It is well known Conclusions that, in chronic viral hepatitis, histological progression from chronic hepatitis to cirrhosis is associated with declining There are two subtypes of classical and severe hepatic forms inflammation in the liver. Such transformation is of special in the primary copper toxicosis of WD. A standard liver test of importance in WD. A large amount of copper is mostly ALT is highly recommended for children between the ages of 4 stored in hepatocyte lysosomes of cirrhosis as detoxified and 8 years-old to detect asymptomatic WD. The selection cuprothioneins.19,20 Age-dependent copper accumulation in criterion is a serum ALT level over 150 IU/L. Increased hepatic

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 293–296 295 Hayashi H. et al: ALT test for Wilson’s disease copper is an etiological test. Serum CP level and ATP7B anal- [11] Shimizu N, Nakazono H, Watanabe A, Yamaguci Y, Hemmi H, Aoki T. Molec- yses are subtype tests. A marked reduction of ALT with anti- ular diagnosis of Wilson’s disease. Lancet 1997;349:1811–1812. doi: 10. copper regimens is diagnostic for WD. 1016/S0140-6736(05)61691-1. [12] Abuduxikuer K, Li LT, Qiu YL, Wang NL, Wang JS. Wilson disease with hepatic presentation in an eight-month-old boy. World J Gastroenterol 2015;21: Conflict of interest 8981–8984. doi: 10.3748/wjg.v21.i29.8981. [13] Kim JW, Kim JH, Seo JK, Ko JS, Chang JY, Yang HR, et al. Genetically con- The authors have no conflict of interests related to this firmed Wilson disease in a 9-month old boy with elevations of aminotrans- ferases. World J Hepatol 2013;5:156–159. doi: 10.4254/wjh.v5.i3.156. publication. [14] Usta J, Wehbeh A, Rida K, El-Rifai O, Estiphan TA, Majarian T, et al. Phenotype-genotype correlation in Wilson disease in a large Lebanese Author contributions family: association of c.2299insC with hepatic and of p. Ala1003Thr with neurologic phenotype. PLoS One 2014;9:e109727. doi: 10.1371/journal. pone.0109727. Conception of study objective and design (HH, AI, TF), [15] Ohura T, Abukawa D, Shiraishi H, Yamaguchi A, Arashima S, Hiyamuta S, collection of patient data (KW, AO, KK), gene analysis (AK, YT). et al. Pilot study of screening for Wilson disease using dried blood spots obtained from children seen at outpatient clinics. J Inherit Metab Dis 1999; 22:74–80. doi: 10.1023/a:1005455401076. References [16] Yuan ZF, Wu W, Yu YL, Shen J, Mao SS, Gao F, et al. Novel mutations of the ATP7B gene in Han Chinese families with pre-symptomatic Wilson’s disease. [1] Tatsumi Y, Kato A, Kato K, Hayashi H. The interactions between iron and World J Pediatr 2015;11:255–260. doi: 10.1007/s12519-015-0031-5. copper in genetic iron overload syndromes and primary copper toxicoses in [17] Sternlieb I, Scheinberg IH. Chronic hepatitis as a first manifestation of Japan. Hepatol Res 2018;48:679–691. doi: 10.1111/hepr.13200. Wilson’s disease. Ann Intern Med 1972;76:59–64. doi: 10.7326/0003- [2] Compston A. Progressive lenticular degeneration: a familial nervous disease 4819-76-1-59. associated with cirrhosis of the liver, by S. A. Kinnier Wilson, (From the [18] Stromeyer FW, Ishak KG. Histology of the liver in Wilson’s disease: a study of National Hospital, and the Laboratory of the National Hospital, Queen 34 cases. Am J Clin Pathol 1980;73:12–24. doi: 10.1093/ajcp/73.1.12. Square, London) Brain 1912: 34; 295–509. Brain 2009;132:1997–2001. [19] Goldfischer S, Sternlieb I. Changes in the distribution of hepatic copper in doi: 10.1093/brain/awp193. relation to the progression of Wilson’s disease (hepatolenticular degenera- [3] Ferenci P, Caca K, Loudianos G, Mieli-Vergani G, Tanner S, Sternlieb I, et al. tion). Am J Pathol 1968;53:883–901. Diagnosis and phenotypic classification of Wilson disease. Liver Int 2003;23: [20] Hanaichi T, Kidokoro R, Hayashi H, Sakamoto N. Electron probe X-ray anal- – 139 142. doi: 10.1034/j.1600-0676.2003.00824.x. ysis on human hepatocellular lysosomes with copper deposits: copper [4] Scheinberg IH, Sternlieb I. Is non-Indian childhood cirrhosis caused by binding to a thiol-protein in lysosomes. Lab Invest 1984;51:592–597. – excess dietary copper? Lancet 1994;344:1002 1004. doi: 10.1016/s0140- [21] Walshe JM. Serum ‘free’ copper in Wilson disease. QJM 2012;105:419–423. 6736(94)91649-7. doi: 10.1093/qjmed/hcr229. [5] Müller T, Müller W, Feichtinger H. Idiopathic copper toxicosis. Am J Clin Nutr [22] Squitti R, Ghidoni R, Simonelli I, Ivanova ID, Colabufo NA, Zuin M, et al. 1998;67:1082S–1086S. doi: 10.1093/ajcn/67.5.1082S. Copper dyshomeostasis in Wilson disease and Alzheimer’s disease as [6] Fujisawa T, Inui A, Kawakami M, editors. Pediatric clinical hepatology. Tokyo: shown by serum and urine copper indicators. J Trace Elem Med Biol 2018; Tokyo Igakusha; 2017. (in Japanese). 45:181–188. doi: 10.1016/j.jtemb.2017.11.005. [7] Hayashi H, Tatsumi Y, Yahata S, Hayashi H, Momose K, Isaji R, et al. Acute [23] Eda K, Mizuochi T, Iwama I, Inui A, Etani Y, Araki M, et al. Zinc monotherapy hepatic phenotype of Wilson disease: Clinical features of acute episodes for young children with presymptomatic Wilson disease: A multicenter study and chronic lesions remaining in survivors. J Clin Transl Hepatol 2015;3: – 239–245. doi: 10.14218/JCTH.2015.00032. in Japan. J Gastroenterol Hepatol 2018;33:264 269. doi: 10.1111/jgh. [8] Katano Y, Hayashi K, Hattori A, Tatsumi Y, Ueyama J, Wakusawa S, et al. 13812. Biochemical staging of the chronic hepatic lesions of Wilson disease. [24] Ramraj R, Finegold MJ, Karpen SJ. Progressive familial intrahepatic choles- Nagoya J Med Sci 2014;76:139–148. tasis type 3: overlapping presentation with Wilson disease. Clin Pediatr [9] Lippi G, Salvagno GL, Montagnana M, Brocco G, Guidi GC. Influence of (Phila) 2012;51:689–691. doi: 10.1177/0009922812451076. hemolysis on routine clinical chemistry testing. Clin Chem Lab Med 2006; [25] Piekuse L, Kreile M, Zarina A, Steinberga Z, Sondore V, Keiss J, et al. Asso- 44:311–316. doi: 10.1515/CCLM.2006.054. ciation between inherited monogenic liver disorders and chronic hepatitis C. [10] Scheinberg IH, Gitlin D. Deficiency of ceruloplasmin in patients with hepato- World J Hepatol 2014;6:92–97. doi: 10.4254/wjh.v6.i2.92. lenticular degeneration (Wilson’s disease). Science 1952;116:484–485. doi: [26] Tahir A, Malik FR, Ahmad I, Akhtar P. Aetiological factors of chronic liver 10.1126/science.116.3018.484. disease in children. J Ayub Med Coll Abbottabad 2011;23:12–14.

296 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 293–296 Original Article

Association of GCKR Gene Polymorphisms with the Risk of Nonalcoholic Fatty Liver Disease and Coronary Artery Disease in a Chinese Northern Han Population

Hui Gao#1, Shousheng Liu#2,4, Zhenzhen Zhao2,4, Xinjuan Yu2, Qun Liu1, Yongning Xin*1,3,4 and Shiying Xuan*1,3,4

1Department of Infectious Disease, Qingdao Municipal Hospital, Qingdao University, Qingdao, Shandong, China; 2Central Laboratories, Qingdao Municipal Hospital, Qingdao University, Qingdao, Shandong, China; 3Department of Gastroenterology, Qingdao Municipal Hospital, Qingdao University, Qingdao, Shandong, China; 4Digestive Disease Key Laboratory of Qingdao, Qingdao, Shandong, China

Abstract patients. Conclusions: GCKR rs780094 and rs1260326 polymorphisms were found to be not associated with the risk of Background and Aims: Accumulated studies have eval- NAFLD nor of CAD in NAFLD patients in this Chinese Northern uated the effects of glucokinase regulatory protein (GCKR) Han population. GCKR rs780094 T and rs1260326 T alleles gene polymorphisms on the risk of nonalcoholic fatty liver could affect the body mass index value and serum fasting disease (NAFLD) and coronary artery disease (CAD), but the plasma glucose and triglyceride levels. association of GCKR polymorphisms with the risk of NAFLD Citation of this article: Gao H, Liu S, Zhao Z, Yu X, Liu Q, Xin and CAD in the Chinese Han population have remained un- Y, et al. Association of GCKR gene polymorphisms with the risk clear. The aim of this study was to investigate the association of nonalcoholic fatty liver disease and coronary artery disease between GCKR gene polymorphisms (rs780094 and in a Chinese Northern Han population. J Clin Transl Hepatol rs1260326) and the risk of NAFLD and CAD in NAFLD patients 2019;7(4):297–303. doi: 10.14218/JCTH.2019.00030. in a Chinese Northern Han population. Methods: GCKR rs780094 and rs1260326 gene polymorphisms were geno- typed by polymerase chain reaction sequencing for B-type Introduction ultrasonography-proven NAFLD patients with (n = 82) or without (n = 142) CAD, and in healthy controls (n = 152). Nonalcoholic fatty liver disease (NAFLD) is the most prevalent Serum lipid profiles’ levels were determined using biochemi- chronic liver disease in the world and is regarded as a severe cal methods. Statistical analyses were conducted using SPSS public health concern.1 The overall prevalence of NAFLD is 22.0 statistical software. Results: As the results showed, sig- approximately 25% worldwide and is 23%, 31% and 27% nificant differences in the serum lipid profiles existed between in Europe, South America and Asia, respectively.2–4 Many each group. No significant differences were observed in the factors contribute to the development of NAFLD, such as distributions of genotypes and alleles of GCKR rs780094 and aging, hyperlipidemia, insulin resistance, type 2 diabetes, rs1260326 in each group. The GCKR rs780094 T and diet, genetics and so on.5–8 Accumulated lines of evidence rs1260326 T allele carriers possessed decreased body mass have suggested that coronary artery disease (CAD) plays an index value, and serum fasting plasma glucose and TG levels important role in the progression of NAFLD and is associated in the overall subjects, respectively. In addition, the GCKR with the severity of NAFLD. rs780094 T allele carriers possessed decreased serum fasting CAD has become the most important cause of mortality plasma glucose level in the controls and NAFLD + CAD among NAFLD patients.9 Although some defects of liver biopsy exist, such as invasive, sample errors and operator Keywords: Non-alcoholic fatty liver disease; Glucokinase regulatory protein dependence, it remains the gold standard for diagnosis of gene; Polymorphism; Coronary artery disease. 10,11 Abbreviations: ALP, alkaline phosphatase; ALT, alanine aminotransferase; AST, NAFLD. In recent years, some noninvasive diagnostic aspartate aminotransferase; BMI, body mass index; BIL, bilirubin; CAD, coronary methods have been developed and their effects have been artery disease; CI, confidence intervals; FPG, fasting plasma glucose; GCK, glu- tested in different countries and populations. Genome-wide cokinase; GCKR, glucokinase regulatory protein; GWAS, genome-wide associa- association studies (GWAS) have identified several important tion studies; HDL, high-density lipoprotein; LDL, low-density lipoprotein; NAFLD, nonalcoholic fatty liver disease; OR, odds ratio; PCR, polymerase chain reaction; single nucleotide polymorphism (SNP) sites which are tightly r-GT, glutamyltransferase; SD, standard deviation; SNP, single nucleotide poly- associated with the risk of development of NAFLD; these morphism; TC, total cholesterol; TG, triglyceride. include PNPLA3 rs738409, TM6SF2 rs58542926, and Received: 22 July 2019; Revised: 7 October 2019; Accepted: 8 December 2019 LYPLAL1 rs12137855.12 In consideration of the differences #Authors contributed equally to this work. *Correspondence to: Yongning Xin, Department of Infectious Disease, Qingdao of genetic background of NAFLD patients, it is necessary to Municipal Hospital, Qingdao University, 1 Jiaozhou Road, Qingdao, Shandong identify the relationship of gene polymorphisms with the risk 266011, China. Tel: +86-532-82789463, Fax: +86-532-85968434, E-mail: of NAFLD in different countries and races. [email protected]; Shiying Xuan, Department of Gastroenterology, Glucokinase (GCK) is a phosphorylating enzyme which can Qingdao Municipal Hospital, Qingdao University, 1 Jiaozhou Road, Qingdao, Shan- dong 266011, China. Tel: +86-532-88905508, Fax: +86-532-88905293, E-mail: regulate hepatic glucose metabolism and activate hepatic 13 [email protected] lipogenesis. The GCKR gene encodes the glucokinase

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 297–303 297

Copyright: © 2019 Authors. This article has been published under the terms of Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0), which permits noncommercial unrestricted use, distribution, and reproduction in any medium, provided that the following statement is provided. “This article has been published in Journal of Clinical and Translational Hepatology at DOI: 10.14218/JCTH.2019.00030 and can also be viewed on the Journal’s website at http://www.jcthnet.com”. Gao H. et al: GCKR with NAFLD and CAD regulatory protein (GCKRP), which can bind to GCK allosteri- Standard clinical evaluations of the NAFLD patients were cally and regulate the activity of GCK.14 In 2011, Speliotes performed according to the criteria of American Association et al.15 identified the GCKR rs780094 by GWAS as a signifi- for the Study of Liver Diseases.21 Subjects with the following cant SNP site associated with the risk of NAFLD. Subsequent symptoms were excluded from the NAFLD cohort: 1) exces- studies in different districts reported controversial conclusive alcohol intake (males >210 g/w, females >140 g/w); 2) sions of GCKR rs780094 in NAFLD. The GCKR rs780094 viral hepatitis, such as hepatitis B or hepatitis C; 3) drug- T allele was found to be tightly associated with susceptibility induced hepatitis; 4) other endocrine and metabolic disor- to NAFLD in Americans and Malaysians, acting as a risk ders, such as renal disease. CAD was diagnosed using the factor for NAFLD.15,16 However, no significant association findings from percutaneous coronary angiogram conducted between GCKR rs780094 and the risk of NAFLD was observed by two experienced interventional cardiologists, and defined in the community-based study in Suzhou and Shanghai, and as the presence of at least 50% stenosis in at least one of the in obese adolescents in Beijing, China.17–20 On the contrary, coronary arteries. The healthy controls were also subjected to Lin et al.21 reported that GCKR rs780094 was associated with the same diagnosis routines for NAFLD and CAD, as well as an increased risk of NAFLD in obese children in Taiwan.22 the laboratory and general examinations. Therefore, large-scale population-based studies in different districts are needed to investigate the effect of GCKR Biochemical analyses rs780094 on the development of NAFLD in China. In addition, GWAS analyses conducted in Finnish, Swedish The basic clinical pathological information (gender, age, and Danish populations found a strong linkage disequilibrium height, and weight) was obtained by a standard study of GCKR rs780094 with GCKR rs1260326.23,24 In Japan, questionnaire. The body mass index (BMI) was calculated as Kawaguchi et al.25 found that GCKR rs1260326 was a signifi- mass weight (kg)/height2 (m2). All the subjects underwent a cant risk factor for the development of NAFLD. In France, Petit 12-h overnight fast, and then the blood samples were col- et al.26 reported that GCKR rs1260326 influences the liver fat lected into ethylene diaminetetraacetic acid-containing content in patients with type 2 diabetes. Di et al.27 reported tubes. Serum fasting plasma glucose (FPG), total cholesterol that GCKR rs1260326 could increase the liver fat content, (TC), triglycerides (TG), high-density lipoprotein (HDL), low- although the detailed role of GCKR rs1260326 in NAFLD density lipoprotein (LDL), alanine aminotransferase (ALT), remains controversial. Besides, Simons et al.28 conducted a aspartate aminotransferase (AST), glutamyltransferase meta-analysis to investigate the relationship of GCKR poly- (r-GT), alkaline phosphatase (ALP), and bilirubin (BIL) were morphisms with the risk of CAD, and a tight association of measured using standard clinical laboratory techniques.30 GCKR variants with the risk of CAD was observed. In consideration of the unclear effects of GCKR rs780094 Genomic DNA extraction and genotyping and rs1260326 on NAFLD in Chinese, and the tight association of CAD and NAFLD, it is of interest to explore the effects Genomic DNA was extracted and stored as previously of GCKR rs780094 and rs1260326 on the development of described.31 The primers for PCR amplification containing NAFLD and the development of CAD in NAFLD patients from rs780094 were 5′-ACGTTGGATGAAGTGGTGGGATTACAGGAG-3′ this population. and 5′-ACGTTGGATGAGGGCCCCAGTTTTTTAGAC-3′, rs1260326 The aim of this study was to explore the relationship between 5′-ACGTTGGATGACCTGGGTCCCTTTGTCAC-3′ and 5′-ACGTTG- GCKR gene polymorphisms (rs780094 and rs1260326) with the GATGAGAAGACCAACCACATCCAG-3′ and were synthesized by risk of NAFLD, the risk of CAD in NAFLD patients in a Chinese Beijing BoMiao Biotech Company (China). The PCR amplifica- Northern Han population, and the effects of GCKR rs780094 and tion profile was as follows: predenaturation at 948Cfor5m, rs1260326 on the levels of serum lipid profiles. followed by 40 cycles of denaturation at 948C for 20 s, anneal- ing at 568C for 30 s, extending at 728C for 60 s, followed by the Methods final extension at 728C for 3 m. The target amplified fragments were detected by 2% gel electrophoresis, with the predicted Subjects fragment length. The genotypes of rs780094 and rs1260326 were detected by direct DNA sequencing using the Applied Bio- This case-control study was conducted according to the systems Inc. (USA) Veriti-384 Prism Sequence Detection principles of the Declaration of Helsinki and its appendices.29 System, and the raw data were analyzed using MassARRAY This study was approved by the ethics committee of Qingdao TYPER4.0 software (Agena Bioscience, USA). Genotyping was Municipal Hospital (Qingdao, China). All the subjects included performed in a blinded fashion and the success rates were in this study were of the Chinese Northern Han population, >95%. and each signed a written informed consent form. From June 2018 to December 2018, 225 unrelated adult Statistical analysis Chinese Northern Han patients with NAFLD, of both genders, that were diagnosed by B-type ultrasonography were Statistical analysis was conducted using the SPSS 22.0 recruited. Among these NAFLD patients, 82 suffered from statistical software (SPSS Inc., USA). Student’s t-test or x2 CAD (41 males, 41 females; mean age of 55.62 ± 7.187 test were used to analyzed the differences in characteristics years) and 142 did not have CAD (75 males, 67 females; mean of different groups. First, continuous variables were tested by age of 50.14 ± 11.842 years). A total of 152 healthy controls, normal distribution and F test; if yes, the baseline character- matched for sex and age (75 males, 77 females; mean age of istics were expressed as the mean ± standard deviation (SD), 51.38 ± 13.276 years) were included. All the patients were and if no, the baseline characteristics were expressed as the recruited from the Departments of Gastroenterology and median ± quartile. Hardy-Weinberg equilibrium was between Cardiology and the Health Examination Center of Qingdao expected and observed genotype distributions and was esti- Municipal Hospital. mated using the x2 test. Genotype and allele frequencies were

298 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 297–303 Gao H. et al: GCKR with NAFLD and CAD

a Table 1. Baseline characteristics of the study participants

Characteristic Controls NAFLD NAFLD + CAD pp1 p2 p3

Age (y) 51.38 6 13.276 50.14 6 11.842 55.62 6 7.187 0.156 - - - Gender (F/M) 77/75 67/75 41/41 0.174 - - - BMI (kg/m2) 23.45 6 3.376 26.74 6 2.805 25.24 6 2.606 <0.001 <0.001 <0.001 0.003 FPG (mmol/L) 4.80 6 1.479 5.00 6 1.343 5.90 6 2.180 <0.001 0.010 <0.001 0.001 TC (mmol/L) 4.99 6 1.192 5.48 6 0.822 4.56 6 1.274 <0.001 0.010 <0.001 <0.001 TG (mmol/L) 1.45 6 1.009 1.90 6 1.903 2.23 6 4.396 <0.001 0.003 <0.001 1.000 HDL (mmol/L) 1.28 6 0.387 1.24 6 0.198 1.05 6 0.294 <0.001 1.000 <0.001 <0.001 LDL (mmol/L) 3.10 6 0.708 3.27 6 0.555 2.75 6 0.972 <0.001 0.034 0.006 <0.001 ALT (U/L) 19.90 6 11.155 34.40 6 26.165 28.90 6 25.044 <0.001 <0.001 0.006 0.019 AST (U/L) 21.93 6 8.877 24.16 6 8.994 44.67 6 69.649 0.067 - - - r-GT (mmol/L) 26.30 6 24.367 39.63 6 26.481 47.89 6 85.986 <0.001 <0.001 0.010 0.019 ALP (mmol/L) 75.14 6 24.661 71.62 6 18.272 85.79 6 21.840 <0.001 0.852 <0.001 <0.001 BIL (mmol/L) 13.20 6 5.358 13.30 6 4.957 14.44 6 7.020 0.475 - - -

Abbreviations: NAFLD, patients with nonalcoholic fatty liver disease; NAFLD + CAD, patients with nonalcoholic fatty liver disease and coronary artery disease; BMI, body mass index; FPG, fasting plasma glucose; TC, total cholesterol; TG, triglycerides; HDL, high-density lipoprotein; LDL, low-density lipoprotein; ALT, alanine aminotransferase; AST, aspartate aminotransferase; r-GT, glutamyltransferase; ALP, alkaline phosphatase; BIL, bilirubin. p, controls vs. NAFLD vs. NAFLD + CAD ; p1, NAFLD vs. controls; p2, NAFLD + CAD vs. controls; p3, NAFLD + CAD vs. NAFLD. a Values are expressed as median ± quartile. assessed by counting the DNA sequencing data, and the dis- suggested that GCKR rs780094 and rs1260326 polymor- tributions in NAFLD versus controls were analyzed by Pear- phisms did not associate with the risk of NAFLD in this North- son’s x2 test or Fisher’s exact test, where appropriate. ern Chinese Han population, and GCKR rs780094 and Association between polymorphisms and NAFLD/CAD were rs1260326 polymorphisms did not increase the risk of CAD analyzed by logistic regression and odds ratios (ORs) with in these NAFLD patients. 95% confidence intervals (CIs). A p-value <0.05 was considered statistically significant. Association of GCKR polymorphisms with clinical parameters in each group Results The effects of GCKR rs780094 T or rs1260326 T allele on the clinical characteristics of NAFLD patients, NAFLD + CAD Characteristics of the study participants patients, and overall series were analyzed. As the results show in Table 4 and Table 5, the BMI value and serum FPG The baseline characteristics of the three groups (NAFLD, and TG levels were significantly lower in the rs780094 T and NAFLD with CAD, and healthy controls) according to the rs1260326 T allele carriers than in the non-carriers in the experimental requirements are shown in Table 1. There overall series (all p < 0.05), respectively. In addition, the were no significant differences in age or gender among the serum level of FPG in rs780094 T carriers was higher than in three groups (all p > 0.05). The increased serum levels of non-carriers in the NAFLD + CAD group (p < 0.05). No BMI, TC, TG, LDL, ALT, AST, and r-GT were observed in the marked difference of other clinical parameters was observed NAFLD patients compared to the healthy controls (all p < between the carriers and non-carriers of the rs780094 T or 0.05). In the NAFLD + CAD group, the BMI value and serum rs1260326 T allele in each group (p > 0.05). Multivariate levels of FPG, TC, TG, LDL, ALT, AST, r-GT, and ALP were higher and the serum HDL level was lower than in healthy

a controls (all p < 0.05). Compared to the NAFLD patients, Table 2. Results of the Hardy-Weinberg equilibrium test the patients with NAFLD and CAD possessed higher serum 2 levels of FPG, AST and ALP, and lower BMI value and serum Gene locus groups CC CT TT x p value levels of TC, HDL, and LDL (all p < 0.05). rs780094 Controls 26 81 45 1.040 0.308 GCKR rs780094 and rs1260326 genotype and allele NAFLD 26 71 45 0.047 0.828 distribution NAFLD + CAD 18 39 25 0.014 0.708 The genotype distributions of GCKR rs780094 and rs1260326 rs1260326 were in accordance with the Hardy-Weinberg equilibrium in Controls 23 82 46 1.890 0.169 each group (all p > 0.05) (Table 2). As the results show in NAFLD 24 73 45 0.372 0.542 Table 3, there were no significant differences in the genotypic NAFLD + CAD 15 41 26 0.028 0.868 and allelic distributions of GCKR rs780094 and rs1260326 between the NAFLD patients and healthy controls, NAFLD + Abbreviations: NAFLD, patients with nonalcoholic fatty liver disease; NAFLD + CAD patients and healthy controls, nor NAFLD + CAD patients CAD, patients with nonalcoholic fatty liver disease and coronary artery disease. a and NAFLD patients, respectively (all p > 0.05). These results Data were compared by chi-square test.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 297–303 299 Gao H. et al: GCKR with NAFLD and CAD

logistic regression analysis showed that GCKR rs780094 is the dependent factor associated with FPG (p < 0.05). 3 p

Discussion 2 x NAFLD can be caused by many risk factors, such as diet, genetics and so on. Genetic susceptibility genes are a type of risk factor that can contribute to the development of NAFLD. 2.642) 0.437 0.509 1.624) 0.247 0.619 1.850) 0.07 0.791 1.516) 0.02 0.887 In this study, we investigated the relationship between GCKR – – – – rs780094 and rs1260326 gene polymorphisms and the risk of NAFLD in a general Chinese Northern Han population, and the risk of CAD in NAFLD patients in this population for the first time. Our results showed that the genotype distributions of OR(95%CI) GCKR rs780094 and rs1260326 were not associated with the risk of NAFLD, and not associated with the risk of CAD in NAFLD patients in this Chinese Northern Han population. In 2 p addition to the BMI value and serum FPG and TG levels, the T allele of GCKR rs780094 and rs1260326 also did not significantly affect the clinical parameters in NAFLD patients and 2

x NAFLD + CAD patients. These data suggest that GCKR ery disease. rs780094 and rs1260326 polymorphisms are not associated with the risk of NAFLD in the general Chinese Northern Han population, and GCKR rs780094 and rs1260326 polymor- 1.473) 0.819 0.365 1.255 (0.640 1.587) 0.169 0.681 1.103 (0.749 2.565) 0.351 0.532 0.908 (0.446 1.541) 0.061 0.804 1.028 (0.698 – – – – phisms do not increase the risk of CAD in NAFLD patients in the Chinese Northern Han population. Accumulated studies have explored the effects of GCKR polymorphisms on the risk of NAFLD but the results have 16–18,32 OR (95%CI) been inconsistent. Some previous studies and our study suggested that GCKR rs780094 and rs1260326 polymorphisms are not associated with the risk of NAFLD, the 1

p probable reason for this finding may be due to the ethnic b

, differences among the NAFLD patients studied. The relation- a ship between GCKR polymorphism and NAFLD or nonalcoholic 2

x steatohepatitis may be modified by the change and conflu- ence of race, therefore the effects of a GCKR polymorphism may become more dominant and significant in European populations than they are in Asian populations. Therefore, the 1.538) 0.073 0.787 0.734 (0.375 1.361) 0.012 0.914 1.083 (0.740 1.636) 0.171 0.679 1.256 (0.615 1.416) 0.015 0.902 1.050 (0.715 – – – – GCKR rs780094 T allele might increase the risk of NAFLD in European populations but not in Asian populations.15,17,18,33 A higher prevalence of NAFLD has been observed in some Indians that possess European ancestry, as opposed to those that possess Asian ancestry, which might explain the above hypothesis.34 The GCKR gene encodes GCKRP, an enzyme that plays a regulatory role in hepatic GCK activity.35,36 The GCKR rs1260326 functional variant P446L increases GCK activity by down-regulating the level of fructose 6-phosphate.37 NAFLD + CAD OR (95%CI) Increased GCK activity is tightly accompanied by elevated hepatic glycolytic flux, de novo lipogenesis, and hepatic TG level.13,38 In this study, our results showed that the GCKR rs780094 and rs1260326 T allele were significantly associated with decreasing serum FPG level among the overall

rs780094 and rs1260326 polymorphisms in the study groups patients and the NAFLD + CAD patients. These results suggest that GCKR rs780094 and rs1260326 may repress

GCKR the activity of GCKRP to decrease the serum FPG level. (%). Previous studies also investigated the association of GCKR n polymorphisms with the risk of CAD. In a dietary intervention study, Shen et al.39 found that the GCKR rs1260326 T allele may increase the risk of atherosclerosis. Lian et al.40 demonstrated the positive association of GCKR rs780093 with the risk of CAD in an older Chinese Han population, through a case-control study. On the contrary, Járomi et al.41 did not CCCT + TT 126 26 (82.9%) (17.1%) 116 (81.7%) 26 (18.3%) 64 (78.0%) 18 (22.0%) 0.837 (0.455 CT 133 (43.8%) 171 (56.2%) 123 (43.3%) 161 (56.7%) 75 (45.7%) 89 (54.3%) 0.982 (0.709 CCCT + TT 129 23 (84.9%) (15.1%) 118 (83.1%) 24 (16.9%) 67 (81.7%) 15 (18.3%) 0.877 (0.470 CT 128 (42.1%) 176 (57.9%) 121 (42.6%) 163 (57.4%) 71 (43.3%) 93 (56.7%) 1.021detect (0.736 any association of GCKR rs1260326 with the suscept- Genotype Controlsrs780094 NAFLD Alleles rs1260326 Alleles Values are expressed as Data were compared by chi-square test. 42 Table 3. Distribution of the Abbreviations: NAFLD, patientsa with nonalcoholic fatty liver disease;b NAFLD + CAD, patients with nonalcoholic fatty liver disease and coronary art ibility of stroke. Bi et al. also found that the GCKR rs780094

300 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 297–303 a H. Gao Table 4. Clinical characteristics of GCKR rs780094 T carriers and non-carriers in the study population

Controls NAFLD NAFLD + CAD tal et Carriers, Non-carriers, Carriers, Non-carriers, Carriers, Non-carriers,

Characteristic n = 126 n =26 p value n = 117 n =26 p value n =64 n =18 p value CAD and NAFLD with GCKR :

# # BMI (kg/m2) 22.91 6 4.36 24.44 6 3.40 0.001 26.35 6 3.50 27.05 6 5.87 0.659 25.29 6 2.422 25.05 6 3.250 0.73 FPG (mmol/L) 4.45 6 1.00 4.74 6 1.00 0.043 4.74 6 0.65 4.86 6 0.60 0.405 5.18 6 1.38 6.04 6 3.16 0.001

# # TC (mmol/L) 5.04 6 1.00 5.09 6 2.00 0.860 5.49 6 1.03 5.48 6 1.28 0.947 4.52 6 1.236 4.69 6 1.430 0.616 TG (mmol/L) 1.08 6 1.00 1.64 6 2.00 0.002 1.53 6 1.01 1.75 6 1.19 0.252 1.49 6 1.20 1.65 6 1.33 0.751 ora fCiia n rnltoa Hepatology Translational and Clinical of Journal HDL (mmol/L) 1.29 6 0.00 1.15 6 0.00 0.297 1.23 6 0.27 1.23 6 0.21 0.325 1.03 6 0.33 1.06 6 0.28 0.396

# # # LDL (mmol/L) 3.07 6 0.697 3.23 6 0.754 0.295 3.25 6 0.547 3.36 6 0.586 0.349 2.74 6 0.948 2.78 6 1.083 0.866 ALT (U/L) 16.83 6 11.00 19.29 6 11.00 0.996 24.81 6 23.66 21.77 6 40.27 0.950 22.78 6 18.59 20.41 6 13.03 0.320 AST (U/L) 22.73 6 6.00 20.33 6 6.00 0.247 22.20 6 8.50 20.38 6 13.11 0.527 21.30 6 18.61 20.34 6 15.07 0.251 r-GT (mmol/L) 18.89 6 13.00 22.02 6 14.00 0.143 29.40 6 26.76 37.35 6 23.79 0.294 25.62 6 16.73 23.19 6 16.95 0.406 ALP (mmol/L) 69.92 6 26.00 77.06 6 36.00 0.417 66.87 6 24.07 74.80 6 22.73 0.056 79.85 6 23.10 84.77 6 28.11 0.809 BIL (mmol/L) 12.70 6 6.00 11.50 6 4.00 0.432 11.90 6 5.10 12.30 6 3.60 0.587 13.05 6 6.15 12.65 6 5.20 0.397

Abbreviations: NAFLD, patients with nonalcoholic fatty liver disease; NAFLD + CAD, patients with nonalcoholic fatty liver disease and coronary artery disease. # Values are expressed as mean ± SD and compared by Student’s t test. a Values are expressed as median ± quartile.

a Table 5. Clinical characteristics of GCKR rs1260326 T carriers and non-carriers in the study population

Controls NAFLD NAFLD + CAD 2019 Carriers, Non-carriers, Carriers, Non-carriers, Carriers, Non-carriers, Characteristic n = 129 n =23 p value n = 119 n =24 p value n =67 n =15 p value o.7|297 | 7 vol. # # BMI (kg/m2) 22.96 6 4.35 24.46 6 3.81 0.003 26.35 6 3.48 27.05 6 6.06 0.665 25.27 6 2.616 25.11 6 2.648 0.836 FPG (mmol/L) 4.49 6 1.00 4.76 6 1.00 0.054 4.75 6 0.67 4.86 6 0.56 0.586 5.22 6 1.43 6.02 6 3.15 0.051

# TC (mmol/L) 5.03 6 1.00 5.13 6 2.00 0.896 5.44 6 1.00 5.89 6 4.33 0.713 4.51 6 1.224 4.76 6 1.506 0.496 TG (mmol/L) 1.08 6 1.00 1.61 6 2.00 0.006 1.54 6 1.00 1.75 6 1.17 0.366 1.46 6 1.23 1.88 6 1.21 0.710 – 0 301 303 HDL (mmol/L) 1.29 6 0.00 1.12 6 0.00 0.260 1.23 6 0.28 1.23 6 0.21 0.369 1.03 6 0.33 1.06 6 0.30 0.564

# # # # LDL (mmol/L ) 3.08 6 0.708 3.20 6 0.709 0.446 3.25 6 0.542 3.37 6 0.61 0.324 2.72 6 0.935 2.85 6 1.15 0.635 ALT (U/L) 16.81 6 11.00 19.70 6 10.00 0.729 24.81 6 22.52 21.77 6 42.58 0.918 22.70 6 18.14 22.63 6 15.31 0.488 AST (U/L) 20.30 6 6.00 22.91 6 6.00 0.085 22.20 6 8.27 20.38 6 14.61 0.647 21.29 6 18.39 20.36 6 20.79 0.358 r-GT (mmol/L) 18.79 6 13.00 22.30 6 13.00 0.135 29.40 6 25.59 37.35 6 27.63 0.296 25.79 6 16.92 22.91 6 14.91 0.503 ALP (mmol/L) 69.71 6 26.00 80.25 6 32.00 0.257 66.87 6 22.96 75.67 6 23.55 0.040 80.20 6 23.36 80.02 6 30.26 0.802 BIL (mmol/L) 12.50 6 6.00 11.70 6 5.00 0.593 11.90 6 6.00 12.30 6 3.80 0.636 13.10 6 5.70 11.10 6 5.20 0.400

SNP had no significant association with the incidence of CAD [3] Ibrahim SH, Hirsova P, Gores GJ. Non-alcoholic steatohepatitis pathogenesis: or stroke. SNPs of GCKR rs780094 showed strong linkage sublethal hepatocyte injury as a driver of liver inflammation. Gut 2018;67: 24,39,40 963–972. doi: 10.1136/gutjnl-2017-315691. disequilibrium with rs1260326. Our results showed [4] Fiorucci S, Biagioli M, Distrutti E. Future trends in the treatment of non-alco- that GCKR rs780094 and rs1260326 genotype distributions holic steatohepatitis. Pharmacol Res 2018;134:289–298. doi: 10.1016/j. were not associated with the occurrence of CAD in NAFLD phrs.2018.07.014. patients but that the rs780094 T allele could decrease the [5] Younossi ZM, Stepanova M, Negro F, Hallaji S, Younossi Y, Lam B, et al. Non- serum FPG levels in the NAFLD + CAD group. These results alcoholic fatty liver disease in lean individuals in the United States. Medicine (Baltimore) 2012;91:319–327. doi: 10.1097/MD.0b013e3182779d49. indicated that GCKR rs780094 and rs1260326 were not the [6] Liangpunsakul S, Chalasani N. Lipid mediators of liver injury in nonalcoholic risk factor of CAD in NAFLD patients but that the rs780094 fatty liver disease. Am J Physiol Gastrointest Liver Physiol 2019;316:G75– T allele could affect serum FPG level in the NAFLD + CAD G81. doi: 10.1152/ajpgi.00170.2018. patients of our Chinese Northern Han population. [7] Feng S, Dai Z, Liu AB, Huang J, Narsipur N, Guo G, et al. Intake of stigmas- terol and b-sitosterol alters lipid metabolism and alleviates NAFLD in mice fed Some limitations of our study should be acknowledged. a high-fat western-style diet. Biochim Biophys Acta Mol Cell Biol Lipids 2018; First, all the subjects in this study were of the Chinese 1863:1274–1284. doi: 10.1016/j.bbalip.2018.08.004. Northern Han population, so that the applicability of our [8] Anstee QM, Day CP. The genetics of NAFLD. Nat Rev Gastroenterol Hepatol conclusion to other ethnic populations requires further study 2013;10:645–655. doi: 10.1038/nrgastro.2013.182. [9] Vilar CP, Cotrim HP, Florentino GS, Bragagnoli G, Schwingel PA, Barreto CP. for confirmation. Second, ultrasonography was used to diag- Nonalcoholic fatty liver disease in patients with coronary disease from a nose NAFLD because of the difficulty in conducting liver Brazil northeast area. Arq Gastroenterol 2015;52:111–116. doi: 10. biopsy. Third, our study did not grade the severity of NAFLD 1590/S0004-28032015000200007. patients. Finally, a greater number of subjects should be [10] Krishan S, Jain D, Bathina Y, Kale A, Saraf N, Saigal S, et al. Non-invasive included in the further studies to verify this conclusion in the quantification of hepatic steatosis in living, related liver donors using dual- echo Dixon imaging and single-voxel proton spectroscopy. Clin Radiol 2016; Chinese Han population and other ethnic populations. 71:58–63. doi: 10.1016/j.crad.2015.10.002. [11] Zhang Q, Zhang HM, Qi WQ, Zhang YG, Zhao P, Jiao J, et al. 3.0T 1H magnetic Conclusions resonance spectroscopy for assessment of steatosis in patients with chronic hepatitis C. World J Gastroenterol 2015;21:6736–6744. doi: 10.3748/wjg. v21.i21.6736. In summary, we investigated the relationship of GCKR [12] León-Mimila P, Vega-Badillo J, Gutiérrez-Vidal R, Villamil-Ramírez H, Villar- rs780094 and rs1260326 gene polymorphisms with the risk eal-Molina T, Larrieta-Carrasco E, et al. A genetic risk score is associated with of NAFLD as well as the risk of CAD in NAFLD patients in the hepatic triglyceride content and non-alcoholic steatohepatitis in Mexicans – Chinese Northern Han population. We concluded that GCKR with morbid obesity. Exp Mol Pathol 2015;98:178 183. doi: 10.1016/j. yexmp.2015.01.012. rs780094 and rs1260326 gene polymorphisms did not asso- [13] Peter A, Stefan N, Cegan A, Walenta M, Wagner S, Königsrainer A, et al. ciate with the risk of NAFLD nor with the risk of CAD in NAFLD Hepatic glucokinase expression is associated with lipogenesis and fatty patients in the Chinese Northern Han population. In addition, liver in humans. J Clin Endocrinol Metab 2011;96:E1126–E1130. doi: 10. the GCKR rs780094 T and rs1260326 T alleles did affect the 1210/jc.2010-2017. [14] Iynedjian PB. Molecular physiology of mammalian glucokinase. Cell Mol Life BMI value and serum FPG and TG levels in the overall subjects Sci 2009;66:27–42. doi: 10.1007/s00018-008-8322-9. and NAFLD + CAD patients. [15] Speliotes EK, Yerges-Armstrong LM, Wu J, Hernaez R, Kim LJ, Palmer CD, et al. Genome-wide association analysis identifies variants associated with nonalcoholic fatty liver disease that have distinct effects on metabolic traits. Acknowledgments PLoS Genet 2011;7:e1001324. doi: 10.1371/journal.pgen.1001324. [16] Tan HL, Zain SM, Mohamed R, Rampal S, Chin KF, Basu RC, et al. Association This study was supported by grants from the National Natural of glucokinase regulatory gene polymorphisms with risk and severity of non- Science Foundation of China (31770837) and the Qingdao alcoholic fatty liver disease: an interaction study with adiponutrin gene. J – ’ Gastroenterol 2014;49:1056 1064. doi: 10.1007/s00535-013-0850-x. People s Livelihood Science and Technology Plan (18-6-1-68-nsh). [17] Yang H, Chen G, Song C, Li D, Ma Q, Chen G, et al. A novel index including SNPs for the screening of nonalcoholic fatty liver disease among elder Conflict of interest Chinese: A population-based study. Medicine (Baltimore) 2018;97:e0272. doi: 10.1097/MD.0000000000010272. [18] Wang X, Liu Z, Wang K, Wang Z, Sun X, Zhong L, et al. Additive effects of the The authors have no conflict of interests related to this risk alleles of PNPLA3 and TM6SF2 on non-alcoholic fatty liver disease publication. (NAFLD) in a Chinese population. Front Genet 2016;7:140. doi: 10. 3389/fgene.2016.00140. [19] Yang Z, Wen J, Tao X, Lu B, Du Y, Wang M, et al. Genetic variation in the GCKR Author contributions gene is associated with non-alcoholic fatty liver disease in Chinese people. Mol Biol Rep 2011;38:1145–1150. doi: 10.1007/s11033-010-0212-1. [20] Shang XR, Song JY, Liu FH, Ma J, Wang HJ. GWAS-identified common var- Study concept and design (YX and SX), acquisition of the data iants with nonalcoholic fatty liver disease in Chinese children. J Pediatr Gas- (HG, SL, ZZ, XY, and QL), analysis and interpretation of the troenterol Nutr 2015;60:669–674. doi: 10.1097/MPG.0000000000000662. data (HG, SL, and ZZ), drafting of the manuscript (HG and [21] Lin YC, Chang PF, Chang MH, Ni YH. Genetic variants in GCKR and PNPLA3 SL), critical revision of the manuscript for important intellec- confer susceptibility to nonalcoholic fatty liver disease in obese individuals. Am J Clin Nutr 2014;99:869–874. doi: 10.3945/ajcn.113.079749. tual content (YX and SX) supervision (YX and SX). All the [22] Kumar A, Shalimar, Walia GK, Gupta V, Sachdeva MP. Genetics of nonalco- authors read and approved the final manuscript. holic fatty liver disease in Asian populations. J Genet 2019;98:29. doi: 10. 1007/s12041-019-1071-8. [23] Saxena R, Voight BF, Lyssenko V, Burtt NP, de Bakker PI, Chen H, et al. References Genome-wide association analysis identifies loci for type 2 diabetes and triglyceride levels. Science 2007;316:1331–1336. doi: 10.1126/science.1142358. [1] Danford CJ, Yao ZM, Jiang ZG. Non-alcoholic fatty liver disease: a narrative [24] Sparsø T, Andersen G, Nielsen T, Burgdorf KS, Gjesing AP, Nielsen AL, et al. review of genetics. J Biomed Res 2018;32:389–400. doi: 10.7555/JBR.32. The GCKR rs780094 polymorphism is associated with elevated fasting serum 20180045. triacylglycerol, reduced fasting and OGTT-related insulinaemia, and reduced [2] Younossi Z, Anstee QM, Marietti M, Hardy T, Henry L, Eslam M, et al. Global risk of type 2 diabetes. Diabetologia 2008;51:70–75. doi: 10.1007/s00125- burden of NAFLD and NASH: trends, predictions, risk factors and prevention. 007-0865-z. Nat Rev Gastroenterol Hepatol 2018;15:11–20. doi: 10.1038/nrgastro. [25] Kawaguchi T, Shima T, Mizuno M, Mitsumoto Y, Umemura A, Kanbara Y, et al. 2017.109. Risk estimation model for nonalcoholic fatty liver disease in the Japanese

302 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 297–303 Gao H. et al: GCKR with NAFLD and CAD

using multiple genetic markers. PLoS One 2018;13:e0185490. doi: 10. [34] Bamshad M, Kivisild T, Watkins WS, Dixon ME, Ricker CE, Rao BB, et al. 1371/journal.pone.0185490. Genetic evidence on the origins of Indian caste populations. Genome Res [26] Petit JM, Masson D, Guiu B, Rollot F, Duvillard L, Bouillet B, et al. GCKR poly- 2001;11:994–1004. doi: 10.1101/gr.gr-1733rr. morphism influences liver fat content in patients with type 2 diabetes. Acta [35] Hayward BE, Dunlop N, Intody S, Leek JP, Markham AF, Warner JP, et al. Diabetol 2016;53:237–242. doi: 10.1007/s00592-015-0766-4. Organization of the human glucokinase regulator gene GCKR. Genomics [27] Di Costanzo A, Pacifico L, D’Erasmo L, Polito L, Martino MD, Perla FM, et al. 1998;49:137–142. doi: 10.1006/geno.1997.5195. Nonalcoholic fatty liver disease (NAFLD), but not its susceptibility gene var- [36] Grimsby J, Coffey JW, Dvorozniak MT, Magram J, Li G, Matschinsky FM, et al. iants, influences the decrease of kidney function in overweight/obese chil- Characterization of glucokinase regulatory protein-deficient mice. J Biol – dren. Int J Mol Sci 2019;20:E4444. doi: 10.3390/ijms20184444. Chem 2000;275:7826 7831. doi: 10.1074/jbc.275.11.7826. [28] Simons PIHG, Simons N, Stehouwer CDA, Schalkwijk CG, Schaper NC, [37] Beer NL, Tribble ND, McCulloch LJ, Roos C, Johnson PR, Orho-Melander M, Brouwers MCGJ. Association of common gene variants in glucokinase regu- et al. The P446L variant in GCKR associated with fasting plasma glucose and latory protein with cardiorenal disease: A systematic review and meta-anal- triglyceride levels exerts its effect through increased glucokinase activity in liver. Hum Mol Genet 2009;18:4081–4088. doi: 10.1093/hmg/ddp357. ysis. PLoS One 2018;13:e0206174. doi: 10.1371/journal.pone.0206174. [38] Rees MG, Wincovitch S, Schultz J, Waterstradt R, Beer NL, Baltrusch S, et al. [29] Rickham PP. Human experimentation. Code of ethics of the world medical Cellular characterisation of the GCKR P446L variant associated with type 2 association. declaration of helsinki. Br Med J 1964;2:177. doi: 10.1136/bmj. diabetes risk. Diabetologia 2012;55:114–122. doi: 10.1007/s00125-011- 2.5402.177. 2348-5. [30] Kotronen A, Westerbacka J, Bergholm R, Pietiläinen KH, Yki-Järvinen H. Liver [39] Shen H, Pollin TI, Damcott CM, McLenithan JC, Mitchell BD, Shuldiner AR. fat in the metabolic syndrome. J Clin Endocrinol Metab 2007;92:3490–3497. Glucokinase regulatory protein gene polymorphism affects postprandial doi: 10.1210/jc.2007-0482. lipemic response in a dietary intervention study. Hum Genet 2009;126: [31] Xu Y, Lu LL, Liu SS, Du SX, Zhu HL, Dong QJ, et al. Apolipoprotein A5 gene 567–574. doi: 10.1007/s00439-009-0700-3. polymorphisms are associated with non-alcoholic fatty liver disease. Hepato- [40] Lian J, Guo J, Chen Z, Jiang Q, Ye H, Huang X, et al. Positive association – biliary Pancreat Dis Int 2018;17:214 219. doi: 10.1016/j.hbpd.2018.04.004. between GCKR rs780093 polymorphism and coronary heart disease in the [32] Flores YN, Velázquez-Cruz R, Ramírez P, Bañuelos M, Zhang ZF, Yee HF Jr, aged Han Chinese. Dis Markers 2013;35:863–868. doi: 10.1155/2013/215407. et al. Association between PNPLA3 (rs738409), LYPLAL1 (rs12137855), [41] Járomi L, Csöngei V, Polgár N, Szolnoki Z, Maász A, Horvatovich K, et al. PPP1R3B (rs4240624), GCKR (rs780094), and elevated transaminase Functional variants of glucokinase regulatory protein and apolipoprotein A5 levels in overweight/obese Mexican adults. Mol Biol Rep 2016;43:1359– genes in ischemic stroke. J Mol Neurosci 2010;41:121–128. doi: 10. 1369. doi: 10.1007/s11033-016-4058-z. 1007/s12031-009-9301-9. [33] Kitamoto A, Kitamoto T, Nakamura T, Ogawa Y, Yoneda M, Hyogo H, et al. [42] Bi M, Kao WH, Boerwinkle E, Hoogeveen RC, Rasmussen-Torvik LJ, Astor BC, Association of polymorphisms in GCKR and TRIB1 with nonalcoholic fatty et al. Association of rs780094 in GCKR with metabolic traits and incident liver disease and metabolic syndrome traits. Endocr J 2014;61:683–689. diabetes and cardiovascular disease: the ARIC Study. PLoS One 2010;5: doi: 10.1507/endocrj.ej14-0052. e11690. doi: 10.1371/journal.pone.0011690.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 297–303 303 Original Article

Minimal Hepatic Encephalopathy in Indians: Psychometric Hepatic Encephalopathy Score and Inhibitory Control Test for Diagnosis and Rifaximin or Lactulose for Its Reversal

Vinay B. Pawar*, Ravindra G. Surude, Nikhil Sonthalia, Vinay Zanwar, Samit Jain, Qais Contractor and Pravin M. Rathi

Department of Gastroenterology, Topiwala National Medical College and BYL Ch Hospital, Mumbai, Maharashtra, India

Abstract Introduction

Background and Aims: Psychometric hepatic encephalop- Hepatic encephalopathy (HE) consists of a spectrum of athy score (PHES) is used widely for diagnosis of minimal neuropsychiatric abnormalities that are seen in patients with hepatic encephalopathy (MHE). This prospective study aimed liver dysfunction and after exclusion of other known neuro- to determine the utility of the inhibitory control test (ICT) for logical diseases.1–3 Minimal hepatic encephalopathy (MHE), a the diagnosis of MHE. Additionally, the efficacy of rifaximin subgroup of HE is a neurocognitive complication of cirrhosis. and lactulose for reversal of MHE was evaluated. Methods: Most of the time, MHE remains undiagnosed and is diagnosed A total of 180 eligible cirrhotic patients underwent testing for in 20–80% of patients with liver cirrhosis with the help of MHE. When PHES was # −5 and ICT lures were $ 14, MHE special diagnostic testing.1 It can progress to overt HE and was diagnosed. The 108 patients with MHE were randomized impaired health-related quality of life (HRQOL).4,5 Further- to three groups for treatment with either lactulose, rifaximin, more, MHE is associated with driving impairment, resulting or placebo. Treatment outcomes were measured at the end of in high risk of road traffic accidents for these patients.6,7 3 months. Results: The 108 patients with MHE diagnosed by MHE includes a specific spectrum of cognitive deficits, PHES and/or ICT accounted for 60%. The diagnosis of MHE which mostly involve the domains of response inhibition, was made by both ICT and PHES positivity in 56 patients, by executive function, attention and vigilance.8–10 Psychometric abnormal ICT and normal PHES in 37 patients, and by abnor- tests are considered the gold standard for the diagnosis of mal PHES and normal ICT in 15 patients. For diagnosis of MHE. These psychometric tests specifically assess the MHE, ICT had sensitivity of 78.87%, specificity of 66.06% above-mentioned cognitive domains.11 The Psychometric with 60.22% positive predictive value and 82.76% negative Hepatic Encephalopathy Score (PHES) is composed of five predictive value. An area under the curve value of 0.724 tests, including the number connection test-A (NCT-A), (95% CI: 0.653–0.788) was obtained for diagnosis of MHE. number connection test-B (NCT-B), serial dotting test Reversal of MHE was seen in 71.42%, 70.27% and 11.11% (SDT), line tracing test (LTT) and digit symbol test (DST). of patients in the rifaximin, lactulose and placebo arms As such, the PHES can be used to assess a number of param- (p < 0.001). Rifaximin showed better tolerability compared eters, like motor speed, motor accuracy, concentration, to lactulose. Conclusions: For the diagnosis of MHE, ICT is a attention, visual perception, visual-spatial orientation, visual simple tool but has lower sensitivity and better specificity construction and memory. These are related to most of the than PHES. Rifaximin is as efficacious as lactulose in the treat- neuropsychological impairments in MHE. ment of MHE and better tolerated. There is also a computer-based test known as the Inhib- Citation of this article: Pawar VB, Surude RG, Sonthalia N, itory Control Test (ICT), which was constructed for the Zanwar V, Jain S, Contractor Q, et al. Minimal hepatic ence- evaluation of attention and response inhibition. It has been phalopathy in Indians: Psychometric hepatic encephalopathy used previously to characterize traumatic brain injury, atten- score and inhibitory control test for diagnosis and rifaximin or tion deficit disorder, and schizophrenia.12 – lactulose for its reversal. J Clin Transl Hepatol 2019;7(4):304 It is an established fact that the clinical manifestations of 312. doi: 10.14218/JCTH.2017.00037. HE in each stage can be reversed when managed by adequate therapeutic management. As mentioned above, MHE is currently underdiagnosed in patients with liver cirrhosis. MHE carries a high potential for progression to overt HE. Mean- Keywords: Minimal hepatic encephalopathy; ICT; PHES; Indians; Lactulose; while, the current management of MHE is directed at reduc- Rifaximin. tion of ammonia production from the gastrointestinal tract. Abbreviations: CTP, Child-Turcotte-Pugh; DST, digit symbol test; HE, hepatic This can be achieved by administration of lactulose, a non- encephalopathy; HRQOL, health-related quality of life; ICT, inhibitory control test; LTT, line tracing test; MELD, model of end-stage liver disease; MHE, absorbable disaccharide. MHE can also be managed with non- 13–16 minimal hepatic encephalopathy; PHES, psychometric hepatic encephalopathy absorbable antibiotic rifaximin. score; SDT, serial dotting test. Randomized controlled studies and a meta-analysis17 Received: 4 June 2017; Revised: 28 May 2018; Accepted: 14 February 2019 have compared probiotics, prebiotics and a symbiotic with *Correspondence to: Vinay B. Pawar, Department of Gastroenterology, Topiwala National Medical College and BYL Ch Hospital, Dr. A.L Nair Road, Mumbai, Maha- lactulose in the treatment of MHE. These studies have rashtra 400008, India. Tel: +22-23021639, E-mail: [email protected] shown that in MHE reversal, lactulose is equally or more

304 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 304–312

Copyright: © 2019 Authors. This article has been published under the terms of Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0), which permits noncommercial unrestricted use, distribution, and reproduction in any medium, provided that the following statement is provided. “This article has been published in Journal of Clinical and Translational Hepatology at DOI: 10.14218/JCTH.2017.00037 and can also be viewed on the Journal’s website at http://www.jcthnet.com”. Pawar V.B. et al: MHE in cirrhotics effective than the pre-pro-synbiotics.18,19 Rifaximin is a broad cirrhotic patients, the PHES score cutoff has been found to spectrum antibiotic; moreover, it is gut-selective and has # −5 for abnormal designation. This cutoff can reliably be minimal systemic absorption. It has a tolerability profile and used for the diagnosis of MHE.20 In the present study, we good safety. The previous studies carried out on MHE patients used PHES scores of # −5 for the diagnosis of MHE. have shown that rifaximin improves cognitive functions and ICT: This test consists of presentation of several letters at HRQOL, as compared to placebo.13,20 a specific interval rate (500-millisecond intervals).21 It is In this study, we compared PHES and ICT for the diagnosis identical to the continuous performance test. X and Y letters of MHE and assessed the efficacy of rifaximin and lactulose in are interspersed between these letters. The subject is asked reversal of the disease. to respond to every X and Y during the initial part of the training run, which shows the pre-potent response. In the latter Methods part of the training run, the subject is asked only to respond when X and Y are alternating (also called as targets) and to Plan of research inhibit responding when X and Y are not alternating (also called ‘lures’). After the training run, 6 test runs, which last The study was conducted in a gastroenterology tertiary care approximately 2 minutes each, are administered. The patient center in Mumbai, India, from May 2015 to March 2017. The is presented with a total of 40 lures, 212 targets, and 1728 PHES and ICT were compared in the diagnosis of MHE. We random letters in between. At the end of the test, the number also compared the efficacy of rifaximin and lactulose in the of lures, target frequency, lure and target response rates, and reversal of MHE. The study was approved by the Institutional target reaction times are automatically calculated. Good psy- Ethics Committee. Written informed consent was obtained chometric performance is indicated by lower lure response from each participant before their enrollment. and higher target response. In the initial studies in the literature, it was revealed that Characteristics of subjects/patients the most important variable is a lure threshold of >5 for the diagnosis of MHE.7 The initial lure threshold of >5 was set too Patient selection (Fig. 1): A total of 250 patients, aged low for our population, mostly reflecting the high educational – 18 65 years and with established diagnosis of liver cirrhosis status of the selected control population. Indian patients without any overt symptoms of HE at the time of testing whom we encounter in our tertiary care center are less famil- (West Haven grade 0), were recruited to the study. The diag- iar with computer use. Since the lure threshold needs to be nosis of cirrhosis was based on clinical history, serologic adjusted according to the local population norms, we took the testing, radiologic imaging and liver histology, when avail- lure threshold of $14 as the diagnostic cut off for diagnosis of able. All the patients were either hospitalized or attending HE.22,23 In our study, we considered only lures and not the Outpatient Clinic of Gastroenterology. After the initial weighted lures as suggested by Amodio et al. (unpublished assessment, 180/250 (72%) of the patients were included study). in the study. Exclusion criteria were: (1) history of recent Treatment groups: In the current study, patients under- alcohol intake or substance abuse in the last 12 weeks; went a comprehensive physical examination and a detailed (2) history of recent gastrointestinal tract bleeding; (3) serious neurological examination that included the Mini Mental infections and use of drugs affecting psychometric performance, Status Examination. This was to rule-out any overt cognitive like antidepressants and sedatives in the last 4 weeks; (4) pre- impairment before administering neuropsychological tests vious transjugular intrahepatic portosystemic shunt or shunt and the ICT. Only those patients with a Mini Mental Status surgery; (5) significant comorbid illness (e.g. heart, respiratory Examination score of >24 were included. We also performed or renal failure); (6) visual or mental impairment; (7) neurologic a battery of laboratory investigations, like complete hemo- diseases, like stroke, epilepsy, dementia, and parkinsonism; (8) hepatocellular carcinoma; (9) prior treatment exposure to gram, liver function tests, renal function tests, ultrasonogra- lactulose or rifaximin, recent antibiotic use, or current probiotics; phy, esophagogastroduodenoscopy and other etiological tests, (10) past history of HE. as and when required. The eligible patients were classified Among the 70 patients who were excluded, a history of according to disease severity, determined by Child-Turcotte- overt HE was present in 10, history of recent alcohol con- Pugh (CTP) classification and model for end-stage liver sumption in 15, recent variceal bleed in 15, acute kidney disease (MELD). Out of the 180 eligible patients who under- injury in 15, and hepatocellular carcinoma in 5. Out of 180 went neuropsychological assessment and ICT, 108 were $ total recruited patients, 30 were hospitalized, 22 of which had positive for MHE by either ICT with lures 14 or PHES score MHE (73.33%). The reasons for hospitalization included new- of #−5, or both. Subsequently patients were randomized to onset ascites, alcoholic hepatitis, and refractory ascites. three groups for treatment with either lactulose (30–60 mL/day to have 2–3 loose stools/day) or rifaximin (550 mg 2 times/day), Study protocol or B-complex tablets as a placebo (2 times/day). We used the titrated dose of lactulose. Neuropsychological assessment: The PHES has been vali- The patients were followed up in our outpatient depart- dated in many populations of various nationalities, including ment every 2 weeks in the first month and then on a once- German, Spanish, Italian, and Indian.20 It includes a battery monthly schedule. At every visit, the patients were assessed of five tests: DST, number connection test-A, number connec- for the development of overt HE. Treatment compliance was tion test-B, LTT for time (t) and error (e), and serial dotting ensured by pill count (checking empty blisters during each test (SDT). The PHES is calculated from the results of these visit), and confirmed by questioning of the patient’s accom- five tests. The clinical significance of the PHES score has been panying relatives. The treatment outcomes were evaluated at previously evaluated in a large number of healthy volunteers, the end of 3 months in the form of improvement in PHES score in addition to patients with liver cirrhosis with MHE. In Indian and ICT findings.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 304–312 305 Pawar V.B. et al: MHE in cirrhotics

Fig. 1. Consort diagram. Abbreviations: ICT, inhibitory control test; MHE, minimal hepatic encephalopathy; OHE, overt hepatic encephalopathy; PHES, psychometric hepatic encephalopathy score.

Outcome/end points: The primary outcome was rever- compare cirrhotic patients with and without MHE. The Wil- sal of MHE at 3 months. Reversal of MHE was defined as PHES coxon sign rank test was used to compare the pre- and post- score of >-5 and or ICT lures <14. treatment results of neuropsychometric tests and of ICT. A Sample size: Previous studies have established MHE p-value of <0.05 was considered significant. reversal in 75.5% of patients treated with rifaximin,24 and a 18 mean of 47.5% of those treated with lactulose, and 10% Result among those receiving no treatment.18 For power of 80%, the total sample size of 105 patients (35 patients in each group) A total of 250 patients with cirrhosis were evaluated for was determined as necessary to obtain accurate results. eligibility, and the 180 patients included in the study (fulfilling Randomization: Patients were randomized by an inde- eligibility criteria) underwent ICT and PHES to establish the pendent observer according to a computer-generated list of diagnosis of MHE. The clinical and personal characteristics of random numbers. The study was carried out double-blinded, the patients are shown in Table 1. with both patients and investigators blinded to the study Out of 180 patients, 108 (60%) were diagnosed with MHE treatment. by PHES and/or ICT. Of the 108 MHE patients, 98 (90.74%) Statistical methods: Data analysis was done by using were males. A total of 33 patients belonged to CTP class A, the Statistical Package for Social Sciences version 20 (IBM while 55 patients were CTP class B and 20 were CTP C. The Corp., Armonk, NY, USA). The chi-square test/Fischer’s baseline MELD score was 10 in patients, with or without MHE. exact test were used to find the association between various Amongst the patients with MHE, alcohol-related chronic liver qualitative data variables within the treatment groups and to disease was the most common diagnosis, being present in determine the association between demographic variables 63 patients, followed by hepatitis B-related cirrhosis in 18 with occurrence of MHE. To find the cutoff value of lures in patients. patients with MHE, receiver operating characteristic curves Diagnosis of MHE was made in 56 patients by abnormal were generated. The Mann-Whitney U test was used to ICT and PHES, by abnormal ICT and normal PHES in

306 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 304–312 Pawar V.B. et al: MHE in cirrhotics

Table 1. Baseline characteristics of 180 patients included in the study

Variables Lactulose arm Rifaximin arm Placebo arm Non-MHE

Age in years, mean6standard deviation 48.468.23 48.8666.75 48.6967.22 48.3268.18 Males, n 31 34 33 60 Education in years, median (min-max) 10 (6–13) 9 (6–15) 10 (6–15) 9 (6–15) Baseline CTP class Class A, n 12 11 10 27 Class B, n 17 20 18 29 Class C, n 66816 MELD score 10 (8–13) 10 (6–14) 10 (8–13) 10 (6–14) Etiology of cirrhosis Alcohol, n 20 21 22 50 NASH, n 3556 Hepatitis B, n 4779 Hepatitis C, n 2334 AIH, n 1122 BCS, n 1011

Abbreviations: AIH, autoimmune hepatitis; BCS, Budd-Chiari syndrome; CTP, Child-Turcotte-Pugh; MELD, model for end-stage liver disease; NASH, nonalcoholic steatohepatitis.

37 patients, and by abnormal PHES and normal ICT in 15 patients in the lactulose arm, 70.27% of patients in the rifax- patients. Sensitivity and specificity of PHES was 65.7% and imin arm, and 11.11% of patients in the placebo arm 100% with cutoff value for PHES of # −5. An area under the (p<0.001). For the diagnosis of MHE by ICT, we considered curve value of 0.841 (95% CI: 0.779–0.891) was obtained for only lures and not the weighted lures. Though the lures diagnosis of MHE by PHES. Sensitivity and specificity of ICT improved in patients with placebo, the improvement was was 78.87% and 66.06% for the diagnosis of MHE, with 60.22% positive predictive value and 82.76% negative predictive value. An area under the curve value of 0.724 (95% CI: 0.653–0.788) was obtained for diagnosis of MHE (Fig. 2). The ICT lures were significantly higher (Table 2) in cirrhotic patients with MHE (17.48±2.62; 95% CI for mean difference: 6.87–8.15), than for those without MHE at baseline (10.01 ±2.01) (p<0.001). The presence of ICT targets was significantly lower in cirrhotic patients with MHE (176.54±6.67; 95%CI for mean difference: −19.54–16.46), than for those without MHE (194.54±3.32) (p<0.001). The mean values (standard deviation) for NCT-A, NCT-B, SDT, LTT, and DST were 92.29 (14.75), 106.84 (24.60), 90.47 (14.83), 103.31(20.88), and 16.47 (9.58) respectively in patients with MHE and were significantly higher compared to patients without MHE (for which the values were 72.88 (5.99), 79.43 (8.62), 75.67 (8.01), 82.75 (9.12), and 18.07 (3.20) respectively; p<0.001) (Table 3). Table 4 shows results of the neuropsychiatric tests and ICT in patients with MHE before and after treatment (Figs. 3 and 4). In the lactulose arm, MHE was present by abnormal ICT in 30 patients, out of which 23 showed reversals, and by abnormal PHES in 23 patients, out of which 14 showed reversal. Reversal of MHE was seen in 25/35 patients in the lactulose arm. In the rifaximin arm, MHE was present by abnormal ICT in 31 patients, out of which 21 showed reversal, and by abnormal PHES in 24 patients, out of which 19 showed reversal. Reversal of MHE was seen in 26/37 patients in the rifaximin arm. In the placebo group, MHE was present by Fig. 2. Receiver operating characteristic for ICT with cutoff value of 14 abnormal ICT in 32 patients, out of which 4 had reversal, had sensitivity of 78.87% and specificity of 66.06% and an area under and by abnormal PHES in 24 patients, out of which 2 the curve value of 0.724 (95% CI: 0.653–0.788) for diagnosis of MHE. showed reversal. Reversal of MHE was seen in 4/36 patients Abbreviations: CI, confidence interval; ICT, inhibitory control test; in the placebo arm. Reversal of MHE was seen in 71.42% of MHE, minimal hepatic encephalopathy.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 304–312 307 Pawar V.B. et al: MHE in cirrhotics

Table 2. Diagnosis of MHE by ICT

MHE BY ICT (Present/Absent) n Mean (SD) p-value Range CI for mean difference

Lures ($14) Present 93 17.48 (2.624) < 0.001 14–24 6.87 to 8.15 Absent 87 10.01 (2.014) 7–13 Correct targets Present 93 176.54 (6.678) < 0.001 162–187 −19.54 to −16.46 Absent 87 194.54 (3.326) 188–200

Abbreviations: CI, confident interval; ICT, inhibitory control test; MHE, minimal hepatic encephalopathy; SD, standard deviation. not significant (p = <0.003) and did not improve to <14 as We did not follow the patients’ serum sodium or ammonia the cutoff considered in our study. This can be due to the levels, nor take them at baseline, but CTP score and MELD difference in diagnostic methods, as different cutoffs can score did not predict MHE positivity in our study. yield different results. Lures in patients with PHES were 15.94±4.18 and targets Discussion were 180.15±10.20, while lures in patients without PHES were 17.21±2.49 and targets were 177.45±6.58 in patients The diagnosis of MHE is an important health issue. It has with MHE. The difference between lures in patients with and significant impact on the HRQOL and ability to drive vehicles, without PHES positivity was not significant (p = 0.09). Sim- and puts the sufferer at increased risk of developing HE.5–7 ilarly, the difference between targets in patients with and The prevalence of MHE is variable and has been reported as without PHES positivity was not significant (p = 0.1463). 30–74% among patients with liver cirrhosis.3,20 In our study, In CTP class A cirrhosis, MHE reversal was seen in 9/12 the prevalence of MHE diagnosed through abnormal PHES patients who were on lactulose, 9/11 patients on rifaximin, and/or ICT was 60%. None of our patients had a past and 2/10 patients on placebo. In CTP class B cirrhosis, MHE episode of overt HE nor did any patient develop an episode reversal was seen in 12/17 patients on lactulose, 13/20 of overt HE during the study period. None of our patients had patients on rifaximin, and 2/18 patients on placebo treat- recent alcohol intake, gastrointestinal bleeding, or history of ment. In cases of CTP class C cirrhosis, MHE reversal was portosystemic shunts, all of which are common risk factors for seen in 4/6 patients on lactulose, 4/6 patients on rifaximin, MHE. and 0/8 patients on placebo. Hence, reversal of MHE was seen Using the cutoff value of 14 lures, 51.66% of our patients in 20/33 (60.6%) with CTP class A, 27/55 (49.1%) with CTP had MHE by ICT. Using the same cutoff value of 14 lures, MHE class B, and 8/20 (40%) with CTP class C (p = 0.456 for was reported by others in 67.5%23 and 40.2%.23 The ICT PHES; p = 0.239 for ICT). ICT and PHES score did not corre- lures were significantly higher and the response to correct late with severity of liver disease, as measured by CTP or targets was significantly lower in patients with MHE than in MELD score. A total of 15 patients out of 35 (42.85%) on those without MHE. The sensitivity of 78.87% and specificity lactulose developed some form of side effects, like abdominal of 66.06% with area under the curve of 0.724 (95% CI: bloating and loose motions, which were corrected by adjust- 0.653–0.788) indicates that, in Indian patients, ICT has a ing the dose of lactulose; none of the patients from the rifax- much lower sensitivity, specificity, and area under the curve imin arm or the placebo group had any drug-related side for MHE diagnosis. In one of the previous Indian studies, sen- effect (p = 0.0005). sitivity, specificity and receiver operating characteristic of the ICT were 92.6%, 78.5% and 0.855 respectively,22 while in another study they were 78%, 65.6% and 0.735 respectively Table 3. Results of neuropsychometric testing and ICT in patients with for the diagnosis of MHE in liver cirrhosis.23 and without MHE For the diagnosis of MHE, the PHES is considered to be a Baseline reliable tool. It is the preferred tool and has been approved by th variables MHE, mean (SD) P-value the working party of the 11 World Congress of Gastroenter- ology.11 The PHES is a battery of paper and pencil tests that Present, Absent, relies heavily on the motor function of the patient. The short- n = 108 n =72 coming of using the PHES for the diagnosis of MHE is that the data needs to be adjusted for the age and education of the NCT-A 92.29 (14.75) 72.88 (5.99) <0.001 person being tested. Secondly, this test can only be adminis- NCT-B 106.84 (24.60) 79.43 (8.62) <0.001 tered by a qualified person. SDT 90.47 (14.83) 75.67 (8.01) <0.001 Initially, the studies established a lure threshold of >5 as the key variable for the diagnosis of MHE,7,25 and ICT LTT 103.31 (20.88) 82.75 (9.12) <0.001 exhibited a sensitivity and a specificity of 88% and 77% DST 16.47 (9.58) 18.07 (3.20) <0.001 respectively for the diagnosis of MHE. This showed good Lures 17.48 (2.62) 10.01 (2.01) <0.001 reproducibility. Furthermore, the results of ICT were comparable to those of the PHES. However, it is believed that the Correct 176.54 (6.67) 194.54 (3.32) <0.001 lure threshold needs to be adjusted to reflect the local pop- targets ulation norms. When compared to the PHES, which examines a wide spectrum of neurocognitive function, ICT examines Abbreviations: DST, digit symbol test; LTT, line tracing test; NCT-A, number connection test-A; NCT-B, number connection test-B; SD, standard deviation; only attention and response inhibition, but ICT is easier to SDT, serial dotting test. administer than the PHES.

308 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 304–312 Pawar V.B. et al: MHE in cirrhotics

In the current study, the PHES and ICT did not correlate with the severity of liver disease, as measured by CTP score and MELD score. This observation matches results in the

-value previous work done by Taneja et al.23 but opposed the find- p ings of Gupta et al.22 who found correlation between ICT and severity of liver disease. Reversal of MHE was seen in 25/35 (71.42%) patients in the lactulose arm, 26/37 (70.27%) in the rifaximin arm, and 4/36 (11.11%) in the placebo arm (p<0.001). Treatment response to lactulose versus placebo and rifaximin versus placebo both showed statistically significant differences. The treatment responses were almost similar with lactulose and rifaximin. A previous study from a tertiary care center in north India also found that there was reversal in 73.7% of patients with MHE in the rifaximin arm and 69.1% in the lactulose arm.26 Improvement in MHE was seen in 57% of patients with lactulose levels at baseline. Low serum sodium and high venous ammonia were highly predictive of nonresponse Rifaximin (C), mean (SD) to lactulose therapy in a study by Sharma et al.27

ation; SDT, serial dotting test. A decrease in the levels of serum endotoxin, proinflammatory markers and arterial ammonia is seen with lactulose treatment. The proinflammatory markers include interleu- -value

p kin-6, interleukin-18, and tumor necrosis factor-alpha. The levels of serum endotoxin, proinflammatory markers and arterial ammonia can be used for the diagnosis of MHE.28 Reduction in their levels is associated with improvement in MHE, as evidenced by improvement in PHES and metabolic parameters on magnetic resonance spectroscopy.28 Review of eight randomized controlled trials that evaluated lactulose or lactitol for patients who had cirrhosis and HE found that there was no difference between the two forms of treatment.29 Rifaximin is a nonsystemic antibiotic. It is, instead, a gut- selective antibiotic, which is used to modulate gut flora, reducing levels of bacteria-derived toxins. These toxins are implicated in the pathophysiology of HE. The psychometric performance and HRQOL in patients with MHE is improved by rifaximin. In addition, rifaximin is well tolerated24 and easily Lactulose (B), mean (SD) available in India, costing around INR 190 for 10 tablets. While the cost of rifaximin is almost the same as that of lactulose, it is better tolerated and has no side effects. -value

p There are few other options for the treatment of MHE. These include probiotics and L-ornithine L-aspartate, which have shown reversal of MHE.18 Previous studies have demonstrated improvement in HRQOL with the reversal of MHE in patients treated with probiotics.18 In the current study, we did not check for improvement of HRQOL. The inaccessibility of magnetic resonance spectroscopy and electroencephalogram has been the main reason for their scarce use by clinicians in our country for the diagnosis of MHE. Procedures such as magnetic resonance spectroscopy and electroencephalogram are not only operator- but also patient-independent tests. Another technique, the critical flicker frequency25 centers on the perception of light as it flickers or fuses as its frequency changes. Sex, level of education and time of day do not significantly affect critical flicker Pretreatment Posttreatment Pretreatment Posttreatment Pretreatment Posttreatment Placebo (A), mean (SD) frequency results. However, there is some evidence of differ- ential effects with age and etiology of the underlying liver disease. This testing requires many prerequisites, like intact binocular vision, careful standardization of operating procedures, and optimization of test runs. Moreover, one of the available commercial machines utilizes a red light for the testing, making it is impossible to use for individuals who

NCT-ANCT-BSDTLTT 92.00 (16.03)DST 108.00 (27.73)Lures 91.39 (15.4) 103.28Correct (92.77) targets 87.56 (15.55) 0.029 177.75 0.829 (8.03) 100.44 (19.74) 89.44 (11.88) 107.03 15.53 92.26 (25.24) (4.32) (15.51) 98.64 17.00 178.67 (21.88) (3.89) (7.73) 96.94 0.145 (16.27) 87.69 (11.50) 15.03 0.584 (3.12) 0.882 91.49 <0.001 15.39 (16.86) (2.75) 0.005 104.80 179.17 (22.74) (9.87) 105.54 (21.12) 0.194 88.43 (14.08) 96.31 0.003 (16.67) 92.59 (13.03) 91.30 186.43 (9.85) 15.57 (7.42) (4.37) 0.180 16.26 0.001 (3.64) <0.001 85.32 (99.6) <0.001 92.35 (11.68) 15.80 104.68 180.73 (2.18) (20.40) (9.77) 12.80 (1.68) <0.001 92.03 86.41 (9.79) (9.56) 0.575 184.95 (7.85) <0.001 0.001 0.006 0.063 18.24 15.89 (15.22) (3.77)25 18.03 (10.33) 13.22 (2.44) 0.010 0.0001

Table 4 Results of Neuropsychometric testing and ICT in patients with MHE Variables Abbreviations: DST, digit symbol test; LTT, line tracing test; NCT-A, number connection test-A; NCT-B, number connection test-B; SD, standard devi are red-green color blind.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 304–312 309 Pawar V.B. et al: MHE in cirrhotics

Fig. 3. Box plot showing (A) baseline and (B) posttreatment neuropsychometric results. Abbreviations: DST, digit symbol test; LTT, line tracing test; NCT-A, number connection test-A; NCT-B, number connection test-B; SDT, serial dotting test.

The primary strengths of our study is that we directly used placebo for lactulose is sorbitol, but it is not an inert compared PHES and ICT in the diagnosis of MHE and also that placebo and exerts purgative action and purges out ammonia, we compared the treatment outcomes in patients with MHE both of which will affect results (favorably). We used B-complex using lactulose and rifaximin as treatment. Finding an exact tablets as placebo to rifaximin, which is readily available and placebo for lactulose is nearly impossible. The most commonly has no purgative action, resulting in minimal confounding.

310 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 304–312 Pawar V.B. et al: MHE in cirrhotics

Fig. 4. Box plot showing baseline and posttreatment inhibitory control test results.

The nonresponse to treatment seen in some of our patients follow-up of this patient population who undergoes these may be attributable to other laboratory features, like serum treatment modalities. sodium, which we did not looked into. Other reasons may be short duration of treatment, and some might have responded Conflict of interest to dual treatment. ICT is a simple tool for the diagnosis of MHE. However, it The authors have no conflict of interests related to this has lower sensitivity and specificity in Indian patients. This publication. study evaluated the efficacy of different treatment modalities in MHE in patients presenting at tertiary care center in India. In this group of MHE patients, rifaximin and lactulose were Author contributions equally efficacious in the treatment of MHE but rifaximin was better tolerated. Hence, we recommend PHES as the primary Conceptualized and designed the study (VBP, NS, VZ, SJ), modality for diagnosis of MHE over ICT in the Indian pop- acquired, analyzed and interpreted the data (VBP, NS), ulation. Moreover, the treatment of MHE with rifaximin should drafted the manuscript (VBP, QC, VZ, SJ), critically revised be considered over lactulose because of its better tolerability the manuscript for important intellectual content and super- vised the study (VBP, RGS, PMR, QC). in this set of patients. It is important to remember that this study was a single- center experience and included patients presenting to one of References the tertiary care centers in India. Taking into consideration the heterogeneity of the Indian population with respect to [1] Iacob S, Gheorghe L, Iacob R, Gheorghe C, Hrehoret¸ D, Popescu I. MELD exceptions and new predictive score of death on long waiting lists for liver genetic makeup, and educational and economic backgrounds, transplantation. Chirurgia (Bucur) 2009;104:267–273. more studies of a similar design will be required at different [2] Li YY, Nie YQ, Sha WH, Zeng Z, Yang FY, Ping L, et al. Prevalence of subclinical centers in India in order to accurately extrapolate these hepatic encephalopathy in cirrhotic patients in China. World J Gastroenterol 2004;10:2397–2401. doi: 10.3748/wjg.v10.i16.2397. results to the entire Indian population. We also recommend [3] Das A, Dhiman RK, Saraswat VA, Verma M, Naik SR. Prevalence and natural further trials based on the design that will include and history of subclinical hepatic encephalopathy in cirrhosis. J Gastroenterol evaluate efficacy of combination therapies, like rifaximin Hepatol 2001;16:531–535. doi: 10.1046/j.1440-1746.2001.02487.x. [4] Romero-Gómez M, Boza F, García-Valdecasas MS, García E, Aguilar-Reina J. plus lactulose for treatment of MHE in patients who are Subclinical hepatic encephalopathy predicts the development of overt nonresponders to monotherapy. Trials looking for recurrence hepatic encephalopathy. Am J Gastroenterol 2001;96:2718–2723. doi: 10. of MHE after stopping treatment could add to our knowledge 1111/j.1572-0241.2001.04130.x. [5] Prasad S, Dhiman RK, Duseja A, Chawla YK, Sharma A, Agarwal R. Lactulose base and give important inputs regarding follow-up of these improves cognitive functions and health-related quality of life in patients with patients. Such trials will have ethical implications that must cirrhosis who have minimal hepatic encephalopathy. Hepatology 2007;45: be considered. Moreover, treatment for longer duration may 549–559. doi: 10.1002/hep.21533. [6] Wein C, Koch H, Popp B, Oehler G, Schauder P. Minimal hepatic encephalop- improve efficacy and compliance. Those trials including long athy impairs fitness to drive. Hepatology 2004;39:739–745. doi: 10. duration of the treatment will give further insights to the 1002/hep.20095.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 304–312 311 Pawar V.B. et al: MHE in cirrhotics

[7] Bajaj JS, Hafeezullah M, Hoffmann RG, Varma RR, Franco J, Binion DG, et al. [19] Shavakhi A, Hashemi H, Tabesh E, Derakhshan Z, Farzamnia S, Meshkinfar S, Navigation skill impairment: Another dimension of the driving difficulties in et al. Multistrain probiotic and lactulose in the treatment of minimal hepatic minimal hepatic encephalopathy. Hepatology 2008;47:596–604. doi: 10. encephalopathy. J Res Med Sci 2014;19:703–708. 1002/hep.22032. [20] Dhiman RK, Kurmi R, Thumburu KK, Venkataramarao SH, Agarwal R, Duseja A, [8] Weissenborn K, Ennen JC, Schomerus H, Rückert N, Hecker H. Neuropsycho- et al. Diagnosis and prognostic significance of minimal hepatic encephalopathy logical characterization of hepatic encephalopathy. J Hepatol 2001;34:768– in patients with cirrhosis of liver. Dig Dis Sci 2010;55:2381–2390. doi: 10. 773. doi: 10.1016/S0168-8278(01)00026-5. 1007/s10620-010-1249-7. [9] Ford JM, Gray M, Whitfield SL, Turken AU, Glover G, Faustman WO, et al. [21] Bajaj JS, Hafeezullah M, Franco J, Varma RR, Hoffmann RG, Knox JF, et al. Acquiring and inhibiting prepotent responses in schizophrenia: event- Inhibitory control test for the diagnosis of minimal hepatic encephalopathy. related brain potentials and functional magnetic resonance imaging. Arch Gastroenterology 2008;135:1591–1600.e1. doi: 10.1053/j.gastro.2008. – Gen Psychiatry 2004;61:119 129. doi: 10.1001/archpsyc.61.2.119. 07.021. [10] Schiff S, Vallesi A, Mapelli D, Orsato R, Pellegrini A, Umiltà C, et al. Impair- [22] Gupta D, Ingle M, Shah K, Phadke A, Sawant P. Prospective comparative ment of response inhibition precedes motor alteration in the early stage of study of inhibitory control test and psychometric hepatic encephalopathy liver cirrhosis: a behavioral and electrophysiological study. Metab Brain Dis score for diagnosis and prognosis of minimal hepatic encephalopathy in cir- – 2005;20:381 392. doi: 10.1007/s11011-005-7922-4. rhotic patients in the Indian subcontinent. J Dig Dis 2015;16:400–407. doi: [11] Ferenci P, Lockwood A, Mullen K, Tarter R, Weissenborn K, Blei AT. Hepatic 10.1111/1751-2980.12248. encephalopathy–definition, nomenclature, diagnosis, and quantification: [23] Taneja S, Dhiman RK, Khatri A, Goyal S, Thumbru KK, Agarwal R, et al. final report of the working party at the 11th World Congresses of Gastro- Inhibitory control test for the detection of minimal hepatic encephalopathy enterology, Vienna, 1998. Hepatology 2002;35:716–721. doi: 10. in patients with cirrhosis of liver. J Clin Exp Hepatol 2012;2:306–314. doi: 1053/jhep.2002.31250. 10.1016/j.jceh.2012.07.001. [12] Epstein JN, Johnson DE, Varia IM, Conners CK. Neuropsychological assess- [24] Sidhu SS, Goyal O, Mishra BP, Sood A, Chhina RS, Soni RK. Rifaximin ment of response inhibition in adults with ADHD. J Clin Exp Neuropsychol improves psychometric performance and health-related quality of life in 2001;23:362–371. doi: 10.1076/jcen.23.3.362.1186. patients with minimal hepatic encephalopathy (the RIME Trial). Am J Gastro- [13] Bajaj JS, Heuman DM, Wade JB, Gibson DP, Saeian K, Wegelin JA, et al. enterol 2011;106:307–316. doi: 10.1038/ajg.2010.455. Rifaximin improves driving simulator performance in a randomized trial of [25] Morgan MY, Amodio P, Cook NA, Jackson CD, Kircheis G, Lauridsen MM, et al. patients with minimal hepatic encephalopathy. Gastroenterology 2011;140: 478–487.e1. doi: 10.1053/j.gastro.2010.08.061. Qualifying and quantifying minimal hepatic encephalopathy. Metab Brain Dis – [14] Foster KJ, Lin S, Turck CJ. Current and emerging strategies for treating 2016;31:1217 1229. doi: 10.1007/s11011-015-9726-5. hepatic encephalopathy. Crit Care Nurs Clin North Am 2010;22:341–350. [26] Sidhu SS, Goyal O, Parker RA, Kishore H, Sood A. Rifaximin vs. lactulose in – doi: 10.1016/j.ccell.2010.04.007. treatment of minimal hepatic encephalopathy. Liver Int 2016;36:378 385. [15] Patidar KR, Bajaj JS. Antibiotics for the treatment of hepatic encephalopathy. doi: 10.1111/liv.12921. Metab Brain Dis 2013;28:307–312. doi: 10.1007/s11011-013-9383-5. [27] Sharma P, Sharma BC, Sarin SK. Predictors of nonresponse to lactulose for [16] Sharma P, Sharma BC. Disaccharides in the treatment of hepatic encephalop- minimal hepatic encephalopathy in patients with cirrhosis. Liver Int 2009; athy. Metab Brain Dis 2013;28:313–320. doi: 10.1007/s11011-013-9392-4. 29:1365–1371. doi: 10.1111/j.1478-3231.2009.02067.x. [17] Shukla S, Shukla A, Mehboob S, Guha S. Meta-analysis: the effects of gut [28] Jain L, Sharma BC, Srivastava S, Puri SK, Sharma P, Sarin S. Serum endo- flora modulation using prebiotics, probiotics and synbiotics on minimal toxin, inflammatory mediators, and magnetic resonance spectroscopy hepatic encephalopathy. Aliment Pharmacol Ther 2011;33:662–671. doi: before and after treatment in patients with minimal hepatic encephalopathy. 10.1111/j.1365-2036.2010.04574.x. J Gastroenterol Hepatol 2013;28:1187–1193. doi: 10.1111/jgh.12160. [18] Mittal VV, Sharma BC, Sharma P, Sarin SK. A randomized controlled trial [29] Gluud LL, Vilstrup H, Morgan MY. Non-absorbable disaccharides versus place- comparing lactulose, probiotics, and L-ornithine L-aspartate in treatment of bo/no intervention and lactulose versus lactitol for the prevention and treat- minimal hepatic encephalopathy. Eur J Gastroenterol Hepatol 2011;23:725– ment of hepatic encephalopathy in people with cirrhosis. Cochrane Database 732. doi: 10.1097/MEG.0b013e32834696f5. Syst Rev 2016;4:CD003044. doi: 10.1002/14651858.CD003044.pub3.

312 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 304–312 Original Article

Liver Stiffness Measurement Can Reflect the Active Liver Necroinflammation in Population with Chronic Liver Disease: A Real-world Evidence Study

Leijie Wang#1, Mingyu Zhu#2, Lihua Cao#3, Mingjie Yao#4, Yiwei Lu5, Xiajie Wen1, Ying Zhang6, Jing Ning1, Huiling Long7, Yueyong Zhu8, Guoxin Hu9, Shuangsuo Dang10, Qingchun Fu11, Liang Chen12, Xinxin Zhang2, Jingmin Zhao13, Zhiliang Gao7, Yuemin Nan*6 and Fengmin Lu*1,14

1Department of Microbiology & Infectious Disease Center, School of Basic Medical Sciences, Peking University Health Science Center, Beijing, China; 2Department of Infectious Diseases, Rui Jin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China; 3Liver Disease Center, Qinhuangdao Third Hospital, Qinhuangdao, Hebei, China; 4Department of Anatomy and Embryology, School of Basic Medical Sciences, Peking University Health Science Center, Beijing, China; 5 Eshelman School of Pharmacy, University of North Carolina, Chapel Hill, NC, USA; 6Department of Traditional and Western Medical Hepatology, The Third Hospital of Hebei Medical University, Shijiazhuang, Hebei, China; 7Department of Infectious Diseases, The Third Affiliated Hospital of Sun Yat-Sen University, Guangzhou, Guangdong, China; 8Liver Research Center, First Affiliated Hospital of Fujian Medical University, Fuzhou, Fujian, China; 9Department of Infectious Diseases, Peking University Shenzhen Hospital, Shenzhen, Guangdong, China; 10Department of Infectious Diseases, Second Affiliated Hospital of Medical College of Xi’an Jiaotong University, Xi’an, Shaanxi, China; 11Shanghai Liver Diseases Research Center, 85th Hospital, Nanjing Military Command, Shanghai, China; 12Department of Liver Disease, Shanghai Public Health Clinical Center, Fudan University, Shanghai, China; 13Department of Pathology and Hepatology, The Fifth Medical Center of PLA General Hospital, Beijing, China; 14Intervention and Cell Therapy Center, Peking University Shenzhen Hospital, Shenzhen, Guangdong, China

Abstract LSM value ascended with the increased severity of liver necroinflammation in patients with the same fibrosis stage. Such Background and Aims: Non-invasive evaluation of liver nec- positive correlation between LSM and liver necroinflammation roinflammation in patients with chronic liver disease is an un- was also found in non-alcoholic steatohepatitis and autoim- met need in clinical practice. The diagnostic accuracy of mune-related liver diseases populations. Furthermore, the transient elastography-based liver stiffness measurement ROC curve exhibited that LSM could identify moderate and se- (LSM) for liver fibrosis could be affected by liver necroinflam- vere inflammation in CHB patients (area under the ROC curve mation, the latter of which could intensify stiffness of the liver. as 0.779 and 0.838) and in non-alcoholic steatohepatitis pa- Such results have prompted us to explore the diagnosis tients (area under the ROC curve as 0.826 and 0.871), respec- potential of LSM for liver inflammation. Methods: Three tively. Such moderate diagnostic value was also found in cross-sectional cohorts of liver biopsy-proven chronic liver dis- autoimmune-related liver diseases patients. In addition, in ease patients were enrolled, including 1417 chronic hepatitis B the longitudinal entecavir treated CHB cohort, a decline of (CHB) patients from 10 different medical centers, 106 non-al- LSM values was observed in parallel with the control of inflam- coholic steatohepatitis patients, and 143 patients with auto- matory activity in liver. Conclusions: Our study implicates a immune-related liver diseases. Another longitudinal cohort of diagnostic potential of LSM to evaluate the severity of liver 14 entecavir treatment patients was also included. The re- necroinflammation in chronic liver disease patients. ceiver operating characteristic (ROC) curve was employed to Citation of this article: Wang L, Zhu M, Cao L, Yao M, Lu Y, explore the diagnostic value of LSM. Results: In CHB patients, Wen X, et al. Liver stiffness measurement can reflect the active liver necroinflammation in population with chronic liver disease: A real-world evidence study. J Clin Transl Hepatol Keywords: Liver stiffness measurement; Liver necroinflammatory grade; Alanine 2019;7(4):313–321. doi: 10.14218/JCTH.2019.00040. aminotransferase; Chronic liver disease. Abbreviations: AUROC, area under the receiver operating characteristic curve; AIH, autoimmune hepatitis; CHB, chronic hepatitis B; LSM, liver stiffness measurement; NASH, non-alcoholic steatohepatitis; PBC, primary biliary cholangitis; ROC, receiver operating characteristic curve; TE, transient electrograph. Introduction Received: 28 August 2019; Revised: 1 November 2019; Accepted: 5 November 2019 Chronic liver diseases, including viral hepatitis and non-viral #These authors contributed equally to this study. *Correspondence to: Yuemin Nan,. Department of Traditional and Western hepatitis, are a heavy health burden worldwide. There are Medical Hepatology, The Third Hospital of Hebei Medical University, 139 Ziqiang approximately 2 million deaths caused by liver disease each Road, Shijiazhuang, Hebei 050017, China. E-mail: [email protected]; year, with 1.16 million due to liver cirrhosis and 0.79 million due Fengmin Lu, Department of Microbiology & Infectious Disease Center, School of to primary liver cancer.1 Chronic hepatitis B (CHB) is a major Basic Medical Sciences, Peking University Health Science Center, 38 Xueyuan Road, Beijing 100191, China. Tel: +86-10-82805136, E-mail: lu.fengmin@hsc. cause of liver cirrhosis and hepatocellular carcinoma. Prompt 2,3 pku.edu.cn anti-viral therapy is able to slow down CHB progression. In

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 313–321 313

Copyright: © 2019 Authors. This article has been published under the terms of Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0), which permits noncommercial unrestricted use, distribution, and reproduction in any medium, provided that the following statement is provided. “This article has been published in Journal of Clinical and Translational Hepatology at DOI: 10.14218/JCTH.2019.00040 and can also be viewed on the Journal’s website at http://www.jcthnet.com”. Wang L. et al: LSM predicts liver necroinflammation in CLD addition, non-alcoholic fatty liver disease, a major causative Methods factor of non-viral hepatitis and cause of 14.1% of hepatocellular carcinoma cases, has gotten much attention in recent years.4,5 Non-alcoholic fatty liver disease can be divided into Patients and ethical consent non-alcoholic fatty liver and non-alcoholic steatohepatitis 6 In this retrospective study, three cross-sectional cohorts were (NASH), based on histologic findings. In those NASH patients, active liver inflammation (A2-A3) is an independent predictive used: CHB cohorts: In order to further examine the LSM’s value factor for non-alcoholic fatty liver disease-related extrahepatic complications.7 The prognosis and treatment strategy of in identifying active liver necroinflammation, liver biopsy- chronic liver diseases depends strongly on the degree of liver confirmed CHB patients were enrolled. CHB patients were inflammatory activity. Thus, to slow down chronic liver disease recruited from the following 10 centers: The Fifth Medical progression and to prevent the occurrence of end-stage liver Center of PLA General Hospital, The Third Hospital of Hebei disease, it is necessary to differentiate those chronic liver Medical University, Rui Jin Hospital, The Third Affiliated Hos- disease patients with active liver necroinflammation for timely pital Sun Yat-Sen University, The Third Hospital of Qinhuang- intervention. dao, Shanghai Public Health Clinical Center, Peking University Non-invasive methods, including surrogate serum Shenzhen Hospital, The First Affiliated Hospital of Fujian th markers and imaging techniques, offer cost effective alter- Medical University, 85 Hospital, and the Second Affiliated ’ natives to liver biopsy. Alanine aminotransferase and aspar- Hospital of Medical College of Xi an Jiaotong University. tate aminotransferase, both indicators of liver injury, are All hospitals followed the same criteria for the patient widely used. But there are limitations in their accuracy for inclusion and exclusion. Inclusion criteria: 1) diagnosed inflammation in CHB, and they are almost completely unus- with CHB, defined as the persistent presence of hepatitis B able in NASH patients.8,9 In NASH patients, studies have surface antigen for at least 6 months, with signs of chronic 16 shown that CK-18, a recently emerging serum biomarker hepatitis; 2) LSM performed within 1 week of when the liver for non-alcoholic fatty liver disease-related inflammation, biopsy was performed; and 3) clinical index data collected on only exhibits limited diagnostic value. Thus, the diagnosis of the same day when the liver biopsy was applied. Exclusion cri- active necroinflammatory liver damage in non-alcoholic fatty teria: 1) patients who were treated with anti-viral treatment liver disease is still solely dependent on liver biopsies.10 Cur- within the past 6 months; or 2) evidence of hepatocellular rently, there is still lack of non-invasive method to diagnose carcinoma. and grade liver necroinflammation in clinical practice. NASH cohort: From 2011 to 2016, liver biopsy-confirmed Liver stiffness measurement (LSM) performed by transient NASH patients were enrolled from The Third Hospital of Hebei electrograph (TE) has been becoming an increasingly com- Medical University. The clinical criteria of NASH were monly used non-invasive method for the diagnosis of liver as follows: NASH was diagnosed based on liver biopsy after cirrhosis. It has been widely noticed that the diagnostic exclusion of, 1) concomitant steatosis-inducing drugs, accuracy of LSM for fibrosis/cirrhosis can be affected by 2) excessive alcohol consumption (>210 g/week in men several factors including liver necroinflammation,11 which or >140 g/week in women), CHB or chronic hepatitis C viral was considered to weaken LSM’s diagnostic accuracy for infection, 3) histological evidence of other concomitant 17 liver fibrosis/cirrhosis.12,13 In several recently published chronic liver disease, and 4) evidence of hepatocellular car- studies, sequential measurements of LSM revealed that LSM cinoma. In addition, the LSMs were performed within 1 week decreases after initiating anti-viral therapy in patients with of when the liver biopsy was performed, and the blood test CHB. One suggested interpretation is that an early steep was conducted on the same day as the liver biopsy. decline of liver stiffness may predict histological reversal of Autoimmune-related liver diseases cohort: Patients 18 fibrosis.14 However, the ‘China HepB-Related Fibrosis Assess- in this cohort partly belonged to a previously study, ment Research Group’ argues that a decline in liver stiffness which included 66 autoimmune hepatitis (AIH) patients and cannot indicate fibrosis regression in patients with CHB.15 In 77 primary biliary cholangitis (PBC) patients. this prospective study, the authors noticed that a greater Longitudinal cohort: This cohort was comprised of decline of LSM was correlated with a change in HAI scores 14 CHB patients enrolled from The Third Hospital of Qin- (r = 0.395, p = 0.000) but only mildly correlated with huangdao, all of who underwent entecavir treatment change in the Ishak fibrosis stage (r = 0.156, p = 0.036). (0.5 mg/d for 78 weeks) and had liver biopsy at baseline and We wondered whether such inconsistencies may be caused at 78 weeks post-treatment initiation. LSM and blood tests by a difference in liver inflammation stages at baseline and were conducted at 13, 26, 52 and 78 weeks. Part of the the therapy received between the two groups. Since the LSM patients’ information have been described in the report from value is closely associated with active liver inflammation, an the China HepB-Related Fibrosis Assessment Research Group, alternative interpretation of an elevated LSM is that it may which was published during our manuscript submission.15 reflect the inflammatory grade of a chronic liver disease All sample tests in this study were performed in qualified patient. From this perspective, it is reasonable to postulate laboratories of each center, and the results were comparable. that LSM may possesses the diagnostic potential for active Ethical consents were obtained by the Central Ethics Com- liver inflammation. mittee of The Third Hospital of Hebei Medical University, Thus, we investigated in this study if LSM could be used as the Ethics Committee of Rui Jin Hospital, and the Medical a non-invasive method to evaluate liver inflammation in Ethics Committee of The Third Affiliated Hospital Sun Yat-Sen chronic liver disease patients with different etiological back- University. The diagnosis and treatment of the patients grounds, including CHB patients, NASH patients, and patients followed the Guideline of Prevention and Treatment for with autoimmune-related liver diseases. Notably, the evalua- Chronic Hepatitis B.19 Informed consent was obtained from tion of LSM’s diagnostic value for liver necroinflammation in all patients in this study and parents’ consent was obtained if CHB cohorts was implemented at multiple centers. the patients were under 18.

314 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 313–321 Wang L. et al: LSM predicts liver necroinflammation in CLD

Liver biopsy Table 1. Clinical characteristic of patients Variable CHB NASH The necroinflammatory grade was evaluated according to the 20 21 Scheuer system in CHB patients and the Brunt system in Number 1417 106 NASH patients respectively. G2 or >G2 was defined as mod- Sex, male/female 961/456 58/48 erate liver necroinflammation and G3 was defined as severe liver necroinflammation in the CHB patients and NASH Age (years) 37 (30,46) 40 (26, 50) patients. PLT (109/L) 178 (142, 218) 213 (180, 250) ALT (U/L) 42 (28, 75) 76 (40, 119) LSM performance AST (U/L) 33 (25, 51) 44 (30, 65) LSM was performed with FibroTouch (HISKY Medical Tech- ALP (IU/L) 72 (59, 90) 95 (74, 120) nologies Co., Ltd., Beijing, China) in Rui Jin Hospital and The GGT (IU/L) 28 (18, 49) 58 (41,93) Third Affiliated Hospital Sun Yat-Sen University, or Fibroscan device (Echosens, Pairs, France) in other medical centers LSM (kPa) 8 (6, 11.9) 6.9 (5.4, 9.2) within 1 week before liver biopsy. LSM was deter-mined in the Liver 784/454/179 70/30/6 right hepatic lobe through the intercostals space (at least 2 necroinflammation, hours after the last meal) with the patient in the supine 0-1/2/3 position and the right arm in maximum abduction. Only valid Fibrosis stage, 656/360/190/ 64/20/13/9 measurements (success rate of more than 60% and inter- 0-1/2/3/4 211 quartile range/median ratio of <0.3) were included. The results were expressed in kPa corresponding to the median Abbreviations: ALP, alkaline phosphatase; ALT, alanine aminotransferase; AST, of 10 determinations. aspartate aminotransferase; CHB, chronic hepatitis B; GGT, gamma-glutamyl transpeptidase; LSM, liver stiffness measurement; NASH, non-alcoholic steato- Statistical analysis hepatitis; PLT, platelet.

Categorical data were presented as a number and percentage, and the results were analyzed by chi-squared test. University (n = 207); and cohort E from the remaining six Covariance analysis was used to adjust the effect of other medical centers (n = 139), with 28 from The Third Hospital variables. The data was analyzed with IBM SPSS statistical of Qinhuangdao, 25 from Shanghai Public Health Clinical software version 22.0 (IBM Corp., New York, NY, USA) and Center, 33 from Peking University Shenzhen Hospital, performed with GraphPad Prism version 5.0 (GraphPad Soft- 23 from The First Affiliated Hospital of Fujian Medical Univer- th ’ ware Inc., La Jolla, CA, USA). The Spearman’s rank correla- sity, 26 from The 85 Hospital, and 4 from Xi an Jiaotong tion test and its coefficient (rho) was used to describe the University (Supplementary Fig. 1 for detail). In addition, association between two variables. The area under the ROC 14 CHB patients were accepted as part of a 78-week treat- curve (AUROC), sensitivity, specificity, positive likelihood ment longitudinal cohort. The clinical characteristics of these ratios, negative likelihood ratios, and diagnostic odds ratios cross-sectional cohorts are shown in Table 1. The difference $ were used to evaluate the diagnostic value. We defined between patients G0-1 and the patients G2 are shown AUROC >0.9 as excellent, 0.9-0.8 as good, and 0.7-0.8 as in Supplementary Table 1. As expected, patients with a fair.22 The positive likelihood ratio is the ratio between true higher liver necroinflammatory grade had a higher level of positive and false positive, while the negative likelihood ratio alanine aminotransferase, aspartate aminotransferase, alka- is the ratio between false negative and true negative. The line phosphatase, gamma-glutamyl transpeptidase and LSM, positive likelihood ratios >10 or negative likelihood ratios while sex and age had no difference between the two groups. <0.1 indicated the high diagnostic value of the index. The The clinical characteristic of AIH and PBC was partly 18 significance of differences between the groups was judged described in previous study. by the Z-test. The ROC curve was made by using Medcalc (15.6.1). All tests of significance were two-tailed and p- LSM value correlated well with liver values <0.05 were considered statistically significant. necroinflammatory grade in a variety of different The data from the five independent cohorts were analyzed chronic liver diseases by using RevMan 5.3, a software usually used for meta- analysis of multicenter data. In summary, ROCs were plotted. First, to validate the quality of our study cohorts, the correlation of LSM to the scores of liver fibrosis/cirrhosis Results was analyzed. Consistent with previous studies, LSM exhibited a gradual increase in parallel with the severity of liver fibrosis, Clinical characteristics of patients in a stepwise manner (Supplementary Fig. 2), with a rho value of 0.548 (p < 0.001) in CHB, 0.613 (p < 0.001) in NASH, As shown in Table 1, a total of 1417 CHB patients from 10 dif- 0.655 (p < 0.001) in AIH, and 0.572 (p < 0.001) in PBC. ferent medical centers and 106 NASH patients from The Fifth Next, the correlation between LSM and the activity of liver Medical Center of PLA General Hospital were enrolled. The necroinflammation was investigated. The rho values were CHB patients were grouped into five independent cohorts: 0.517 (p < 0.001) in CHB, 0.547 (p < 0.001) in NASH, cohort A from The Fifth Medical Center of PLA General Hospital 0.742 (p < 0.001) in AIH, and 0.459 (p < 0.001) in PBC (n = 269); cohort B from The Third Hospital of Hebei Medical respectively (Fig. 1A & B). Such strong correlation with the University (n = 253); cohort C from Rui Jin Hospital (n =549); degree of liver necroinflammatory activity may implicate a cohort D from The Third Affiliated Hospital Sun Yat-Sen linearly dependent increase of LSM values to the worsening

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 313–321 315 Wang L. et al: LSM predicts liver necroinflammation in CLD

Fig. 1. The LSM value associated with liver necroinflammation. LSM value in (A) chronic hepatitis B patients with different liver necroinflammatory grades and (B) the patients with same fibrosis stage. (C) LSM value and (D) alanine aminotransferase value in NASH patients with different liver necroinflammatory grades in NASH patients. Abbreviations: LSM, liver stiffness measurement; NASH, non-alcoholic steatohepatitis. of liver necroinflammatory severity, as shown in Fig. 1A tion, CHB patients were divided into subgroups according for CHB patients and Fig. 1C for NASH patients. Based on to their liver fibrotic scores. As shown in Fig. 1B, gradual the fact that the occurrence of hepatic fibrosis usually increases of LSM with liver inflammation severity were still accompanied active liver inflammation, it was reasonable observed, though the severity of fibrosis in each subgroup to suspect that in the current study, the stepwise increase was the same. of LSM with liver necroinflammation could be simply due to To further confirm that the correlation between LSM and its direct correlation with fibrosis. In order to exclude liver liver necroinflammation was independent to liver fibrosis, fibrosis as the common responding variable for LSM eleva- covariance analysis was performed. It revealed that the

Fig. 2. Diagnostic performance of LSM for the non-invasive assessment of moderate liver necroinflammatory grade. (A) The summary receiver operating characteristic curves of LSM, alanine aminotransferase and aspartate aminotransferase in chronic hepatitis B patients. (B) Diagnostic performance of LSM and alanine aminotransferase for the non-invasive assessment of moderate liver necroinflammatory grade in NASH patients. Abbreviations: LSM, liver stiffness measurement; NASH, non-alcoholic steatohepatitis.

316 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 313–321 Wang L. et al: LSM predicts liver necroinflammation in CLD

Table 2. Diagnostic performance of LSM and ALT for the non-invasive assessment of liver necroinflammation

CHB, n = 1417 NASH, n = 106

G0-1 vs. G2-3 G0-2 vs. G 3 G0-1 vs. G2-3 G0-1 vs. G2-3 Necroinflammation (784 vs. 633) (1238 vs. 179) (70 vs. 36) (70 vs. 36)

LSM AUROC (95% CI) 0.779 (0.756–0.800) 0.838 (0.817–0.856) 0.826 (0.740–0.893) 0.871 (0.792–0.928) Cut-off values 10.2 10.4 8.7 8.9 Sensitivity/ 55/86 78/75 61/87 100/78 specificity (%) PPV/NPV (%) 75.8/70.4 31.2/95.9 71/81 18.5/98.7 Positive/ 3.66/0.52 3.13/0.30 4.75/0.45 3.79/0.21 negative LR ALT AUROC (95% CI) 0.694 (0.669–0.718) 0.676 (0.651–0.701) 0.521 (0.422–0.619) 0.538 (0.438–0.635) Cut-off values 41 37 122 81 Sensitivity/ 66/63 78/47 33/81 67/57 specificity (%) PPV/NPV (%) 59.8/68.7 18.1/93.3 48/70 9/97 Positive/ 1.80/0.54 1.47/0.47 1.79/0.82 1.55/0.58 negative LR

ALT, alanine transaminase; AUROC, area under the ROC curve; CHB, chronic hepatitis B; LR, likelihood ratio; LSM, liver stiffness measurement; NASH, non-alcoholic steatohepatitis; NPV, negative predictive value; PPV, positive predictive value. correlation between LSM and liver necroinflammation was C or cohort E (performed by FibroTouch) would not affect the independent of liver fibrosis. All these results verified that the overall diagnostic accuracy of LSM, which means different latent capability of LSM to reflect liver necroinflammation is devices would not affect the diagnostic accuracy of LSM. independent of liver fibrosis. Considering that moderate liver necroinflammation and significant liver fibrosis were both necessary indications of LSM showed diagnostic potential for moderate and anti-viral treatment in clinical use, the diagnostic value of LSM severe liver necroinflammation in the CHB patients for moderate necroinflammation and/or fibrosis was also explored by a strategy mimicking meta-analysis. The results – – As shown in Table 2, with cut-off value set at 10.2 kPa, LSM showed that with 59% (55 62%) sensitivity, 88% (85 91%) exhibited good diagnostic value to distinguish CHB patients specificity and 10.259 (7.340, 14.338) diagnostic odds ratio, with moderate to severe liver necroinflammation from LSM had moderate diagnostic value. The positive likelihood those with only mild or without liver inflammation. Moreover, ratio and negative likelihood ratio were 4.9 (3.8, 6.4) and it could even identify if a CHB patient had severe liver inflam- 0.47 (0.43, 0.52) respectively. mation (G3) when the best cut-off value was set as 10.4 kPa. Such diagnosis power for moderate to severe liver necroin- Decline of LSM value was found to accompany the flammation (G0-1 vs. G2-3) was further validated separately improvement of liver inflammatory activity in the CHB in all five independent cohorts, with AUROCs that varied from patients after receiving anti-viral treatment 0.780 (cohort C) to 0.829 (cohort E) (Supplementary Table 2). To further confirm that the value of LSM was associated with By use of RevMan 5.3, the data from five independent the activity of liver necroinflammation, the dynamic changes centers were analyzed. Summary ROC curves were also of LSM among a small longitudinal cohort of CHB patients who plotted to evaluate the overall inflammation diagnostic value underwent anti-viral treatment was analyzed. As expected, of LSM, alanine aminotransferase, and aspartate aminotrans- after receiving entecavir treatment, alanine aminotransferase ferase respectively. As shown in Fig. 2B, compared to alanine activity returned to normal in 13 weeks and was maintained aminotransferase and aspartate aminotransferase, LSM per- at a low level thereafter. Though slower than the rapid decline formed statistically significant higher AUROC. LSM and aspar- of alanine aminotransferase activity, the values of LSM also tate aminotransferase both exhibited reliable diagnostic value dramatically declined in the first 26 weeks post-treatment in different cohorts, while alanine aminotransferase showed a initiation, and no obvious change was observed thereafter large difference in five cohorts. Taken together, LSM in the (Fig. 3). To uncover the driving factors that contributed to the CHB cohorts exhibited good diagnostic value for moderate rapid decline in LSM value seen in CHB patients after receiving to severe liver necroinflammation. In order to explore anti-viral therapy, histopathological evaluation was con- whether the different devices would affect the diagnostic ducted via liver biopsy at week 78 post-treatment initiation. accuracy of LSM for liver necroinflammation, a subgroup anal- In contrast to the largely unchanged fibrotic scores (2.36 ± ysis was performed. The results showed that excluding cohort 0.75 vs. 2.14 ± 0.95), the liver micro-environmental

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 313–321 317 Wang L. et al: LSM predicts liver necroinflammation in CLD necroinflammation severity grades at 78 weeks post-initia- Table 3. Correlation between liver necroinflammation and clinical tion of anti-viral treatment was significantly improved, declin- characteristics ing from 2.79 ± 0.89 at baseline to 1.86 ± 0.77 (p < 0.001). CHB, n = 1417 NASH, n = 106 Thus, an early sharp drop of LSM post-initiation of treatment likely implicated the improvement of liver necroinflammation, Variables rho p-value rho p-value same as what the rapid decline of alanine aminotransferase indicated. These results further supported our hypothesis that ALT (U/L) 0.342 <0.001 0.037 0.704 increased LSM values represent active liver necroinflamma- AST (U/L) 0.413 <0.001 0.194 0.047 tion in patients with chronic liver disease, at least in some TBIL 0.106 <0.001 −0.053 0.585 cases. (umol/L) DBIL 0.266 <0.001 0.067 0.495 LSM diagnostic value for moderate and severe liver (umol/L) necroinflammation in the NASH population GGT (U/L) 0.459 <0.001 0.135 0.169 As shown in Table 3, alanine aminotransferase failed to ALP (U/L) 0.272 <0.001 0.239 0.014 display a correlation with the severity of liver necroinflamma- TBA 0.312 <0.001 0.059 0.549 tion in the NASH patient group (rho = 0.037, p = 0.704), (umol/L) which was in contrast to its strong correlations with the LSM (kPa) 0.517 <0.001 0.547 <0.001 severity of necroinflammation (rho = 0.342, p < 0.001) in the CHB cohort. Concordant with alanine aminotransferase, Fibrosis 0.531 <0.001 0.541 <0.001 direct bilirubin (rho = 0.067, p = 0.495), gamma-glutamyl stage transpeptidase (rho = 0.135, p = 0.169), total bile acid (rho = 0.059, p = 0.549), alkaline phosphatase (rho = Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; DBIL, direct bilirubin; ALP, alkaline phosphatase; TBA, total bile acid; GGT, gamma 0.239, p = 0.014) and aspartate aminotransferase (rho = glutamyl transpeptidase; LSM, liver stiffness measurement. 0.194, p = 0.047), the currently commonly used indexes, did not exhibit a strong correlation with the severity of liver necroinflammation in the NASH patient group. Obviously, LSM exhibited a good diagnostic value for moderate liver there is a great need to develop additional markers or indica- inflammation ($G2) and severe inflammation ($G3). In tors to reflect liver necroinflammation for NASH patients. NASH patients, the AUROC curves of LSM for diagnosing mod- TE-based controlled attenuation parameter has become a erate and severe liver inflammation were significantly higher reliable method for fatty liver diagnosis, which indicates the than that of alanine aminotransferase (0.826 vs. 0.528, multiple potentials for a TE-based method to reflect different p < 0.0001 and 0.871 vs. 0.538, p = 0.0011), respectively. histological features of the liver.23 Meanwhile, previous studies The data here further confirmed the poor performance have already revealed the liver inflammation diagnostic poten- of alanine aminotransferase as a surrogate to identify liver tial of magnetic resonance elastography, an ultrasound-based necroinflammatory grade in NASH patients, as previously technique.24 As shown above, LSM values showed a strong reported.25 correlation with the degree of liver inflammatory activity. Since LSM is also an ultrasound-based technique, it may be LSM could distinguish moderate and severe liver worthwhile to explore the potential use of TE-based LSM in necroinflammation in the autoimmune-related liver the diagnosis of active liver necroinflammation in NASH disease cohort patients, and if possible to utilize LSM to better score the degree of inflammatory activity. Following this idea, the ROC In order to further demonstrate that the diagnostic potential curves were plotted. As shown in Table 2 and Fig. 2A, of LSM for liver necroinflammation was etiology-independent, the ROC curves were also plotted for AIH and PBC patients, respectively. As shown in Table 4, LSM exhibited good diagnostic performances for moderate liver inflammation ($G2) and severe inflammation ($G3) not only in the AIH patient group but also in the PBC patient group. In AIH patients, the AUROC curves of LSM for diagnosing moderate and severe liver inflammation were 0.880 and 0.907 respectively. In PBC patients, the AUROCs of LSM were 0.815 and 0.732 for diagnosing moderate and severe liver inflammation respectively. These data further confirmed the capability of LSM to discriminate liver necroinflammatory was etiology- independent.

Discussion

Rather than consider the existence of liver inflammation as a noise factor for the accurate diagnosis of cirrhosis by TE- Fig. 3. In longitudinal cohort LSM and alanine aminotransferase (median based LSM, here in this study, we tried to explore the potential ±interquartile range) level change during treatment with entecavir. application of TE to reflect the severity of liver necroinflam- Abbreviation: LSM, liver stiffness measurement. mation in chronic liver diseases with different etiologies. To

318 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 313–321 Wang L. et al: LSM predicts liver necroinflammation in CLD

Table 4. Diagnostic performance of LSM for the non-invasive assessment of liver necroinflammation

AIH, n = 66 PBC, n =77

0-1 vs. G2-3 G0-2 vs. G 3 G0-1 vs. G2-3 G0-2 vs. G 3 Necroinflammation (11 vs. 55) (33 vs. 33) (14 vs. 63) (37 vs. 40)

LSM AUROC (95% CI) 0.880 (0.777–0.947) 0.907 (0.810–0.965) 0.815 (0.710–0.894) 0.732 (0.619–0.827) Cut-off values 6.1 10.1 6.8 6.8 Sensitivity/specificity 80.00/90.91 81.82/90.91 71.43/85.71 78.38/55.00 in % PPV/NPV in % 97.8/47.6 90.0/83.3 95.7/40.0 59.1/66.7 Positive/negative LR 8.80/0.22 9.00/0.20 5.00/0.33 1.56/0.54

Abbreviations: AIH, auto-immune hepatitis; AUROC, area under the ROC curve; LR, likelihood ratio; LSM, liver stiffness measurement; NPV, negative predictive value; PBC, primary biliary cholangitis; PPV, positive predictive value. provide supporting evidence to our hypothesis, the relation- numerous studies support the idea that liver inflammation ship between values of LSM and degree of liver inflammation drives the rise of fibrosis by accelerating hepatocyte necrosis was first assessed through various aspects. As expected, the and apoptosis. Since the presence of moderate liver necroin- aberrant elevation of LSM was consistent with the severity of flammation and fibrosis was considered as the index of pro- liver necroinflammation seen in all chronic liver diseases, gressive liver disease,33 it has been recommended by major including CHB, NASH and autoimmune-related liver diseases guidelines that, in clinical practice, moderate liver necroin- (autoimmune disease and PBC) populations (Fig. 1A, 1C & flammation and/or moderate liver fibrosis are indicators for Supplementary Fig. 1). Importantly, after divid-ing the immediate initiation of anti-viral therapy in CHB patients2 or large-sample sized group of CHB patients into subgroups need of strong intervention for NASH patients. To identify according to each individual’s fibrotic scores, the correlation those chronic liver disease patients with ongoing liver nec- between elevated LSM values and the severity of liver necroinflammatory damage, alanine aminotransferase level has roinflammation still existed in every subgroup of CHB patients been recommended by all major guidelines as a surrogate with the same fibrotic scores. To further validate our findings, marker for hepatic damage in and indicator for anti-viral by use of the idea of meta-analysis, we verified LSM’s capa- treatment of CHB patients.2,3,16 As is known, alanine amino- bility to respond to inflammatory activities in five independent transferase might remain normal or only mildly elevated (up cohorts composed of 1417 CHB patients in total (Fig. 3). We to 13–47%) in CHB patients with active liver inflammatory believe that this meta-analysis-mimicking approach is more activity,8,9,34,35 or even worse has almost no value for the authentic and more reliable than traditional meta-analysis diagnosis of inflammation damage among NASH patients, as due to the fact that the raw data was obtained from each previously reported36–38 and shown in the current study. research hospital. Serum CK-18 was found to be elevated in the NASH popula- Another important piece of evidence in support of our idea tion compared to the non-alcoholic fatty liver population. is that increased LSM could respond to active liver inflamma- However, few studies have evaluated whether CK-18 could tion came from the observation of the dynamic change of LSM act as a non-invasive biomarker to reflect NASH-related nec- in CHB patients who underwent anti-viral therapy. As roinflammatory activity.39 With the impressive prevalence revealed in this longitudinal cohort, LSM decreased rapidly of non-alcoholic fatty liver disease population in China and during the first 13–26 weeks of treatment in these CHB worldwide in recent decades, there is an urgent need to patients, after the recovery of liver inflammation achieved develop a novel method to meet this clinical need. Currently, by anti-viral treatment (indicated by the decline of alanine LSM’s potential diagnostic value for liver necroinflammation aminotransfer-ase). As is known, the fast drop in LSM values deserves more attention, particularly for NASH patients. in such a short time could not be contributed to the regression Therefore, the potential use of LSM as a supplementary of liver fibrosis, and this was confirmed by the pathologic indicator for liver necroinflammation was explored in the evaluation of inflammation activity and status of fibrotic current study. Indeed, in the current chronic liver disease lesions. We suggest that after the necroinflammation was cohort with etiologies including CHB, NASH and autoimmune controlled by anti-viral treatment, the equilibrium value liver diseases, LSM exhibited certain diagnostic performance thereafter might represent the true fibrotic scores of the for moderate liver necroinflammation. In addition, LSM patients. provided a potential means to survey whether anti-viral As important histological features, necroinflammatory therapy had improved necroinflammation in CHB patients. grade and liver fibrosis were the main indicators for the For example, a constant decline in LSM value might implicate prediction of disease prognosis and evaluation of treatment the efficient control of liver necroinflammation in CHB benefit.26–28 The former represents ongoing liver injury and patients who have undergone anti-viral therapy. From this the latter stands for accumulated lesions. It has been respect, LSM provides a more accurate assessment of the acknowledged that in general, patients with no or mild inflam- disease progression than we thought previously, which mation usually have slower disease progression,29–31 while reflects ongoing liver necroinflammation in addition to cumu- those with active liver necroinflammation display more pro- lative liver damage (liver fibrosis/cirrhosis) over time. Mean- gressive disease processes.32 In line with this, findings from while, the dual characteristics of LSM might be helpful to a

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 313–321 319 Wang L. et al: LSM predicts liver necroinflammation in CLD clinician to decide whether or not to initiate anti-viral treat- [8] Göbel T, Erhardt A, Herwig M, Poremba C, Baldus SE, Sagir A, et al. High ment for individuals with chronic HBV infection, rather than prevalence of significant liver fibrosis and cirrhosis in chronic hepatitis B patients with normal ALT in central Europe. J Med Virol 2011;83:968–973. setting a different cut-off value to distinguish inflammation doi: 10.1002/jmv.22048. from fibrosis, or vice versa, for CHB patients with different [9] Nguyen K, Pan C, Xia V, Hu J, Hu KQ. Clinical course of chronic hepatitis B alanine aminotransferase levels. (CHB) presented with normal ALT in Asian American patients. J Viral Hepat The relatively small overall number of NASH participants is 2015;22:809–816. doi: 10.1111/jvh.12388. also a limitation of this study. Other limitations of this study [10] Kleiner DE, Makhlouf HR. Histology of nonalcoholic fatty liver disease and nonalcoholic steatohepatitis in adults and children. Clin Liver Dis 2016;20: are that the chronic liver disease patients with etiologies such 293–312. doi: 10.1016/j.cld.2015.10.011. as hepatitis C virus, alcoholic liver disease and drug-induced [11] Zeng X, Xu C, He D, Zhang H, Xia J, Shi D, et al. Influence of hepatic inflam- liver disease were not collected and analyzed. mation on fibroscan findings in diagnosing fibrosis in patients with chronic In summary, results in this study suggest that TE-based hepatitis B. Ultrasound Med Biol 2015;41:1538–1544. doi: 10.1016/j.ultra- LSM can identify moderate and severe liver inflammation with smedbio.2015.01.011. [12] Boursier J, Vergniol J, Guillet A, Hiriart JB, Lannes A, Le Bail B, et al. Diag- high accuracy in CHB, NASH, and autoimmune-related liver nostic accuracy and prognostic significance of blood fibrosis tests and liver disease patients. We suggest that rather than search for a stiffness measurement by FibroScan in non-alcoholic fatty liver disease. novel non-invasive biomarker, taking full advantage of the J Hepatol 2016;65:570–578. doi: 10.1016/j.jhep.2016.04.023. present TE-based LSM for the diagnosis of active liver [13] Yan LB, Zhu X, Bai L, Liang LB, Chen EQ, Du LY, et al. Impact of mild to moderate elevations of alanine aminotransferase on liver stiffness measure- necroinflammation may be more practical. ment in chronic hepatitis B patients during antiviral therapy. Hepatol Res 2013;43:185–191. doi: 10.1111/j.1872-034X.2012.01068.x. Acknowledgments [14] Kong Y, Sun Y, Zhou J, Wu X, Chen Y, Piao H, et al. Early steep decline of liver stiffness predicts histological reversal of fibrosis in chronic hepatitis B patients treated with entecavir. J Viral Hepat 2019;26:576–585. doi: 10. This work was supported by the National S & T Major Project for 1111/jvh.13058. Infectious Diseases (No.2017ZX10201201, 2017ZX10202202, [15] Dong XQ, Wu Z, Li J, Wang GQ, Zhao H. Declining in liver stiffness cannot 2017ZX10302201 and 2017ZX10202203), the project indicate fibrosis regression in patients with chronic hepatitis B: A 78-week – from Beijing Municipal Science & Technology Commission prospective study. J Gastroenterol Hepatol 2019;34:755 763. doi: 10. 1111/jgh.14498. (No.Z161100000116047), Project funded by China Postdoc- [16] Liaw YF, Kao JH, Piratvisuth T, Chan HL, Chien RN, Liu CJ, et al. Asian-Pacific toral Science Foundation (No. 2017M620544, 2018T110014) consensus statement on the management of chronic hepatitis B: a 2012 and Project funded by Shenzhen Municipal Health Commission update. Hepatol Int 2012;6:531–561. doi: 10.1007/s12072-012-9365-4. (No. SZSM201612071). [17] Ratziu V, Bellentani S, Cortez-Pinto H, Day C, Marchesini G. A position statement on NAFLD/NASH based on the EASL 2009 special conference. J Hepatol 2010;53:372–384. doi: 10.1016/j.jhep.2010.04.008. [18] Yao M, Wang L, Leung PSC, Li Y, Liu S, Wang L, et al. The clinical significance Conflict of interest of GP73 in immunologically mediated chronic liver diseases: Experimental data and literature review. Clin Rev Allergy Immunol 2018;54:282–294. The authors have no conflict of interests related to this doi: 10.1007/s12016-017-8655-y. publication. [19] Hou J, Wang G, Wang F, Cheng J, Ren H, Zhuang H, et al. Guideline of prevention and treatment for chronic hepatitis B (2015 update). J Clin Transl Hepatol 2017;5:297–318. doi: 10.14218/JCTH.2016.00019. [20] Scheuer P. Primary biliary cirrhosis. Proc R Soc Med 1967;60:1257–1260. Author contributions doi: 10.1177/003591576706001205. [21] Brunt EM, Janney CG, Di Bisceglie AM, Neuschwander-Tetri BA, Bacon BR. Conceived the idea and designed the research (FL, MY, LW, Nonalcoholic steatohepatitis: a proposal for grading and staging the histological lesions. Am J Gastroenterol 1999;94:2467–2474. doi: 10.1111/j.1572- YL), collected data and performed the research (XZ, YN, LC, 0241.1999.01377.x. ZG, HL,GH, LW, MY, MZ, XW, YZ, SD, YZ, LC,QF, JN), analyzed [22] Herrmann E, de Lédinghen V, Cassinotto C, Chu WC, Leung VY, Ferraioli G, the data and wrote the paper (LW, XW, MY), and revised the et al. Assessment of biopsy-proven liver fibrosis by two-dimensional shear paper (FL, JZ, YL). wave elastography: An individual patient data-based meta-analysis. Hepa- tology 2018;67:260–272. doi: 10.1002/hep.29179. [23] Eddowes PJ, Sasso M, Allison M, Tsochatzis E, Anstee QM, Sheridan D, et al. References Accuracy of fibroscan controlled attenuation parameter and liver stiffness measurement in assessing steatosis and fibrosis in patients with nonalcoholic fatty liver disease. Gastroenterology 2019;156:1717–1730. doi: [1] Asrani SK, Devarbhavi H, Eaton J, Kamath PS. Burden of liver diseases in the 10.1053/j.gastro.2019.01.042. world. J Hepatol 2019;70:151–171. doi: 10.1016/j.jhep.2018.09.014. [24] Yin M, Glaser KJ, Manduca A, Mounajjed T, Malhi H, Simonetto DA, et al. [2] EASL 2017 Clinical Practice Guidelines on the management of hepatitis B Distinguishing between hepatic inflammation and fibrosis with MR elastog- virus infection. J Hepatol 2017;67:370–398. doi: 10.1016/j.jhep.2017.03.021. raphy. Radiology 2017;284:694–705. doi: 10.1148/radiol.2017160622. [3] Terrault NA, Bzowej NH, Chang KM, Hwang JP, Jonas MM, Murad MH. [25] Younossi ZM, Loomba R, Anstee QM, Rinella ME, Bugianesi E, Marchesini G, AASLD guidelines for treatment of chronic hepatitis B. Hepatology 2016; et al. Diagnostic modalities for nonalcoholic fatty liver disease, nonalcoholic 63:261–283. doi: 10.1002/hep.28156. steatohepatitis, and associated fibrosis. Hepatology 2018;68:349–360. doi: [4] Wang FS, Fan JG, Zhang Z, Gao B, Wang HY. The global burden of liver disease: the major impact of China. Hepatology 2014;60:2099–2108. doi: 10.1002/hep.29721. 10.1002/hep.27406. [26] Sanyal AJ, Chalasani N, Kowdley KV, McCullough A, Diehl AM, Bass NM, et al. [5] Younossi ZM, Otgonsuren M, Henry L, Venkatesan C, Mishra A, Erario M, et al. Pioglitazone, vitamin E, or placebo for nonalcoholic steatohepatitis. N Engl – Association of nonalcoholic fatty liver disease (NAFLD) with hepatocellular J Med 2010;362:1675 1685. doi: 10.1056/NEJMoa0907929. carcinoma (HCC) in the United States from 2004 to 2009. Hepatology [27] Neuschwander-Tetri BA, Clark JM, Bass NM, Van Natta ML, Unalp-Arida A, 2015;62:1723–1730. doi: 10.1002/hep.28123. Tonascia J, et al. Clinical, laboratory and histological associations in adults [6] Younossi ZM, Koenig AB, Abdelatif D, Fazel Y, Henry L, Wymer M. Global with nonalcoholic fatty liver disease. Hepatology 2010;52:913–924. doi: 10. epidemiology of nonalcoholic fatty liver disease-Meta-analytic assessment 1002/hep.23784. of prevalence, incidence, and outcomes. Hepatology 2016;64:73–84. doi: [28] Younossi ZM, Stepanova M, Rafiq N, Makhlouf H, Younoszai Z, Agrawal R, 10.1002/hep.28431. et al. Pathologic criteria for nonalcoholic steatohepatitis: interprotocol [7] Ito T, Ishigami M, Ishizu Y, Kuzuya T, Honda T, Hayashi K, et al. Utility and agreement and ability to predict liver-related mortality. Hepatology 2011; limitations of noninvasive fibrosis markers for predicting prognosis in biopsy- 53:1874–1882. doi: 10.1002/hep.24268. proven Japanese non-alcoholic fatty liver disease patients. J Gastroenterol [29] McMahon BJ. The natural history of chronic hepatitis B virus infection. Semin Hepatol 2019;34:207–214. doi: 10.1111/jgh.14448. Liver Dis 2004;24 Suppl 1:17–21. doi: 10.1055/s-2004-828674.

320 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 313–321 Wang L. et al: LSM predicts liver necroinflammation in CLD

[30] Hoofnagle JH, Doo E, Liang TJ, Fleischer R, Lok AS. Management of ALT less than twice ULN and high HBV-DNA levels in Indonesia. J Dig Dis hepatitis B: summary of a clinical research workshop. Hepatology 2007; 2011;12:476–480. doi: 10.1111/j.1751-2980.2011.00540.x. 45:1056–1075. doi: 10.1002/hep.21627. [36] Browning JD, Szczepaniak LS, Dobbins R, Nuremberg P, Horton JD, Cohen JC, [31] Fattovich G. Natural history and prognosis of hepatitis B. Semin Liver Dis et al. Prevalence of hepatic steatosis in an urban population in the 2003;23:47–58. doi: 10.1055/s-2003-37590. United States: impact of ethnicity. Hepatology 2004;40:1387–1395. doi: [32] Czaja AJ. Hepatic inflammation and progressive liver fibrosis in chronic liver 10.1002/hep.20466 disease. World J Gastroenterol 2014;20:2515–2532. doi: 10.3748/wjg.v20. [37] Sorrentino P, Tarantino G, Conca P, Perrella A, Terracciano ML, Vecchione R, i10.2515. et al. Silent non-alcoholic fatty liver disease-a clinical-histological study. [33] Asselah T, Boyer N, Guimont MC, Cazals-Hatem D, Tubach F, Nahon K, et al. J Hepatol 2004;41:751–757. doi: 10.1016/j.jhep.2004.07.010. Liver fibrosis is not associated with steatosis but with necroinflammation in [38] Mofrad P, Contos MJ, Haque M, Sargeant C, Fisher RA, Luketic VA, et al. French patients with chronic hepatitis C. Gut 2003;52:1638–1643. doi: 10. Clinical and histologic spectrum of nonalcoholic fatty liver disease associated 1136/gut.52.11.1638 with normal ALTvalues. Hepatology 2003;37:1286–1292. doi: 10.1053/jhep. [34] Sarin SK, Kumar M. Should chronic HBV infected patients with normal ALT 2003.50229. treated: debate. Hepatol Int 2008;2:179–184. doi: 10.1007/s12072-008- [39] Feldstein AE, Wieckowska A, Lopez AR, Liu YC, Zein NN, McCullough AJ. 9065-2. Cytokeratin-18 fragment levels as noninvasive biomarkers for nonalco- [35] Lesmana CR, Gani RA, Hasan I, Simadibrata M, Sulaiman AS, Pakasi LS, et al. holic steatohepatitis: a multicenter validation study. Hepatology 2009; Significant hepatic histopathology in chronic hepatitis B patients with serum 50:1072–1078. doi: 10.1002/hep.23050.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 313–321 321 Original Article

Baseline Characteristics and Treatment Patterns of the Patients Recruited to the China Registry of Hepatitis B

Shan Shan1, Hong You1, Junqi Niu2, Jia Shang3, Wen Xie4, Yuexin Zhang5, Xun Li6, Hong Ren7, Hong Tang8, Huiguo Ding9, Xihong Wang10, Yuemin Nan11, Xiaoguang Dou12, Tao Han13, Lingyi Zhang14, Xiaoqing Liu15, Cunliang Deng16, Jilin Cheng17, Xiaozhong Wang18, Qing Xie19, Shumei Lin20, Yan Huang21, Youqing Xu22, Yong Xiong23,WuLi24, Xuebing Yan25, Hongxin Piao26, Wenxiang Huang27, Qinghua Lu28, Weijin Gong29, Shiping Li30, Xiaoxuan Hu31, Xiaolan Zhang32, Shourong Liu33, Yufang Li34, Dongliang Yang35, Hai Li36, Caixia Yang37, Mingliang Cheng38, Liaoyun Zhang39, Huanwei Zheng40, Xinhua Luo41, Feng Lin42, Lei Wang43, Guanghua Xu44, Xiaoyuan Xu45, Lai Wei46, Jinlin Hou47, Zhongping Duan48, Hui Zhuang49, Xizhong Yang50, Yuanyuan Kong51*, and Jidong Jia1* for the CR-HepB study group, Beijing, China

1Liver Research Center, Beijing Friendship Hospital, Capital Medical University, Beijing Key Laboratory of Translational Medicine on Liver Cirrhosis, National Clinical Research Center for Digestive Diseases, Beijing, China; 2Department of Hepatology, First Hospital of Jilin University, Changchun, Jilin, China; 3Department of Infectious Diseases, Henan Provincial People’s Hospital, Zhengzhou, Henan, China; 4Center of Liver Diseases, Beijing Ditan Hospital, Capital Medical University, Beijing, China; 5Department of Infectious Diseases, First Affiliated Hospital of Xinjiang Medical University, Urumqi, Xinjiang, China; 6Department of General Surgery, The First Hospital of Lanzhou University, Lanzhou, Gansu, China; 7Department of Infectious Diseases, The Second Affiliated Hospital of Chongqing Medical University, Chongqing, China; 8Center of Infectious Diseases, West China Hospital of Sichuan University, Chengdu, Sichuan, China; 9Department of Gastroenterology and Hepatology, Beijing Youan Hospital, Capital Medical University, Beijing, China; 10Center of Infectious Diseases, The Third People’s Hospital of Taiyuan, Taiyuan, Shanxi, China; 11Department of Traditional and Western Medical Hepatology, The Third Hospital of Hebei Medical University, Shijiazhuang, Hebei, China; 12Department of Infectious Diseases, Shengjing Hospital of China Medical University, Shenyang, Liaoning, China; 13Department of Hepatology, Tianjin Third Central Hospital, Tianjin Medical University, Tianjin, China; 14Department of Hepatology, Lanzhou University Second Hospital, Lanzhou, Gansu, China; 15Department of Infectious Diseases, Peking Union Medical College Hospital, Beijing, China; 16Department of Infectious Diseases, The Affiliated Hospital of Southwest Medical University, Luzhou, Gansu, China; 17Department of Gastroenterology, Shanghai Public Health Clinical Center, Shanghai, China; 18Department of Hepatology, Xinjiang Uygur Autonomous Region Traditional Chinese Medicine Hospital, Urumqi, Xinjiang, China; 19Department of Infectious Diseases, Shanghai Ruijin Hospital, Jiao Tong University School of Medicine, Shanghai, China; 20Department of Infectious Diseases, The First Affiliated Hospital of Xian Jiao Tong University, Xi’an, Shaanxi, China; 21Department of Infectious Diseases, Xiangya Hospital Central South University, Changsha, Hunan, China; 22Department of Digestive System, Beijing Tiantan Hospital, Capital Medical University, Beijing, China; 23Department of Infectious Diseases, Zhongnan Hospital of Wuhan University, Wuhan, Hubei, China; 24Department of Infectious Diseases, First Affiliated Hospital of Kunming Medical University, Kunming, Yunnan, China; 25Department of Infectious Diseases, The Affiliated Hospital of Xuzhou Medical University, Xuzhou, Henan, China; 26Department of Infectious Diseases, Yanbian University Hospital, Yanji, Jilin, China; 27Department of Infectious Diseases, The First Affiliated Hospital of Chongqing Medical University, Chongqing, China; 28Department of Hepatology, The Fourth People’s Hospital of Qinghai Province, Xining, Qinghai, China; 29Department of Infectious Diseases, Xinjiang Changji Prefecture People’s Hospital, Changji, Xinjiang, China; 30Department of Hepatopancreatobiliary Surgery, Affiliated Hospital of Weifang Medical University, Weifang, Shandong, China; 31Department of Hepatology, Hunan Provincial People’s Hospital, Changsha, Hunan, China; 32Department of Gastroenterology, The Second Hospital of Hebei Medical University, Shijiazhuang, Hebei, China; 33Department of Hepatology, Xixi Hospital of Hangzhou, Hangzhou, Zhejiang, China; 34Department of Infectious Diseases, General Hospital of Ningxia Medical University, Yinchuan, Ningxia, China; 35Department of Infectious Diseases, Union Hospital, Tongji Medical College of Huazhong University of Science and Technology, Wuhan, Hubei, China; 36Department of Hepatopancreatobiliary and Splenic Medicine, The Affiliated Hospital, Logistics University of People’s Armed Police Force, Tianjin, China; 37Department of Infectious Diseases, Infectious Disease Hospital of Wuhai, Wuhai, Inner Mongolia, China; 38Department of Infectious Diseases, The Affiliated Hospital of Guizhou Medical University, Guiyang, Guizhou, China; 39Department of Infectious Diseases, The First Affiliated Hospital of Shanxi Medical University, Taiyuan, Shanxi, China; 40Department of Infectious Diseases, The Fifth Hospital of Shijiazhuang, Shijiazhuang, Hebei, China; 41Department of Infectious Diseases, Guizhou Provincial People’s Hospital, Guiyang, Guizhou, China; 42Department of Infectious Diseases, Hainan General Hospital, Haikou, Hainan, China; 43Department of Infectious Diseases, The Second Hospital of Shandong University, Jinan, Shandong, China; 44Department of Infectious Diseases, Yanan University Affiliated Hospital, Yan’an, Shaanxi, China; 45Department of Infectious Diseases, Peking University First Hospital, Beijing, China; 46Peking University Hepatology Institute, Peking University People’s Hospital, Beijing, China; 47Institute of Hepatology and Department of Infectious Disease, Nanfang Hospital, Southern Medical University, Guangzhou, Guangdong, China; 48Artificial Liver Center, Beijing Youan Hospital, Capital

322 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 322–328

Medical University, Beijing, China; 49Department of Microbiology and Infectious Disease Center, Peking University Health Science Center, Beijing, China; 50The China Foundation of Hepatitis Prevention and Control, Beijing China; 51Center for Clinical Epidemiology and EBM, Beijing Friendship Hospital, Capital Medical University, National Clinical Research Center for Digestive Diseases, Beijing, China

Abstract ciated with a considerable burden of liver morbidity and mortality, and can lead to cirrhosis, decompensation and Background and Aims: Chronic hepatitis B virus (HBV) in- hepatocellular carcinoma (HCC).6 fection remains a major public health problem globally. Here, In China, with high coverage of HBV vaccination in infants, we describe the baseline characteristics and treatment pro- the estimated prevalence of HBsAg declined to 6.1% in the files of HBV-infected patients recruited to the China Registry general population.7,8 However, historical HBV endemicity of Hepatitis B. Methods: Inclusion criteria were patients with built a large reservoir of chronically infected persons. It is different stages of chronic HBV infection and complete key estimated that there are more than 70 million persons with data. Exclusion criteria were patients with hepatocellular car- chronic HBV infection in China.3 To facilitate real-world clinical cinoma. The baseline clinical, laboratory and treatment pro- study of chronic HBV infection, we have established a national files were analyzed. Results: Finally, 40,431 patients were HBV registry platform, the China Registry of Hepatitis B included. The median age was 43 years, with 65.2% being (known as the CR-HepB),9 which was launched in July 2012. men and 51.3% being positive for hepatitis B e antigen Currently, it consists of 47 tertiary hospitals in mainland China (HBeAg). The most common initial diagnosis was chronic hep- (ClinicalTrials.gov registry number: NCT03108794).9 atitis B (81.0%), followed by cirrhosis (9.3%), inactive carrier In the present cross-sectional study, we described the of hepatitis B surface antigen (HBsAg) (6.7%), and immune demographic, baseline characteristics, and treatment profiles tolerant phase of hepatitis B infection (3.0%). Among the of patients recruited in CR-HepB from June 2012 through June 21,228 patients who were on treatment, 88.0%, 10.0% and 2017. 2.0% received nucleos(t)ide analogues (NAs), interferon or combination of NAs and interferon, respectively. The propor- Methods tion of patients who received preferred NAs (entecavir or tenofovir disoproxil fumarate) had increased from 13.5% in Data sources 2003 to 79.7% in 2016. Conclusions: We concluded that middle-aged men accounted for most of the patients with The CR-HepB was launched in June 2012 but retrospectively chronic hepatitis B in this cross-sectional study. About half captured data of patients from 2000. The current study of the patients were HBeAg-positive. NAs were the most com- retrieved data from CR-HepB registrants prospectively or monly used therapy, and use of the preferred NAs had steadily retrospectively from June 2012 to June 2017. The key infor- increased in the past decade. mation includes patients’ age, gender, diagnosis, laboratory Citation of this article: Shan S, You H, Niu J, Shan J, Xie W, results, liver biopsy results, and antiviral treatment profiles. Zhang Y, et al. Baseline characteristics and treatment patterns of the patients recruited to the China Registry of Hep- Patient population atitis B. J Clin Transl Hepatol 2019;7(4):322–328. doi: 10.14218/JCTH.2019.00052. Inclusion criteria were patients with different stages of chronic HBV infection and available information on hepatitis B e-antigen (HBeAg) status and HBV DNA and alanine transaminase levels. Exclusion criteria were patients with HCC. Introduction The diagnostic criteria for immune tolerant phase, HBeAg- positive chronic hepatitis B (CHB), HBeAg-negative CHB, Universal vaccination against hepatitis B virus (HBV) in inactive HBsAg carriers, cirrhosis, and HCC were in line with infants has achieved great success but chronic HBV infection major international and national guidelines10 and described in remains a major public health problem globally.1 The 2017 our previous paper.9 World Health Organization (WHO) Global Hepatitis Report estimates that 257 million persons, or 3.5% of the popula- Statistical analyses tion, are chronically infected by HBV,2 with the highest hepatitis B surface antigen (HBsAg) prevalence (6.2%) being in – We use proportions and percentages to describe the demo- the Western Pacific region.3 5 Chronic HBV infection is asso- graphic and clinical characteristics of the patients. We present the proportions of patients by their age, sex, HBeAg status, Keywords: Hepatitis B; Treatment; Registry. diagnosis, liver biopsy results, and type(s) of treatment Abbreviations: CHB, chronic hepatitis B; CR-HepB, China Registry of Hepatitis B; HBeAg, hepatitis B e-antigen; HBsAg, hepatitis B surface antigen; HBV, hepatitis B received. Descriptive statistics are expressed as medians, virus; HCC, hepatocellular carcinoma; NAs, nucleos(t)ide analogues; WHO, World lower quartiles, and upper quartiles, or as a number and Health Organization. percentage of patients. All statistical analyses were per- Received: 13 October 2019; Revised: 3 December 2019; Accepted: 11 December formed using SPSS v19.0. 2019 *Correspondence to: Jidong Jia, Liver Research Center, Beijing Friendship Hos- pital, Capital Medical University, 95 Yong-an Road, Xicheng District, Beijing Results 100050, China. Tel & Fax: +86-10-63139246, E-mail: [email protected]; Yuanyuan Kong, Center for Clinical Epidemiology and EBM, Beijing Friendship Hos- pital, Capital Medical University, 95 Yong-an Road, Xicheng District, Beijing After excluding 530 individuals with HCC, 40,431 patients 100050, China. E-mail: [email protected] with confirmed diagnoses of immune tolerant phase hepatitis B,

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 322–328 323 Shan S. et al: Baseline and treatment for CR-HepB

Fig. 1. Flowchart of selection of patients.

CHB, inactive HBsAg carrier status, and cirrhosis were included been diagnosed with CHB, 9.3% with cirrhosis, 6.7% as inac- in the present study (Fig. 1). tive HBsAg carriers, and 3.0% with immune tolerant phase hepatitis B. Demographic and clinical characteristics of patients Age distribution of the 40,431patients with chronic The demographic and clinical characteristics of the patients HBV infection (by sex) are shown in Table 1. The median age was 43 years, with a men-to-women ratio of 1.9. Overall, 51.3% were HBeAg- Among the 40,431 patients included in the present study, the positive. Approximately 81.0% of the patients had initially 30–49 years-old age group was the most predominant in both

Table 1. Demographic and baseline data of 40,431 patients with hepatitis B virus-related diseases

Immune Inactive Overall, tolerance HBsAg carrier, Chronic hepatitis Cirrhosis, n = 40,431 phase, n = 1,214 n = 2,725 B, n = 32,740 n = 3,752

Age in years 43 (33, 53) 33 (28, 41) 39 (31, 49) 43 (33, 52) 55 (48, 63) Sex Men, n (%) 26,347 (65.2) 610 (50.2) 1522 (55.9) 21,472 (65.6) 2743 (73.1) Women, n (%) 14084 (34.8) 604 (49.8) 1203 (44.1) 11,268 (34.4) 1,009 (26.9) HBeAg-positive, 20740 (51.3) 1214 (100.0) 0 (0) 17936 (54.8) 1590 (42.4) n (%) HBV DNA 3.9 (2.3, 6.6) 7.6 (6.3, 8.1) 0 (0, 2.6) 4.2 (2.7, 6.9) 3.9 (2.0, 5.6)

(log10 IU/mL) ALT (IU/mL) 41.7 (24.6, 87.0) 27.0 (21.0, 34.8) 24.0 (18.0, 33.0) 46.0 (26.0, 99.0) 42.0 (27.0, 76.0) AST (IU/mL) 34.0 (23.0, 63.0) 22.0 (17.0, 26.0) 23.0 (19.0, 28.0) 35.6 (24.0 67.2) 47.0 (29.5.0,83.6) ALP (U/L) 78.0 (62.0, 101.0) 69.0 (56.0, 85.0) 70.0 (58.0, 86.0) 77.7 (62.0, 99.0) 97.8 (73.0, 133.0) GGT (U/L) 27.0 (16.0, 55.0) 15.0 (12.0, 21.0) 17.0 (12.0, 25.0) 28.0 (17.0, 55.2) 50.9 (27.9, 99.0) Bilirubin 14.8 (10.9, 21.5) 12.1 (9.3, 15.8) 12.8 (9.7, 17.2) 14.4 (10.8, 20.3) 24.5 (15.9, 41.3) (mmol/L) ALB (g/L) 44.0 (39.7, 46.7) 45.2 (43.2, 47.3) 45.4 (43.4, 47.4) 44.2 (40.4, 46.9) 34.8 (29.4, 41.2) PLT count 165.0 (115.0, 200.0 (173.0, 188.0 (154.0, 171.0 (129.0, 81.0 (53.0, 123.0) (3109/L) 208.3) 236.0) 225.0) 211.0)

Data are expressed as median (range) or n (%). Abbreviations: ALB, albumin; ALT, alanine aminotransferase; AST, aspartate aminotransferase; HBeAg, hepatitis B e antigen; HBsAg, hepatitis B surface antigen; HBV, hepatitis B virus; PLT, platelet.

324 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 322–328 Shan S. et al: Baseline and treatment for CR-HepB

Fig. 2. Age distribution of chronic hepatitis B virus infection by sex in 40,431 patients. men and women, followed by the 50–59 years-old age group. Among the 485 patients who underwent liver biopsy, the pro- Patients aged between 30–59 years-old accounted for 71.8% portion of patients with liver inflammation grade $2 or the of all patients (Fig. 2). stage of liver fibrosis $2 increased with age (Fig. 4).

Disease distribution in different age groups among the Treatment profiles of 21,228 patients and the changing 40,431 patients prescription of different nucleos(t)ide analogues

The proportion of patients diagnosed with cirrhosis was increased Nucleos(t)ide analogues (NAs) were the most common with increasing ages, whereas the proportion of patients diag- therapy among the 21,228 patients with prescription infor- nosed with immune tolerance phase was decreased with increas- mation. A much smaller proportion of patients were treated ing age (Fig. 3). with interferon (10.0%) or a combination of interferon and NAs (2.0%). Liver histology of 485 patients Lamivudine (15.3%) and adefovir dipivoxil (18.4%) were widely used before 2011, whereas the use of entecavir Necroinflammation activity and fibrosis stage were assessed (51.4%) and tenofovir disoproxil fumarate (2.1%) dramati- according to the Scheuer grading and staging system.11 cally increased after 2011 (Fig. 5).

Fig. 3. Disease distribution in different age groups among the 40,431 patients.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 322–328 325 Shan S. et al: Baseline and treatment for CR-HepB

Fig. 4. Distribution of (A) the necroinflammation grade and (B) fibrosis stage in 485 patients.

Discussion In our study, the middle-aged group was the most predominant in both men and women. This is in line with the In the present study, with large number of patients, we found recent reports that the prevalence of HBsAg in childbearing- that middle-aged men represented the major proportion of aged men and women still being around 6% in rural and this cohort. About half of the patients were HBeAg-positive. endemic areas in China.14 Therefore, prevention of mother- The most common initial diagnosis was CHB, followed by to-child transmission is still of paramount importance.3,15,16 cirrhosis, inactive HBsAg carrier, and immune tolerant phase Not surprisingly, the proportion of patients diagnosed with of hepatitis B infection. The proportion of patients diagnosed cirrhosis was increased with increasing age. Similarly, among with cirrhosis was increasing with increasing age. Among the the 485 patients who underwent liver biopsy, the majority of patients with prescription information, nearly 90% received these patients had mild to moderate necroinflammation and NAs and the use of preferred NAs have increased dramatically fibrosis. This may be due to the fact that patients with more in the past decade. disease activity and advanced fibrosis could be identified easily Our study showed that men accounted for a significantly by noninvasive modalities, making them under-repre- higher proportion (65.2%) than women (34.8%), and about sented among patients who received liver biopsy. half of patients were HBeAg-positive. This is similar to the In our study, more than half of the patients were pre- result of a recent multicenter, real-world study conducted in scribed treatment, and nearly 90% of them received NAs due tier-2 city hospitals in China, which showed that 74% of 3,408 to their favorable efficacy and safety as well as ease of patients with CHB were men, with an overall mean age of administration. All major international guidelines recommend 40 years, and that 60% of patients were HBeAg-positive.12 highly potent entecavir and tenofovir disoproxil fumarate as Not surprisingly, patients with HBeAg-negative infection were preferred therapy,10,17–19 since accumulating evidence indi- older than those with HBeAg-positive infection, also similar to cates that long-term therapy with entecavir or tenofovir dis- that reported from the USA.13 oproxil fumarate can prevent or reverse liver fibrosis and

Fig. 5. Changes in proportion of each nucleos(t)ide analogue use, 2003–2016.

326 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 322–328 Shan S. et al: Baseline and treatment for CR-HepB reduce risk of HCC.20–22 However, in real-world practice, lots coordination, technical assistance, and unrestricted grant to of patients had been treated with lamivudine, adefovir dipi- the CR-HepB and this paper. voxil and telbivudine, which are not preferred therapy, due to their low antiviral potency and low genetic barrier. This dis- Conflict of interest crepancy between guideline recommendations and real-world clinical practice may be influenced by many factors, including The authors have no conflict of interests related to this ’ ’ doctors knowledge, reimbursement policy, and patients eco- publication. nomic status and compliance.22–24 Fortunately, this study showed the prescription of different NAs has favorably changed in the past years, with Author contributions entecavir prescription increased from less than one-third to more than half. This trend may reflect the following facts: 1) Designed the study (JJ, HY, HZ, LW, JH, ZD), drafted the evidence from clinical trial and real-world studies convinc- manuscript (SS), data management (YK). Served as the ingly demonstrates the efficacy and safety of antiviral project leader and extensively and critically revised this therapy,12,25,26 2) update of evidence-based national guide- manuscript (JJ). The other authors are the team members. lines recommends entecavir and tenofovir disoproxil fuma- All authors have read and approved the final version of the rate as first-line therapy,10 3) evolving national and local manuscript. reimbursement policy offers more potent antiviral therapy for people who are covered by basic social medical insur- References ance. All these improved the standard of care in clinical practice for CHB treatment. Tenofovir disoproxil fumarate was [1] Naghavi M, Wang H, Lozano R, Davis A, Liang X, Zhou M, et al. Global, used only in less than 10%, simply because it had not been regional, and national age-sex specific all-cause and cause-specific mortality proved for HBV until mid-2014 in mainland China. for 240 causes of death, 1990–2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet 2015;385:117–171. doi: 10. We hope this large nationwide database could provide a 1016/S0140-6736(14)61682-2. point of view of clinical profiles of chronic HBV infection and [2] World Health Organization. Global hepatitis report, 2017. Available from: the treatment landscape in mainland China. However, several https://apps.who.int/iris/bitstream/han- limitations in our study need to be mentioned. First, since CR- dle/10665/255016/9789241565455-eng.pdf;jsessionid=7DA50E712691- FED09098855C3796A3A7?sequence=1. HepB is a hospital-based registry system, the proportion of [3] Schweitzer A, Horn J, Mikolajczyk RT, Krause G, Ott JJ. Estimations of world- inactive HBsAg carriers may be underestimated, as these wide prevalence of chronic hepatitis B virus infection: a systematic review of patients are usually asymptomatic and may not seek medical data published between 1965 and 2013. Lancet 2015;386:1546–1555. doi: service. Second, the cross-sectional design made it difficult to 10.1016/S0140-6736(15)61412-X. [4] Ott JJ, Stevens GA, Groeger J, Wiersma ST. Global epidemiology of hepatitis identify factors associated with disease progression or regres- B virus infection: new estimates of age-specific HBsAg seroprevalence and sion. However, CR-HepB registrants are advised to received endemicity. Vaccine 2012;30:2212–2219. doi: 10.1016/j.vaccine.2011.12. follow-up visits every 3 to 6 months, so we could expect this 116. limitation may be solved in the future. A final limitation is [5] Tian Q, Jia J. Hepatitis B virus genotypes: epidemiological and clinical rele- vance in Asia. Hepatol Int 2016;10:854–860. doi: 10.1007/s12072-016- potential selection bias, as the majority of patients in the CR- 9745-2. HepB are from tertiary hospitals, therefore not necessarily [6] Stanaway JD, Flaxman AD, Naghavi M, Fitzmaurice C, Vos T, Abubakar I, reflecting the clinical practice in secondary or primary medical et al. The global burden of viral hepatitis from 1990 to 2013: findings from care settings where the resources and expertise are far less the Global Burden of Disease Study 2013. Lancet 2016;388:1081–1088. doi: 10.1016/S0140-6736(16)30579-7. privileged. [7] Liang X, Bi S, Yang W, Wang L, Cui G, Cui F, et al. Reprint of: Epidemiological serosurvey of Hepatitis B in China–declining HBV prevalence due to Hepatitis B vaccination. Vaccine 2013;31 Suppl 9:J21–J28. doi: 10.1016/j.vaccine. Conclusions 2013.08.012. [8] Cui F, Shen L, Li L, Wang H, Wang F, Bi S, et al. Prevention of chronic hepatitis In conclusion, this hospital-based cross-sectional study pro- B after 3 decades of escalating vaccination policy, China. Emerg Infect Dis 2017;23:765–772. doi: 10.3201/eid2305.161477. vides a snapshot of demographic and baseline profiles of [9] Shan S, Wei W, Kong Y, Niu J, Shang J, Xie W, et al. China Registry of Hepatitis Chinese patients with different stage of chronic HBV infection, B (CR-HepB): Protocol and implementation of a nationwide hospital-based as well as the landscape of clinical management. registry of hepatitis B. Scand J Public Health 2018:1403494818772188. doi: 10.1177/1403494818772188. [10] Hou J, Wang G, Wang F, Cheng J, Ren H, Zhuang H, et al. Guideline of pre- Ethics Approval vention and treatment for chronic hepatitis B (2015 update). J Clin Transl Hepatol 2017;5:297–318. doi: 10.14218/JCTH.2016.00019. [11] Scheuer PJ. Classification of chronic viral hepatitis: a need for reassessment. The registry protocol was reviewed and approved by the J Hepatol 1991;13:372–374. doi: 10.1016/0168-8278(91)90084-o. ethics committee of Beijing Friendship Hospital, Capital [12] Jia J, Tang H, Ning Q, Jiang J, Dou X, Zhang M, et al. Real-world evidence for Medical University (Approval Number: BJFH-EC/2014-044). nucleoside/nucleotide analogues in a 5-year multicentre study of antiviral- naive chronic hepatitis B patients in China: 52-week results. Antivir Ther Each participating institution also obtained approval from its 2018;23:201–209. doi: 10.3851/IMP3205. institutional ethics committee. [13] Spradling PR, Xing J, Rupp LB, Moorman AC, Gordon SC, Teshale ET, et al. Distribution of disease phase, treatment prescription and severe liver disease among 1598 patients with chronic hepatitis B in the Chronic Hepatitis Cohort Study, 2006–2013. Aliment Pharmacol Ther 2016;44:1080–1089. Acknowledgments doi: 10.1111/apt.13802. [14] He T, Jia J. Chronic HBV: which pregnant women should be treated? Liver Int We thank the China Foundation for Hepatitis Prevention and 2016;36 Suppl 1:105–108. doi: 10.1111/liv.13010. Control, Chinese Society of Hepatology, Chia Tai-Tianqing [15] Cui F, Liang X, Gong X, Chen Y, Wang F, Zheng H, et al. Preventing hepatitis B though universal vaccination: reduction of inequalities through the GAVI Pharmaceutical Group Co., Ltd and Shanghai Ashermed China project. Vaccine 2013;31 Suppl 9:J29–J35. doi: 10.1016/j.vaccine. Healthcare Communications, Ltd, for their administrative 2012.07.048.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 322–328 327 Shan S. et al: Baseline and treatment for CR-HepB

[16] Liu J, Zhang S, Wang Q, Shen H, Zhang M, Zhang Y, et al. Seroepidemiology [21] Papatheodoridis GV, Chan HL, Hansen BE, Janssen HL, Lampertico P. Risk of of hepatitis B virus infection in 2 million men aged 21–49 years in rural hepatocellular carcinoma in chronic hepatitis B: assessment and modifica- China: a population-based, cross-sectional study. Lancet Infect Dis 2016; tion with current antiviral therapy. J Hepatol 2015;62:956–967. doi: 10. 16:80–86. doi: 10.1016/S1473-3099(15)00218-2. 1016/j.jhep.2015.01.002. [17] Terrault NA, Lok ASF, McMahon BJ, Chang KM, Hwang JP, Jonas MM, et al. [22] Sun Y, Zhou J, Wang L, Wu X, Chen Y, Piao H, et al. New classification of liver Update on prevention, diagnosis, and treatment of chronic hepatitis B: biopsy assessment for fibrosis in chronic hepatitis B patients before and after AASLD 2018 hepatitis B guidance. Hepatology 2018;67:1560–1599. doi: treatment. Hepatology 2017;65:1438–1450. doi: 10.1002/hep.29009. 10.1002/hep.29800. [23] Lim SG, Amarapurkar DN, Chan HL, Crawford DH, Gane EJ, Han KH, et al. [18] Lampertico P, Agarwal K, Berg T, Buti M, Janssen HLA, Papatheodoridis G, Reimbursement policies in the Asia-Pacific for chronic hepatitis B. Hepatol Int et al. EASL 2017 Clinical Practice Guidelines on the management of hepatitis 2015;9:43–51. doi: 10.1007/s12072-014-9593-x. B virus infection. J Hepatol 2017;67:370–398. doi: 10.1016/j.jhep.2017.03. [24] Shan S, Cui F, Jia J. How to control highly endemic hepatitis B in Asia. Liver 021. Int 2018;38 Suppl 1:122–125. doi: 10.1111/liv.13625. [19] Sarin SK, Kumar M, Lau GK, Abbas Z, Chan HL, Chen CJ, et al. Asian-Pacific [25] Wei L, Jia JD, Weng XH, Dou XG, Jiang JJ, Tang H, et al. Treating chronic clinical practice guidelines on the management of hepatitis B: a 2015 update. hepatitis B virus: Chinese physicians’ awareness of the 2010 guidelines. Hepatol Int 2016;10:1–98. doi: 10.1007/s12072-015-9675-4. World J Hepatol 2016;8:762–769. doi: 10.4254/wjh.v8.i18.762. [20] Varbobitis I, Papatheodoridis GV. The assessment of hepatocellular carci- [26] Zeng N, Zou C, He Z, Ma H, Ou X, You H, et al. Systematic review on the noma risk in patients with chronic hepatitis B under antiviral therapy. Clin reporting quality of randomized controlled trials in patients with hepatitis B or Mol Hepatol 2016;22:319–326. doi: 10.3350/cmh.2016.0045. C in China. Int J Infect Dis 2018;67:58–64. doi: 10.1016/j.ijid.2017.11.011.

328 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 322–328 Original Article

Outcomes and Toxicology of Herbal Drugs in Alcoholic Hepatitis – A Single Center Experience from India

Cyriac Abby Philips*1, Sasidharan Rajesh2, Tom George2, Rizwan Ahamed3, Sandeep Kumbar3 and Philip Augustine3

1The Liver Unit and Monarch Liver Lab, Cochin Gastroenterology Group, Ernakulam Medical Center, Kochi, Kerala, India; 2Interventional Radiology, Department of Gastroenterology and Hepatology, Cochin Gastroenterology Group, Ernakulam Medical Center, Kochi, Kerala, India; 3Gastroenterology and Advanced G.I. Endoscopy, Cochin Gastroenterology Group, Ernakulam Medical Center, Kochi, Kerala, India

Abstract unregulated traditional health practices can help ease the burden of liver-related death. Background and Aims: We aimed to study clinical out- Citation of this article: Philips CA, Rajesh S, George T, comes and liver biopsy features of alcoholic hepatitis (AH) Ahamed R, Kumbar S, Augustine P. Outcomes and toxicology patients on complementary and alternative medicines (CAMs) of herbal drugs in alcoholic hepatitis – A single center experi- and to analyze the retrieved drugs for chemical and toxic ence from India. J Clin Transl Hepatol 2019;7(4):329–340. components linked to drug-induced liver injury. Methods: doi: 10.14218/JCTH.2019.00043. We retrospectively assessed clinical, biochemical and liver biopsy features of AH patients on CAM with drug-induced liver injury (AH-CAM, n = 27) and compared them to a control Introduction group (classical AH, n = 29) on standard of care. Patients without liver biopsy evaluation and other causes for liver dis- Alcoholic hepatitis (AH) is a catastrophic entity associated ease were excluded. Samples of the CAMs (n = 42) from pa- with high mortality, occurring in the natural history of tients were retrieved and assessed for chemical and toxins. alcoholic liver disease in persons with continued heavy or Results: All were males, and significantly worse clinical pre- binge alcohol use. In patients with underlying alcoholic sentation, biochemical severity, and liver disease scores cirrhosis, it can present as acute-on-chronic liver failure were notable in patients with AH-CAM. Traditional Ayurve- (ACLF), with very poor prognosis and high mortality in the dic-polyherbal formulations were the most commonly used absence of best medical standard of care or liver trans- CAM. On liver histology, varying grades of severe-necrosis, plantation.1,2 Recommended treatment options for AH, even severe hepatocellular, canalicular, cholangiolar cholestasis though beneficial in select patients in the short term, are not with predominant lymphocytic-portal-inflammation and without risk – such as sepsis and multiorgan failure with cor- varying grades of interface-hepatitis were noted in AH- ticosteroid use, rapidly progressive extrahepatic organ fail- CAM. Analysis of CAMs revealed presence of heavy metals ures and high risk of alcohol relapse that foretell poor up to 100,000 times above detectable range and adulter- outcomes with liver transplantation.3,4 ants, such as antibiotics, chemotherapy agents, nonsteroi- In regions entrenched in complementary and alternative dal anti-inflammatory drugs, alcohols, antidepressants, medicine (CAM) medical practices, such as India, the use of anxiolytics, and recreational drugs. On follow up, a signifi- alternative systems of medicines, such as Ayurveda, Siddha, cantly higher number of patients with AH on CAM died at end Homeopathy, Unani and proprietary herbal drugs for manage- of 1, 3- and-6-months compared to controls (37% vs. 83%, ment of acute as well as chronic liver diseases, is on the rise. 29% vs. 62%, 18% vs. 52% respectively; p <0.001). Multiple studies encompassing single and as well as multiple Conclusions: Patients with AH and CAM-related drug- centers have shed light on hepatotoxicity associated with induced liver injury have extremely poor short-term survival CAM drugs. Drug-induced liver injury (DILI) and the toxicol- in the absence of liver transplantation compared to those ogy of implicated Ayurvedic drugs among the healthy general patients with AH on evidence-based management. Early population and patients with cirrhosis, leading to the syn- transplant referral and educating on and curbing of CAM drome of ACLF, have been demonstrated recently.5,6 The false useinsevereliverdiseasethroughstrictmonitoringof belief of a safer drug profile, cheaper treatment cost and easy approachability and availability of CAM-related health serv- Keywords: Hepatitis; Alcohol; CAM; DILI; Ayurveda. ices, along with rampant advertisements regarding the Abbreviations: AH, alcoholic hepatitis; ACLF, acute-on-chronic liver failure; CAM, same on social media, has facilitated the growth of CAM- complementary and alternative medicine; DILI, drug-induced liver injury; GC-MS/ related health seeking practices, especially among patient MS, gas chromatography coupled to tandem mass spectrometry method; SD, standard deviation. populations with chronic diseases. However, this has only Received: 5 September 2019; Revised: 4 October 2019; Accepted: 24 October added to the liver disease burden, leading to higher resource 2019 utilization and exhaustive treatments with greater financial *Correspondence to: Cyriac Abby Philips, The Liver Unit and Monarch Liver Lab, implications.7–10 Cochin Gastroenterology Group, Ernakulam Medical Center, Symphony, Automo- bile Road, Palarivattom, Cochin 682025, India. Tel/Fax: +91-484-2907000, Clinical outcomes with CAM use among patients with acute E-mail: [email protected] viral hepatitis and chronic liver disease, including alcoholic

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 329–340 329

Copyright: © 2019 Authors. This article has been published under the terms of Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0), which permits noncommercial unrestricted use, distribution, and reproduction in any medium, provided that the following statement is provided. “This article has been published in Journal of Clinical and Translational Hepatology at DOI: 10.14218/JCTH.2019.00043 and can also be viewed on the Journal’s website at http://www.jcthnet.com”. Philips C.A. et al: CAM use in alcoholic hepatitis and outcomes liver disease, have been reported before;11–16 outcomes Methodology of drug chemical analysis and toxicology associated with CAM use among patients with AH have not. In this novel study, we determined the clinical outcomes, liver Heavy metal contamination, presence of potential hepato- histopathological patterns and toxicology of CAM-related DILI toxic volatile organic compounds, adulterants, and insecti- among patients with AH compared to a group of severe AH cides and pesticides were analyzed in the retrieved drug patients on standard of care. Our findings are noteworthy in samples as per previously published standard methodol- improving education with regards to CAM-related health- ogy.6,18–22 Heavy metal concentration was determined by seeking behavior among patients with severe liver disease, inductively coupled plasma-atomic emission spectrometer so as to curb unwarranted resource utilization, impart timely (IRIS Intrepid II XSP Duo; Thermo Electron Corp., Munich, and sensible therapeutic options, and reduce mortality Germany). Methodology, chemical standards, reagents, and directly related to liver-related events. vials were acquired as per standards set by the United States Environmental Protection Agency, methods 5021A, 8015, Methods 8021, and 8260. Hepatotoxic volatile organic compounds’ qualitative analyses were performed using gas chromatogra- Patients phy coupled to tandem mass spectrometry method (GC-MS/ MS; Thermo Fisher Scientific, Waltham, MA, USA). Pesticide We conducted a retrospective analysis of patients with residue analysis was also performed using the triple quadru- alcoholic liver disease in a dedicated liver disease unit of a ple GC-MS/MS (GC TRACE 1300 with TSQ EVO 8000 MS). tertiary care center to identify those fulfilling the clinical Briefly, the required quantity of sample was extracted and definition of AH.17 Those who developed jaundice (clinically homogenized. Extract weighing 10 g was admixed (according defined as yellowish discoloration of the eyes and urine) with to the sample weight, analytical chemicals proportionally continuous heavy drinking or binge alcohol use in the preced- used as per standardized guidelines) with 10 mL of acetoni- ing 2 months were screened for inclusion. Patients were trile. Thereafter, 10 g of magnesium sulphate sodium acetate grouped into those with and without liver tests at the onset mixture was prepared and vortexed, followed by centrifuga- of symptoms. In the former, those with serum total bilirubin tion at 2000 rpm for 5 m. A 5 mL aliquot was taken from the >3.5 mg/dL and aspartate aminotransferase to alanine ami- supernatant and cleaned up using a PSA, C18 & GCB sorbent notransferase ratio of >2 were considered further for liver removal kit to exclude all of the matrix interfering materials in biopsy. Those undergoing liver biopsy and fulfilling liver his- the sample, during the dispersive solid phase extraction. Vor- tology criteria for AH were included as the control group. The texing and centrifugation at 10000 rpm was further done for former as well as latter patients, with probable AH, consum- 5 m. To detect and analyze organochlorine and organophos- ing CAMs for treatment of liver disease and presenting to the phorus pesticide residue, 1 mL of the supernatant was taken m liver unit after consumption of CAM were advised to undergo and 1 L was injected into the gas chromatograph. Identifi- liver biopsy evaluation. Those patients with probable AH and cation and quantification were carried out by mass spectrom- CAM use who consented to liver biopsy were placed into the eter using organochlorine and organophosphorus pesticide AH-CAM group. standards available. For qualitative corticosteroid analysis, 1 m Prior to liver biopsy evaluation, all patients underwent L of the extract was injected into the gas chromatograph and extensive evaluation for other causes of acute hepatitis and qualitative identification of all possible peaks of steroids avail- were excluded from the study at specified time periods, in the able in the GC-MS/MS temperature programming method event of alternate diagnosis. Etiology workup included eval- (3500C max) was made and noted. Further augmentation of uation for acute viral hepatitis A, B, C and E, herpes zoster, detection process was done using the Salkowski and Lieber- m herpes simplex, Epstein-Barr virus, cytomegalovirus, Dengue man Burchardt tests. Further to these tests, 1 L of the virus, malaria parasite, human immunodeficiency virus, anti- sample extract was injected into the gas chromatograph bodies to Leptospira in those clinically indicated [acute febrile and qualitative identification of all possible peaks of organic illness and jaundice associated with severe headache, compounds available in the GC-MS/MS temperature pro- myalgia (particularly calf muscle) and prostration associated gramming method was performed so as not to miss all iden- with conjunctival suffusion, bleeding diathesis, renal and pul- tifiable substances in the drug sample. For alcohol analysis monary involvement with or without central nervous system and quantification, approximately 1 mL of sample was symptoms and signs], autoantibodies for autoimmune hepa- added to a sample vial and mixed for at least 2 m by mechanical shaking. The vial was then placed in the head space ana- titis (including antinuclear, anti-smooth muscle, anti-liver- 8 kidney-microsomal antibodies) serum ceruloplasmin, 24 h lyzer and the sample was heated to 100 C for 25 m with urine for copper, ophthalmology evaluation for Kayser- periodic mechanical shaking. A representative sample of 1 Fleischer ring, use of other known hepatotoxic agents other mL of this prepared mixture was then injected to the gas than CAMs, imaging for primary sclerosing cholangitis, gall chromatograph and identification and quantification of bladder and bile duct diseases, and pyogenic and amoebic detected alcohols were done by mass spectrophotometer liver abscesses when clinically indicated. Patients with hepa- using procured and defined standards. Similarly, using the tocellular carcinoma, portal vein thrombosis and those who GC-MS/MS standardized technique, volatile organic com- did not provide consent for liver biopsy were excluded. Ulti- pounds were detected and identified using the equilibrium- mately, only those patients with biopsy proven definite AH based static head space method for solid samples and the and those with probable AH with additional features of CAM- purge-and-trap method for liquid samples. DILI on liver biopsy were included in the comparative study. All procedures performed in the study were in accordance Statistical analysis with the ethical standards of the institutional research committee and with the 1964 Helsinki Declaration and its later Statistical analysis was performed using MedCalc Statistical amendments or comparable ethical standards. Software (Ostend, Belgium). Data are presented as mean and

330 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 329–340 Philips C.A. et al: CAM use in alcoholic hepatitis and outcomes standard deviation (SD) or as median and range between vs. n = 3/29, 10.3%, p = 0.07). The presence of hepatic brackets as applicable. Shapiro-Wilk test was used to test encephalopathy at admission was not significantly different normality and Levene’s test was utilized to check for equality between the groups (51.7% AH vs. 59.3% AH-CAM). of variances. Chi-square and Fisher’s exact tests were used to However, a higher proportion of patients in the AH-CAM compare nominal variables. Mann-Whitney’s U test was used group had more severe grades of hepatic encephalopathy to evaluate continuous variables. One-way analysis of var- (grade 3/4) at presentation (22.2%, n = 6/27) when com- iance (ANOVA) was used to test for differences at baseline pared to patients with classical AH (13.7%, n = 4/29) but between the means of investigational variables of groups and the difference did not reach statistical significance. Lower Kruskal-Wallis test with pairwise comparisons on post hoc (0 + 1) and higher (2 + 3) grades of ACLF were comparable analysis between categorical, noncontinuous variables. A between groups at baseline, even though lower grades were p-value <0.05 was considered significant. The probability of prominent in patients with classical AH compared to the AH- patients surviving up to the study end-point was calculated CAM group (65.5% vs. 51.8%). The comparative patient using the Kaplan–Meier method and graphically represented characteristics between groups are shown in Table 1. by the survival time curve. Comparison between the survival curves was made using the log-rank test and a p-value <0.05 Investigational characteristics and severity of liver was considered significant. disease

Results In patients in the AH-CAM group, the mean hemoglobin was lower (possibly due to the higher number of patients with Patient screening and inclusion variceal bleeding) and leukocytosis was higher at presentation, even though this difference did not reach statistical From November 2017 to February 2019, 108 patients with significance. Significant thrombocytopenia was notable jaundice and continuous heavy alcohol use or binge alcohol among patients in the AH-CAM group compared to classical drinking 2 months prior to onset of jaundice were screened for AH patients (median, 81.0 vs. 101.0 3 109/L respectively; inclusion. A total of 72 patients were found to have total p = 0.02). Hyperbilirubinemia was significantly more severe bilirubin >3.5 mg/dL with aspartate aminotransferase: in patients with AH consuming CAM (18.8 ± 9.7 mg/dL vs. alanine aminotransferase ratio >2 on liver function test 12.6 ± 6.7 mg/dL; p = 0.007). Presence of coagulation evaluation after onset of jaundice. Nine of these patients failure, hyponatremia, acute kidney injury and hypoalbumi- were excluded in view of alternative causes of acute icteric nemia were comparable between groups. The severity of AH hepatitis. Of the remaining 51 patients, 29 consented to liver (defined as the discriminant function) was significantly higher biopsy and were included in the control (AH) group and 12 in patients on CAM compared to classical AH patients patients consumed CAMs after clinical diagnosis of probable (median, 80.8 vs. 61.0 respectively, p = 0.01). The presence AH, of which 8 consented to liver biopsy. A total of 36 patients of autoantibodies was notable in patients on CAM therapy (n = 108) did not undergo liver function tests after onset of (23.6%) in comparison to classical AH patients (2.8%, p = jaundice and directly consumed CAMs as the treatment for 0.02). The severity of liver disease (Child Pugh score and the jaundice. Among these, 7 patients were excluded in view of Chronic Liver Failure score) were significantly higher in patients identification of alternate causes for acute icteric hepatitis. Of with AH on CAM therapy. However, the model for end-stage the remaining 29 patients, 19 patients consented to liver liver disease sodium score, ACLF grades and Chronic Liver biopsy, giving a total of 27 patients in the AH-CAM group. Failure-C-ACLF scores were comparable between groups. The The complete patient screening, exclusion causes, and group- comparative investigational and liver disease severity charac- ing are shown in Fig. 1. teristics between groups are shown in Table 2.

Patient characteristics Comparative liver histopathology of patients between groups In both groups, all patients were males with mean age (±SD) 49.4 ± 9.8 and 45.8 ± 9.7 years in AH and AH-CAM group Steatosis of predominantly mixed type, ballooning of hep- respectively. Seven patients (26%, n = 27) in the AH-CAM atocytes, and pericellular fibrosis were present in the liver group were obese alcoholics. In the AH-CAM group, 3 histology of all patients in both groups. However, milder (11%), 1 (4%), 11 (41%) and 6 (22.2%) patients had grades of steatosis significantly predominated liver histology chronic disease, such as systemic hypertension, hypothyroid- of patients with AH on CAM (n = 13/27, 48.1%) compared to ism and diabetes mellitus respectively, which was comparable patients with classical AH, in whom a moderate degree of to those in the AH group; there were none in the AH group, steatosis was noted (n = 19/29, 65.5%; p = 0.01). The pre- while 6 (22%) in the AH-CAM group had dyslipidemia dominant region of inflammation was lobular in 75.9% of (p = 0.007). Ascites at presentation was comparable at base- patients with classical AH, while it was portal-based in line, in 79% (n = 23/29) and 70% (n = 19/27) of patients in 77.8% of patients in the AH-CAM group (p < 0.001). In the AH and AH-CAM group respectively, while acute variceal 63% of patients in the AH-CAM group, the predominant bleeding at presentation was significantly higher among type of inflammation was lymphocytic, while in 89.7% of patients in the AH-CAM group (40.7% vs. 10.3%, p = patients with classical AH, it was neutrophilic (p < 0.001). A 0.009). On follow up, at the end of 180 days, a trend trend towards higher occurrence for neutrophilic satellitosis towards higher variceal bleeding events was notable among was notable in the classical AH group (55.2%) compared to patients in the AH-CAM group compared to the classical AH patients in the AH-CAM group (29.6%, p = 0.05). Mallory- patients (n = 13/27, 48% vs. n = 7/29, 24%; p = 0.06). A Denk bodies were noted in patients in both groups, with sig- trend towards higher severity of ascites (grade 3, tense) at nificantly higher predominance in classical AH patients admission was noted in the AH-CAM group (n = 8/27, 29.6% (100%), compared to patients with AH on CAMs (77.8%,

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 329–340 331 Philips C.A. et al: CAM use in alcoholic hepatitis and outcomes

Fig. 1. Enrolment and inclusion of patients in the study.

p = 0.008). An increased presence of megamitochondria with common in patients with AH consuming CAMs compared to trend towards significance was noted among patients with classical AH (37% vs. 6.9%; p = 0.006). Interface hepatitis classical AH compared to AH patients consuming CAMs was not seen in the liver biopsy of patients with classical AH (27.6% vs. 7.4%, p = 0.05). Interestingly, the presence of but was significantly higher in patients with AH consuming foamy degeneration of hepatocytes were found to be higher in CAMs (n = 19/27, 70.4%; p < 0.001), with almost one- patients of AH on CAMs (n = 11/27, 40.7%) compared to third of the patients demonstrating severe grades of interface those with classical AH (n = 5/29, 17.2%; p = 0.05). Under- hepatitis. The presence of hepatocyte necrosis was more lying cirrhosis was seen in 81.5% and 89.7% of patients in commonly noted in patients with severe AH consuming the AH-CAM and classical AH groups respectively (p = 0.38). CAMs (19/27, 70.4% vs. 5/29, 17.2%; p < 0.001). Among Apart from the classical features seen on liver histopathol- patterns of necrosis, periportal necrosis (p = 0.003), bridging ogy in AH patients, variable additional findings were notable necrosis (p = 0.004), confluent necrosis (p = 0.003) and sub- in patients with AH consuming CAMs. Eosinophilic infiltration massive necrosis (p = 0.01) were significantly more conspic- of the portal and lobular areas was striking in patients with AH uous in patients who consumed CAMs, while spotty necrosis on CAMs only compared to those with classical AH (44.4% vs. and perivenular necrosis patterns of liver injury were compa- 0%, p < 0.001). Plasma cell infiltrates were significantly more rable between both groups. Significantly larger proportion of

332 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 329–340 Philips C.A. et al: CAM use in alcoholic hepatitis and outcomes

Table 1. Patient characteristics between groups

Descriptive, (n,%) AH–CAM group, n = 27 Classical AH group, n =29 p value

Obesity 7 (25.9%) 0 (0%) 0.003 Systemic hypertension 3 (11.1%) 2 (6.9%) 0.58 Hypothyroidism 1 (3.7%) 0 (0%) 0.30 Diabetes mellitus 11 (40.7%) 9 (31%) 0.45 Dyslipidemia 6 (22.2%) 0 (0%) 0.007 Hepatic encephalopathy (any grade) 16 (59.3%) 15 (51.7%) 0.57 Hepatic encephalopathy Grade 1 + 2 9 (33.3%) 13 (44.8%) 0.38 Grade 3 + 4 6 (22.2%) 4 (13.7%) 0.41 Ascites (any grade) 19 (70.4%) 23 (79.3%) 0.44 Grade 1 3 (11.1%) 7 (24.1%) 0.20 Grade 2 11 (40.7%) 14 (48.3%) 0.57 Grade 3 8 (29.6%) 3 (10.3%) 0.07 Acute variceal bleeding 11 (40.7%) 3 (10.3%) 0.009 Acute variceal bleeding on follow up (at 180 days) 13 (48.1%) 7 (24.1%) 0.06 ACLF grades 0 + 1 14 (51.8%) 19 (65.5%) 0.30 2 + 3 13 (48.1%) 10 (34.4%) 0.31

Abbreviations: AH, alcoholic hepatitis; ACLF, acute-on-chronic liver failure; CAM, complementary and alternative medicine. patients in the CAM group had cholestasis on liver biopsy 87 CAMs were consumed by 27 patients, of which identifica- (n = 20/27, 74.1% vs. n = 6/29, 20.7%; p < 0.001). Regard- tion of at least one component in the CAM was possible only in ing the type of cholestasis, cholangiolar cholestasis was seen 40.7% of patients. Eight patients ingested four CAMs, seven- only in patients of AH on CAM therapy (44.4% vs. 0%, p < teen ingested three CAMs, and two patients consumed two 0.001), while hepatocellular and canalicular cholestasis, even CAM products each. The mean duration of CAM intake was though observed among both groups, was significantly higher 10.6 ± 5.7 days (min-max, 3–21 days). A detailed list of in AH patients consuming CAMs (74.1% vs. 20.7%; p < 0.001 identifiable CAMs confirmed retrospectively is shown in Sup- and 40.7% vs. 15.4%; p = 0.003 respectively). Sinusoidal plementary Table 2. We could retrieve 42 samples of CAM fibrosis and sinusoidal dilatation were significantly more fre- from the patients in retrospect. Complete toxicology analysis quent in patients with AH on CAM therapy. Liver histopathol- of the retrieved CAM products revealed unsettling data ogy of none of the patients revealed classical features of regarding quality of manufacturing, adulteration practices autoimmune hepatitis. Table 3 shows the detailed liver and presence of a multitude of liver and other organ biopsy comparisons between groups, and the pertinent and toxicity-promoting organic as well as inorganic components. specific liver histology features associated with AH consuming A comprehensive list of these CAMs, along with the complete CAMs patients is shown in Fig. 2. chemical analysis findings, are shown in Supplementary Table 3. Representational images of some of the retrieved Patient outcomes CAMs are shown in Fig. 4. Disclosure of components as well as potentially toxic ingredients were available in less than At the end of 1-, 3- and 6-months follow-up, the proportion of 40% of the retrieved samples. For example, the presence of patients surviving in the AH consuming CAMs group in ‘excreta of baby elephant’ as well as ‘semen of the Civet cat’ comparison to the classical AH group were 37.04% vs. as a CAM constituent was disclosed in a drug (‘Dhanwantaram 82.76% (p < 0.001), 29.63% vs. 62.1% and 18.52% Gulika’, see Fig. 4 for more detail). The pertinent findings on (p = 0.002) vs. 51.72% (p = 0.001) respectively. A signifi- chemical analysis and toxicology included adulteration with cantly higher proportion of patients died early on, during the topical antifungal agents, nonsteroidal anti-inflammatory course of AH when exposed to CAM treatments (Fig. 3). drugs (like aspirin and diclofenac), alcohols, antimicrobials Sepsis was the most common cause of death among both (such as amoxycillin, lincomycin and salinomycin), chemo- groups (AH consuming CAMs 45.5% vs. classical AH 64.3%) therapeutic agents, hazardous industrial solvents, pesticides, followed by progressive liver failure (Supplementary Table 1). herbicides, insecticide components, neuropsychiatric drugs, recreational drug components, anticoagulants, mood CAMs and chemical analysis of retrieved drugs enhancers, mutagens, diuretics (spironolactone), sedatives, antidepressants and anticonvulsant drugs, as well as poison- Only six (n = 27, 22.2%) patients with suspected AH-related ous sesquiterpenes. With regards to heavy metal identifica- jaundice approached a registered Ayurveda practitioner for tion, lead was the most commonly detected (36 samples, treatments, while the majority took treatments from unregis- 85.7%, max 73.3 mg/kg) followed by nickel (33, 78.6%, tered, unregulated local traditional Ayurvedic healers or max 73.2 mg/kg), arsenic (29, 69.1%, max 119.4 mg/kg), underwent self-treatment with herbal medicines. A total of mercury (25, 59.5%, max 8 mg/kg) and manganese (20,

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 329–340 333 Philips C.A. et al: CAM use in alcoholic hepatitis and outcomes

Table 2. Patient investigational characteristics between groups

Group descriptive Group n Mean Median SD SE p value

Age (years) AH-CAM 27 45.85 44.0 9.77 1.88 0.18 AH 29 49.41 51.0 9.84 1.83 Hemoglobin (g/L) AH-CAM 27 9.74 9.8 1.87 0.36 0.78 AH 29 10.01 10.1 2.41 0.45 WBC count (x1000/mm3) AH-CAM 27 13.21 14.7 5.18 1.00 0.29 AH 29 12.63 10.9 7.68 1.43 Platelet count (x109/L) AH-CAM 27 99.85 81.0 64.36 12.39 0.02 AH 29 115.55 101.0 47.68 8.85 Total bilirubin (mg/dL) AH-CAM 27 18.80 18.6 9.73 1.87 0.007 AH 29 12.61 10.2 6.77 1.26 AST (IU/L) AH-CAM 27 129.04 120.0 87.95 16.93 0.44 AH 29 143.38 92.0 119.26 22.15 ALT (IU/L) AH-CAM 27 63.59 46.0 46.52 8.95 0.84 AH 29 81.34 48.0 93.73 17.41 ALP (IU/L) AH-CAM 27 146.33 142.0 58.73 11.30 0.90 AH 29 169.38 134.0 106.56 19.79 GGT (IU/L) AH-CAM 27 222.89 184.0 156.32 30.08 0.18 AH 29 168.83 141.0 116.12 21.56 Serum albumin (mg/dL) AH-CAM 27 2.67 2.8 0.41 0.08 0.43 AH 29 2.84 2.7 0.55 0.10 INR AH-CAM 27 2.53 2.2 0.83 0.16 0.71 AH 29 2.61 2.4 0.83 0.15 Sodium (mmol/L) AH-CAM 27 128.74 130.0 4.74 0.91 0.22 AH 29 130.00 131.0 5.87 1.09 Potassium (mmol/L) AH-CAM 27 4.11 4.1 0.78 0.15 0.95 AH 29 4.09 4.2 0.86 0.16 Blood urea (mg/dL) AH-CAM 27 39.78 29.0 30.03 5.78 0.87 AH 29 47.62 27.0 49.38 9.17 Creatinine (mg/dL) AH-CAM 27 1.31 1.0 0.71 0.14 0.93 AH 29 1.53 1.1 1.16 0.22 DF AH-CAM 27 88.74 80.8 31.14 5.99 0.01 AH 29 68.58 61.0 30.23 5.61 CTP AH-CAM 27 12.63 13.0 1.11 0.21 0.01 AH 29 11.62 12.0 1.54 0.29 MELD sodium AH-CAM 27 31.41 30.0 4.61 0.89 0.65 AH 29 30.31 31.1 5.87 1.09 ACLF grade AH-CAM 27 1.44 1.0 1.25 0.24 0.08 AH 29 0.90 0.0 1.11 0.21 CLIF C ACLF score AH-CAM 27 60.41 58.0 15.23 2.93 0.38 AH 29 57.38 57.0 9.86 1.83 CLIF score AH-CAM 27 11.30 11.0 2.64 0.51 0.001 AH 29 9.21 9.0 2.14 0.40

Abbreviations: ACLF, acute-on-chronic liver failure; AH, alcoholic hepatitis; ALP, alkaline phosphatase; ALT, alanine aminotransferase; AST, aspartate aminotransferase; CAM, complementary and alternative medicine; CLIF, chronic liver failure; CTP, Child Turcotte Pugh score; DF, discriminant function; GGT, gamma glutamyl transpeptidase; INR, international normalized ratio; MELD, model for end-stage liver disease score; SD, standard deviation; SE, standard error.

334 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 329–340 Philips C.A. et al: CAM use in alcoholic hepatitis and outcomes

Table 3. Liver biopsy characteristics between groups

Histology findings AH–CAM group, n = 27 Classical AH group, n =29 p value

Steatosis 100% 100% Grading of steatosis Mild 48.1% Mild 13.8% 0.02 Moderate 44.4% Moderate 65.5% Severe 7.4% Severe 20.7% Predominant inflammation Lymphocyte 63% Lymphocytes 10.3% <0.001 Neutrophils 37% Neutrophils 89.7% Predominant site of inflammation Lobular 0% Lobular 75.9% Portal 77.8% Portal 0% <0.001 Lobular + portal 22.2% Lobular + portal 24.1% Ballooning 100% 100% Interface hepatitis 70.4% 0% <0.001 Mallory Denk bodies 77.8% 100% 0.007 Neutrophilic satellitosis 29.6% 55.2% 0.05 Megamitochondria 7.4% 27.6% 0.05 Pericellular fibrosis 100% 100% Cirrhosis 81.5% 89.7% 0.38 Plasma cells (portal, lobular or interface) 37% 6.9% 0.006 Eosinophilic infiltration (portal, lobular or 44.4% 0% <0.001 interface) Cholestasis 74.1% 20.7% <0.001 Type of cholestasis Hepatocellular 74.1% Hepatocellular 20.7% <0.001 Canalicular 40.7% Canalicular 6.9% 0.003 Cholangiolar 44.4% Cholangiolar 0% <0.001 Foamy degeneration 40.7% 17.2% 0.05 Necrosis (any type) 70.4% 17.2% <0.001 Patterns of necrosis Periportal necrosis 25.9% Periportal necrosis 0% 0.004 Perivenular necrosis Perivenular necrosis 0.23 22.2% 10.3% Spotty necrosis 14.8% Spotty necrosis 6.9% 0.90 Bridging necrosis 25.9% Bridging necrosis 0% 0.003 Confluent necrosis 25.9% Confluent necrosis 0% 0.004 Submassive necrosis Submassive necrosis 0% 0.02 18.5% Hepatocyte rossetting 11.1% 0% 0.07 Sinusoidal dilatation 33.3% 0% <0.001 Sinusoidal fibrosis 44.4% 7% 0.001

Abbreviations: AH, alcoholic hepatitis; CAM, complementary and alternative medicine.

47.6%, max 824.2 mg/kg). The complete quantification important aspects in CAM-seeking health behavior and its levels of heavy metals and the permissible limits of daily outcomes among patients with acute liver disease, such as oral exposure is shown in Table 4. AH. Foremost in this regard is the lack of proper investigational acumen among CAM practitioners to understand and stratify patients with severe AH. Only one-third of the patients Discussion approaching a CAM practitioner underwent baseline liver In this novel study, we determined the clinical outcomes and function tests, mostly driven by patient request. This is in liver histopathology of AH patients undergoing complemen- agreement with previous publications on CAM that impress on tary and alternative therapy with Ayurvedic drugs, both the fact that adequate clinical evaluation, quality of treatment proprietary and traditional. Our study brings forth many options as well as services provided are in the absence of

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 329–340 335 Philips C.A. et al: CAM use in alcoholic hepatitis and outcomes

Fig. 2. Liver histopathology features of alcoholic hepatitis patients on Ayurvedic and herbal medications. A. Extensive liver cell necrosis in a patient with severe alcoholic hepatitis associated with ballooned hepatocytes, Mallory-Denk bodies and mixed inflammatory reaction predominated with lymphocytes and eosinophils (hematoxylin and eosin, x40). B. Confluent necrosis in a patient with multiple unknown polyherbal Ayurvedic medications (hematoxylin and eosin, x100). C. Clusters of ballooned hepatocytes surrounded by submassive necrosis and pericellular fibrosis due to consumptions of multiple Ayurvedic products after diagnosis of alcoholic liver disease (Masson-trichrome stain, x100). D. Extensive ballooned hepatocytes with Mallory-Denk bodies, with hepatocellular and cholangiolar cholestasis associated with herbal drug intake (hematoxylin and eosin, x400). E. Mixed inflammatory cells of lymphocytes and eosinophils in the portal areas with surrounding ballooned hepatocytes in a patient with severe alcoholic hepatitis with herbal drug-induced liver injury (hematoxylin and eosin, x400). F. Severe canalicular cholestasis associated with liver cell necrosis and ballooning of hepatocytes seen in a patient consuming Ayurvedic supplements for alcoholic hepatitis (hematoxylin and eosin, x400). G. Extensive bridging necrosis associated with mixed cellular inflammation and pericellular fibrosis in a patient with severe alcoholic liver disease after consuming traditional Ayurvedic medicines (Masson trichrome stain, x100).

Fig. 3. Kaplan-Meier survival analysis’ graphical representation between groups at end of 1-, 3- and 6-months.

336 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 329–340 Philips C.A. et al: CAM use in alcoholic hepatitis and outcomes

Fig. 4. Example of retrieved Ayurvedic and herbal medicines. A. Traditional Ayurvedic medications prepared and marketed by various pharmaceutical companies, with some lacking proper ingredient labelling, adverse effect notification and maximal dosing for use; labelled ingredients in an Ayurvedic formulation called ‘Dhanwantaram Gulika’, which was given for nausea, anorexia and overall ‘liver health’. B. The components mentioned include fresh excreta of baby elephant (yellow lined) as well as semen of the Civet cat (red lined). C. Some of the formulations came as powders contained in plain papers and unlabeled capsules, tablets and pills. D, E. Freshly plucked (D) and dried herbs (E) were a major part of CAMs utilized by alcoholic hepatitis patients, prescribed by traditional unregistered healers. F, G. Traditional Ayurvedic healer-prepared thick syrups (F) and unlabeled, unknown semi-solid material (G), made out of multiple herbs and powders. identifiable benefits.23 Even though we looked at outcomes their clinical symptoms. On analysis of these agents, we were only in patients with AH, it was already shown that alcoholic able to clearly identify a multitude of liver toxic ingredients liver disease and alcohol use were among the important and undisclosed poisonous substances and multiple adulter- factors associated with CAM-seeking behavior among ants (mostly modern medicine related). Antimicrobials, 15 patients with cirrhosis. including topical agents, pain killers, poisonous hazardous In our study, the AH consuming CAMs group of patients chemicals, chemotherapeutic agents, mood enhances, anti- had more severe liver disease, higher grades of portal hyper- depressants, sedatives and heavy metals were found in a tensive complications and poorer clinical outcomes. The large proportion of CAMs analyzed as per state-of-the-art survival rates in the short and long term were appalling, due methodology. This finding agrees with previous studies which to the fact that liver-related events were higher in the have showcased toxic levels of heavy metals, mislabeling of absence of strong recommendations to treatment for severe herbal products and volatile organic compounds, and heavy DILI in an already sick cohort of patients. Several authors metal-related higher death rates among patients on uniden- have already cautioned against and implored on the sensible tified and complex CAMs.6,9,25,26 Our study also confirms the use of natural medicine for liver diseases.24 The liver disease fact that certain well known and currently heavily utilized pro- severity scores and AH severity were higher among patients prietary Ayurvedic agents (such as Himalaya® Liv.52) is asso- in the CAM group. In the current study, we found that several specific findings predominated in AH patients on CAM therapy. ciated with high death rate among patients with advanced Notable among these was the presence of severe grades of liver disease. This particular product has undergone extensive necrosis, interface hepatitis, eosinophilic infiltration, and preclinical study, mostly from the Indian subcontinent, from lesser degrees of steatosis and Mallory-Denk bodies. The the year 1966 up until 2012, without any quality trials to presence of necrosis in patients with CAM induced DILI, confirm its true efficacy in human related liver diseases. The among both the general population and cirrhosis patients, only high quality multicenter randomized controlled trial of ® 27 has been previously documented to promote worse Liv.52 for cirrhosis in humans conducted by Fleig et al., outcomes.5,6 was abruptly stopped in view of higher mortality seen in To the best of our abilities, we were able to retrieve a good Child Pugh class C patients, leading to the drug’s withdrawal proportion of the CAMs ingested by the patients at the start of in the United States.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 329–340 337 Philips C.A. et al: CAM use in alcoholic hepatitis and outcomes

Table 4. Heavy metal** analysis and quantification in retrieved Ayurvedic samples

Total samples, n = 42 Detected, n (%) Minimum1Ã Maximum2Ã Mean3Ã Median4Ã SD

Aluminum 7 (16.6%) 18.560 418.400 130.380 68.850 142.2248 Arsenic 29 (69.1%) 0.0100 119.430 13.963 0.400 33.3589 Gold 13 (30.9%) 0.0400 1.440 0.662 0.480 0.5816 Mercury 25 (59.5%) 0.0100 8.010 0.673 0.190 1.5854 Manganese 20 (47.6%) 0.640 824.170 74.031 19.720 182.0916 Nickel 33 (78.6%) 0.200 73.290 7.461 2.150 14.2936 Lead 36 (85.7%) 0.0800 73.370 7.129 1.465 13.7904 Thallium 18 (42.8%) 0.0100 2.400 0.432 0.250 0.5826

* values in mg/kg. ** the level below which there is low risk for human ill health for arsenic is 0.07 mg/kg; the permissible tolerable weekly intake (PTWI) for mercury is 1.6 mg/kg body weight; PTWI of aluminum is 1 mg/kg of body weight; tolerable upper limit of manganese is 11 mg per day; tolerable daily intake for nickel is 5 mg/kg body weight; permissible daily oral exposure (PDE) of gold is 134 mg/kg; PDE of lead is 5 mg per day and of thallium is 8 mg per day. From U. S. Department of Health and Human Services Food and Drug Administration, Center for Drug Evaluation and Research (CDER) and Center for Biologics Evaluation and Research (CBER), Guidance for Industry 2015 update. Retrieved from https://www.fda.gov/regulatory-information/search-fda-guidance-documents/q3d-elemental- impurities on 3/10/2019.

The toxicity of Ayurvedic preparations, even leading to patient and family needs to be educated with respect to sensible cluster-poisoning, with respect to poor manufacturing practi- health-seeking behavior for possible severe liver diseases, so as ces and unregulated quality checks has been published to reduce morbidity and mortality, and two being the educating immensely in literature.28,29 Previous studies have also shown and integrating of CAM practitioners to identify patients with that heavy metal contamination and exposure could potentiate potential severe liver disease who require early referrals, so a multitude of liver disease, both acute and chronic.30,31 as to initiate proper evidence-based treatment to improve clin- Our study is important from two perspectives – one, the ical outcomes. Our findings also substantiate the use of liver

Fig. 5. Visual abstract summary infographic of the study.

338 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 329–340 Philips C.A. et al: CAM use in alcoholic hepatitis and outcomes biopsy in patients with severe AH to identify occult CAM use to as per clinical criteria, the most possible diagnosis in these tailor specific medical treatments and to define specific patterns patients, prior to CAM use, could have been AH itself. of injury associated with high mortality to potentiate early liver Patients with severe AH and associated CAM intake have transplant referral. Nickel, lead, mercury, arsenic and manga- severely augmented liver injury leading to higher morbidity nese have been shown to induce acute as well as chronic and mortality than seen with classical AH patients on con- hepatic injury in animal models and human subjects. The inju- servative and evidence-based management. The presence of ries have been demonstrated to range from transaminase ele- adulterants such as antimicrobials, topical agents, poisons as vations to hepatocyte necrosis and in the case of arsenic, well as chemotherapeutic agents and psychiatry prescription occurrence of portal hypertension.32–39 Food- and drug-grade drugs in CAM from the Indian subcontinent is clarified (Info- substances are recommended to be devoid such contaminants graphic Fig. 5). The use of CAMs in patients with acute as well since continued exposure associated with long-term consump- as chronic liver diseases must be regulated, and the general tion could lead to potential liver injury. The recommended and patient population educated with regards to sensible maximal exposures to heavy metals are for environmental soil health-seeking behavior. The need for close monitoring and and water sources and not for food or drug components and stringent Good Manufacturing Practices in CAM needs an must not be referenced synonymously. immense upheaval to prevent adulterant-related organ tox- Our study has strengths as well as weaknesses. This is the icity. Physicians prescribing CAMs need to be educated, with first study to describe and analyze a large group of patients regards to sensible evaluation, timely referrals and incorpo- with AH on CAM treatments from clinical, histopathological ration of evidence-based medicine into practice and taught and toxicology points of view, underlining important clinical research methodology for improving identification of benefi- and liver histological aspects that can help differentiate dual cial CAMs that may become an important component in the armamentarium of modern medicine. insults in AH patients as well as prognosticate regarding transplant-free survival. All our patients, in both groups, underwent liver biopsy, leaving no room for doubt on the Conflict of interest type and patterns of liver injury in this study. We also identified significant adulteration and presence of undisclosed The authors have no conflict of interests related to this components in a multitude of CAM agents analyzed that calls publication. for proper regulation from the government agencies con- cerned, in close participation with the industry. Our study Author contributions limitations are that it was retrospective in nature and from a single center. Even though we systematically compiled and Prepared the manuscript, wrote the initial draft, collated data, analyzed our patient groups, other rare or atypical causes performed analysis (CAP), revised the manuscript, performed for liver disease apart from the suspected alcohol and additional analysis (SR), collected data and provided critical CAM-related DILI could have been overseen. The presence review of the manuscript (TG), collected data, performed of autoantibodies in patients with CAM use could possibly additional analysis and provided critical revisions (RA), col- mean an underlying autoimmune phenomenon related to lected data and provided critical manuscript revisions (SK), liver injury or autoimmune hepatitis. However, even in the provided critical revisions to the manuscript (PA). All authors presence of autoantibodies, the evidence for autoimmune finalized the current version of manuscript for submission. hepatitis was negated in all patients on liver biopsy. The presence of autoantibodies associated with herbal medicines References has been previously well described, which agrees with our 6,25,40 findings. We did not apply the Roussel Uclaf Causality [1] Im GY. Acute alcoholic hepatitis. Clin Liver Dis 2019;23:81–98. doi: 10. Assessment Method (commonly known as RUCAM) to assess 1016/j.cld.2018.09.005. DILI causality, since this tool has not been validated in [2] Woolbright BL, Jaeschke H. Alcoholic hepatitis: Lost in translation. J Clin Transl Hepatol 2018;6:89–96. doi: 10.14218/JCTH.2017.00054. patients with concurrent acute liver injury. However, our [3] Wu T, Morgan TR, Klein AS, Volk ML, Saab S, Sundaram V. Controversies in biopsy findings inarguably confirmed a second insult, most early liver transplantation for severe alcoholic hepatitis. Ann Hepatol 2018; probably CAM-related DILI due to the heterogeneity in 17:759–768. doi: 10.5604/01.3001.0012.3134. biopsy findings, that skewed from the expected findings. We [4] Weeks SR, Sun Z, McCaul ME, Zhu H, Anders RA, Philosophe B, et al. Liver transplantation for severe alcoholic hepatitis, updated lessons from the did not quantify the organic, inorganic and volatile organic world’s largest series. J Am Coll Surg 2018;226:549–557. doi: 10.1016/j. compounds in the samples, due to the high costs associated jamcollsurg.2017.12.044. with such analysis as well as difficulties in procuring stand- [5] Philips CA, Paramaguru R, Augustine P, Rajesh S, Ahamed R, George T, et al. A single-center experience on outcomes of complementary and alternative ards. However, the presence of some of these compounds has medicine use among patients with cirrhosis. Hepatol Commun 2019;3:1001– been linked to liver injury the in literature, which agrees with 1012. doi: 10.1002/hep4.1355. our study findings.6,41 Analysis of body fluids for toxins and [6] Philips CA, Paramaguru R, Joy AK, Antony KL, Augustine P. Clinical outcomes, chemicals were not undertaken in view of differences in time histopathological patterns, and chemical analysis of Ayurveda and herbal medicine associated with severe liver injury-A single-center experience to presentation to the hospital, heterogeneity in duration of from southern India. Indian J Gastroenterol 2018;37:9–17. doi: 10. consumption, and time to consumption and stopping CAMs in 1007/s12664-017-0815-8. the study population. Even though the analytical lab facility [7] Sharp D, Lorenc A, Morris R, Feder G, Little P, Hollinghurst S, et al.Comple- mentary medicine use, views, and experiences: a national survey in England. was very close to our hospital premises, volatility of drug BJGP Open 2018;2:bjgpopen18X101614. doi: 10.3399/bjgpopen18X101614. components and associated changes on analysis could have [8] Fernández A, Simian D, Quera R, Flores L, Ibáñez P, Lubascher J, et al. Com- occurred. Since most of the patients who consumed CAMs did plementary and alternative medicine in patients with inflammatory bowel not undergo liver tests at baseline, the presence of AH at disease: A survey performed in a tertiary center in Chile. Complement Ther Med 2018;40:77–82. doi: 10.1016/j.ctim.2018.07.016. baseline in these are doubtful and the acute liver injury [9] Navarro V, Avula B, Khan I, Verma M, Seeff L, Serrano J, et al. The contents of could have been purely due to the ingested CAMs. However, herbal and dietary supplements implicated in liver injury in the United States

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 329–340 339 Philips C.A. et al: CAM use in alcoholic hepatitis and outcomes

are frequently mislabeled. Hepatol Commun 2019;3:792–794. doi: 10. [25] Devarbhavi H. Ayurvedic and herbal medicine-induced liver injury: It is time 1002/hep4.1346. to wake up and take notice. Indian J Gastroenterol 2018;37:5–7. doi: 10. [10] Lin NH, Yang HW, Su YJ, Chang CW. Herb induced liver injury after using 1007/s12664-018-0820-6. herbal medicine: A systemic review and case-control study. Medicine (Balti- [26] Stolz A, Navarro V, Hayashi PH, Fontana RJ, Barnhart HX, Gu J, et al. Severe more) 2019;98:e14992. doi: 10.1097/MD.0000000000014992. and protracted cholestasis in 44 young men taking bodybuilding supple- [11] Bernuau JR, Durand F. Herbal medicines in acute viral hepatitis: a ticket for ments: assessment of genetic, clinical and chemical risk factors. Aliment more trouble. Eur J Gastroenterol Hepatol 2008;20:161–163. doi: 10. Pharmacol Ther 2019;49:1195–1204. doi: 10.1111/apt.15211. 1097/MEG.0b013e3282f2bbf7. [27] Schuppan D, Jia JD, Brinkhaus B, Hahn EG. Herbal products for liver dis- [12] Henson JB, Brown CL, Chow SC, Muir AJ. Complementary and alternative eases: a therapeutic challenge for the new millennium. Hepatology 1999; medicine use in United States adults with liver disease. J Clin Gastroenterol 30:1099–1104. doi: 10.1002/hep.510300437. 2017;51:564–570. doi: 10.1097/MCG.0000000000000617. [28] Manohar PR. Toxicity of Ayurveda medicines and safety concerns: The need [13] Mullin GE. The use of complementary and alternative medicine for liver to revive the branch of toxicology in Ayurveda. Anc Sci Life 2014;34:1–2. disease: part II. Nutr Clin Pract 2013;28:277–279. doi: 10. doi: 10.4103/0257-7941.150761. 1177/0884533612475133. [29] Mikulski MA, Wichman MD, Simmons DL, Pham AN, Clottey V, Fuortes LJ. [14] Mullin GE. The use of complementary and alternative medicine for liver disease: Toxic metals in ayurvedic preparations from a public health lead poisoning – part I. Nutr Clin Pract 2013;28:138–139. doi: 10.1177/0884533612471766. cluster investigation. Int J Occup Environ Health 2017;23:187 192. doi: 10. [15] Ferrucci LM, Bell BP, Dhotre KB, Manos MM, Terrault NA, Zaman A, et al. 1080/10773525.2018.1447880. Complementary and alternative medicine use in chronic liver disease [30] Jan AT, Azam M, Siddiqui K, Ali A, Choi I, Haq QM. Heavy metals and human patients. J Clin Gastroenterol 2010;44:e40–e45. doi: 10.1097/MCG. health: Mechanistic insight into toxicity and counter defense system of anti- – 0b013e3181b766ed. oxidants. Int J Mol Sci 2015;16:29592 29630. doi: 10.3390/ijms161226183. [16] Batey RG, Salmond SJ, Bensoussan A. Complementary and alternative med- [31] Jaishankar M, Tseten T, Anbalagan N, Mathew BB, Beeregowda KN. Toxicity, icine in the treatment of chronic liver disease. Curr Gastroenterol Rep 2005; mechanism and health effects of some heavy metals. Interdiscip Toxicol 2014;7:60–72. doi: 10.2478/intox-2014-0009. 7:63–70. doi: 10.1007/s11894-005-0068-z. [32] Kim JJ, Kim YS, Kumar V. Heavy metal toxicity: An update of chelating ther- [17] Singal AK, Louvet A, Shah VH, Kamath PS. Grand rounds: Alcoholic hepatitis. apeutic strategies. J Trace Elem Med Biol 2019;54:226–231. doi: 10.1016/j. J Hepatol 2018;69:534–543. doi: 10.1016/j.jhep.2018.05.001. jtemb.2019.05.003. [18] Song NE, Lee JY, Mansur AR, Jang HW, Lim MC, Lee Y, et al. Determination of [33] Fiati Kenston SS, Su H, Li Z, Kong L, Wang Y, Song X, et al. The systemic 60 pesticides in hen eggs using the QuEChERS procedure followed by LC- toxicity of heavy metal mixtures in rats. Toxicol Res (Camb) 2018;7:396– MS/MS and GC-MS/MS. Food Chem 2019;298:125050. doi: 10.1016/j.food- 407. doi: 10.1039/c7tx00260b. chem.2019.125050. [34] Lin X, Gu Y, Zhou Q, Mao G, Zou B, Zhao J. Combined toxicity of heavy metal [19] Mazumder S, Ahamed RA, McGahee E, Wang L, Seyler TH. A new automated mixtures in liver cells. J Appl Toxicol 2016;36:1163–1172. doi: 10.1002/jat. method for the analysis of aromatic amines in human urine by GC-MS/MS. J 3283. Anal Toxicol 2019;43:25–35. doi: 10.1093/jat/bky045. [35] Tchounwou PB, Yedjou CG, Patlolla AK, Sutton DJ. Heavy metal toxicity and [20] Chang QY, Pang GF, Fan CL, Chen H, Yang F, Li J, et al. High-throughput the environment. Exp Suppl 2012;101:133–164. doi: 10.1007/978-3-7643- analytical techniques for the determination of the residues of 653 multiclass 8340-4_6. pesticides and chemical pollutants in tea, Part VII: A GC-MS, GC-MS/MS, and [36] Zeinali T, Salmani F, Naseri K. Dietary intake of cadmium, chromium, copper, LC-MS/MS study of the degradation profiles of pesticide residues in green nickel, and lead through the consumption of meat, liver, and kidney and – tea. J AOAC Int 2016;99:1619 1627. doi: 10.5740/jaoacint.16-0158. assessment of human health risk in Birjand, Southeast of Iran. Biol Trace [21] Bernardi G, Kemmerich M, Ribeiro LC, Adaime MB, Zanella R, Prestes OD. An Elem Res 2019;191:338–347. doi: 10.1007/s12011-019-1637-6. effective method for pesticide residues determination in tobacco by GC- [37] Yu S, Liu F, Wang C, Zhang J, Zhu A, Zou L, et al. Role of oxidative stress in MS/MS and UHPLC-MS/MS employing acetonitrile extraction with low- liver toxicity induced by nickel oxide nanoparticles in rats. Mol Med Rep – temperature precipitation and d-SPE clean-up. Talanta 2016;161:40 47. 2018;17:3133–3139. doi: 10.3892/mmr.2017.8226. doi: 10.1016/j.talanta.2016.08.015. [38] Adjroud O. The toxic effects of nickel chloride on liver, erythropoiesis, and [22] Chen X, Li Y, Chang QY, Hu XY, Pang GF, Fan CL. High-throughput analytical development in Wistar albino preimplanted rats can be reversed with sele- techniques for determination of residues of 653 multiclass pesticides and nium pretreatment. Environ Toxicol 2013;28:290–298. doi: 10.1002/tox. chemical pollutants in tea–Part V: A comparative study of the influence of 20719. tea hydration on the efficiency of pesticide multiresidue determination using [39] Huang P, Chen C, Wang H, Li G, Jing H, Han Y, et al. Manganese effects in the three sample preparation methods and GC/MS/MS. J AOAC Int 2015;98: liver following subacute or subchronic manganese chloride exposure in rats. 149–159. doi: 10.5740/jaoacint.13-279. Ecotoxicol Environ Saf 2011;74:615–622. doi: 10.1016/j.ecoenv.2010.08.011. [23] Bloom BS, Retbi A, Dahan S, Jonsson E. Evaluation of randomized controlled [40] Sebode M, Schulz L, Lohse AW. “Autoimmune(-like)” drug and herb induced trials on complementary and alternative medicine. Int J Technol Assess liver injury: New insights into molecular pathogenesis. Int J Mol Sci 2017;18: Health Care 2000;16:13–21. doi: 10.1017/s0266462300016123. 1954. doi: 10.3390/ijms18091954. [24] Xiong F, Guan YS. Cautiously using natural medicine to treat liver problems. [41] Lang AL, Beier JI. Interaction of volatile organic compounds and underlying World J Gastroenterol 2017;23:3388–3395. doi: 10.3748/wjg.v23.i19. liver disease: a new paradigm for risk. Biol Chem 2018;399:1237–1248. doi: 3388. 10.1515/hsz-2017-0324.

340 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 329–340 Original Article

The Association between Meteorological Factors and the Prevalence of Acute-on-chronic Liver Failure: A Population-based Study, 2007–2016

Su Lin1, Lifen Han2, Dongliang Li3, Ting Wang1, Zimu Wu1, Haoyang Zhang4, Zhansong Xiao5, Yinlian Wu1, Jiaofeng Huang1, Mingfang Wang1 and Yueyong Zhu*1

1Liver Research Center of the First Affiliated Hospital of Fujian Medical University, Fuzhou, Fujian, China; 2Department of Infectious Disease, Mengchao Hepatobiliary Hospital of Fujian Medical University, Fuzhou, Fujian, China; 3Department of Hepatobiliary Disease, 900 Hospital of PLA, Fuzhou, Fujian, China; 4School of Public Health, Sun Yat-sen University, Guangzhou, Guangdong, China; 5Fujian Medical University, Fuzhou, Fujian, China

Abstract Introduction

Background and Aims: The aim of this study was to inves- Acute-on-chronic liver failure (ACLF) is a critical clinical tigate the effect(s) of meteorological factors on the preva- syndrome characterized by rapid development of hepatocel- lence of acute-on-chronic liver failure (ACLF) based on lular dysfunction.1,2 Epidemiological and clinical data show 10-years’ worth of population data. Methods: We retrospec- that although the prevalence of ACLF is relatively low, it has tively collected ACLF case data from January 2007 to Decem- an extremely poor survival rate (28-day mortality rate of 30– ber 2016 from three major hospitals in Fuzhou City, China. 40%).2–4 A widely accepted concept is that ACLF results from Climatic data, including rainfall, mean temperature, differen- the acute deterioration of preexisting chronic liver diseases.5 ces in temperature (delta temperature) and mean humidity for Bacterial infection,6 viral infection,7,8 and alcohol intake9 each month were downloaded from the China Climatic Data have been shown to contribute to the initiation of ACLF. Service Center. Following data collection, Poisson regression However, there are still 40% of patients with ACLF whose analysis was used to estimate the effect(s) of climatic factors precipitating factors remain unknown.10 on the risk of the prevalence of ACLF. Results: The population Meteorological factors, such as temperature, humidity and consisted of a total of 3510 cases, with a mean age of 44.7 ± air pollution, can affect the health of human beings. Previous 14.8 years-old and with 79.8% being male. Upon analyzing studies have proven the existence of seasonal variation in the population data, we found a growing trend and seasonal several critical diseases, and climatic factors are considered pattern of monthly counts of ACLF-related hospitalization to play an important role in these seasonal patterns.11–16 In throughout the past decade. Specifically, the primary peak of accordance with this knowledge, we also noticed that the ACLF prevalence was in January and the secondary peak was in number of patients hospitalized for ACLF has varied over July. Poisson regression showed mean temperature (risk ratio time. However, we found few studies that focused on the = 0.991, 95%CI = 0.986–0.996) and mean humidity (risk ra- effect of climatic factors on the risk of ACLF. Therefore, we tio = 1.011, 95%CI = 1.006–1.017) to be independently cor- developed this study to explore the seasonal patterns contri- related with the monthly cases of ACLF. The results suggest buting to ACLF risk in a provincial city in Southeastern China that every unit increase of mean temperature (1°C) and mean and to identify the possible meteorological risk factors of ACLF humidity (1%) are associated with 0.991- and 1.011-fold based on 10-years’ worth of population data. The results of changes of ACLF cases, respectively. Rainfall and delta temper- our study may provide clues as to the underlying mechanisms ature did not appear to affect the prevalence of this disease. that condition patterns of ACLF hospitalizations and may con- Conclusions: The hospitalization for ACLF peaks in January tribute to the health and medical fields by pointing to ways in and July. Low temperature and high humidity appear to func- which ACLF-related interventions might be better tailored in tion as factors contributing to this seasonal pattern. public health. Citation of this article: Lin S, Han L, Li D, Wang T, Wu Z, Zhang H, et al. The association between meteorological Methods factors and the prevalence of acute-on-chronic liver failure: A population-based study, 2007–2016. J Clin Transl Hepatol Study population 2019;7(4):341–345. doi: 10.14218/JCTH.2019.00044. Hospitalization data for patients with ACLF as the main Keywords: Acute-on-chronic liver failure; Seasonal variation; Temperature; diagnosis were collected from administrative databases in Humidity. Abbreviations: ACLF, acute–on-chronic liver failure; CI, confidential interval; three tertiary hospitals located in Fuzhou City, Fujian Prov- HBV, hepatitis B virus. ince, China from January 2007 to December 2016. The Received: 6 September 2019; Revised: 13 November 2019; Accepted: 6 Decem- hospitals included in this study were the First Affiliated ber 2019 Hospital of Fujian Medical University, the Mengchao Hepato- *Correspondence to: Yueyong Zhu, Department of Liver Research Center, First Affiliated Hospital of Fujian Medical University, Fuzhou, Fujian 350001, China. Tel: biliary Hospital of Fujian Medical University and the Fuzhou +86-591-87981656, Fax: +86-591-87982526, E-mail: [email protected] PLA General Hospital. Liver failure is an extremely dangerous

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 341–345 341

Four major climatic factors were explored, including rainfall Statistical analyses (in milliliters), mean monthly temperature (°C), delta temperature, and mean humidity (%). Univariate analysis To explore the trends and seasonal patterns of monthly ACLF showed that rainfall was not associated with the prevalence cases, an automatic exponential model with trends and of ACLF, but that mean temperature and delta temperature seasonal components was used to fit the time series of were negatively related to ACLF, while humidity was positively ACLF cases. Univariate and multivariate Poisson regressions associated with ACLF cases (Table 2). After including these were used to measure the effect of climatic factors on ACLF three variables in a multivariate analysis, only mean temper- prevalence. Statistical analysis was performed using R soft- ature (risk ratio = 0.991, 95% confidential interval (CI) = ware (version 3.3.2, https://cran.r-project.org). All tests 0.986–0.996) and mean humidity (risk ratio = 1.011, 95% were two-sided and p-value <0.05 was considered statistically significant.

Ethical statement

This study was approved by the Ethical Review Committee of the First Affiliated Hospital of Fujian Medical University (Permission number: FYDYFY-0014211). All methods were performed in accordance with the Declaration of Helsinki, regarding ethical standards for research involving human subjects.

Patient and public involvement

No patient was involved in the design of this study, including in the development of the research question and outcome Fig. 1. Distribution of different etiologies of ACLF in each year. measures, in the recruitment and conduct of the study, nor in Abbreviation: ACLF, acute-on-chronic liver failure.

342 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 341–345 Lin S. et al: Meteorological factors and ACLF

Table 1. Demographic characteristics of 3510 ACLF patients

Overall, First Affiliated Fuzhou PLA General Mengchao Hepatobiliary Characteristics n = 3510 Hospital, n = 823 Hospital, n = 697 Hospital, n = 1990

Age (years), mean 6 SD 44.7 6 14.8 46.5 6 14.9 46.9 6 15.6 42.5 6 14.0 Male, n (%) 2802 (79.8%) 638 (77.5%) 533 (76.4%) 1631 (82.0%) Etiology, n (%) HBV alone 2593 (73.8%) 611 (74.2%) 407 (58.4%) 1575 (79.1%) Alcohol alone 158 (4.5%) 56 (6.8%) 40 (5.8%) 62 (3.1%) HBV & alcohol 213 (6.1%) 30 (3.6%) 28 (4.0%) 155 (7.8%) Others 546 (15.6%) 126 (15.3%) 222 (31.9%) 198 (9.9%) Years 2007 197 (5.6%) 62 (7.5%) 56 (8.0%) 79 (4.0%) 2008 221 (6.3%) 75 (9.1%) 55 (7.9%) 91 (4.6%) 2009 358 (10.2%) 79 (9.6%) 56 (8.0%) 223 (11.2%) 2010 339 (9.7%) 74 (9.0%) 74 (10.6%) 191 (9.6%) 2011 309 (8.8%) 83 (10.1%) 56 (8.0%) 170 (8.5%) 2012 337 (9.6%) 102 (12.4%) 50 (7.2%) 185 (9.3%) 2013 349 (9.9%) 82 (10.0%) 84 (12.1%) 183 (9.2%) 2014 375 (10.7%) 85 (10.3%) 84 (12.1%) 206 (10.4%) 2015 512 (14.6%) 89 (10.8%) 87 (12.5%) 336 (16.9%) 2016 513 (14.6%) 92 (11.2%) 95 (13.6%) 326 (16.4%)

Abbreviations: ACLF, acute-on-chronic liver failure; HBV, hepatitis B virus; SD, standard deviation.

CI = 1.006–1.017) were found to be independently correlated ary.21 Similarly, this study found seasonal variance associated with the prevalence of ACLF cases. The results suggested that with the incidence of ACLF, a fatal liver disease. These results every unit increase of mean temperature (1°C) and mean revealed again the existence of seasonal pattern of liver humidity (1%) were associated with 0.991- and 1.011-fold diseases. changes in ACLF cases, respectively. We found the number of ACLF hospital admissions peaked Smoothing plots of monthly mean temperature and mean in winter. The winter season has frequently been shown to relative humidity in relation to ACLF cases are shown in Fig. 3. increase the risks of critical diseases.11–14 Part of the reason Despite some fluctuations of the curve, it can be clearly seen for this could be related to lower temperatures since low envi- that the risk of ACLF increased with mean humidity. The peak ronmental temperatures induce immunosuppression in verte- of the relative risk of ACLF was at a mean temperature of 12° brates. In response, animals have to increase their immune C, which was the average temperature of Fuzhou City in activities to maintain immune balance.22 Proinflammatory January. The lowest risk of ACLF was associated with the factors’ expressions as well as macrophage and T-cell infiltra- humidity of 61% and a temperature of 16°C. tion are up-regulated in animals after cold exposure, causing systemic inflammation in lower temperatures. Given that excessive systemic inflammation has been known to play a Discussion crucial role in the development of ACLF,10 changes in the immune system due to low temperature may also contribute ACLF has raised an increasing concern because of its high 3,9,18 to its seasonality. mortality rate and heavy disease burden. A growing From a different perspective, it is also the case that the trend in the incidence of this fatal disease throughout the 2,18 decreased sun exposure in winter which has been shown to be past decade has been observed in the USA. In this 23 study, although a different diagnostic criterion for ACLF was associated with higher probabilities of vitamin D deficiency. used, we also demonstrated a rise in the prevalence of ACLF Low vitamin D serum concentration significantly correlates 24 in Southeastern China. Additionally, a correlation between the with liver diseases, including nonalcoholic fatty liver disease 25,26 prevalence of ACLF and low temperature as well as high and hepatitis B. Vitamin D deficiency leads to higher levels humidity was demonstrated in this study. of HBV replication, which increases the risk of infection and 25,26 The results showed that, although ACLF occurs throughout often leads to poor outcomes in patients with liver cirrhosis. the year, cold seasons are associated with higher prevalence. Moreover, temperature and light have also been shown to play Seasonal variance has also been found in less severe liver aroleinregulatingDNAmethylation.27 Fitting for our study, it diseases. This pattern seems to be dependent on the etiology is also shown that aberrant GSTP1 promoter methylation is of liver diseases: hepatitis A and hepatitis E have prominent associated with the development of ACLF and its prognosis.28 peaks in the spring and summer,19 primary biliary cirrhosis These factors, especially when they are combined, might help peaks in June,20 and hepatitis B peaks in January and Febru- explain why ACLF peaks in winter. However, further research is

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 341–345 343 Lin S. et al: Meteorological factors and ACLF

Fig. 2. The seasonal variations of ACLF over 10 years. The time series is divided into seasonal, trend, and irregular components. The total ACLF is demon- Fig. 3. Smoothing plots of monthly mean temperature and mean relative strated on the upper panel and HBV-ACLF is shown on the lower panel. The grey bars humidity against ACLF. The risk of ACLF can be seen to have increased with higher on the right represent the same magnitude. The time trend of ACLF prevalence is humidity and lower temperature. The peak of the relative risk of ACLF was at a shown to have been increasing over the last decade. The seasonal effect suggests the temperature of 12°C, which was the average temperature of Fuzhou City in January. peak prevalence of ACLF is in January. No difference in seasonal pattern is found Abbreviation: ACLF, acute-on-chronic liver failure. between the total ACLF and HBV-ACLF. Abbreviations: ACLF, acute-on-chronic liver failure; HBV, hepatitis B virus. explain the small peak of ACLF in July, as we found in our cohort study. still needed to elucidate the mechanistic links between temper- The major advantage of this study is that it is the first ature and ACLF. population-based study to investigate the correlation Our study found a positive relationship between high between climatic factors and ACLF risk. Even so, it also has humidity and the prevalence of ACLF, which is not surprising limitations. The major weakness of this study is that we only considering that moisture environment accelerates bacteria evaluated the condition of a single city. Therefore, the results growth15 and that warm-humid conditions increase the risk of of our study still need to be verified in other regions. infection.29 Infection is one of the most important precipitat- Furthermore, we only focused on the climate factors that ing factors of ACLF.6,30,31 Therefore, humidity likely has an influence ACLF onset; however, there are various confounding indirect effect on ACLF onset and high humidity might also factors which may have impacted this liver disease, such as

Table 2. Poisson regression for the risk of ACLF

Univariate Multivariate

Meteorological factor Risk ratio 95%CI p Risk ratio 95%CI p

Mean temperature in °C 0.991 0.986–0.996 <0.001 0.991 0.986–0.996 0.001 Mean humidity, % 1.012 1.007–1.018 <0.001 1.011 1.006–1.017 <0.001 Delta temperature in °C 0.945 0.920–0.979 <0.001 0.973 0.945–1.002 0.066 Rainfall in milliliter 1 1–1 0.719

Abbreviation: ACLF, acute-on-chronic liver failure.

344 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 341–345 Lin S. et al: Meteorological factors and ACLF alcohol consumption or upper respiratory infections. These disease in the young population in the United States. Am J Gastroenterol factors are prevalent in winter and summer as well, but it is 2019. doi: 10.14309/ajg.0000000000000411. [10] Clària J, Stauber RE, Coenraad MJ, Moreau R, Jalan R, Pavesi M, et al. Sys- difficult to collect such information in a retrospective study. temic inflammation in decompensated cirrhosis: Characterization and role in Additionally, well-designed, prospective research is required acute-on-chronic liver failure. Hepatology 2016;64:1249–1264. doi: 10. to address these issues. 1002/hep.28740. [11] Cheng J, Xu Z, Bambrick H, Su H, Tong S, Hu W. Impacts of heat, cold, and – Conclusions temperature variability on mortality in Australia, 2000 2009. Sci Total Environ 2019;651:2558–2565. doi: 10.1016/j.scitotenv.2018.10.186. [12] Sun Z, Chen C, Xu D, Li T. Effects of ambient temperature on myocardial A seasonal variation exists in ACLF hospitalization, which infarction: A systematic review and meta-analysis. Environ Pollut 2018; peaks in January and July. Low temperature and high humid- 241:1106–1114. doi: 10.1016/j.envpol.2018.06.045. [13] Morral-Puigmal C, Martínez-Solanas È, Villanueva CM, Basagaña X. Weather ity appear to be the primary contributing factors to this and gastrointestinal disease in Spain: A retrospective time series regression seasonal pattern. study. Environ Int 2018;121:649–657. doi: 10.1016/j.envint.2018.10.003. [14] Chen R, Yin P, Wang L, Liu C, Niu Y, Wang W, et al. Association between ambient temperature and mortality risk and burden: time series study in Acknowledgments 272 main Chinese cities. BMJ 2018;363:k4306. doi: 10.1136/bmj.k4306. [15] Redfern J, Verran J. Effect of humidity and temperature on the survival of This study was funded by Fujian Provincial Health Technology Listeria monocytogenes on surfaces. Lett Appl Microbiol 2017;64:276–282. Project (2016-CX-33), Fujian Science and Technology Project doi: 10.1111/lam.12714. [16] Kim BI, Ki H, Park S, Cho E, Chun BC. Effect of climatic factors on hand, foot, (2017J01187), Fujian Province Science and Technology and mouth disease in South Korea, 2010–2013. PLoS One 2016;11: Program (2019D002) and Fuzhou Science and Technology e0157500. doi: 10.1371/journal.pone.0157500. Program (2018-S-103-1). They supported the analysis of the [17] Sarin SK, Kedarisetty CK, Abbas Z, Amarapurkar D, Bihari C, Chan AC, et al. data and writing of the manuscript. Acute-on-chronic liver failure: consensus recommendations of the Asian Pacific Association for the Study of the Liver (APASL) 2014. Hepatol Int 2014;8:453–471. doi: 10.1007/s12072-014-9580-2. Conflict of interest [18] Kim D, Cholankeril G, Li AA, Kim W, Tighe SP, Hameed B, et al. Trends in hospitalizations for chronic liver disease-related liver failure in the United – – The authors have no conflict of interests related to this States, 2005 2014. Liver Int 2019;39:1661 1671. doi: 10.1111/liv.14135. [19] Fares A. Seasonality of hepatitis: a review update. J Family Med Prim Care publication. 2015;4:96–100. doi: 10.4103/2249-4863.152263. [20] McNally RJ, James PW, Ducker S, James OF. Seasonal variation in the patient Author contributions diagnosis of primary biliary cirrhosis: further evidence for an environmental component to etiology. Hepatology 2011;54:2099–2103. doi: 10.1002/hep. 24597. Designed the study (SL and YZ), collected and cleaned the [21] Gan R, Chen N, Huang D. Comparisons of forecasting for hepatitis in Guangxi data (TW, HL, DL, YW, JH, ZX and MW), analyzed the data (SL, Province, China by using three neural networks models. PeerJ 2016;4: HZ and ZW), wrote the manuscript (SL), contributed to e2684. doi: 10.7717/peerj.2684. [22] Martin LB, Weil ZM, Nelson RJ. Seasonal changes in vertebrate immune supervising the project and checked the final version of the activity: mediation by physiological trade-offs. Philos Trans R Soc Lond B manuscript (YZ). All authors read and approved the final Biol Sci 2008;363:321–339. doi: 10.1098/rstb.2007.2142. version of the manuscript. [23] Rosecrans R, Dohnal JC. Seasonal vitamin D changes and the impact on health risk assessment. Clin Biochem 2014;47:670–672. doi: 10.1016/j. clinbiochem.2014.02.004. References [24] Liu S, Liu Y, Wan B, Zhang H, Wu S, Zhu Z, et al. Association between vitamin D status and non-alcoholic fatty liver disease: A population-based study. – [1] Hernaez R, Solà E, Moreau R, Ginès P. Acute-on-chronic liver failure: an J Nutr Sci Vitaminol (Tokyo) 2019;65:303 308. doi: 10.3177/jnsv.65.303. update. Gut 2017;66:541–553. doi: 10.1136/gutjnl-2016-312670. [25] Farnik H, Bojunga J, Berger A, Allwinn R, Waidmann O, Kronenberger B, et al. [2] Allen AM, Kim WR, Moriarty JP, Shah ND, Larson JJ, Kamath PS. Time trends Low vitamin D serum concentration is associated with high levels of hepatitis in the health care burden and mortality of acute on chronic liver failure in the B virus replication in chronically infected patients. Hepatology 2013;58: – United States. Hepatology 2016;64:2165–2172. doi: 10.1002/hep.28812. 1270 1276. doi: 10.1002/hep.26488. [3] Allen AM, Kim WR. Epidemiology and healthcare burden of acute-on-chronic [26] Konstantakis C, Tselekouni P, Kalafateli M, Triantos C. Vitamin D deficiency in – liver failure. Semin Liver Dis 2016;36:123–126. doi: 10.1055/s-0036- patients with liver cirrhosis. Ann Gastroenterol 2016;29:297 306. doi: 10. 1583201. 20524/aog.2016.0037. [4] Lin S, Zhang K, Zhang J, Wang M, Velani B, Zhu Y. Long-term outcomes of [27] Alvarado S, Fernald RD, Storey KB, Szyf M. The dynamic nature of DNA patients with hepatitis B virus-related acute on chronic liver failure: An methylation: a role in response to social and seasonal variation. Integr observational cohort study. Liver Int 2019;39:854–860. doi: 10.1111/liv. Comp Biol 2014;54:68–76. doi: 10.1093/icb/icu034. 14072. [28] Gao S, Sun FK, Fan YC, Shi CH, Zhang ZH, Wang LY, et al. Aberrant GSTP1 [5] Anand AC, Dhiman RK. Acute on chronic liver failure-what is in a ‘definition’? promoter methylation predicts short-term prognosis in acute-on-chronic J Clin Exp Hepatol 2016;6:233–240. doi: 10.1016/j.jceh.2016.08.011. hepatitis B liver failure. Aliment Pharmacol Ther 2015;42:319–329. doi: [6] Fernández J, Acevedo J, Wiest R, Gustot T, Amoros A, Deulofeu C, et al. 10.1111/apt.13271. Bacterial and fungal infections in acute-on-chronic liver failure: prevalence, [29] Gao J, Sun Y, Lu Y, Li L. Impact of ambient humidity on child health: a sys- characteristics and impact on prognosis. Gut 2018;67:1870–1880. doi: 10. tematic review. PLoS One 2014;9:e112508. doi: 10.1371/journal.pone. 1136/gutjnl-2017-314240. 0112508. [7] Frias M, López-López P, Rivero A, Rivero-Juarez A. Role of hepatitis E virus [30] Verma N, Singh S, Taneja S, Duseja A, Singh V, Dhiman RK, et al. Invasive infection in acute-on-chronic liver failure. Biomed Res Int 2018;2018: fungal infections amongst patients with acute-on-chronic liver failure at high 9098535. doi: 10.1155/2018/9098535. risk for fungal infections. Liver Int 2019;39:503–513. doi: 10.1111/liv. [8] Cao Z, Liu Y, Wang S, Lu X, Yin S, Jiang S, et al. The impact of HBV flare on the 13981. outcome of HBV-related decompensated cirrhosis patients with bacterial [31] Mücke MM, Rumyantseva T, Mücke VT, Schwarzkopf K, Joshi S, Kempf VAJ, infection. Liver Int 2019;39:1943–1953. doi: 10.1111/liv.14176. et al. Bacterial infection-triggered acute-on-chronic liver failure is associated [9] Singal AK, Arora S, Wong RJ, Satapathy SK, Shah VH, Kuo YF, et al. Increas- with increased mortality. Liver Int 2018;38:645–653. doi: 10.1111/liv. ing burden of acute-on-chronic liver failure among alcohol-associated liver 13568.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 341–345 345 Review Article

Highlights for Dental Care as a Hepatitis C Risk Factor: A Review of Literature

Leon D. Averbukh* and George Y. Wu

Department of Medicine, Division of Gastroenterology-Hepatology, University of Connecticut Health Center, CT, USA

Abstract Europe, South America, and North America, genotype 4 found in the Middle East, Egypt, and central Africa, genotype 5 found Hepatitis C (HCV) is a viral infection that affects an estimated exclusively in South Africa, and genotype 6 found in South- 71 million people worldwide, with over 1 million new infec- east Asia.2 In the USA, 70% of all HCV infections are caused tions yearly. While medical treatments exist, HCV continues by genotype 1.3 to be a significant public health concern. Primary prevention HCV, initially known as the non-A, non-B hepatitis, was and transmission risk factor identification remain key in first discovered in the 1970s in patients that had previously helping decrease disease prevalence. While intravenous drug undergone blood transfusions.4 However, it was not until the use, healthcare exposure (i.e. blood transfusions and surgical late 1980s that the virus itself was formally identified. Since care), and body modification (i.e. tattooing and piercings) are the advent of interferon as a treatment for HCV in 1991, rapid well accepted risk factors for HCV transmission, others progress has been made in the production of both genotype- remain controversial. Because dental practice is often asso- specific and general therapies for HCV infection. Even with ciated with procedures and bleeding, the possibility of HCV rapidly evolving treatment options, HCV remains without a transmission seemed reasonable to investigate. Here, we preventative vaccine. Thus, HCV persists as a public health review the evidence for dental care as a potential risk factor burden and primary prevention is key. for HCV transmission. We identified a total of 1,180 manu- Many studies have explored the risk factors involved in scripts related to HCV and dental care, of which 26 manu- viral transmission, and at present data has correlated signifi- scripts were included in the study after exclusionary criteria cant transmission risk with intravenous drug use (IVDU), were applied. As per our review of the available literature, in healthcare exposure (i.e. blood transfusions and surgical the developing world, the improper use of sterile technique care), body modification (i.e. tattooing and piercings), as and lack of provider education likely increases the risk of HCV well as from infected mothers to their fetuses.5–8 Because transmission during dental care. In developed nations, on the dental practice is often associated with bleeding, the possibil- other hand, general dental care does not appear to be a ity of HCV transmission seemed reasonable to investigate. significant risk factor for HCV transmission in non-intravenous At present, few studies have investigated the overall risk of drug user patients; although, the improper use and reuse of dental care as a risk factor for HCV transmission. A review by anesthetics during procedures poses a rare potential risk for Mahboobi et al.9 identified a weak all-time risk of HBV and HCV viral transmission. infection during dental treatment.Forthisreview,weexamined Citation of this article: Averbukh LD, Wu GY. Highlights for literature from around the world describing investigations of dental care as a hepatitis C risk factor: A review of literature. dental care as a potential risk factor for HCV transmission. J Clin Transl Hepatol 2019;7(4):346–351. doi: 10.14218/ JCTH.2019.00023. Methods

We conducted a complete literature search using Medline/ Introduction PubMed, Scopus, and Google Scholar for case-control studies, investigative before and after studies, surveys, and observa- Hepatitis C virus (HCV) is a single-stranded RNA virus of the tional studies examining dental care as a risk factor for HCV Flaviviridae family that globally affects an estimated 71 transmission. Search terms used consisted of: risk factors, million people chronically, with nearly 1.75 million new infec- HCV, hepatitis C, oral surgery, transmission, infection, dental tions reported in 2015 alone.1 With variable infection rates care, and oral care. Search hits were limited to studies around the world, HCV has the highest prevalence in the published during or after the year 2000. The year 2000 was eastern Mediterranean, with 2.3% of the population selected as a starting point to limit studies to those with dental affected.1 The virus is subdivided into at least six major gen- care practices most similar to present day and to minimize otypes, with genotypes 1, 2, and 3 being the most common in historical confounding variables. Studies that analyzed co-infectivity with other viruses, such as the human immuno- Keywords: Hepatitis C; Oral care; Dental care; Risk factors; Hepatitis. deficiency virus, were excluded from analysis. On initial Abbreviations: HCV, hepatitis C; IVDU, intravenous drug use. literature search, 1,180 manuscripts were identified and sub- Received: 28 June 2019; Revised: 22 September 2019; Accepted: 8 October 2019 sequently narrowed to a final count of 34 manuscripts after *Correspondence to: Leon D. Averbukh, Department of Medicine, Division of search term integration. An additional eight manuscripts were Gastroenterology-Hepatology, University of Connecticut Health Center, 236 Farm- ington Ave., Farmington, CT 06030, USA. Tel: +1-347-306-4752, Fax: +1-860- removed after exclusionary criteria were applied. All studies 679-1025, E-mail: [email protected] were further subdivided into geographic regions of origin.

346 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 346–351

Results risk factors. This method of data collection has a high risk for social desirability bias, a subcategory of response bias We found 40% of the USA/European studies, 37.5% of the where survey respondents answer questions in a manner Middle Eastern studies, 33.3% of the African studies, 100% of that they believe will be viewed favorably by others. Given the South American studies, and 60% of the Asian studies to the cultural taboos of topics such as intravenous drug use have found clinically significant HCV transmission risk with (IVDU) and sexual history in many parts of the world, dental care. In total, 12 of the 26 studies that met the Medhat et al.’s30 data has a high risk for respondent bias described search criteria found an association between skew. The investigators themselves admit that given the cul- dental care and HCV infection.10–35 tural taboos of Pakistan, they avoided directly asking partic- Five of the studies queried nationwide data (generally ipants whether they practiced IVDU. As IVDU is a known risk those of European or American origin), twelve examined factor for HCV transmission, the potential for confounding urban populations, seven examined rural populations, and variable presence in the analyzed data is high. To offset the two investigated a mixed urban/rural population.9–34 Sixteen potential confounding effect of not knowing participant IVDU of the studies used adult populations in the sample pool (two history, Medhat et al.30 questioned use of shared syringes, a of which examined strictly adult pregnant women, one common practice in the region. Unfortunately, this does not queried only adults over the age of 60, and one study exclude IVDU as a confounding variable, as at present there looked exclusively at adult healthcare workers), six used a are no data showing whether the sharing of syringes within a mixed pool of adults and children, and four investigated only family nucleus is a risk factor for HCV transmission and/or if it children.9–34 Eighteen of the 26 studies did not define “dental correlates to transmission risks associated with IVDU. care”, and one clarified that all dental care was provided by Unfortunately, the lack of identification and exclusion of IVDU uncertified providers (Table 1).12,14–19,21,23–28,31,33–35 was a common shortcoming across studies, whether because of omission of participant questioning or non-differentiation Discussion of IVDU from the population pool. Only three of the studies either excluded participants who admitted to IVDU or per- The results of the reviewed studies were conflicting. Close to formed repeat statistical analyses accounting for IVDU as a 10,23,24 half of the studies showed dental care as a significant risk known variable. Dental care was found not be a signifi- factor for HCV transmission, while the rest of the studies cant risk factor for HCV transmission in two of the three studies, 10,23 10 showed no significant association. Furthermore, no world when IVDU was excluded. Murphy et al. demonstrated region proved to be uniform in study findings (Table 1).10–35 the confounding nature of IVDU, as when IV drug users were Findings and conclusions varied even within the same coun- excluded from the study population there was no significant tries. For instance, in Pakistan, three of seven studies found a difference between dental care and HCV transmission, while significant correlation between dental care and HCV transmis- the opposite was true when the same population was included. sion; in Egypt, one of five and in China, one of two.16–24,29–33 While it is unclear why IVDU is a confounding variable for HCV The studies examined do not appear to suffer from a lack of transmission in dental care, it is possible that because data is power as sample sizes were relatively large, involving hun- obtained by check box questionnaires, individuals who practice dreds, if not thousands of respondents. Limitations of study IVDU require more dental care on average than their non-IVDU design and the presence of confounding variables highlights counterparts due to drug side effects and lifestyle. how these studies may have resulted in inconsistent data. Another type of bias which may have limited the findings of the presented studies is Berkson bias. Berkson bias is a type of Study limitations selection bias generally found when study samples are taken from a subpopulation rather than the general population. As dental care was not the only variable examined as a Nationwide data for sample analysis was used in only five of 10–12,14,27 potential risk factor for HCV transmission in all reviewed the twenty-six studies (Table 1). Instead, most studies, dental care did not receive a high level of detail or studies either investigated specific environments, i.e. urban subcategorization and was rarely well-defined. Most studies or rural regions, or certain demographics, such as pregnant asked participants whether they had any kind of dental women, healthcare workers, or children. In addition to increas- care over the span of their respective lifetimes.12,16,31 Some ing the risk of Berkson bias with the aforementioned subpopu- studies did qualify the question of previous dental care with lations, the inclusion of children also increased the probability 33 specific procedures that defined “dental care”, but these were of social desirability bias. In Barakat et al., parents were ’ of a broad range of invasiveness, including tooth fillings, asked to answer questionnaires about their children s potential filings, extractions, and oral surgeries.22,29,32 As participants risk factors for HCV infection. Based on the child age groups were not asked to specify the type of dental work they had being studied, parents may have either refrained from being done, it opened the possibility of data skew given that the risk entirely truthful so as not paint themselves in a bad light given of HCV transmission is different in cases involving blood expo- cultural taboos or may have answered in a manner they sure, such as during root canal placement versus tooth filings. thought was culturally appropriate, even if they were not com- ’ Additionally, most articles did not define anesthesia use in pletely aware of all of their children sexposures. their definition of dental care, a potential confounding variable as the use of syringes and anesthetic solutions may Age as a clue increase HCV transmission risk. Other shortcomings of the evaluated studies were the An interesting finding noted in two studies28,29 was the signifi- methods for data collection. For instance, Medhat et al.30 cance of age in determining whether dental care was a risk went door to door, checking blood samples from local citizenry factor for HCV transmission. Habib et al.29 noted that with in Pakistan for HCV antibodies, while simultaneously asking adjustment for age, there was no increased risk for HCV trans- the volunteers questions regarding potential HCV exposure mission in dental care, while the opposite was true when no

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 346–351 347 4 ora fCiia n rnltoa Hepatology Translational and Clinical of Journal 348 Table 1. Studies investigating dental care as a variable for HCV transmission risk from the year 2000 to present

Significance of dental care for First author (year)Ref Country Setting Population Dental care type HCV transmission

Europe and US Murphy (2000)10 United Nationwide Adults Tooth extraction and Not significant States tooth cleaning (analyzed separately) Mele (2001)11 Italy Nationwide Adults Oral surgery Significant Karmochkine France Nationwide Adults Not defined Not significant (2006)12 Lionis (2000)13 Greece Rural Pregnant women Tooth extractions and Significant fillings Gheorghe (2010)14 Romania Nationwide Adults Not defined Not significant Asia Hajiani (2002)15 Iran Urban Adults Not defined Not significant Bari (2001)16 Pakistan Urban Adults Not defined Not significant Akhtar (2002)17 Pakistan Urban Adults & children Not defined Not significant Jafri (2006)18 Pakistan Urban Children Not defined Not significant Khan (2008)19 Pakistan Urban Pregnant women Not defined Significant Qureshi (2009)20 Pakistan Urban Adults Extract/ filing/ scaling Significant Janjua (2010)21 Pakistan Peri-urban Adults Note defined Significant Ahmed (2012)22 Pakistan Rural Adults Filling/ extraction/root Not significant

canal/ dental injections L.D. Averbukh Zhang (2009)23 China Urban Adults over 60 Not defined Not significant 2019 years old He (2011)24 China Rural Adults Not defined Significant o.7|346 | 7 vol. Nguyen (2007)25 Vietnam Rural Adults Not defined Not significant Davaalkham (2006)26 Mongolia Urban/Rural Children (7–12 Not defined Significant al et years old) etlcr sa C ikfactor risk HCV an as care Dental : Ruzibakiev (2001)27 Uzbekistan Nationwide Adults Not defined Significant – 351 Africa Ramarokoto (2008)28 Madagascar Urban Adults Not defined Significant Habib (2001)29 Egypt Rural Adults & children Filling, extraction, Not significant anesthesia, gum treatment Medhat (2002)30 Egypt Rural Adults & children Any, extraction, gum Not significant ** treatment, filling

(continued) Averbukh L.D. et al: Dental care as an HCV risk factor

such adjustment was made. Ramarokoto et al.28 found that the percentage of HCV seroprevalent individuals with a history of dental care, when categorized by age groups (0–24, 25–44, 45–64, and those over >64 years), varied, with those aged between 45 and 64 having the highest seroprevalence. This finding may be explained by the quality of sterile technique employed or available to the healthcare providers performing the dental procedures. Poor sterilization technique and unsafe injections, factors that would put patients at a higher potential risk of HCV transmission during dental work, Significant Not significant Significance of dental care for HCV transmission may have improved over time to the present, whether secondary to provider education and/or tool accessibility. As to why Ramarokoto et al.28 found peak HCV seroprevalence in the 45–64 age group with subsequent decrease, it may be due to survivorship bias, as those that were HCV-infected were less likely to survive to more advanced age.

The risks involved

In developing countries, the risks associated with HCV transmission during dental care are likely related to improper sterile technique as well as the use of poorly trained or legally practicing but uncertified practitioners.34 Studies in Not DefinedNot defined Not significant Not defined but donenon-certified by providers Significant anesthesia/ gum treatment the West have generally concluded that dental care is not a risk factor for HCV transmission (Table 1).10,12,14 However, the use of anesthesia during dental care, a practice more prevalent in developed nations, remains an important, though not well identified, potential source for HCV transmission. In the USA state of Oklahoma, in 2013, a vial of propo-

9 years 15 fol, initially used during a multitooth extraction from an – – HCV-seropositive patient was subsequently reused for a patient undergoing a similar procedure.36 Unlike general data currently available which identifies statistical association between HCV and transmission risk factors, the Oklahoma

old years old workers case represents proven HCV transmission, and is to date the only such example in the USA. While the re-use of anesthetic may increase the risk of HCV exposure, the type of anesthetic used may also contribute. Propofol, an injected anesthetic medication used for procedural sedation is lipid rich in content, a property that is beneficial for prolonged HCV particle survival. One study showed high viral particle load in propofol emulsion after 35 days, a duration over which viral particles became undetectable even in optimal cell medium.37 While studies investigating HCV transmission between physicians and patients are limited, it remains an important complicating factor in medical care. Exposure-prone procedures, such as surgery, have shown an association between HCV transmission from seropositive surgical providers to their patients. Studies have found the risk of patient acquisition of HCV during an operation performed by an infected surgeon to CountryEgyptEgypt SettingEgypt Urban Rural PopulationBrazilPeru Urban/Rural Children 1 Adults & children Children 6 Urban Dental care type Urban Filling/extraction/ bebetween Healthcare Adults & children 0 and 3.7%, Not defined a risk similar to thatof Significant acquiring HCV after a blood transfusion.38–40 One retrospective study investigating cardiac surgeons demonstrated a surgeon-to-patient HCVtransmissionrateofashighas6%duringopenheart surgery.41 The risks of HCV transmission from dental care pro- Ref

31 viders to their patients has not been well studied, and at present 35 32 33 ) 34 no data exists on dentist to patient HCV transmission rates. Unlike provider-to-patient HCV transmission, the rate of patient-to-surgeon HCV transmission appears to be signifi- continued ( cantly lower. In a British study investigating surgical interventions and HCV transmission, rates of HCV infection amongst surgeons ranged between 0.035% and 1.12% Risk assessment done for each type of dental intervention individually. First author (year) Sanchez (2000) Mostafa (2009) South America Parana (2007) Barakat (2010) El-Raziky (2007) 42 Table 1. ** over the span of a 35-year professional career. The data

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 346–351 349 Averbukh L.D. et al: Dental care as an HCV risk factor on transmission of HCV from patients to dentist remains limited. References Examining USA data from between the years of 2000 and 2005, the prevalence of HCV antibodies among the general USA pop- [1] World Health Organization. Hepatitis C. Available from: https://www.who. ulation ranged between 1.3% and 1.6%, while being only int/news-room/fact-sheets/detail/hepatitis-c, Accessed 9 July 2019. [2] Nainan OV, Alter MJ, Kruszon-Moran D, Gao FX, Xia G, McQuillan G, et al. 0.47% amongst dentists, signifying a low occupational risk of Hepatitis C virus genotypes and viral concentrations in participants of a 43 HCV infection amongst USA dentists. general population survey in the United States. Gastroenterology 2006; While HCV exposure during cardiac surgery is significantly 131:478–484. doi: 10.1053/j.gastro.2006.06.007. greater than during dental care, the identification of viral [3] Nguyen MH, Keeffe EB. Prevalence and treatment of hepatitis C virus genotypes 4, 5, and 6. Clin Gastroenterol Hepatol 2005;3:S97–S101. doi: 10. particles in human blood and saliva reported in cases of HCV 1016/S1542-3565(05)00711-1. transmission following splashes of infected blood on broken [4] Alter H. Discovery of non-A, non-B hepatitis and identification of its etiology. skin, toothbrush sharing, and viral infectivity makes dental care Am J Med 1999;107:16S–20S. doi: 10.1016/s0002-9343(99)00375-7. an important issue. Identification and reduction of potential [5] Xia X, Luo J, Bai J, Yu R. Epidemiology of hepatitis C virus infection among injection drug users in China: systematic review and meta-analysis. Public HCV transmission risk factors during regular and interventional – 44–47 Health 2008;122:990 1003. doi: 10.1016/j.puhe.2008.01.014. care remains crucial for both patients and providers. [6] Wilkins T, Malcolm JK, Raina D, Schade RR. Hepatitis C: diagnosis and treatment. Am Fam Physician 2010;81:1351–1357. Conclusions [7] Jafari S, Copes R, Baharlou S, Etminan M, Buxton J. Tattooing and the risk of transmission of hepatitis C: a systematic review and meta-analysis. Int J Infect Dis 2010;14:e928–e940. doi: 10.1016/j.ijid.2010.03.019. At present, dental care does pose a risk for HCV transmission, [8] Alter MJ. Epidemiology of hepatitis C virus infection. World J Gastroenterol although the extents vary worldwide. In the developing world, 2007;13:2436–2441. doi: 10.3748/wjg.v13.i17.2436. the improper use of sterile technique and lack of provider [9] Mahboobi N, Porter SR, Karayiannis P, Alavian SM. Dental treatment as a risk education likely increase the risk of HCV transmission secon- factor for hepatitis B and C viral infection. A review of the recent literature. J Gastrointestin Liver Dis 2013;22:79–86. dary to dental care. In developed nations, general dental care [10] Murphy EL, Bryzman SM, Glynn SA, Ameti DI, Thomson RA, Williams AE, does not appear to be a significant risk factor for HCV trans- et al. Risk factors for hepatitis C virus infection in United States blood mission when IVDU is an excluded variable. However, anes- donors. NHLBI Retrovirus Epidemiology Donor Study (REDS) Hepatology – thesia, more commonly used in developed nations, appears to 2000;31:756 762. doi: 10.1002/hep.510310329. [11] Mele A, Spada E, Sagliocca L, Ragni P, Tosti ME, Gallo G, et al. Risk of parent- be a potential risk for viral transmission and should be closely erally transmitted hepatitis following exposure to surgery or other invasive monitored. Unfortunately, current studies on dental HCV trans- procedures: results from the hepatitis surveillance system in Italy. J Hepatol mission contain limitations, including methods of data procure- 2001;35:284–289. doi: 10.1016/s0168-8278(01)00111-8. ment, lack of IVDU exclusion, Berkson bias, and social [12] Karmochkine M, Carrat F, Dos Santos O, Cacoub P, Raguin G. A case-control study of risk factors for hepatitis C infection in patients with unexplained desirability bias, which make their findings more difficult to routes of infection. J Viral Hepat 2006;13:775–782. doi: 10.1111/j.1365- interpret accurately. In the future, studies should focus on 2893.2006.00742.x. large, nationwide samples in which participants with known [13] Lionis C, Vlachonikolis IG, Skliros S, Symeonidis A, Merkouris BP, Kourouma- risk factors for HCV transmission should be excluded. Ques- lis E. Do undefined sources of hepatitis C transmission exist? The Greek study in General Practice. J Viral Hepat 2000;7:218–224. doi: 10.1046/j. tionnaires regarding risk factors should be done using methods 1365-2893.2000.00217.x. that best preserve participant anonymity to reduce the risk of [14] Gheorghe L, Csiki IE, Iacob S, Gheorghe C, Smira G, Regep L. The prevalence social desirability bias. While treatment options for HCV con- and risk factors of hepatitis C virus infection in adult population in Romania: a tinue to evolve and improve, primary prevention remains nationwide survey 2006-2008. J Gastrointestin Liver Dis 2010;19:373–379. paramount for disease eradication. [15] Hajiani E, Masjedizadeh R, Hashemi J, Azmi M, Rajabi T. Hepatis C virus transmission and its risk factors within families of patients infected with hep- Standard precautions for dental care remain key in reducing atitis C virus in southern Iran: Khuzestan. World J Gastroenterol 2006;12: the risk of HCV transmission to patients and providers alike. 7025–7028. doi: 10.3748/wjg.v12.i43.7025. These precautions include hand hygiene, proper use of pro- [16] Bari A, Akhtar S, Rahbar MH, Luby SP. Risk factors for hepatitis C virus infec- tective equipment, such as masks and gloves, proper sharps’ tion in male adults in Rawalpindi-Islamabad, Pakistan. Trop Med Int Health 2001;6:732–738. doi: 10.1046/j.1365-3156.2001.00779.x. handling and injection practices, environmental cleaning, [17] Akhtar S, Moatter T, Azam SI, Rahbar MH, Adil S. Prevalence and risk factors proper use of reusable and single-use medical equipment for intrafamilial transmission of hepatitis C virus in Karachi, Pakistan. J Viral and drugs, and respiratory hygiene.48 Special care should be Hepat 2002;9:309–314. doi: 10.1046/j.1365-2893.2002.00350.x. taken when treating high-risk populations, such as prison [18] Jafri W, Jafri N, Yakoob J, Islam M, Tirmizi SF, Jafar T, et al. Hepatitis B and C: prevalence and risk factors associated with seropositivity among children in inmates and those with human immunodeficiency virus, as Karachi, Pakistan. BMC Infect Dis 2006;6:101. doi: 10.1186/1471-2334-6-101. these populations have higher rates of HCV infection. [19] Khan UR, Janjua NZ, Akhtar S, Hatcher J. Case-control study of risk factors associated with hepatitis C virus infection among pregnant women in hospitals of Karachi-Pakistan. Trop Med Int Health 2008;13:754–761. doi: 10. Acknowledgments 1111/j.1365-3156.2008.02075.x. [20] Qureshi H, Arif A, Riaz K, Alam SE, Ahmed W, Mujeeb SA. Determination of The support of the Herman Lopata Chair for Hepatitis risk factors for hepatitis B and C in male patients suffering from chronic hepatitis. BMC Res Notes 2009;2:212. doi: 10.1186/1756-0500-2-212. Research is gratefully acknowledged. [21] Janjua NZ, Hamza HB, Islam M, Tirmizi SF, Siddiqui A, Jafri W, et al. Health care risk factors among women and personal behaviours among men explain Conflict of interest the high prevalence of hepatitis C virus infection in Karachi, Pakistan. J Viral Hepat 2010;17:317–326. doi: 10.1111/j.1365-2893.2009.01230.x. [22] Ahmed F, Irving WL, Anwar M, Myles P, Neal KR. Prevalence and risk factors The authors have no conflict of interests related to this for hepatitis C virus infection in Kech District, Balochistan, Pakistan: most publication. infections remain unexplained. A cross-sectional study. Epidemiol Infect 2012;140:716–723. doi: 10.1017/S0950268811001087. [23] Zhang M, Fan J, Li H, Cui J, Qiao Y, Sung J, et al. Alternative risk factors of Author contributions HCV infection in a rural community in China. Epidemiol Infect 2010;138: 1032–1035. doi: 10.1017/S0950268809991269. [24] He Y, Zhang J, Zhong L, Chen X, Liu HM, Wan LK, et al. Prevalence of and risk Writing the manuscript (LDA), developing the idea for the factors for hepatitis C virus infection among blood donors in Chengdu, China. article and critically revising it (GYW). J Med Virol 2011;83:616–621. doi: 10.1002/jmv.22010.

350 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 346–351 Averbukh L.D. et al: Dental care as an HCV risk factor

[25] Nguyen VT, McLaws ML, Dore GJ. Prevalence and risk factors for hepatitis C [37] Steinmann E, Ciesek S, Friesland M, Erichsen TJ, Pietschmann T. Prolonged infection in rural north Vietnam. Hepatol Int 2007;1:387–393. doi: 10. survival of hepatitis C virus in the anesthetic propofol. Clin Infect Dis 2011; 1007/s12072-007-9008-3. 53:963–964. doi: 10.1093/cid/cir530. [26] Davaalkham D, Ojima T, Nymadawa P, Uehara R, Watanabe M, Oki I, et al. [38] Gunson RN, Shouval D, Roggendorf M, Zaaijer H, Nicholas H, Holzmann H, Prevalence and risk factors for hepatitis C virus infection in Mongolian chil- et al. Hepatitis B virus (HBV) and hepatitis C virus (HCV) infections in health dren: Findings from a nationwide survey. J Med Virol 2006;78:466–472. doi: care workers (HCWs): guidelines for prevention of transmission of HBV and 10.1002/jmv.20563. HCV from HCW to patients. J Clin Virol 2003;27:213–230. doi: 10. [27] Ruzibakiev R, Kato H, Ueda R, Yuldasheva N, Hegay T, Avazova D, et al. Risk 1016/S1386-6532(03)00087-8. factors and seroprevalence of hepatitis B virus, hepatitis C virus, and human [39] Olsen K, Dahl PE, Paulssen EJ, Husebekk A, Widell A, Busund R. Increased immunodeficiency virus infection in uzbekistan. Intervirology 2001;44:327– risk of transmission of hepatitis C in open heart surgery compared with vas- 332. doi: 10.1159/000050066. cular and pulmonary surgery. Ann Thorac Surg 2010;90:1425–1431. doi: [28] Ramarokoto CE, Rakotomanana F, Ratsitorahina M, Raharimanga V, Razafin- 10.1016/j.athoracsur.2010.06.053. dratsimandresy R, Randremanana R, et al. Seroprevalence of hepatitis C and [40] Ross RS, Viazov S, Roggendorf M. Risk of hepatitis C transmission from associated risk factors in urban areas of Antananarivo, Madagascar. BMC infected medical staff to patients: model-based calculations for surgical set- Infect Dis 2008;8:25. doi: 10.1186/1471-2334-8-25. tings. Arch Intern Med 2000;160:2313–2316. doi: 10.1001/archinte.160. [29] Habib M, Mohamed MK, Abdel-Aziz F, Magder LS, Abdel-Hamid M, Gamil F, 15.2313. et al. Hepatitis C virus infection in a community in the Nile Delta: risk factors [41] Puro V, De Carli G, Scognamiglio P, Porcasi R, Ippolito G. Risk of HIV and for seropositivity. Hepatology 2001;33:248–253. doi: 10.1053/jhep.2001. other blood-borne infections in the cardiac setting: patient-to-provider and 20797. provider-to-patient transmission. Ann N Y Acad Sci 2001;946:291–309. doi: [30] Medhat A, Shehata M, Magder LS, Mikhail N, Abdel-Baki L, Nafeh M, et al. 10.1111/j.1749-6632.2001.tb03918.x. Hepatitis c in a community in Upper Egypt: risk factors for infection. Am J [42] Thorburn D, Roy K, Cameron SO, Johnston J, Hutchinson S, McCruden EA, Trop Med Hyg 2002;66:633–638. doi: 10.4269/ajtmh.2002.66.633. et al. Risk of hepatitis C virus transmission from patients to surgeons: model [31] El-Raziky MS, El-Hawary M, Esmat G, Abouzied AM, El-Koofy N, Mohsen N, based on an unlinked anonymous study of hepatitis C virus prevalence in et al. Prevalence and risk factors of asymptomatic hepatitis C virus infection hospital patients in Glasgow. Gut 2003;52:1333–1338. doi: 10.1136/gut. in Egyptian children. World J Gastroenterol 2007;13:1828–1832. doi: 10. 52.9.1333. 3748/wjg.v13.i12.1828. [43] Cleveland JL, Gray SK, Harte JA, Robison VA, Moorman AC, Gooch BF. Trans- [32] Mostafa A, Taylor SM, el-Daly M, el-Hoseiny M, Bakr I, Arafa N, et al.Isthe mission of blood-borne pathogens in US dental health care settings: 2016 hepatitis C virus epidemic over in Egypt? Incidence and risk factors of new update. J Am Dent Assoc 2016;147:729–738. doi: 10.1016/j.adaj.2016.03. hepatitis C virus infections. Liver Int 2010;30:560–566. doi: 10.1111/j. 020. 1478-3231.2009.02204.x. [44] Dusheiko GM, Smith M, Scheuer PJ. Hepatitis C virus transmitted by human [33] Barakat SH, El-Bashir N. Hepatitis C virus infection among healthy Egyptian bite. Lancet 1990;336:503–504. doi: 10.1016/0140-6736(90)92049-n. children: prevalence and risk factors. J Viral Hepat 2011;18:779–784. doi: [45] Toda T, Mitsui T, Tsukamoto Y, Ebara T, Hirose A, Masuko K, et al. Molecular 10.1111/j.1365-2893.2010.01381.x. analysis of transmission of hepatitis C virus in a nurse who acquired acute [34] Paraná R, Paiva T, Leite MR, Oliveira FN, Kali N, Lobato C, et al. Infection with hepatitis C after caring for a viremic patient with epistaxis. J Med Virol 2009; hepatitis C virus among health care workers in the Brazilian Western Amazon 81:1363–1370. doi: 10.1002/jmv.21537. region (Rio Branco, State of Acre). Am J Trop Med Hyg 2007;76:165–169. [46] Lock G, Dirscherl M, Obermeier F, Gelbmann CM, Hellerbrand C, Knöll A, et al. doi: 10.4269/ajtmh.2007.76.165. Hepatitis C - contamination of toothbrushes: myth or reality? J Viral Hepat [35] Sanchez JL, Sjogren MH, Callahan JD, Watts DM, Lucas C, Abdel-Hamid M, 2006;13:571–573. doi: 10.1111/j.1365-2893.2006.00735.x. et al. Hepatitis C in Peru: risk factors for infection, potential iatrogenic trans- [47] Recommendations for prevention and control of hepatitis C. virus (HCV) mission, and genotype distribution. Am J Trop Med Hyg 2000;63:242–248. infection and HCV-related chronic disease. Centers for Disease Control and doi: 10.4269/ajtmh.2000.63.242. Prevention. MMWR Recomm Rep 1998;47:1–39. doi: 10.1037/e547602006- [36] Bradley KK. Dental healthcare-associated transmission of hepatitis C: 001. Final report of public health investigation and response, 2013. Available [48] Klevens RM, Moorman AC. Hepatitis C virus: an overview for dental health from: https://www.ok.gov/health2/documents/Dental%20Healthcare_ care providers. J Am Dent Assoc 2013;144:1340–1347. doi: 10.14219/jada. Final%20Report_2_17_15.pdf. archive.2013.0069.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 346–351 351 Review Article

The Direct Contribution of Astrocytes and Microglia to the Pathogenesis of Hepatic Encephalopathy

Victoria Jaeger1, Sharon DeMorrow2,3,4,5 and Matthew McMillin*2,3,4

1Baylor Scott & White Health, Department of Internal Medicine, Temple, TX, USA; 2Texas A&M University Health Science Center, Department of Medical Physiology, Temple, TX, USA; 3Central Texas Veterans Health Care System, Temple, TX, USA; 4University of Texas at Austin, Dell Medical School, Department of Internal Medicine, Austin, TX, USA; 5University of Texas at Austin, College of Pharmacy, Austin, TX, USA

Abstract to as hepatic encephalopathy (HE). HE manifests in several different ways and while the severity of symptoms is Hepatic encephalopathy is a neurological complication result- variable, patients often present with altered mental status, ing from loss of hepatic function and is associated with poor disturbances in sleep-wake cycles and functional impair- clinical outcomes. During acute liver failure over 20% of ment.1,2 Hospitalizations due to HE in the USA increased mortality can be associated with the development of hepatic over the period of 2010 to 2014, from 25,039 to 31,182, encephalopathy. In patients with liver cirrhosis, 1-year sur- though in-hospital mortality decreased from 13.4% to vival for those that develop overt hepatic encephalopathy is 12.3%.3 In addition to this, HE in-patient charges increased under 50%. The pathogenesis of hepatic encephalopathy is from 8.15 billion USD to 11.9 billion USD over the same time complicated due to the multiple disruptions in homeostasis period.3 Therefore, it is evident that HE is a significant finan- that occur following a reduction in liver function. Of these, cial burden on health care systems and the development of elevations of ammonia and neuroinflammation have been this syndrome is associated with poor patient prognosis and shown to play a significant contributing role to the develop- mortality. ment of hepatic encephalopathy. Disruption of the urea cycle The current understanding of HE pathogenesis is that following liver dysfunction leads to elevations of circulating disruption of hepatic function leads to increased ammonia ammonia, which enter the brain and disrupt the functioning of concentrations and impaired urea metabolism, which allows astrocytes. This results in dysregulation of metabolic path- ammonia to accumulate in the blood and the brain. The ways in astrocytes, oxidative stress and cerebral edema. downstream effects of ammonia accumulation in the brain Besides ammonia, circulating chemokines and cytokines are are astrocyte swelling and dysregulated neurotransmission, increased following liver injury, leading to activation of micro- which lead to the cognitive deficits present during HE. In glia and a subsequent neuroinflammatory response. The addition to this, systemic inflammation, elevation of pro- combination of astrocyte dysfunction and microglia activation inflammatory cytokines and microglia activation have been are significant contributing factors to the pathogenesis of observed during HE in both patients and rodent models and hepatic encephalopathy. are associated with worse outcomes. Current treatments are Citation of this article: Jaeger V, DeMorrow S, McMillin M. primarily focused on ammonia-lowering agents, optimization The direct contribution of astrocytes and microglia to the of nutrition and treatment of comorbid disease processes. pathogenesis of hepatic encephalopathy. J Clin Transl Hepatol While these have been shown to have some efficacy, deter- 2019;7(4):352–361. doi: 10.14218/JCTH.2019.00025. mining all of the pathophysiological mechanisms that contribute to HE progression is important if novel therapeutic approaches are to be developed. In this review, we will discuss the clinical definitions and Introduction description of HE, further examine the available research regarding the pathophysiology of HE within the central Liver disease is a significant cause of morbidity and mortality nervous system (CNS), and more specifically discuss the worldwide. One of the most severe consequences of acute role of astrocytes and microglia within the context of HE liver failure and decompensated chronic liver disease is the pathology. development of neurological complications, a state referred

Keywords: Acute liver failure; Microglia; Neuroinflammation; Ammonia; Clinical epidemiology and description of HE Astroglia. Abbreviations: AQP-4, aquaporin-4; BDL, bile duct-ligated; CCL2, chemokine The neurological and cognitive manifestations that present in ligand 2; CNS, central nervous system; CX3CL1, C-X3-C ligand 1; GFAP, glial patients with liver disease or portosystemic shunting in the fibrillary acidic protein; HE, hepatic encephalopathy; IL, interleukin; IBA1, 4 ionized calcium binding adaptor molecule 1; LPS, lipopolysaccharide; TGFb1, absence of liver disease is termed HE. Two categories of HE transforming growth factor beta 1; TGFbR2, transforming growth factor beta exist: covert HE (also known as minimal HE) and overt HE; receptor 2; TNFa, tumor necrosis factor alpha; WHC, West Haven Criteria. these are distinguished by the severity of the patient presen- Received: 5 July 2019; Revised: 7 October 2019; Accepted: 24 October 2019 tation.5 Patients with minimal HE have abnormal results with *Correspondence to: Matthew McMillin, University of Texas at Austin Dell Medical School, 1601 Trinity Street, Building B, Austin, TX 78701, USA. Tel: +1-512-495- psychometric testing and subtle personality changes that 5037, Fax: +1-512-495-5839, E-mail: [email protected] may be evident to caregivers or by those familiar with the

352 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 352–361

Copyright: © 2019 Authors. This article has been published under the terms of Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0), which permits noncommercial unrestricted use, distribution, and reproduction in any medium, provided that the following statement is provided. “This article has been published in Journal of Clinical and Translational Hepatology at DOI: 10.14218/JCTH.2019.00025 and can also be viewed on the Journal’s website at http://www.jcthnet.com”. Jaeger V. et al: Astrocytes and microglia in encephalopathy patient. Individuals with minimal HE can have a reduced When hepatocytes are damaged as a result of liver ability to drive, impairment at work, and inability or difficulty disease, detoxification of ammonia into urea is impaired and – performing other complex tasks.6 8 Diagnosing minimal HE is portosystemic shunting leads to ammonia accumulation in important clinically because it best predicts future diagnosis the blood. In addition, dysbiosis of the microbiome has been of overt HE and is associated with higher rates of hospital- reported during cirrhosis, leading to increased urase activity 6,9 ization and lower survival rates. Overt HE is more evident and even greater production of ammonia in the gut.14 These upon clinical presentation, with increased severity of symp- combined effects lead to a significant increase of circulating toms, but is diagnosed only after exclusion of other causes of ammonia concentrations, which are then metabolized by altered mental status. muscle and other extrahepatic organs. Ammonia also can The West Haven Criteria (WHC) have traditionally been cross the blood-brain barrier, generating a state of hyperam- used for grading the severity of overt HE based upon clinical monemia in the brain. It is not clear whether the increased findings, with higher WHC grades being associated with more severe HE. The mortality rate is high for overt HE patients concentrations of ammonia found in the brain and cerebrospi- regardless of grade. During acute liver failure, 20–25% of nal fluid are due to active or passive transport but increased mortality results from increased intracranial pressure and the systemic levels of ammonia are associated with increased 15 development of HE.10 In cirrhotic patients, survival rates are ammonia concentrations in the brain. The highest eleva- estimated to be 42% at 1 year and 23% at 3 years.11 The tions of serum ammonia following liver dysfunction are incidence of overt HE is 8% annually in decompensated cir- observed in Type A and Type B HE.16,17 Arterial and venous rhotic patients12 and is estimated to be attributed to 0.33% of ammonia have been reported to be elevated during Type C HE hospitalizations in the USA.9 compared to controls, though the values observed are lower The American Association for the Study of Liver Diseases than the other types of HE, including in patients with acute- (commonly known as AASLD) together with the European on-chronic liver failure.16 Association for the Study of the Liver (commonly known as Besides ammonia, acute liver injury or chronic liver EASL) established guidelines in 2014 consisting of four axes disease is associated with inflammation that contributes to for better categorization and management of HE in the setting this syndrome, with neuroinflammation being a pathological 6 of chronic liver disease. These guidelines are focused on eti- contributor to overt HE during both acute liver failure and ology, severity, time course, and inciting factors. Axis I is chronic liver disease. Elevation of circulating pro-inflamma- categorized via etiology as Type A (acute liver failure), Type tory cytokines has been observed during acute liver failure, B (portosystemic shunting), or Type C (cirrhosis). Axis II is – cirrhosis, and acute-on-chronic liver failure.18 20 Associated determined by clinical severity of symptoms and is catego- with increased systemic inflammation, is the development of rized as unimpaired (normal mentation), WHC Grade I neuroinflammation and HE. In the brain, neuroinflammation (abnormal psychometric testing without apparent mental status changes), WHC Grade II (evidence of disorientation is regulated by microglia, the resident macrophage-like cells or asterixis), WHC Grade III (gross disorientation, somno- of the brain. While normally present in a quiescent state, in lence), and WHC Grade IV (coma).4,6 Axis III is based on response to pro-inflammatory signals, they can become acti- time of the clinical course and is categorized as episodic vated, producing cytokines generating oxidative stress and (more than 6 months between episodes), recurrent (episodes contributing to aspects of pathology, and activation of micro- occurring within 6 months), or persistent (always present).6,9 glia is observed in patients with both Type A and Type C HE.21 Axis IV is categorized as spontaneous or precipitated. Docu- In addition to increased ammonia and inflammation are a mentation of a patient with HE should include all four axis variety of other dysregulated metabolites that can contribute components and can aid clinicians in consistent categoriza- to this syndrome. The concentrations of various amino acids tion of this patient population. have been shown to be outside the normal range when assessed in plasma, cerebrospinal fluid and frontal cortex of patients with Grade 3 or 4 HE.22 Similar findings have been Pathogenesis of HE observed for bile acids, where concentrations are increased in serum, cerebrospinal fluid and tissue from patients with ful- The pathogenesis of HE is complex and the exact disease minant hepatic failure.23 Manganese is normally excreted mechanisms have yet to be fully delineated. The most from the body through the hepatobiliary system but during common theory is that the manifestations of HE are due to cirrhosis has been shown to accumulate in both the circulation alteration of the urea cycle and the resulting increase of 4 and brain, and can inhibit dopaminergic neurotransmission in systemic ammonia levels. Glutamine is used by the gut as a the basal ganglia.24–26 primary energy source and the resulting by-products of its In addition to this, a variety of comorbidities, including metabolism by glutaminase are glutamate and ammonia. In infections, gastrointestinal bleeding, diarrhea, hyponatremia, addition, bacteria in the gastrointestinal tract use urase to hypokalemia, hyperkalemia, benzodiazepine and diuretic produce ammonia from urea. Ammonia from the gut diffuses use, and others contribute to this syndrome.27 These comor- through the intestinal mucosa, enters the mesenteric circulation and is transported to the liver. In the liver, periportal bidities are most often present in patients with liver cirrhosis. hepatocytes detoxify ammonia via the ornithine cycle, result- In acute-on-chronic liver failure patients, elevated creatinine, ing in the production of urea. Urea passes into the systemic as well as altered jugular venous oxygen saturation ranges, 28 circulation via the hepatic veins, where it enters the kidney were associated with higher grades of HE. This gives and is ultimately excreted in urine. Ammonia that does not support that multiorgan complications during cirrhosis and enter the urea cycle is metabolized by glutamine synthetase acute-on-chronic liver failure can exacerbate the develop- in the liver, kidney, muscle and brain to produce glutamine ment of HE. A diagram of the complex interactions that are from glutamate.13 involved in the pathophysiology of HE is provided in Fig. 1.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 352–361 353 Jaeger V. et al: Astrocytes and microglia in encephalopathy

Fig. 1. Pathophysiology of HE. Following the development of liver failure or decompensated liver cirrhosis, there is an increase of circulating bile acids, amino acids, serum ammonia, and toxic metabolites as well as an increase of systemic inflammation. The systemic elevation of these factors leads to an increase of their concentrations in the brain, leading to metabolic and oxidative stress as well as increased neuroinflammation. Elevation of circulating cytokines and chemokines is associated with increased activation of microglia in the brain and subsequent neuroinflammation. The increase of neuroinflammation as well as the metabolic and oxidative stress present under these conditions promote the development of HE.

Astrocyte dysfunction GFAP has been shown to be important for cell volume regulation and therefore could contribute to the development of Astrocytes are the most abundant glial cell in the CNS and are cerebral edema during HE. This has been demonstrated by responsible for trophic and metabolic support of neurons and the presence of increased brain water content when there neuronal-glial communication, and are essential for blood- was reduced GFAP protein levels during HE.32 A proposed brain barrier function.29 In response to increased ammonia mechanism for the down-regulation of GFAP during HE is concentrations during HE, astrocytes undergo a change in that ammonia can interfere with metabolism, leading to morphology, developing large pale nuclei known as Alzheimer decreased ATP levels and thus reducing GFAP production.36 Type II astrocytosis.30 A large number of Alzheimer Type II In support of ammonia inhibiting GFAP expression, cultured astrocytic nuclei have been detected by autoradiography in astrocytes treated with ammonium chloride showed an 85% hyperammonemic rats induced by methionine sulfoxamine.30 reduction in GFAP mRNA.37 GFAP expression has also been Hyperammonemia also leads to increased production of glu- shown to be reduced during chronic liver disease and tamine in astrocytes, resulting in increased osmotic pressure HE.38,39 Loss of GFAP in chronic hepatitis was found to be via astrocyte swelling. Swelling of astrocytes can lead to cer- related to astrocyte swelling, with subsequent loss of their ebral edema and a worsening of HE pathology. Research to stellate shape.40 Similarly, in bile duct-ligated (BDL) rats, date involving astrocytes has primarily focused on the effects altered astrocytes have been found but with no changes in of HE on regulation of oxidative stress, glutamine and gluta- the total number of astrocytes themselves.41 mine synthase, brain water balance, and lactate metabolism. Studies have shown a direct relationship between astrocyte swelling and oxidative stress with p47phox-dependent Glial fibrillary acidic protein (GFAP) and oxidative activation of NADPH oxidase, contributing to reactive oxygen stress species generation from ammonia, in cultured rat astrocytes and cortical mouse brain slices.42 Oxidative and nitrosative GFAP is an intermediate filament found in astrocytes that is stress can lead to an induction of astrocyte swelling. used for both motility and maintenance of structural stabil- However, recent studies have shown the reverse is also ity.31 Multiple studies have shown that GFAP is decreased true, as astrocyte swelling induces oxidative and nitrosative during acute liver failure, thereby influencing the morphology stress—creating a self-amplifying cycle.43 Hyperammonemia and function of astrocytes.32,33 Investigating individual brain can lead to both oxidative and nitrosative stress, causing regions found that GFAP expression was greatly decreased in protein modifications such as nitration of tyrosine residues the corpus callosum34 and in the hippocampus during HE.35 along with oxidation of RNA.44 That being said, oxidative

354 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 352–361 Jaeger V. et al: Astrocytes and microglia in encephalopathy stress, astrocyte swelling and down-regulation of GFAP likely pore opening, subsequent oxidative stress, and further astro- contribute together to induce HE symptomology. Treatments cyte swelling.56,57 aimed at reducing oxidative stress in astrocytes, such as treating cells with curcumin, reduce mitochondrial dysfunc- Aquaporin-4 (AQP-4) tion and lead to increased GFAP levels.45 Therefore, many of the studies that are tying oxidative stress to changes in astro- AQP-4 is a bidirectional transmembrane water channel cyte swelling in the context of HE could be a result of the protein found on astrocyte end-feet and it plays a role in down-regulation of GFAP due to oxidative stress. maintaining brain water homeostasis.58,59 Direct evidence for a causal role of AQP-4 in brain edema has been demonstrated 60 Glutamine & glutamine synthetase in AQP-4 knockout mice. Studies have also shown the time course of cerebral edema development differs among brain 59 Currently, astrocytes are considered the primary target of regions, reflecting differences in AQP-4 distribution. Simi- ammonia toxicity because they are responsible for ammonia larly, silencing AQP-4 gene expression in cultured astrocytes metabolism in the CNS via glutamine synthetase.46 Ammonia was shown to reduce water permeability under hypoosmotic 61 can freely cross the blood-brain barrier as perivascular pro- conditions. cesses of astrocytes rapidly metabolize ammonia from the AQP-4 is dysregulated during acute liver failure and HE. circulation to prevent neurotoxicity using glutamine synthe- Acute liver failure is associated with up-regulation of AQP-4, 62,63 tase.47 Currently, few reports describe investigations into the as evidenced in thioacetamide-treated rats as well as 64 role of glutamine synthetase on progression of HE due to during galactosamine and LPS-induced liver failure. This acute liver failure. phenomenon is also present in patients, as there was an up- One study using the acute liver failure model of lipopoly- regulation of AQP-4 expression in the cerebral cortex from 8 33 saccharide (LPS) and D-galactosamine injection determined patients with acute liver failure. In thioacetamide-treated that administration of methionine sulfoximine, a glutamine rats, AQP-4 expression was positively correlated with brain 63 synthetase inhibitor, led to improved survival and reduced water content. Exposure of rat cortical astrocytes cultures plasma tumor necrosis factor alpha (TNFa) and interferon to interleukin (IL)-1b but not ammonia resulted in up-regu- gamma levels.48 However, the effect of chronic HE on gluta- lation of AQP-4, showing that brain edema may be worsened 65 mine synthetase has been more extensively studied and the by inflammation. results are inconsistent with each other. After 4-week porta- During chronic liver disease it was thought that cerebral caval anastomosis, glutamine synthetase activity was signifi- edema was not a pathological characteristic of HE, as cantly decreased in the hippocampus, cerebellum, and increased intracranial pressure is rare in this patient popula- cerebral cortex but unchanged in other brain regions.49 Con- tion. In addition, there is a decrease in grey matter and an 66 sistent with these findings, portacaval anastomosis rats have increase in white matter in patients with cirrhosis, and low a 15% decrease in glutamine synthetase activity in the cere- grade cerebral edema has been observed in patients with cir- 67 bral cortex.50 In addition, portacaval anastomosis rats at 6 rhosis and minimal HE. In 4-week BDL rats, there is an months had regional variation of glutamine synthetase activ- increase of AQP-4 expression in the cortex, hippocampus, ity as well as a strong increase in glutamine synthetase in striatum and cerebellum that was associated with increased 68 astrocyte end-feet processes.51 This regional variation was water content in all brain regions. This was validated in 4- also shown in BDL rats 30 days after surgery, where gluta- week BDL rats, that also had elevations of AQP-4 in the 69 mine synthetase activity was significantly decreased in the cortex. Interestingly, these researchers also used a galac- liver but there was no change in the frontal cortex.52 In addi- tosamine and high ammonia diet model in rats where AQP-4 tion, the administration of ornithine phenylacetate to reduce was not changed following LPS injection but brain water per- 69 ammonia concentrations significantly reduced glutamine syn- centage was significantly increased. Therefore, there may thetase activity in the frontal cortex in BDL rats, indicating be other factors outside AQP-4 that need to be better charac- that treatments aimed at reducing ammonia may influence terized to understand the control of brain water content glutamine or glutamine synthetase activity.52 In post- during chronic liver disease. mortem cortical brain tissue from cirrhotic patients with and without HE, glutamine synthase activity was significantly Lactate decreased whereas glutamine synthetase protein expression was not affected, which the authors surmise was a conse- Lactate is made in every cell from pyruvate, following a quence of tyrosine nitration of the enzyme.53 This is sup- reaction catalyzed by lactate dehydrogenase. The liver will ported, as cultured astrocytes treated with ammonium metabolize lactate into glucose, where it can serve as an chloride have reduced glutamine synthetase activity and energy source for all organs. In the brain, neurons use lactate increased tyrosine nitration of glutamine synthetase.54 as their preferred oxidative energy source and lactate is The elevation of astrocyte glutamine due to increased primarily synthesized by astrocytes, where it can be trans- ammonia concentrations can lead to changes in brain water ported to neurons via the astrocyte-neuron lactate shuttle.70 content. Glutamine is osmotically active and is a cause of In HE, there is possible derangement of this relationship brain edema and increased intracranial pressure.55 Elevated leading to hyperlactatemia. Increased concentrations of glutamine inside of astrocytes creates a hypertonic state lactate, considered a marker of energy failure, are a mecha- leading to increased water accumulation in astrocytes and nism of generating brain edema and neuronal dysfunction cytotoxic astrocyte swelling. In addition, glutamine enters during HE.71 Lactate can induce swelling of cultured and the mitochondria of astrocytes where it is metabolized, yield- primary astrocytes, as determined through in vitro ing glutamate and ammonia.56 The mitochondrial compart- studies.72 However, lactate has been associated with ment is small, and the high levels of ammonia that increased brain edema during acute liver failure in patients accumulate can lead to mitochondrial permeability transition and rodents and during chronic liver disease in BDL rats.73–75

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 352–361 355 Jaeger V. et al: Astrocytes and microglia in encephalopathy

Hyperlactatemia has been suggested as a prognostic signals leading to their activation, chemokine and cytokine marker of acetaminophen-induced acute liver failure, as regulation, and oxidative stress. increased arterial lactate correlated with the severity of HE and was present at significantly higher concentrations in non- Ammonia and microglia survivors.76 The use of 1H and 13C NMR spectroscopy on the frontal cortex of rats with acute liver failure secondary to Studies have investigated if hyperammonemia causes micro- hepatic devascularization determined that lactate was glia activation in both acute and chronic HE. The exposure of increased 169.2% compared to controls.77 Likewise, use of hepatic devascularization to model acute liver failure deter- primary cell cultures of microglia to ammonia led to an a mined that there were significant increases in lactate levels, increase in both synthesis and release of IL-6 and TNF 90 with a 166% increase at 6 h and an increase in 3293% at compared to basal microglia. Likewise, in the azoxyme- coma.78 Also, nuclear magnetic resonance spectroscopy was thane model of acute liver failure, the investigators found used to examine lactate usage by cells and determined that microglia to be activated; however, they found microglia to 91 increased brain lactate synthesis along with impaired glucose not be activated in mice injected with ammonium chloride. oxidation were the major contributing factors to brain edema In rats fed an ammonia-containing diet for 4 weeks to induce rather than accumulation of intracellular glutamine.74 It a state of hyperammonemia, microglia activation was should be mentioned that not every study investigating observed in the hippocampus, that could be reversed by lactate has found increases, as the use of 1H and 31P magnetic removing the ammonia-containing diet for 2 or 4 weeks.92 resonance spectroscopy found essentially no change of brain Interestingly, in a co-culture of rat astrocytes and microglia lactate in BDL rats at 4 weeks or 8 weeks following surgery.79 that were treated with ammonium chloride and LPS, it was found that ammonia treatment did not up-regulate the gene expression of IL-1a,IL-1b, IL-6 or TNFa in microglia or co- Microglia activation cultured astrocytes and microglia.93 The investigators also found that astrocytes reduced the up-regulation of microglia Microglia are cells of myeloid origin, whose main function is to activation markers induced by LPS.93 As elevated brain 80 control the immune response of the CNS. Additionally, acti- ammonia concentrations and neuroinflammation are patho- vated microglia are known to induce the inflammatory logical characteristics of HE, it is evident that gaining greater response in the brain by releasing proinflammatory cytokines, understanding into the exact influences of ammonia on a b a 81 such as IL-1 ,IL-1 and TNF . Evidence of neuroinflamma- microglia activation in the different contexts of HE are tion has been shown in HE patients with acute liver failure and necessary. chronic liver disease. In patients with acute liver failure, microglia activation occurs as shown by increased immunos- taining for human leukocyte antigen DR (CR3/43) when com- Chemokines pared to controls.21 In post-mortem cortical brain tissue from patients with liver cirrhosis and overt HE, there is up-regula- Outside of ammonia, chemokines are the primary contrib- tion of the microglia marker ionized calcium binding adaptor utors to activating microglia during states of neuroinflamma- molecule 1 (known as IBA1) when compared to cirrhotic tion. Chemokines are involved with cell-cell communication patients without HE.82 and regulate neuroinflammation by influencing migration and Multiple reports have shown microglia activation in the activation of immune cells.94 Most studies have focused on BDL model of chronic HE.34,83,84 Interestingly, one study investigating the pro-inflammatory chemokine ligand 2 found that BDL triggered alternative activation of microglia.34 (CCL2) and anti-inflammatory C-X3-C ligand 1 (CX3CL1). In Instead of the classical microglial markers OX6, ED1 and IBA1 mice that had undergone bile duct ligation, the release of along with pro-inflammatory markers IL-1b and inducible CCL2 in the brain triggered recruitment of infiltrating mono- nitric oxide synthase were not elevated but transforming cytes, leading to neurological decline.83 The investigators b growth factor beta 1 (known as TGF 1) was found to be also found that microglia have increased levels of CCL2 and 34 increased. Another study using post-mortem tissue from intraperitoneal injection of anti-TNFa serum led to reduced cirrhotic patients with HE observed activated microglia with CCL2 expression in microglia.83 In a model of portal hyper- hypertrophied cell bodies and thickened processes along with tension using triple calibrated portal vein ligation for 1 month higher levels of IL-6.85 Outside of cytokines, microglia activa- in rats, the investigators found that the CX3CL1 expression tion can be assessed by 11C-PK11195, which is a positive was not changed in the hippocampus or cerebellum, but its emission tomography ligand for translocator protein.86 In receptor CX3CR1 was significantly up-regulated in both the context of acute HE, 11C-PK1195 and 18F-DPA-714 have regions while stromal cell-derived factor 1 alpha and C-X-C been used and found to detect neuroinflammation in thioace- 87 chemokine receptor type 4 were up-regulated in only the hip- tamide-treated rats by binding to translocator protein. 95 Interestingly, translocator protein has been deleted from pocampus. Specific targeting of CCL2 activity through intra- astrocytes, demonstrating an increase of mitochondria per- peritoneal injection of chemokine receptor 2 and chemokine meability transition and cell volume in response to ammonia, receptor 4 inhibitors during acute liver failure in azoxyme- indicating that this protein is involved in more processes than thane- or thioacetamide-treated mice was found to reduce 96,97 just neuroinflammation.88 microglia activation and improve neurological function. That being said, not all evidence shows induction of a pro- In the azoxymethane model of acute liver failure, injection inflammatory phenotype during HE as microglia polarization of soluble CX3CL1 was found to reduce microglia activation occurs in cirrhotic patients, with both pro-inflammatory M1 and improve time taken to reach coma, indicating that an and anti-inflammatory M2 phenotypes being present.89 imbalance of CCL2 and CX3CL1 expression may be driving Research involving microglia has primarily focused on the activation of microglia during HE.98

356 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 352–361 Jaeger V. et al: Astrocytes and microglia in encephalopathy

Cytokines microglia activation.107 In azoxymethane-treated mice, cortex IL-1b, IL-6, TNFa and CCL2 protein expression were In patients and animal models of HE, systemic inflammation increased.108 The use of anti-TGFb1 antibodies or genetic causes worsening of neurological function and it has been ablation of transforming growth factor beta receptor 2 proposed that pro-inflammatory signals act in concert with (known as TGFbR2) in neurons of azoxymethane-treated ammonia to generate the neurological complications of acute mice led to reduced microglia activation and normalized liver failure and chronic liver disease.21,99,100 TNFa is a potent levels of IL-1b, IL-6, TNFa and CCL2, giving support that pro-inflammatory cytokine that has been shown to activate TGFbR2-mediated signaling contributes to neuroinflamma- microglia in a number of experimental models of neuroinflam- tion during HE.108 Interestingly, the use of the gamma ami- mation.101,102 Circulating levels of TNFa are increased as a nobutyric acid antagonist bicuculine was found to reduce the function of the severity of HE in both patients103 and exper- expression of IL-1b, but not effect microglia activation in imental animals104 with liver failure. Moreover, the presence ammonium chloride-fed rats.109 This gives support that the of TNFa gene polymorphisms is known to influence the out- pro-inflammatory state during elevated levels of ammonia in comes of patients with acute liver failure.105 During HE, sys- the brain may not be directly linked to microglia activation, temic levels of TNFa are increased in the azoxymethane though more studies are need in this area to better charac- model of acute liver failure.106 Inhibition of TNFa signaling terize this. by systemic treatment with etanercept reduced systemic inflammation, attenuated the neurological decline, and pre- Clinical management of HE vented microglial activation in the cerebral cortex.106 These data support the hypothesis that peripherally-derived TNFa, The most efficacious treatment for patients with HE is liver at least in part, contributes to the microglial activation and transplantation, with cognitive measures and metabolites in subsequent neurological decline of liver failure. In further the brain becoming normalized after transplantation.110 That support of this concept, neurological complications occurring being said, in comparison to cirrhotic patients without HE that in BDL mice were shown to be the consequence of monocyte undergo liver transplantation, those with HE that undergo recruitment in response to TNFa signaling and to occur via liver transplantation have impaired cognitive function, with microglial activation.83 Specifically, peripheral TNFa signaling one study showing that 13% of the cohort of patients main- stimulates microglia to produce CCL2, which subsequently tained mild cognitive impairment 6–12 months after liver mediates monocyte recruitment into the brain.83 transplantation.111 Besides liver transplantation, there are Pro-inflammatory mediators other than TNFa can contrib- therapeutics commonly used, including lactulose or nonab- ute to HE pathogenesis as well. IL-1b and microglia activation sorbable antibiotics such as rifaximin. Lactulose is a nondi- are increased in rats after portacaval shunt and administer- gestible disaccharide that is metabolized by bacteria in the ing sildenafil was found to reduce neuroinflammation and colon, decreasing pH, reducing bacterial ammonia production,

Fig. 2. Summary of the involvement of microglia and astrocytes in HE pathology. The liver can lose its ability to function during acute liver failure or chronic liver disease. When this occurs, toxins, such as ammonia, enter the circulation and can enter the brain. After accumulating in neural tissue, they cause a disruption of astrocyte and microglia cellular function. Astrocytes metabolize ammonia, leading to an increase of glutamine, cell swelling, cerebral edema, oxidative stress and hyperlactatemia. Microglia become activated leading to increased neuroinflammation due to pro-inflammatory cytokine release and oxidative stress. Together, these changes in neural cellular function lead to increased HE pathology, resulting in increased morbidity and mortality.

Abbreviations: HE, hepatic encephalopathy; NH3, ammonia.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 352–361 357 Jaeger V. et al: Astrocytes and microglia in encephalopathy and producing more ammonium from ammonia, which cannot by NIH R01 awards (DK082435 and DK112803) and a VA be absorbed into the intestine.112 Rifaximin is used to elimi- Merit award (BX002638) from the United States Department nate bacteria with a specific efficacy for anaerobic bacteria and of Veterans Affairs Biomedical Laboratory Research and can reduce ammonia production, endotoxin generation, and Development Service to Dr. Sharon DeMorrow. This work inflammation.113 There are a variety of other treatments that was also supported by the resources of the Central Texas have been investigated, including L-ornithine-L-aspartate, Veterans Health Care System Research Service, Temple, ornithine phenylacetate, glycerol phenylbutyrate, various pro- Texas. The content is the responsibility of the author(s) biotics, AST-120 and others that are generally aimed at reduc- alone and does not necessarily reflect the views or policies ing ammonia levels in HE patients that have had limited of the Department of Veterans Affairs or the United States improvement when compared to lactulose or rifaximin.5,114 Government. Other than these direct treatments of HE, clinical management strategies for HE patients include alleviating precipitat- Conflict of interest ing factors that contribute to the syndrome and management of comorbidities. In HE patients that develop hepatorenal The authors have no conflict of interests related to this syndrome there is a complete disruption of the urea cycle, publication. leading to even greater concentrations of ammonia in the circulation due to both loss of hepatic function and kidney Author contributions failure. Other conditions can lead to inhibition of ammonia metabolism, including sarcopenia in cirrhotic patients, as Contributed to draft and critically revise the document (VJ, these have reduced ability to metabolize ammonia into SD), draft the document, create the figures, and critically glutamine due to muscle loss. The presence of sarcopenia is revise the document (MM). All authors approved the final associated with development of both minimal and overt version of this manuscript. HE.115 Infections are also a risk factor for HE with bacterial overgrowth, spontaneous bacterial peritonitis, and other infections contributing to morbidity and HE progression.116 References Controlling these systemic complications that contribute to HE while lowering circulating ammonia concentrations are [1] Martino ME, Fernández-Lorente J, Romero-Vives M, Bárcena R, Gaztelu JM. Brain oscillatory activity during sleep shows unknown dysfunctions in early the most effective strategies currently used for management encephalopathy. J Physiol Biochem 2014;70:821–835. doi: 10. of this syndrome. 1007/s13105-014-0351-2. [2] Nardelli S, Allampati S, Riggio O, Mullen KD, Prakash R, Gioia S, et al. Hepatic encephalopathy is associated with persistent learning impairments Summary and future perspectives despite adequate medical treatment: A multicenter, international study. Dig Dis Sci 2017;62:794–800. doi: 10.1007/s10620-016-4425-6. HE is a devastating consequence of acute liver failure and [3] Hirode G, Vittinghoff E, Wong RJ. Increasing burden of hepatic encephalopathy among hospitalized adults: An analysis of the 2010-2014 national chronic liver disease that is associated with increased morbid- inpatient sample. Dig Dis Sci 2019;64:1448–1457. doi: 10.1007/s10620- ity and mortality and has few effective treatments besides 019-05576-9. liver transplantation. The elevation of circulating ammonia [4] Suraweera D, Sundaram V, Saab S. Evaluation and management of hepatic levels that occurs following the disruption of the urea cycle encephalopathy: Current status and future directions. Gut Liver 2016;10: 509–519. doi: 10.5009/gnl15419. leads to accumulation of ammonia in astrocytes. This gen- [5] Patidar KR, Bajaj JS. Covert and overt hepatic encephalopathy: Diagnosis erates a myriad of complications, including dysregulation of and management. Clin Gastroenterol Hepatol 2015;13:2048–2061. doi: glutamine concentrations, elevation of lactate, generation of 10.1016/j.cgh.2015.06.039. oxidative and nitrosative stress, and cerebral edema. In [6] Dharel N, Bajaj JS. Definition and nomenclature of hepatic encephalopathy. J Clin Exp Hepatol 2015;5:S37–S41. doi: 10.1016/j.jceh.2014.10.001. addition, hepatic injury and loss of liver function leads to [7] Grover VP, Tognarelli JM, Massie N, Crossey MM, Cook NA, Taylor-Robinson systemic inflammation, which results in activation of micro- SD. The why and wherefore of hepatic encephalopathy. Int J Gen Med 2015; glia in the brain due to both increases of ammonia and 8:381–390. doi: 10.2147/IJGM.S86854. dysregulation of chemokine concentrations. This activation [8] NeSmith M, Ahn J, Flamm SL. Contemporary understanding and management of overt and covert hepatic encephalopathy. Gastroenterol Hepatol of microglia leads to the production of pro-inflammatory (N Y) 2016;12:91–100. cytokines that ultimately work in concert with astrocyte [9] Liu A, Perumpail RB, Kumari R, Younossi ZM, Wong RJ, Ahmed A. Advances dysfunction to further increase oxidative stress and cerebral in cirrhosis: Optimizing the management of hepatic encephalopathy. World J – edema. Collectively, this results in increased cerebral edema, Hepatol 2015;7:2871 2879. doi: 10.4254/wjh.v7.i29.2871. [10] Bernal W, Auzinger G, Dhawan A, Wendon J. Acute liver failure. Lancet intracranial pressure, and worse HE pathology. A working 2010;376:190–201. doi: 10.1016/S0140-6736(10)60274-7. model of these aspects of HE pathology are presented in [11] Bustamante J, Rimola A, Ventura PJ, Navasa M, Cirera I, Reggiardo V, et al. Fig. 2. Ultimately, if we are to develop new treatments for the Prognostic significance of hepatic encephalopathy in patients with cirrhosis. – management of HE, gaining a greater understanding into the J Hepatol 1999;30:890 895. doi: 10.1016/s0168-8278(99)80144-5. [12] Romeiro FG, Augusti L. Nutritional assessment in cirrhotic patients with specific cellular contributions to neurological dysfunction hepatic encephalopathy. World J Hepatol 2015;7:2940–2954. doi: 10. associated with HE is necessary to identify new targets that 4254/wjh.v7.i30.2940. allow for better management of this syndrome. [13] Hakvoort TB, He Y, Kulik W, Vermeulen JL, Duijst S, Ruijter JM, et al. Pivotal role of glutamine synthetase in ammonia detoxification. Hepatology 2017; 65:281–293. doi: 10.1002/hep.28852. [14] Hansen BA, Vilstrup H. Increased intestinal hydrolysis of urea in patients Acknowledgments with alcoholic cirrhosis. Scand J Gastroenterol 1985;20:346–350. doi: 10. 3109/00365528509091662. This study was funded by a VA Career Development award [15] Sørensen M. Update on cerebral uptake of blood ammonia. Metab Brain Dis – (BX003486) from the United States Department of Veterans 2013;28:155 159. doi: 10.1007/s11011-013-9395-1. [16] Clemmesen JO, Kondrup J, Ott P. Splanchnic and leg exchange of amino Affairs Biomedical Laboratory Research and Development acids and ammonia in acute liver failure. Gastroenterology 2000;118: Service to Dr. Matthew McMillin. This study was also funded 1131–1139. doi: 10.1016/s0016-5085(00)70366-0.

358 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 352–361 Jaeger V. et al: Astrocytes and microglia in encephalopathy

[17] Riggio O, Nardelli S, Pasquale C, Pentassuglio I, Gioia S, Onori E, et al.No [40] Fomenko OZ, Ushakova HO, Piierzhynovs’kyĭ SH. Astroglia proteins in the effect of albumin infusion on the prevention of hepatic encephalopathy after rat brain in experimental chronic hepatitis and 2-oxoglutarate effect. Ukr transjugular intrahepatic portosystemic shunt. Metab Brain Dis 2016;31: Biokhim Zh (1999) 2011;83:69–76. 1275–1281. doi: 10.1007/s11011-015-9713-x. [41] Rivera-Mancía S, Montes S, Méndez-Armenta M, Muriel P, Ríos C. Morpho- [18] Clària J, Stauber RE, Coenraad MJ, Moreau R, Jalan R, Pavesi M, et al. Sys- logical changes of rat astrocytes induced by liver damage but not by man- temic inflammation in decompensated cirrhosis: Characterization and role ganese chloride exposure. Metab Brain Dis 2009;24:243–255. doi: 10. in acute-on-chronic liver failure. Hepatology 2016;64:1249–1264. doi: 10. 1007/s11011-009-9138-5. 1002/hep.28740. [42] Reinehr R, Görg B, Becker S, Qvartskhava N, Bidmon HJ, Selbach O, et al. [19] Moore JK, MacKinnon AC, Man TY, Manning JR, Forbes SJ, Simpson KJ. Hypoosmotic swelling and ammonia increase oxidative stress by NADPH Patients with the worst outcomes after paracetamol (acetaminophen)- oxidase in cultured astrocytes and vital brain slices. Glia 2007;55:758– induced liver failure have an early monocytopenia. Aliment Pharmacol 771. doi: 10.1002/glia.20504. Ther 2017;45:443–454. doi: 10.1111/apt.13878. [43] Lachmann V, Görg B, Bidmon HJ, Keitel V, Häussinger D. Precipitants of [20] Dirchwolf M, Podhorzer A, Marino M, Shulman C, Cartier M, Zunino M, et al. hepatic encephalopathy induce rapid astrocyte swelling in an oxidative Immune dysfunction in cirrhosis: Distinct cytokines phenotypes according stress dependent manner. Arch Biochem Biophys 2013;536:143–151. to cirrhosis severity. Cytokine 2016;77:14–25. doi: 10.1016/j.cyto.2015. doi: 10.1016/j.abb.2013.05.004. 10.006. [44] Häussinger D, Görg B. Interaction of oxidative stress, astrocyte swelling and [21] Butterworth RF. Hepatic encephalopathy: a central neuroinflammatory dis- cerebral ammonia toxicity. Curr Opin Clin Nutr Metab Care 2010;13:87–92. order? Hepatology 2011;53:1372–1376. doi: 10.1002/hep.24228. doi: 10.1097/MCO.0b013e328333b829. [22] Record CO, Buxton B, Chase RA, Curzon G, Murray-Lyon IM, Williams R. [45] Daverey A, Agrawal SK. Curcumin alleviates oxidative stress and mitochon- Plasma and brain amino acids in fulminant hepatic failure and their relation- drial dysfunction in astrocytes. Neuroscience 2016;333:92–103. doi: 10. ship to hepatic encephalopathy. Eur J Clin Invest 1976;6:387–394. doi: 10. 1016/j.neuroscience.2016.07.012. 1111/j.1365-2362.1976.tb00533.x. [46] Bobermin LD, Wartchow KM, Flores MP, Leite MC, Quincozes-Santos A, [23] Bron B, Waldram R, Silk DB, Williams R. Serum, cerebrospinal fluid, and Gonçalves CA. Ammonia-induced oxidative damage in neurons is prevented brain levels of bile acids in patients with fulminant hepatic failure. Gut by resveratrol and lipoic acid with participation of heme oxygenase 1. Neu- 1977;18:692–696. doi: 10.1136/gut.18.9.692. rotoxicology 2015;49:28–35. doi: 10.1016/j.neuro.2015.05.005. [24] Spahr L, Butterworth RF, Fontaine S, Bui L, Therrien G, Milette PC, et al. [47] Suárez I, Bodega G, Fernández B. Glutamine synthetase in brain: effect of Increased blood manganese in cirrhotic patients: relationship to pallidal ammonia. Neurochem Int 2002;41:123–142. doi: 10.1016/s0197-0186 magnetic resonance signal hyperintensity and neurological symptoms. (02)00033-5. Hepatology 1996;24:1116–1120. doi: 10.1002/hep.510240523. [48] Jambekar AA, Palma E, Nicolosi L, Rasola A, Petronilli V, Chiara F, et al. [25] Pomier-Layrargues G, Spahr L, Butterworth RF. Increased manganese con- A glutamine synthetase inhibitor increases survival and decreases cytokine centrations in pallidum of cirrhotic patients. Lancet 1995;345:735. doi: 10. response in a mouse model of acute liver failure. Liver Int 2011;31:1209– 1016/s0140-6736(95)90909-5. 1221. doi: 10.1111/j.1478-3231.2011.02553.x. [26] Butterworth RF, Spahr L, Fontaine S, Layrargues GP. Manganese toxicity, [49] Girard G, Giguère JF, Butterworth RF. Region-selective reductions in activ- dopaminergic dysfunction and hepatic encephalopathy. Metab Brain Dis ities of glutamine synthetase in rat brain following portacaval anastomosis. 1995;10:259–267. doi: 10.1007/bf02109357. Metab Brain Dis 1993;8:207–215. doi: 10.1007/bf01001062. [27] Sharma P, Sharma BC. Management of overt hepatic encephalopathy. J Clin [50] Butterworth RF, Girard G, Giguère JF. Regional differences in the capacity for Exp Hepatol 2015;5:S82–S87. doi: 10.1016/j.jceh.2014.04.004. ammonia removal by brain following portocaval anastomosis. J Neurochem [28] Sawhney R, Holland-Fischer P, Rosselli M, Mookerjee RP, Agarwal B, Jalan R. 1988;51:486–490. doi: 10.1111/j.1471-4159.1988.tb01064.x. Role of ammonia, inflammation, and cerebral oxygenation in brain dysfunc- [51] Suárez I, Bodega G, Arilla E, Fernández B. Region-selective glutamine syn- tion of acute-on-chronic liver failure patients. Liver Transpl 2016;22:732– thetase expression in the rat central nervous system following portocaval 742. doi: 10.1002/lt.24443. anastomosis. Neuropathol Appl Neurobiol 1997;23:254–261. doi: 10. [29] Blackburn D, Sargsyan S, Monk PN, Shaw PJ. Astrocyte function and role in 1111/j.1365-2990.1997.tb01209.x. motor neuron disease: a future therapeutic target? Glia 2009;57:1251– [52] Jover-Cobos M, Noiret L, Lee K, Sharma V, Habtesion A, Romero-Gomez M, 1264. doi: 10.1002/glia.20848. et al. Ornithine phenylacetate targets alterations in the expression and [30] Brumback RA, Lapham LW. DNA synthesis in Alzheimer type II astrocytosis. activity of glutamine synthase and glutaminase to reduce ammonia levels The question of astrocytic proliferation and mitosis in experimentally in bile duct ligated rats. J Hepatol 2014;60:545–553. doi: 10.1016/j.jhep. induced hepatic encephalopathy. Arch Neurol 1989;46:845–848. doi: 10. 2013.10.012. 1001/archneur.1989.00520440027016. [53] Görg B, Qvartskhava N, Bidmon HJ, Palomero-Gallagher N, Kircheis G, Zilles [31] Butterworth RF. Pathogenesis of hepatic encephalopathy and brain edema K, et al. Oxidative stress markers in the brain of patients with cirrhosis and in acute liver failure. J Clin Exp Hepatol 2015;5:S96–S103. doi: 10.1016/j. hepatic encephalopathy. Hepatology 2010;52:256–265. doi: 10.1002/hep. jceh.2014.02.004. 23656. [32] Bélanger M, Desjardins P, Chatauret N, Butterworth RF. Loss of expression [54] Schliess F, Görg B, Fischer R, Desjardins P, Bidmon HJ, Herrmann A, et al. of glial fibrillary acidic protein in acute hyperammonemia. Neurochem Int Ammonia induces MK-801-sensitive nitration and phosphorylation of 2002;41:155–160. doi: 10.1016/s0197-0186(02)00037-2. protein tyrosine residues in rat astrocytes. FASEB J 2002;16:739–741. [33] Thumburu KK, Dhiman RK, Vasishta RK, Chakraborti A, Butterworth RF, doi: 10.1096/fj.01-0862fje. Beauchesne E, et al. Expression of astrocytic genes coding for proteins [55] Tofteng F, Hauerberg J, Hansen BA, Pedersen CB, Jørgensen L, Larsen FS. implicated in neural excitation and brain edema is altered after acute liver Persistent arterial hyperammonemia increases the concentration of gluta- failure. J Neurochem 2014;128:617–627. doi: 10.1111/jnc.12511. mine and alanine in the brain and correlates with intracranial pressure in [34] Wright GA, Sharifi Y, Newman TA, Davies N, Vairappan B, Perry HV, et al. patients with fulminant hepatic failure. J Cereb Blood Flow Metab 2006;26: Characterisation of temporal microglia and astrocyte immune responses in 21–27. doi: 10.1038/sj.jcbfm.9600168. bile duct-ligated rat models of cirrhosis. Liver Int 2014;34:1184–1191. doi: [56] Albrecht J, Norenberg MD. Glutamine: a Trojan horse in ammonia neuro- 10.1111/liv.12481. toxicity. Hepatology 2006;44:788–794. doi: 10.1002/hep.21357. [35] Suarez I, Bodega G, Fernandez B. Changes in glial fibrillary acidic protein [57] Norenberg MD, Rao KV, Jayakumar AR. Mechanisms of ammonia-induced immunoreactivity in response to experimental hepatic encephalopathy in astrocyte swelling. Metab Brain Dis 2005;20:303–318. doi: 10. the rat hippocampus. Restor Neurol Neurosci 1998;12:49–54. 1007/s11011-005-7911-7. [36] Haghighat N, McCandless DW, Geraminegad P. Responses in primary astro- [58] Nielsen S, Nagelhus EA, Amiry-Moghaddam M, Bourque C, Agre P, Ottersen cytes and C6-glioma cells to ammonium chloride and dibutyryl cyclic-AMP. OP. Specialized membrane domains for water transport in glial cells: high- Neurochem Res 2000;25:277–284. doi: 10.1023/a:1007535922977. resolution immunogold cytochemistry of aquaporin-4 in rat brain. J Neuro- [37] Neary JT, Whittemore SR, Zhu Q, Norenberg MD. Destabilization of glial sci 1997;17:171–180. doi: 10.1523/JNEUROSCI.17-01-00171.1997. fibrillary acidic protein mRNA in astrocytes by ammonia and protection by [59] Amiry-Moghaddam M, Xue R, Haug FM, Neely JD, Bhardwaj A, Agre P, et al. extracellular ATP. J Neurochem 1994;63:2021–2027. doi: 10.1046/j.1471- Alpha-syntrophin deletion removes the perivascular but not endothelial 4159.1994.63062021.x. pool of aquaporin-4 at the blood-brain barrier and delays the development [38] Arias N, Méndez M, Alcalde I, Íñigo-Portugués A, Merayo-Lloves J, Arias J, of brain edema in an experimental model of acute hyponatremia. FASEB J et al. Assessing the brain through the eye: New ways to explore hepatic 2004;18:542–544. doi: 10.1096/fj.03-0869fje. encephalopathy. Physiol Behav 2017;173:263–271. doi: 10.1016/j. [60] Manley GT, Binder DK, Papadopoulos MC, Verkman AS. New insights into physbeh.2017.02.022. water transport and edema in the central nervous system from phenotype [39] Arias N, Méndez M, Gómez-Lázaro E, Azpiroz A, Arias JL. Main target of analysis of aquaporin-4 null mice. Neuroscience 2004;129:983–991. doi: minimal hepatic encephalopathy: Morphophysiological, inflammatory and 10.1016/j.neuroscience.2004.06.088. metabolic view. Physiol Behav 2015;149:247–254. doi: 10.1016/j. [61] Nicchia GP, Frigeri A, Liuzzi GM, Svelto M. Inhibition of aquaporin-4 expres- physbeh.2015.06.019. sion in astrocytes by RNAi determines alteration in cell morphology, growth,

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 352–361 359 Jaeger V. et al: Astrocytes and microglia in encephalopathy

and water transport and induces changes in ischemia-related genes. FASEB [84] Rodrigo R, Cauli O, Gomez-Pinedo U, Agusti A, Hernandez-Rabaza V, J 2003;17:1508–1510. doi: 10.1096/fj.02-1183fje. Garcia-Verdugo JM, et al. Hyperammonemia induces neuroinflammation [62] Yang JH, Song ZJ, Liao CD, You LY, Xu Y, Yang J, et al. The relationship of the that contributes to cognitive impairment in rats with hepatic encephalop- expression of aquaporin-4 and brain edema in rats with acute liver failure. athy. Gastroenterology 2010;139:675–684. doi: 10.1053/j.gastro.2010. Zhonghua Gan Zang Bing Za Zhi 2006;14:215–216. 03.040. [63] Wang LQ, Zhu SM, Zhou HJ, Pan CF. Expression of aquaporin-4 during brain [85] Dennis CV, Sheahan PJ, Graeber MB, Sheedy DL, Kril JJ, Sutherland GT. edema in rats with thioacetamide-induced acute encephalopathy. Zhong- Microglial proliferation in the brain of chronic alcoholics with hepatic ence- hua Yi Xue Za Zhi 2011;91:2573–2577. doi: 10.3760/cma.j.issn.0376- phalopathy. Metab Brain Dis 2014;29:1027–1039. doi: 10.1007/s11011- 2491.2011.36.016. 013-9469-0. [64] Eefsen M, Jelnes P, Schmidt LE, Vainer B, Bisgaard HC, Larsen FS. Brain [86] Cui Y, Takashima T, Takashima-Hirano M, Wada Y, Shukuri M, Tamura Y, et al. expression of the water channels aquaporin-1 and -4 in mice with acute 11C-PK11195 PET for the in vivo evaluation of neuroinflammation in the rat liver injury, hyperammonemia and brain edema. Metab Brain Dis 2010; brain after cortical spreading depression. J Nucl Med 2009;50:1904–1911. 25:315–323. doi: 10.1007/s11011-010-9213-y. doi: 10.2967/jnumed.109.066498. [65] Chastre A, Jiang W, Desjardins P, Butterworth RF. Ammonia and proinflam- [87] Luo S, Kong X, Wu JR, Wang CY, Tian Y, Zheng G, et al. Neuroinflammation in matory cytokines modify expression of genes coding for astrocytic proteins acute hepatic encephalopathy rats: imaging and therapeutic effectiveness implicated in brain edema in acute liver failure. Metab Brain Dis 2010;25: evaluation using 11C-PK11195 and 18F-DPA-714 micro-positron emission 17–21. doi: 10.1007/s11011-010-9185-y. tomography. Metab Brain Dis 2018;33:1733–1742. doi: 10.1007/s11011- [66] Zhang LJ, Qi R, Zhong J, Xu Q, Zheng G, Lu GM. The effect of hepatic 018-0282-7. encephalopathy, hepatic failure, and portosystemic shunt on brain volume [88] Panickar KS, Jayakumar AR, Rama Rao KV, Norenberg MD. Downregulation of cirrhotic patients: a voxel-based morphometry study. PLoS One 2012;7: of the 18-kDa translocator protein: effects on the ammonia-induced mito- e42824. doi: 10.1371/journal.pone.0042824. chondrial permeability transition and cell swelling in cultured astrocytes. [67] Rai R, Ahuja CK, Agrawal S, Kalra N, Duseja A, Khandelwal N, et al. Reversal Glia 2007;55:1720–1727. doi: 10.1002/glia.20584. of low-grade cerebral edema after lactulose/rifaximin therapy in patients [89] Görg B, Bidmon HJ, Häussinger D. Gene expression profiling in the cerebral with cirrhosis and minimal hepatic encephalopathy. Clin Transl Gastroen- cortex of patients with cirrhosis with and without hepatic encephalopathy. terol 2015;6:e111. doi: 10.1038/ctg.2015.38. Hepatology 2013;57:2436–2347. doi: 10.1002/hep.26265. [68] Dhanda S, Sandhir R. Blood-brain barrier permeability is exacerbated in [90] Rao KV, Brahmbhatt M, Norenberg MD. Microglia contribute to ammonia- experimental model of hepatic encephalopathy via MMP-9 activation and induced astrocyte swelling in culture. Metab Brain Dis 2013;28:139–143. downregulation of tight junction proteins. Mol Neurobiol 2018;55:3642– doi: 10.1007/s11011-012-9339-1. 3659. doi: 10.1007/s12035-017-0521-7. [91] Rangroo Thrane V, Thrane AS, Chang J, Alleluia V, Nagelhus EA, Nedergaard [69] Wright G, Soper R, Brooks HF, Stadlbauer V, Vairappan B, Davies NA, et al. M. Real-time analysis of microglial activation and motility in hepatic and Role of aquaporin-4 in the development of brain oedema in liver failure. J hyperammonemic encephalopathy. Neuroscience 2012;220:247–255. Hepatol 2010;53:91–97. doi: 10.1016/j.jhep.2010.02.020. doi: 10.1016/j.neuroscience.2012.06.022. [70] Bélanger M, Magistretti PJ. The role of astroglia in neuroprotection. Dia- [92] Balzano T, Dadsetan S, Forteza J, Cabrera-Pastor A, Taoro-Gonzalez L, logues Clin Neurosci 2009;11:281–295. Malaguarnera M, et al. Chronic hyperammonemia induces peripheral [71] Bosoi CR, Rose CF. Elevated cerebral lactate: Implications in the pathogen- inflammation that leads to cognitive impairment in rats: reversal by anti- esis of hepatic encephalopathy. Metab Brain Dis 2014;29:919–925. doi: 10. tnfa treatment. J Hepatol 2019. doi: 10.1016/j.jhep.2019.01.008. 1007/s11011-014-9573-9. [93] Karababa A, Groos-Sahr K, Albrecht U, Keitel V, Shafigullina A, Görg B, et al. [72] Staub F, Baethmann A, Peters J, Weigt H, Kempski O. Effects of lactacidosis Ammonia attenuates LPS-induced upregulation of pro-inflammatory cyto- on glial cell volume and viability. J Cereb Blood Flow Metab 1990;10:866– kine mrna in co-cultured astrocytes and microglia. Neurochem Res 2017; 876. doi: 10.1038/jcbfm.1990.143. 42:737–749. doi: 10.1007/s11064-016-2060-4. [73] Tofteng F, Jorgensen L, Hansen BA, Ott P, Kondrup J, Larsen FS. Cerebral [94] Wright G, Swain M, Annane D, Saliba F, Samuel D, Arroyo V, et al. Neuro- microdialysis in patients with fulminant hepatic failure. Hepatology 2002; inflammation in liver disease: sessional talks from ISHEN. Metab Brain Dis 36:1333–1340. doi: 10.1053/jhep.2002.36944. 2016;31:1339–1354. doi: 10.1007/s11011-016-9918-7. [74] Zwingmann C, Chatauret N, Leibfritz D, Butterworth RF. Selective increase [95] Merino J, Aller MA, Rubio S, Arias N, Nava MP, Loscertales M, et al. Gut-brain of brain lactate synthesis in experimental acute liver failure: results of a chemokine changes in portal hypertensive rats. Dig Dis Sci 2011;56:2309– [H-C] nuclear magnetic resonance study. Hepatology 2003;37:420–428. 2317. doi: 10.1007/s10620-011-1625-y. doi: 10.1053/jhep.2003.50052. [96] McMillin M, Frampton G, Thompson M, Galindo C, Standeford H, Whittington E, [75] Bosoi CR, Zwingmann C, Marin H, Parent-Robitaille C, Huynh J, Tremblay M, et al. Neuronal CCL2 is upregulated during hepatic encephalopathy and con- et al. Increased brain lactate is central to the development of brain edema in tributes to microglia activation and neurological decline. J Neuroinflammation rats with chronic liver disease. J Hepatol 2014;60:554–560. doi: 10.1016/j. 2014;11:121. doi: 10.1186/1742-2094-11-121. jhep.2013.10.011. [97] Zhang L, Tan J, Jiang X, Qian W, Yang T, Sun X, et al. Neuron-derived CCL2 [76] Schmidt LE, Larsen FS. Prognostic implications of hyperlactatemia, multiple contributes to microglia activation and neurological decline in hepatic ence- organ failure, and systemic inflammatory response syndrome in patients phalopathy. Biol Res 2017;50:26. doi: 10.1186/s40659-017-0130-y. with acetaminophen-induced acute liver failure. Crit Care Med 2006;34: [98] McMillin M, Grant S, Frampton G, Andry S, Brown A, DeMorrow S. Fractalkine 337–343. doi: 10.1097/01.ccm.0000194724.70031.b6. suppression during hepatic encephalopathy promotes neuroinflammation [77] Chatauret N, Zwingmann C, Rose C, Leibfritz D, Butterworth RF. Effects of in mice. J Neuroinflammation 2016;13:198. doi: 10.1186/s12974-016- hypothermia on brain glucose metabolism in acute liver failure: a H/C- 0674-8. nuclear magnetic resonance study. Gastroenterology 2003;125:815–824. [99] Shawcross DL, Davies NA, Williams R, Jalan R. Systemic inflammatory doi: 10.1016/s0016-5085(03)01054-0. response exacerbates the neuropsychological effects of induced hyperam- [78] Chavarria L, Oria M, Romero-Gimenez J, Alonso J, Lope-Piedrafita S, monemia in cirrhosis. J Hepatol 2004;40:247–254. doi: 10.1016/j.jhep. Cordoba J. Diffusion tensor imaging supports the cytotoxic origin of brain 2003.10.016. edema in a rat model of acute liver failure. Gastroenterology 2010;138: [100] Butterworth RF. The liver-brain axis in liver failure: neuroinflammation and 1566–1573. doi: 10.1053/j.gastro.2009.10.003. encephalopathy. Nat Rev Gastroenterol Hepatol 2013;10:522–528. doi: [79] Rackayova V, Braissant O, McLin VA, Berset C, Lanz B, Cudalbu C. 1H and 31P 10.1038/nrgastro.2013.99. magnetic resonance spectroscopy in a rat model of chronic hepatic ence- [101] Lambertsen KL, Biber K, Finsen B. Inflammatory cytokines in experimental phalopathy: in vivo longitudinal measurements of brain energy metabolism. and human stroke. J Cereb Blood Flow Metab 2012;32:1677–1698. doi: 10. Metab Brain Dis 2016;31:1303–1314. doi: 10.1007/s11011-015-9715-8. 1038/jcbfm.2012.88. [80] Kabba JA, Xu Y, Christian H, Ruan W, Chenai K, Xiang Y, et al. Microglia: [102] Rubio-Perez JM, Morillas-Ruiz JM. A review: inflammatory process in Housekeeper of the central nervous system. Cell Mol Neurobiol 2018;38: Alzheimer’s disease, role of cytokines. ScientificWorldJournal 2012;2012: 53–71. doi: 10.1007/s10571-017-0504-2. 756357. doi: 10.1100/2012/756357. [81] Garden GA, Möller T. Microglia biology in health and disease. J Neuroim- [103] Odeh M. Pathogenesis of hepatic encephalopathy: the tumour necrosis mune Pharmacol 2006;1:127–137. doi: 10.1007/s11481-006-9015-5. factor-alpha theory. Eur J Clin Invest 2007;37:291–304. doi: 10.1111/j. [82] Zemtsova I, Görg B, Keitel V, Bidmon HJ, Schrör K, Häussinger D. Microglia 1365-2362.2007.01778.x. activation in hepatic encephalopathy in rats and humans. Hepatology 2011; [104] Jiang W, Desjardins P, Butterworth RF. Cerebral inflammation contributes to 54:204–215. doi: 10.1002/hep.24326. encephalopathy and brain edema in acute liver failure: protective effect of [83] D’Mello C, Le T, Swain MG. Cerebral microglia recruit monocytes into the minocycline. J Neurochem 2009;109:485–493. doi: 10.1111/j.1471-4159. brain in response to tumor necrosis factoralpha signaling during peripheral 2009.05981.x. organ inflammation. J Neurosci 2009;29:2089–2102. doi: 10.1523/JNEUR- [105] Bernal W, Donaldson P, Underhill J, Wendon J, Williams R. Tumor necrosis OSCI.3567-08.2009. factor genomic polymorphism and outcome of acetaminophen

360 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 352–361 Jaeger V. et al: Astrocytes and microglia in encephalopathy

(paracetamol)-induced acute liver failure. J Hepatol 1998;29:53–59. doi: [110] Chavarria L, Cordoba J. Encephalopathy and liver transplantation. Metab 10.1016/s0168-8278(98)80178-5. Brain Dis 2013;28:285–292. doi: 10.1007/s11011-012-9350-6. [106] Chastre A, Bélanger M, Beauchesne E, Nguyen BN, Desjardins P, [111] Garcia-Martinez R, Rovira A, Alonso J, Jacas C, Simón-Talero M, Chavarria L, Butterworth RF. Inflammatory cascades driven by tumor necrosis factor- et al. Hepatic encephalopathy is associated with posttransplant cognitive alpha play a major role in the progression of acute liver failure and its function and brain volume. Liver Transpl 2011;17:38–46. doi: 10.1002/lt. neurological complications. PLoS One 2012;7:e49670. doi: 10. 22197. 1371/journal.pone.0049670. [112] Gerber T, Schomerus H. Hepatic encephalopathy in liver cirrhosis: patho- [107] Agusti A, Hernández-Rabaza V, Balzano T, Taoro-Gonzalez L, Ibañez-Grau A, genesis, diagnosis and management. Drugs 2000;60:1353–1370. doi: 10. Cabrera-Pastor A, et al. Sildenafil reduces neuroinflammation in cerebellum, 2165/00003495-200060060-00008. restores GABAergic tone, and improves motor in-coordination in rats with [113] Patidar KR, Bajaj JS. Antibiotics for the treatment of hepatic encephalopathy. hepatic encephalopathy. CNS Neurosci Ther 2017;23:386–394. doi: 10. Metab Brain Dis 2013;28:307–312. doi: 10.1007/s11011-013-9383-5. 1111/cns.12688. [114] Zacharias HD, Zacharias AP, Gluud LL, Morgan MY. Pharmacotherapies that [108] McMillin M, Grant S, Frampton G, Petrescu AD, Williams E, Jefferson B, et al. specifically target ammonia for the prevention and treatment of hepatic Elevated circulating TGFb1 during acute liver failure activates TGFbR2 on encephalopathy in adults with cirrhosis. Cochrane Database Syst Rev cortical neurons and exacerbates neuroinflammation and hepatic encephal- 2019;6:CD012334. doi: 10.1002/14651858.CD012334.pub2. opathy in mice. J Neuroinflammation 2019;16:69. doi: 10.1186/s12974- [115] Wijarnpreecha K, Werlang M, Panjawatanan P, Kroner PT, Cheungpasitporn 019-1455-y. W, Lukens FJ, et al. Association between sarcopenia and hepatic encephal- [109] Malaguarnera M, Llansola M, Balzano T, Gómez-Giménez B, Antúnez-Muñoz opathy: A systematic review and meta-analysis. Ann Hepatol 2019. doi: 10. C, Martínez-Alarcón N, et al. Bicuculline reduces neuroinflammation in hip- 1016/j.aohep.2019.06.007. pocampus and improves spatial learning and anxiety in hyperammonemic [116] Piotrowski D, Boron-Kaczmarska A. Bacterial infections and hepatic ence- rats. Role of glutamate receptors. Front Pharmacol 2019;10:132. doi: 10. phalopathy in liver cirrhosis-prophylaxis and treatment. Adv Med Sci 2017; 3389/fphar.2019.00132. 62:345–356. doi: 10.1016/j.advms.2016.11.009.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 352–361 361 Review Article

Pathogenesis of Insulin Resistance and Atherogenic Dyslipidemia in Nonalcoholic Fatty Liver Disease

Daud H. Akhtar1, Umair Iqbal*2, Luis Miguel Vazquez-Montesino3, Brittany B. Dennis4,5 and Aijaz Ahmed5

1Department of Medicine, University of British Columbia Faculty of Medicine, Vancouver BC, Canada; 2Department of Medicine, Geisinger Commonwealth School of Medicine, Danville, PA, USA; 3Department of Medicine, Stanford University School of Medicine, Stanford, CA, USA; 4Department of Medicine, Michael G. DeGroote School of Medicine, McMaster University, Hamilton ON, Canada; 5Division of Gastroenterology and Hepatology, Stanford University School of Medicine, Stanford, CA, USA

Abstract chronic liver disease in the developed world, with an estimated global prevalence of around 25%.1–3 NAFLD may be Nonalcoholic fatty liver disease (NAFLD) is the most common associated with features of metabolic syndrome, including cause of chronic liver disease in the developed world, with a obesity, insulin resistance, and atherogenic dyslipidemia, pre- global prevalence of around 25%. NAFLD is considered to be disposing to an increased risk of progressive liver damage the hepatic manifestation of metabolic syndrome and is that can lead to nonalcoholic steatohepatitis (commonly strongly associated with obesity, insulin resistance and dysli- known as NASH), cirrhosis, diabetes and premature cardio- pidemia. Insulin resistance plays a pivotal role in the develop- vascular disease (CVD).4–9 Insulin resistance plays a major ment of NAFLD-related dyslipidemia, which ultimately role in the development of atherogenic dyslipidemia in NAFLD increases the risk of premature cardiovascular diseases, a and the subsequent development of premature CVD.10 In this leading cause of morbidity and mortality in patients with article, we will review the role of insulin in hepatic glucose and NAFLD. Insulin affects hepatic glucose and lipid metabolism lipid metabolism, the mechanisms that may lead to NAFLD- by hepatic or extrahepatic pathways. Aside from insulin related dyslipidemia and ectopic lipid deposition, and how resistance, several other factors also contribute to the patho- these factors contribute to cardiovascular morbidity and genesis of atherogenic dyslipidemia in patients with NAFLD. mortality. These include diet composition, gut microbiota and genetic factors, to name a few. The identification of potentially Hepatic glucose and lipid metabolism modifiable risk factors of NAFLD is of importance, so as to target those who may benefit from lifestyle changes and to Liver plays a central role in the metabolism of carbohydrates, help develop targeted therapies that decrease the risk of lipids and proteins under states of caloric excess and starva- cardiovascular diseases in patients with NAFLD. tion. A key regulator of these homeostatic processes is Citation of this article: Akhtar DH, Iqbal U, Vazquez- insulin, which exerts its effects on hepatic glucose and lipid Montesino LM, Dennis BB, Ahmed A. Pathogenesis of insulin metabolism through a variety of different hepatic (direct) and resistance and atherogenic dyslipidemia in nonalcoholic extrahepatic (indirect) pathways. Fig. 1 illustrates these fatty liver disease. J Clin Transl Hepatol 2019;7(4):362– important pathways. 370. doi: 10.14218/JCTH.2019.00028. Hepatic glucose metabolism

Direct insulin-mediated regulation of hepatic glucose metab- Introduction olism begins with the activation of insulin receptor tyrosine kinase (IRTK), phosphoinositide-dependent kinase-1 Nonalcoholic fatty liver disease (NAFLD) is a clinical spectrum (referred to as PDK1) and the mammalian target of rapamy- of disease severity, ranging from isolated fatty infiltration of cin (referred to as mTORC)-2, and ends with protein kinase B the liver (simple hepatic steatosis) to lobular inflammation, (known as AKT) phosphorylation.11–13 Insulin, via this balloon degeneration, hepatic fibrosis, and cirrhosis. With pathway, decreases hepatic gluconeogenesis by activating epidemiological shifts due to lifestyle changes and modern- hepatocyte insulin receptors, which in turn trigger hepatic ization, NAFLD has become the most common cause of glycogen synthesis and downregulate the transcription of glu- coneogenic enzymes through forkhead box protein O1 Keywords: Nonalcoholic fatty liver disease; Insulin resistance; Cardiovascular (referred to as FOXO1) phosphorylation.14,15 Indirectly, diseases; Dyslipidemia. insulin further inhibits lipolysis in white adipose tissue, result- Abbreviations: ChREBP, carbohydrate response element binding protein; CVD, cardiovascular disease; DAG, diacylglycerol; DNL, de novo lipogenesis; IMCL, ing in decreased fatty acid delivery to the liver by reducing intramyocellular lipid; IRTK, insulin receptor tyrosine kinase; NAFLD, nonalcoholic hepatic acetyl-CoA content, an allosteric activator of pyruvate fatty liver disease; PKC, protein kinase C. carboxylase. This occurs in conjunction with insulin-mediated Received: 8 July 2019; Revised: 8 October 2019; Accepted: 8 November 2019 suppression of glycerol conversion to glucose and delivery to *Correspondence to: Umair Iqbal, Department of Medicine, Geisinger Common- 16–18 wealth School of Medicine, Danville, PA 17821, USA. Tel: +1-570-271-6211, the liver. The direct and indirect actions of insulin may E-mail: [email protected] play a role in different metabolic states, with glycogen replete

362 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 362–370

Table 1. Summary of factors associated with atherogenic dyslipidemia in NAFLD

Factors involved in atherogenic dyslipidemia Summary

Dietary content Increases DNL and increases levels of hepatic inflammation Skeletal insulin Increases hepatic DNL resistance Liver insulin resistance Increases hepatic DNL and ectopic lipid deposition Adipose insulin Increases inflammation, resistance lipolysis, and incomplete fatty acid oxidation Genetic factors Unclear, but thought to be affected by genetic alterations that predispose to pro- inflammatory and profibrogenic mechanisms

Fig. 1. Effects of Insulin on Glucose and Lipid metabolism. Gut microbiota Increases inflammation and increases production and absorption of gut-derived fatty states favoring direct insulin action and glycogen deplete acids, while also being – states favoring the indirect effects of insulin.19 20 associated with SIBO

Hepatic lipid metabolism Abbreviations: DNL, de novo lipogenesis; NAFLD, nonalcoholic fatty liver disease; SIBO, small intestinal bacterial overgrowth. Insulin directly regulates hepatic lipid metabolism by increasing mRNA expression of nuclear transcription factors, such as hepatic insulin resistance, as studies have shown that sterol regulatory element-binding protein-1c (referred to as hepatic triglyceride synthesis is dependent on the rate of SREBP-1c), which play a key role in the regulation of de novo fatty acid delivery and independent of hepatocellular insulin lipogenesis (DNL) and triglyceride synthesis. This effect is signaling; on the contrary, DNL is insulin-dependent.27 amplified by an insulin-mediated increase in mTORC1 activity, In addition to excess caloric consumption and high-fat which also activates SREBP1c mRNA expression and uncou- diets, simple sugars, such as fructose, can also promote ples lipogenesis from gluconeogenesis. Indirectly, insulin acts lipogenesis. Fructose is exclusively metabolized by the liver peripherally by stimulating the uptake of fatty acids into and thus, is converted into triglycerides via DNL. At the cellular muscle and adipose tissue that are produced by lipoprotein level, fructose and glucose both increase SREBP1c expression 21 lipase-mediated triglyceride hydrolysis. and enlist lipogenesis-favoring transcription factors, such as carbohydrate response element binding protein (ChREBP), Atherogenic dyslipidemia and ectopic lipid deposition peroxisome proliferator-activated receptor (referred to as PPARg) coactivator 1-b, and liver X-receptor. Fructose- Atherogenic dyslipidemia and ectopic lipid deposition in containing beverages have also been associated with the 28,29 NAFLD are multifactorial and occur broadly as a result of development of hyperlipidemia and insulin resistance. a diet, skeletal muscle, hepatic and adipose insulin resistance, High-fructose diets also reduce PPAR activity and stimulate and genetic factors. Table 1 summarizes these factors. the expression of inflammatory regulators, such as nuclear factor kappa-light-chain-enhancer of activated B cells (referred to as NF-kB) and c-Jun N-terminal kinases (referred Dietary content to as JNKs) which inhibit the phosphorylation of insulin receptor substrate-1 (referred to as IRS-1). This leads to insulin Diet may strongly influence the development of atherogenic resistance and intrahepatic inflammation. Indeed, this is dyslipidemia. Specifically, injurious effects of excess caloric evident in rats that develop hepatic steatosis and hepatic fib- intake are implicated in NAFLD, with one study demonstrating rosis when fed high-fructose diets.30–32 To this end, multiple that patients with obesity and NAFLD may consume more studies have demonstrated that diets with high saturated fat calories than healthy lean individuals; in fact, a study exam- and simple carbohydrate content can significantly predispose ining patients with metabolic syndrome demonstrated a 3- individuals to a higher risk for ectopic lipid deposition and athe- fold higher rate of DNL.22,23 Furthermore, studies have shown rogenic dyslipidemia despite hepatic insulin resistance.33–35 that short-term caloric restriction can normalize hepatic insulin sensitivity and reverse NAFLD, thereby suggesting a causal role of ectopic lipids in insulin resistant states, such as Insulin resistance NAFLD.24–26 It should be noted that excess fatty acid reser- voirs (due to high-fat diets or increased caloric intake) The human body encompasses complex metabolic pathways promote hepatic lipid synthesis despite the presence of that are fundamental to the interplay between glucose and

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 362–370 363 Akhtar D.H. et al: NAFLD, insulin resistance and CVD insulin at the molecular level. Dysfunction of this interplay in blocks the AKT pathway to cause impaired insulin signal- states of nutritional imbalance is evident in metabolic syn- ing.46–48 This has been validated in genetic mouse models drome. The inability of insulin to promote glucose uptake in with IRTK and glucose transporter type 4 (referred to as muscle and the suppression of gluconeogenesis in the liver GLUT4) mutations, which demonstrate a predisposition for leads to a progressive reduction in beta cell function and hepatic steatosis and increased adiposity.49,50 ultimately insulin resistance. Insulin resistance plays a major Furthermore, progression to atherogenic dyslipidemia role in the development of NAFLD and atherogenic dyslipide- through increased liver triglyceride synthesis and plasma mia, and can manifest in multiple tissues, including skeletal triglyceride concentrations with reduced high-density lipo- muscle, liver, and adipose.36 Fig. 2 illustrates the mechanism of protein concentrations were observed in young, lean healthy insulin resistance in skeletal muscle, liver and adipose tissue. humans with skeletal muscle insulin resistance. This illustrates that ingested glucose is diverted to the liver and not utilized by skeletal muscle in insulin-resistant states.49 These Skeletal muscle insulin resistance results suggest that in insulin responsive states, glucose is stored as muscle and liver glycogen; however, in insulin Skeletal muscle insulin resistance is caused by the accumu- resistant states, defects in glycogen storage in muscle result lation of intramyocellular lipid (IMCL) content, which impairs in glucose being diverted to the liver. This resultant hyper- insulin-stimulated glucose transport and glycogen synthesis. glycemia and hyperinsulinemia further stimulate the Studies examining skeletal muscle in the context of obesity enzymes involved in hepatic DNL, thereby promoting the con- have demonstrated that IMCL content predicts insulin resist- version of glucose into fat. Ultimately, this increased plasma ance more accurately than fat mass in varying sedentary – triglyceride level is the precursor to atherogenic dyslipidemia populations.37 39 Additional studies have further demonstrated that leads to premature CVD and NAFLD.10,19,50 that glucose transport is the rate-limiting step responsible for impaired insulin-stimulated glycogen synthesis and that IMCL accumulates when there is a mismatch between lipid oxidation Liver insulin resistance and lipid delivery in muscle cells.16,40–44 These findings suggest an IMCL metabolite, such as diacylglycerol (DAG), acts as a Many different models hypothesize on the mechanisms of potential mediator of insulin resistance, with high-fat fed and hepatic insulin resistance. Initial studies involving patients lipid-infused rats demonstrating DAG accumulation.45 Specifi- with generalized lipodystrophy and lipodystrophic mouse cally, DAG accumulation has been shown to promote increased models showed significant hepatic insulin resistance and protein kinase C (PKC) translocation in skeletal muscle, result- recovery of insulin action post- leptin therapy and fat trans- ing in the phosphorylation of IRS-1 at serine 1101, and this plantation respectively.51,52 These findings were augmented

Fig. 2. Mechanism of insulin resistance in skeletal muscle, liver and adipose tissue.

364 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 362–370 Akhtar D.H. et al: NAFLD, insulin resistance and CVD by Kim et al.,53 who demonstrated that the overexpression of reactive oxygen species due to the incomplete oxidation of lipoprotein lipase and short-term high-fat diets in mouse fatty acids, which impairs insulin signaling.81 This resultant models resulted in hepatic steatosis and hepatic insulin resist- subclinical inflammation plays a role in the pathogenesis of ance without muscle lipid accumulation or peripheral insulin NAFLD and may contribute to the development of atherogenic resistance. This highlights the specific association between dyslipidemia with high-sensitivity C-reactive protein, a strong ectopic lipid deposition in the liver and hepatic insulin cardiovascular risk predictor, also shown to be elevated in resistance. patients with NAFLD.82 In combination, these factors all play At the cellular level, current evidence indicates hepatic a role in disrupting the effect of insulin on adipose tissue and insulin resistance may share similarities with skeletal muscle may predispose to atherogenic-favoring metabolic states that insulin resistance, with evidence supporting a role for DAG precede NAFLD. accumulation and PKC activation.10,54 Specifically, studies examining fat-fed rats have shown that the intrahepatic accumulation of DAG promotes PKCe translocation in the liver and Genetic factors results in the increased phosphorylation of ISR-1 at threonine (Thr) 1160 (Thr1150 in mice). This subsequently results in Genetic variations in different populations have been shown impaired IRTK activation.55–59 Furthermore, these findings to influence the risk of NAFLD. Genome-wide association have been validated in studies in human patients undergoing studies have identified multiple genes with numerous poly- bariatric surgery. Lipidomic analysis in these patients demon- morphisms that may be responsible for predisposing individ- strate increases in specific lipid metabolites, such as DAGs, uals to NAFLD and atherogenic dyslipidemia. The mechanism while other studies show that hepatic DAG content and PKCe by which these genes influence the pathogenesis of NAFLD is activity are the strongest predictors of insulin sensitivity. unclear and may include inflammatory, profibrogenic, or These studies did not demonstrate any association energy regulating pathways. A recent systematic review between insulin sensitivity and other factors implicated in highlighted six genes that were replicated in more than one hepatic insulin resistance, such as ceramide content, endo- histologically characterized cohort: patatin-like phospholi- plasmic reticulum stress markers, or inflammatory cytokine pase domain-containing protein 3 (referred to as PNPLA3), concentrations.60–63 Studies with mice lacking glycogen syn- apolipoprotein E (referred to as APOE), superoxide dismutase thase demonstrate increased lipogenesis and NAFLD in the 2, mitochondrial (referred to as SOD2), TNF, transmembrane presence of hepatic insulin resistance.19,64 These findings 6 superfamily member 2 (referred to as TM6SF2), and may suggest a mechanism by which hepatic insulin resistance glucokinase regulator (referred to as GCKR). Only PNPLA3 impairs glycogen synthesis and directs glucose into lipogenic (rs738409 variant) was consistently associated with NAFLD pathways that lead to atherogenic dyslipidemia and the pro- 83 motion of NAFLD. susceptibility, from among these genes. Polymorphisms in APOC3 have also been associated with NAFLD, which may predispose to dyslipidemia and insulin resistance through the Adipose insulin resistance – inhibition of lipoprotein lipase activity.84 86 The membrane- bound O-acyltransferase domain-containing 7 gene and Insulin primarily functions to suppress lipolysis and promote transmembrane channel-like 4 gene (referred to as lipid and glucose uptake in adipocytes. Adipose insulin MBOAT7-TMC4) and SCL16A11 are other examples of poly- resistance may be explained by reduced IRTK activity in 87–91 combination with decreased plasma membrane insulin recep- morphisms that may be associated with NAFLD. Further tor content. Interestingly, weight loss corrected both adipose research is warranted to elucidate the role of genomic varia- insulin resistance and defective adipocyte insulin receptor tions in NAFLD. kinase activity. However, further research is warranted to 54 better understand these mechanisms. An important regu- Gut microbiota lator of adipose metabolism is ChREBP. ChREBP is directly related to insulin sensitivity in humans and has been impli- The gut microbiota is a growing area of research with regards cated in the synthesis of fatty acid esters that are associated to the development of metabolic diseases and associations 65,66 with improved glucose tolerance. Furthermore, patients with insulin resistance. Alterations in normal gut flora can lead with NASH and severe insulin resistance have lower levels of to the increased production and absorption of gut short-chain ChREBP on liver biopsy.67 Additionally, adipocytes can secrete fatty acids, altered dietary choline metabolism, altered bile adipocytokines, such as leptin and resistin (inflammatory), acid pools, and changes in gut permeability.92 These changes and adiponectin (anti-inflammatory), which exert their are evident in disease states such as small intestinal bacterial affects through cytokine signaling pathways. These adipocy- overgrowth (commonly known as SIBO), where there is a tokines are associated with increased insulin resistance and 32,68–71 higher prevalence of NAFLD. Specifically, SIBO induces sensitivity respectively. Other factors that impact ectopic lipid deposition and hepatic expression of toll-like receptor 4 (referred to as 93–95 affect the role of insulin in adipose lipolysis suppression TLR4), and release of IL-8 and TNF-alpha. 96 include inflammatory agents involved in chemokine signaling, Pederson et al. demonstrated an increased prevalence of such as tumor necrosis factor (referred to as TNF)-alpha, Prevotella copri (P. copri) and Bacteriodes vulgatus in insulin- interleukin (referred to as IL)-1B and interferon gamma resistant subjects that was supported by mouse models in (referred to as IFNg), central nervous system-mediated path- which P. copri had induced insulin resistance. These findings ways involving the hypothalamus, and reactive oxygen implicate gut microbiota with insulin resistance and suggest species produced as a result of increased mitochondrial fat that gut microbiomes can influence disease severity in oxidation.72–80 Conversely, abnormal hepatic mitochondrial NAFLD. Further research is underway to better elucidate the function demonstrated in steatohepatitis can also result in role of gut microbiota on the pathogenesis of NAFLD.97,98

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 362–370 365 Akhtar D.H. et al: NAFLD, insulin resistance and CVD

Table 2. Summary of studies examining the associations between NAFLD and cardiovascular disease

Number of studies in meta- Study Study description analysis Results

Ampuero et al. Meta-analysis showing an 14 studies were included, of NAFLD showed a higher (2015)99 association between NAFLD and which 10 were aimed at the prevalence of pathological carotid subclinical atherosclerosis and presence of subclinical intima-media thickness and CAD atherosclerosis and 4 studies at carotid plaques; with regards to the presence of CAD CAD, subjects with NAFLD were more likely to have CAD Zhou et al. Meta-analysis showing that 26 studies were included in this The presence of NAFLD (2018)100 NAFLD contributes to subclinical meta-analysis, with a total demonstrated a higher risk of atherosclerosis participant size of 85,395 increased carotid intima-media (including 29,493 with NAFLD) thickness/plaques, arterial stiffness, coronary artery calcification, and endothelial dysfunction Jaruvongvanich Meta-analysis showing an 12 studies involving 42,410 Mean coronary artery calcium et al. (2016)101 association between NAFLD and subjects were included in this scoring was significantly higher in coronary artery calcification meta-analysis subjects with NAFLD and higher aspartate aminotransferase levels were also associated with increased subclinical atherosclerosis Fan et al. Meta-analysis showing an 11 observational studies were NAFLD was associated with a (2016)102 association between NAFLD and included, in which endothelial reduction in brachial artery flow- impaired endothelial function function was compared between mediated dilation NAFLD patients and healthy controls Cai et al. Meta-analysis showing an 9 studies were included The carotid intima-media (2015)103 association between NAFLD and involving 2446 subjects (925 thickness was increased 0.16 mm carotid atherosclerosis with NAFLD) with a 95% confidence interval (0.11, 0.21) with a 3.73-times greater risk for carotid plaque Valbusa et al. Prospective study showing Cohort study with a total sample Patients with NAFLD had (2016)104 NAFLD is an independent risk of 314 elderly patients with remarkably higher 1-year all- factor for acute heart failure confirmed consecutive cause and cardiac readmissions admissions for acute heart rehospitalizations compared with failure their counterparts without NAFLD Hung et al. Cross-sectional analysis Observational study of 31,116 Mild, moderate, and severe NAFLD (2015)108 showing an association between patients were included was associated with increases in NAFLD and QT prolongation in QTc interval (2.55, 6.59 and 12.14 the general population ms) compared with no NAFLD (p < 0.001); NAFLD was also associated with an increased risk of QTc prolongation Mantovani et al. Retrospective cohort study Retrospective study evaluating a Overall, patients with NAFLD had a (2017)109 showing an independent cohort of 751 patients with type higher prevalence of persistent association between NAFLD and 2 diabetes mellitus during the heart block than those without increased risk of heart block in years 2007-2014 NAFLD (31.3% vs. 16.7%, p < hospitalized patients with type 2 0.001). NAFLD was associated diabetes mellitus with a three-fold increased risk of prevalent heart block Targher et al. Retrospective cohort study 702 patients with type 2 NAFLD was associated with an (2013)110 showing an association between diabetes mellitus were increased risk of prevalent atrial NAFLD and increased risk of evaluated for the prevalence of fibrillation atrial fibrillation in hospitalized atrial fibrillation during 2007- patients with type 2 diabetes 2011 mellitus

(continued)

366 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 362–370 Akhtar D.H. et al: NAFLD, insulin resistance and CVD

Table 2. (continued ) Number of studies in meta- Study Study description analysis Results

Targher et al. Meta-analysis showing an 16 observational prospective Patients with NAFLD had a higher (2016)113 association between NAFLD and and retrospective studies with risk of fatal and/or non-fatal CVD an increased incidence of CVD 34,043 individuals over a events that those without NAFLD, median period of 6.9 years were with patients with more severe assessed in this meta-analysis NAFLD more likely to develop fatal and non-fatal events Wu et al. Meta-analysis showing an 34 studies with a total of NAFLD was not associated with (2016)114 association of NAFLD with major 164,494 participants were overall mortality and CVD adverse CVD events included; specifically, 21 cross- mortality but was, however, sectional studies and 13 cohort associated with an increased risk studies were examined in this of prevalence and incidence of meta-analysis CVD Hagström et al. Cohort study showing patients A large cohort of 603 biopsy- Patients with NAFLD are at an (2019)116 with NAFLD are at an increased proven NAFLD patients free of increased risk for CVD compared risk for CVD baseline CVD were cross-linked to matched controls but to national registries to evaluate histological parameters do not histological and traditional seem to independently predict this cardiovascular risk factors as risk predictors of cardiovascular outcomes

Abbreviations: CAD, cardiac artery disease; CVD, cardiovascular disease; NAFLD, nonalcoholic fatty liver disease.

Cardiovascular risk adipose tissue. Graner et al.105 illustrated this phenomenon in nondiabetic patients with NAFLD who exhibited higher Patients with NAFLD are at an increased risk of mortality from amounts of epicardial adipose tissue. This increased epicar- cardiovascular events, including atherosclerosis, systolic or dial adipose tissue is associated with sympathovagal imbal- diastolic cardiac dysfunction, conduction abnormalities, and ances that can predispose to cardiac arrhythmias, such as arrhythmias. Subclinical atherosclerosis begins as a result of first degree heart block prolongation and atrial fibrilla- 106–111 endothelial dysfunction and can eventually manifest as tion. A prospective cohort study following diabetic carotid disease. Carotid disease can manifest as increased patients for 10 years showed that patients with NAFLD devel- 112 carotid intima-media thickness and the presence of carotid oped atrial fibrillation more often than the controls. These plaque on ultrasound, coronary artery calcification evident on findings suggest a potential role for screening and surveil- computerized tomography, left ventricular hypertrophy seen lance of atherogenic dyslipidemia in patients with NAFLD. on electrocardiogram and echocardiogram, and peripheral Table 2 summarizes these studies mentioned above. arterial disease seen on ankle-brachial pressure index.99 A plethora of studies report an association between NAFLD Cardiovascular outcomes and mortality in NAFLD and these markers of preclinical atherosclerosis, with multiple systematic reviews and meta-analyses highlighting CVD has been incriminated as the leading cause of mortality the association between NAFLD and carotid-artery intima- among patients with NAFLD, even surpassing the mortality media-thickness, coronary artery calcification and endothelial associated with chronic liver disease. Recently, a meta- – dysfunction, independent of traditional risk factors.100 103 analysis by Targher et al.113 demonstrated that NAFLD is A recent systematic review and meta-analysis by Zhou associated with an increased risk of both fatal and nonfatal et al.100 examining 26 observational studies further validated CVD events. Another meta-analysis by Wu et al.114 showed that the presence of NAFLD predisposes to a significantly that NAFLD was associated with an increased prevalence of higher risk of increased carotid artery intima-media thick- adverse cardiovascular events but did not show an associa- ness/plaques, arterial stiffness, coronary artery calcification tion with overall or CVD mortality; similarly, NASH was not and endothelial dysfunction, with odds ratios (95% confi- associated with overall or CVD mortality but was associated dence intervals) of 1.74 (1.47–2.06), 1.56 (1.24–1.96), 1.4 with an increased incidence of CVD. This increase in incidence (1.22–1.60), and 3.73 (0.99–14.09) respectively. Case- may correlate with another systematic review by Dulai et al. control studies have also reported strong associations of that demonstrated an increased risk of mortality with increas- NAFLD with early changes in left ventricular morphology ing fibrosis severity in NAFLD.115 A recent Swedish cohort and/or diastolic dysfunction with insulin resistance a likely study in liver biopsy-diagnosed NAFLD with the longest docu- key contributor. A prospective one-year follow-up study dem- mented follow-up time (of over 18 years) supported these onstrated that NAFLD is an independent risk factor for acute findings by showing that patients with NAFLD are at an heart failure readmission rates.104 increased risk for CVD.116 Besides cardiac dysfunction, significant ectopic lipid dep- Although more studies are needed to confirm a causal osition seen in NAFLD can present as an increase in epicardial relationship between NAFLD and CVD, these findings highlight

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 362–370 367 Akhtar D.H. et al: NAFLD, insulin resistance and CVD the importance of risk factor modification and interventions [7] Firneisz G. Non-alcoholic fatty liver disease and type 2 diabetes mellitus: the aimed at preventing CVD progression in NAFLD. Table 2 sum- liver disease of our age? World J Gastroenterol 2014;20:9072–9089. doi: 10.3748/wjg.v20.i27.9072. marizes below the current literature evaluating the associa- [8] de Alwis NM, Day CP. Non-alcoholic fatty liver disease: the mist gradually tion between NAFLD and CVD. clears. J Hepatol 2008;48 Suppl 1:S104–S112. doi: 10.1016/j.jhep.2008. 01.009. Conclusions [9] Targher G, Marra F, Marchesini G. Increased risk of cardiovascular disease in non-alcoholic fatty liver disease: causal effect or epiphenomenon? Diabeto- logia 2008;51:1947–1953. doi: 10.1007/s00125-008-1135-4. The pathophysiological pathways that link NAFLD to prema- [10] Samuel VT, Shulman GI. The pathogenesis of insulin resistance: integrating ture CVD events result from the development of atherogenic signaling pathways and substrate flux. J Clin Invest 2016;126:12–22. doi: dyslipidemia and ectopic lipid deposition. It involves an 10.1172/JCI77812. [11] Guertin DA, Sabatini DM. Defining the role of mTOR in cancer. Cancer Cell altered hepatic metabolism of lipids and glucose due to 2007;12:9–22. doi: 10.1016/j.ccr.2007.05.008. insulin resistance and a constellation pathogenetic factors [12] Alessi DR, James SR, Downes CP, Holmes AB, Gaffney PR, Reese CB, et al. associated with the gut microbiome, genetics, and individual Characterization of a 3-phosphoinositide-dependent protein kinase which lifestyle. The significant association between NAFLD and CVD phosphorylates and activates protein kinase Balpha. Curr Biol 1997;7: 261–269. doi: 10.1016/s0960-9822(06)00122-9. warrants screening of NAFLD and its associated risk factors in [13] Stephens L, Anderson K, Stokoe D, Erdjument-Bromage H, Painter GF, high-risk patients in order to effectively intervene, with a Holmes AB, et al. Protein kinase B kinases that mediate phosphatidylinositol focus on novel approaches that target insulin resistance and 3,4,5-trisphosphate-dependent activation of protein kinase B. Science dyslipidemia. These novel approaches encompass both non- 1998;279:710–714. doi: 10.1126/science.279.5351.710. [14] Petersen KF, Laurent D, Rothman DL, Cline GW, Shulman GI. Mechanism by pharmacological approaches, such as weight loss, lifestyle which glucose and insulin inhibit net hepatic glycogenolysis in humans. J modifications and bariatric surgery, as well as pharmacologic Clin Invest 1998;101:1203–1209. doi: 10.1172/JCI579. targets aimed at improving insulin sensitivity. A better under- [15] Nakae J, Kitamura T, Silver DL, Accili D. The forkhead transcription factor standing of the pathogenesis of insulin resistance will allow Foxo1 (Fkhr) confers insulin sensitivity onto glucose-6-phosphatase expression. J Clin Invest 2001;108:1359–1367. doi: 10.1172/JCI12876. the development of newer treatment options that can address [16] Previs SF, Cline GW, Shulman GI. A critical evaluation of mass isotopomer the growing health concerns posed by NAFLD and insulin distribution analysis of gluconeogenesis in vivo. Am J Physiol 1999;277: resistance. E154–E160. doi: 10.1152/ajpendo.1999.277.1.E154. [17] Perry RJ, Zhang XM, Zhang D, Kumashiro N, Camporez JP, Cline GW, et al. Leptin reverses diabetes by suppression of the hypothalamic-pituitary- Acknowledgments adrenal axis. Nat Med 2014;20:759–763. doi: 10.1038/nm.3579. [18] Perry RJ, Camporez JG, Kursawe R, Titchenell PM, Zhang D, Perry CJ, et al. Hepatic acetyl CoA links adipose tissue inflammation to hepatic insulin The authors would like to acknowledge everyone who con- resistance and type 2 diabetes. Cell 2015;160:745–758. doi: 10.1016/j. tributed to the completion of this project. cell.2015.01.012. [19] Samuel VT, Shulman GI. Nonalcoholic fatty liver disease as a nexus of metabolic and hepatic diseases. Cell Metab 2018;27:22–41. doi: 10.1016/j. cmet.2017.08.002. Conflict of interest [20] Wan M, Leavens KF, Hunter RW, Koren S, von Wilamowitz-Moellendorff A, Lu M, et al. A noncanonical, GSK3-independent pathway controls postpran- The authors have no conflict of interests related to this dial hepatic glycogen deposition. Cell Metab 2013;18:99–105. doi: 10. publication. 1016/j.cmet.2013.06.001. [21] Donnelly KL, Smith CI, Schwarzenberg SJ, Jessurun J, Boldt MD, Parks EJ. Sources of fatty acids stored in liver and secreted via lipoproteins in patients with nonalcoholic fatty liver disease. J Clin Invest 2005;115:1343–1351. Author contributions doi: 10.1172/JCI23621. [22] Wehmeyer MH, Zyriax BC, Jagemann B, Roth E, Windler E, Schulze Zur Wiesch J, et al. Nonalcoholic fatty liver disease is associated with excessive Study concept and design (AA and BBD), literature search and calorie intake rather than a distinctive dietary pattern. Medicine (Baltimore) drafting of the manuscript (DHA, UI, and LMV), critical 2016;95:e3887. doi: 10.1097/MD.0000000000003887. revision of manuscript for important intellectual content, [23] Lambert JE, Ramos-Roman MA, Browning JD, Parks EJ. Increased de novo lipogenesis is a distinct characteristic of individuals with nonalcoholic fatty senior authorship guidance and supervision (AA and BBD). liver disease. Gastroenterology 2014;146:726–735. doi: 10.1053/j.gastro. 2013.11.049. [24] Petersen KF, Dufour S, Befroy D, Lehrke M, Hendler RE, Shulman GI. Rever- References sal of nonalcoholic hepatic steatosis, hepatic insulin resistance, and hyperglycemia by moderate weight reduction in patients with type 2 diabetes. [1] Younossi Z, Henry L. Contribution of alcoholic and nonalcoholic fatty liver Diabetes 2005;54:603–608. doi: 10.2337/diabetes.54.3.603. disease to the burden of liver-related morbidity and mortality. Gastroenter- [25] Taylor R, Leslie WS, Barnes AC, Brosnahan N, Thom G, McCombie L, et al. ology 2016;150:1778–1785. doi: 10.1053/j.gastro.2016.03.005. Clinical and metabolic features of the randomised controlled Diabetes [2] Younossi Z, Anstee QM, Marietti M, Hardy T, Henry L, Eslam M, et al. Global Remission Clinical Trial (DiRECT) cohort. Diabetologia 2018;61:589–598. burden of NAFLD and NASH: trends, predictions, risk factors and preven- doi: 10.1007/s00125-017-4503-0. tion. Nat Rev Gastroenterol Hepatol 2018;15:11–20. doi: 10.1038/nrgas- [26] Perry RJ, Peng L, Cline GW, Wang Y, Rabin-Court A, Song JD, et al. Mecha- tro.2017.109. nisms by which a very-low-calorie diet reverses hyperglycemia in a rat [3] Younossi ZM, Stepanova M, Afendy M, Fang Y, Younossi Y, Mir H, et al. model of type 2 diabetes. Cell Metab 2018;27:210–217.e3. doi: 10. Changes in the prevalence of the most common causes of chronic liver dis- 1016/j.cmet.2017.10.004. eases in the United States from 1988 to 2008. Clin Gastroenterol Hepatol [27] Vatner DF, Majumdar SK, Kumashiro N, Petersen MC, Rahimi Y, Gattu AK, 2011;9:524–530.e1. doi: 10.1016/j.cgh.2011.03.020. et al. Insulin-independent regulation of hepatic triglyceride synthesis by [4] Marchesini G, Moscatiello S, Di Domizio S, Forlani G. Obesity-associated fatty acids. Proc Natl Acad Sci U S A 2015;112:1143–1148. doi: 10. liver disease. J Clin Endocrinol Metab 2008;93:S74–S80. doi: 10.1210/jc. 1073/pnas.1423952112. 2008-1399. [28] Hudgins LC, Parker TS, Levine DM, Hellerstein MK. A dual sugar challenge [5] Lallukka S, Yki-Järvinen H. Non-alcoholic fatty liver disease and risk of type test for lipogenic sensitivity to dietary fructose. J Clin Endocrinol Metab 2 diabetes. Best Pract Res Clin Endocrinol Metab 2016;30:385–395. doi: 10. 2011;96:861–868. doi: 10.1210/jc.2010-2007. 1016/j.beem.2016.06.006. [29] Lecoultre V, Egli L, Carrel G, Theytaz F, Kreis R, Schneiter P, et al. Effects of [6] Yki-Järvinen H. Non-alcoholic fatty liver disease as a cause and a conse- fructose and glucose overfeeding on hepatic insulin sensitivity and intra- quence of metabolic syndrome. Lancet Diabetes Endocrinol 2014;2:901– hepatic lipids in healthy humans. Obesity (Silver Spring) 2013;21:782– 910. doi: 10.1016/S2213-8587(14)70032-4. 785. doi: 10.1002/oby.20377.

368 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 362–370 Akhtar D.H. et al: NAFLD, insulin resistance and CVD

[30] Spruss A, Bergheim I. Dietary fructose and intestinal barrier: potential risk [52] Kim JK, Gavrilova O, Chen Y, Reitman ML, Shulman GI. Mechanism of insulin factor in the pathogenesis of nonalcoholic fatty liver disease. J Nutr Biochem resistance in A-ZIP/F-1 fatless mice. J Biol Chem 2000;275:8456–8460. 2009;20:657–662. doi: 10.1016/j.jnutbio.2009.05.006. doi: 10.1074/jbc.275.12.8456. [31] Roglans N, Vilà L, Farré M, Alegret M, Sánchez RM, Vázquez-Carrera M, et al. [53] Kim JK, Fillmore JJ, Chen Y, Yu C, Moore IK, Pypaert M, et al. Tissue-specific Impairment of hepatic Stat-3 activation and reduction of PPARalpha activity overexpression of lipoprotein lipase causes tissue-specific insulin resistin fructose-fed rats. Hepatology 2007;45:778–788. doi: 10.1002/hep. ance. Proc Natl Acad Sci U S A 2001;98:7522–7527. doi: 10.1073/pnas. 21499. 121164498. [32] Tilg H, Moschen AR. Inflammatory mechanisms in the regulation of insulin [54] Petersen MC, Shulman GI. Mechanisms of insulin action and insulin resist- resistance. Mol Med 2008;14:222–231. doi: 10.2119/2007-00119.Tilg. ance. Physiol Rev 2018;98:2133–2223. doi: 10.1152/physrev.00063. [33] Assy N, Nasser G, Kamayse I, Nseir W, Beniashvili Z, Djibre A, et al. 2017. Soft drink consumption linked with fatty liver in the absence of traditional [55] Qu X, Seale JP, Donnelly R. Tissue and isoform-selective activation of risk factors. Can J Gastroenterol 2008;22:811–816. doi: 10.1155/ protein kinase C in insulin-resistant obese Zucker rats - effects of feeding. 2008/810961. J Endocrinol 1999;162:207–214. doi: 10.1677/joe.0.1620207. [34] Ouyang X, Cirillo P, Sautin Y, McCall S, Bruchette JL, Diehl AM, et al. Fructose [56] Samuel VT, Liu ZX, Qu X, Elder BD, Bilz S, Befroy D, et al. Mechanism of consumption as a risk factor for non-alcoholic fatty liver disease. J Hepatol hepatic insulin resistance in non-alcoholic fatty liver disease. J Biol Chem 2008;48:993–999. doi: 10.1016/j.jhep.2008.02.011. 2004;279:32345–32353. doi: 10.1074/jbc.M313478200. [35] Zelber-Sagi S, Nitzan-Kaluski D, Goldsmith R, Webb M, Blendis L, Halpern Z, [57] Samuel VT, Liu ZX, Wang A, Beddow SA, Geisler JG, Kahn M, et al. Inhibition et al. Long term nutritional intake and the risk for non-alcoholic fatty liver of protein kinase Cepsilon prevents hepatic insulin resistance in nonalco- – disease (NAFLD): a population based study. J Hepatol 2007;47:711 717. holic fatty liver disease. J Clin Invest 2007;117:739–745. doi: 10. doi: 10.1016/j.jhep.2007.06.020. 1172/JCI30400. [36] Shulman GI. Cellular mechanisms of insulin resistance. J Clin Invest 2000; [58] Raddatz K, Turner N, Frangioudakis G, Liao BM, Pedersen DJ, Cantley J, – 106:171 176. doi: 10.1172/JCI10583. et al. Time-dependent effects of Prkce deletion on glucose homeostasis [37] Sinha R, Dufour S, Petersen KF, LeBon V, Enoksson S, Ma YZ, et al. Assess- and hepatic lipid metabolism on dietary lipid oversupply in mice. Diabeto- ment of skeletal muscle triglyceride content by (1)H nuclear magnetic res- logia 2011;54:1447–1456. doi: 10.1007/s00125-011-2073-0. onance spectroscopy in lean and obese adolescents: relationships to insulin [59] Petersen MC, Madiraju AK, Gassaway BM, Marcel M, Nasiri AR, Butrico G, – sensitivity, total body fat, and central adiposity. Diabetes 2002;51:1022 et al. Insulin receptor Thr1160 phosphorylation mediates lipid-induced 1027. doi: 10.2337/diabetes.51.4.1022. hepatic insulin resistance. J Clin Invest 2016;126:4361–4371. doi: 10. [38] Perseghin G, Scifo P, De Cobelli F, Pagliato E, Battezzati A, Arcelloni C, et al. 1172/JCI86013. Intramyocellular triglyceride content is a determinant of in vivo insulin [60] Kumashiro N, Erion DM, Zhang D, Kahn M, Beddow SA, Chu X, et al. Cellular resistance in humans: a 1H-13C nuclear magnetic resonance spectroscopy mechanism of insulin resistance in nonalcoholic fatty liver disease. Proc Natl assessment in offspring of type 2 diabetic parents. Diabetes 1999;48: Acad Sci U S A 2011;108:16381–16385. doi: 10.1073/pnas.1113359108. 1600–1606. doi: 10.2337/diabetes.48.8.1600. [61] Magkos F, Su X, Bradley D, Fabbrini E, Conte C, Eagon JC, et al. Intrahepatic [39] Krssak M, Falk Petersen K, Dresner A, DiPietro L, Vogel SM, Rothman DL, diacylglycerol content is associated with hepatic insulin resistance in obese et al. Intramyocellular lipid concentrations are correlated with insulin sen- subjects. Gastroenterology 2012;142:1444–1446.e2. doi: 10.1053/j. sitivity in humans: a 1H NMR spectroscopy study. Diabetologia 1999;42: gastro.2012.03.003. 113–116. doi: 10.1007/s001250051123. [62] Luukkonen PK, Zhou Y, Sädevirta S, Leivonen M, Arola J, Oresi cM, et al. [40] Dresner A, Laurent D, Marcucci M, Griffin ME, Dufour S, Cline GW, et al. Hepatic ceramides dissociate steatosis and insulin resistance in patients Effects of free fatty acids on glucose transport and IRS-1-associated phos- with non-alcoholic fatty liver disease. J Hepatol 2016;64:1167–1175. doi: phatidylinositol 3-kinase activity. J Clin Invest 1999;103:253–259. doi: 10. 10.1016/j.jhep.2016.01.002. 1172/JCI5001. [63] Ter Horst KW, Gilijamse PW, Versteeg RI, Ackermans MT, Nederveen AJ, la [41] Roden M, Price TB, Perseghin G, Petersen KF, Rothman DL, Cline GW, et al. Fleur SE, et al. Hepatic diacylglycerol-associated protein kinase Ce trans- Mechanism of free fatty acid-induced insulin resistance in humans. J Clin location links hepatic steatosis to hepatic insulin resistance in humans. Cell Invest 1996;97:2859–2865. doi: 10.1172/JCI118742. Rep 2017;19:1997–2004. doi: 10.1016/j.celrep.2017.05.035. [42] Petersen KF, Hendler R, Price T, Perseghin G, Rothman DL, Held N, et al. [64] Irimia JM, Meyer CM, Segvich DM, Surendran S, DePaoli-Roach AA, Morral 13C/31P NMR studies on the mechanism of insulin resistance in obesity. N, et al. Lack of liver glycogen causes hepatic insulin resistance and stea- Diabetes 1998;47:381–386. doi: 10.2337/diabetes.47.3.381. tosis in mice. J Biol Chem 2017;292:10455–10464. doi: 10.1074/jbc. [43] Cline GW, Petersen KF, Krssak M, Shen J, Hundal RS, Trajanoski Z, et al. M117.786525. Impaired glucose transport as a cause of decreased insulin-stimulated [65] Herman MA, Peroni OD, Villoria J, Schön MR, Abumrad NA, Blüher M, et al.A muscle glycogen synthesis in type 2 diabetes. N Engl J Med 1999;341: novel ChREBP isoform in adipose tissue regulates systemic glucose metab- 240–246. doi: 10.1056/NEJM199907223410404. – [44] Perseghin G, Price TB, Petersen KF, Roden M, Cline GW, Gerow K, et al. olism. Nature 2012;484:333 338. doi: 10.1038/nature10986. Increased glucose transport-phosphorylation and muscle glycogen synthe- [66] Yore MM, Syed I, Moraes-Vieira PM, Zhang T, Herman MA, Homan EA, et al. sis after exercise training in insulin-resistant subjects. N Engl J Med 1996; Discovery of a class of endogenous mammalian lipids with anti-diabetic and – 335:1357–1362. doi: 10.1056/NEJM199610313351804. anti-inflammatory effects. Cell 2014;159:318 332. doi: 10.1016/j.cell. [45] Yu C, Chen Y, Cline GW, Zhang D, Zong H, Wang Y, et al. Mechanism by which 2014.09.035. fatty acids inhibit insulin activation of insulin receptor substrate-1 (IRS-1)- [67] Benhamed F, Denechaud PD, Lemoine M, Robichon C, Moldes M, Bertrand- associated phosphatidylinositol 3-kinase activity in muscle. J Biol Chem Michel J, et al. The lipogenic transcription factor ChREBP dissociates hepatic 2002;277:50230–50236. doi: 10.1074/jbc.M200958200. steatosis from insulin resistance in mice and humans. J Clin Invest 2012; – [46] Itani SI, Ruderman NB, Schmieder F, Boden G. Lipid-induced insulin resist- 122:2176 2194. doi: 10.1172/JCI41636. ance in human muscle is associated with changes in diacylglycerol, protein [68] Tilg H. The role of cytokines in non-alcoholic fatty liver disease. Dig Dis – kinase C, and IkappaB-alpha. Diabetes 2002;51:2005–2011. doi: 10. 2010;28:179 185. doi: 10.1159/000282083. 2337/diabetes.51.7.2005. [69] Kershaw EE, Flier JS. Adipose tissue as an endocrine organ. J Clin Endocrinol [47] Szendroedi J, Yoshimura T, Phielix E, Koliaki C, Marcucci M, Zhang D, et al. Metab 2004;89:2548–2556. doi: 10.1210/jc.2004-0395. Role of diacylglycerol activation of PKCu in lipid-induced muscle insulin [70] Asano T, Watanabe K, Kubota N, Gunji T, Omata M, Kadowaki T, et al. Adi- resistance in humans. Proc Natl Acad Sci U S A 2014;111:9597–9602. ponectin knockout mice on high fat diet develop fibrosing steatohepatitis. J doi: 10.1073/pnas.1409229111. Gastroenterol Hepatol 2009;24:1669–1676. doi: 10.1111/j.1440-1746. [48] Li Y, Soos TJ, Li X, Wu J, Degennaro M, Sun X, et al. Protein kinase C Theta 2009.06039.x. inhibits insulin signaling by phosphorylating IRS1 at Ser(1101). J Biol Chem [71] Nawrocki AR, Rajala MW, Tomas E, Pajvani UB, Saha AK, Trumbauer ME, 2004;279:45304–45307. doi: 10.1074/jbc.C400186200. et al. Mice lacking adiponectin show decreased hepatic insulin sensitivity [49] Petersen KF, Dufour S, Savage DB, Bilz S, Solomon G, Yonemitsu S, et al. and reduced responsiveness to peroxisome proliferator-activated receptor The role of skeletal muscle insulin resistance in the pathogenesis of the gamma agonists. J Biol Chem 2006;281:2654–2660. doi: 10.1074/jbc. metabolic syndrome. Proc Natl Acad Sci U S A 2007;104:12587–12594. M505311200. doi: 10.1073/pnas.0705408104. [72] Weisberg SP, McCann D, Desai M, Rosenbaum M, Leibel RL, Ferrante AW Jr. [50] Flannery C, Dufour S, Rabøl R, Shulman GI, Petersen KF. Skeletal muscle Obesity is associated with macrophage accumulation in adipose tissue. J insulin resistance promotes increased hepatic de novo lipogenesis, hyper- Clin Invest 2003;112:1796–1808. doi: 10.1172/JCI19246. lipidemia, and hepatic steatosis in the elderly. Diabetes 2012;61:2711– [73] Xu H, Barnes GT, Yang Q, Tan G, Yang D, Chou CJ, et al. Chronic inflamma- 2717. doi: 10.2337/db12-0206. tion in fat plays a crucial role in the development of obesity-related insulin [51] Petersen KF, Oral EA, Dufour S, Befroy D, Ariyan C, Yu C, et al. Leptin resistance. J Clin Invest 2003;112:1821–1830. doi: 10.1172/JCI19451. reverses insulin resistance and hepatic steatosis in patients with severe [74] Laurencikiene J, van Harmelen V, Arvidsson Nordström E, Dicker A, Blomqv- lipodystrophy. J Clin Invest 2002;109:1345–1350. doi: 10.1172/JCI15001. ist L, Näslund E, et al. NF-kappaB is important for TNF-alpha-induced

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 362–370 369 Akhtar D.H. et al: NAFLD, insulin resistance and CVD

lipolysis in human adipocytes. J Lipid Res 2007;48:1069–1077. doi: 10. dysbiosis and shift in the metabolic function of the gut microbiota. Hepatol- 1194/jlr.M600471-JLR200. ogy 2016;63:764–775. doi: 10.1002/hep.28356. [75] Ranjit S, Boutet E, Gandhi P, Prot M, Tamori Y, Chawla A, et al. Regulation of [96] Pedersen HK, Gudmundsdottir V, Nielsen HB, Hyotylainen T, Nielsen T, Jensen fat specific protein 27 by isoproterenol and TNF-a to control lipolysis in BA, et al. Human gut microbes impact host serum metabolome and insulin murine adipocytes. J Lipid Res 2011;52:221–236. doi: 10.1194/jlr. sensitivity. Nature 2016;535:376–381. doi: 10.1038/nature18646 M008771. [97] Abdel-Razik A, Mousa N, Shabana W, Refaey M, Elzehery R, Elhelaly R, et al. [76] Scherer T, O’Hare J, Diggs-Andrews K, Schweiger M, Cheng B, Lindtner C, Rifaximin in nonalcoholic fatty liver disease: hit multiple targets with a et al. Brain insulin controls adipose tissue lipolysis and lipogenesis. Cell single shot. Eur J Gastroenterol Hepatol 2018;30:1237–1246. doi: 10. Metab 2011;13:183–194. doi: 10.1016/j.cmet.2011.01.008. 1097/MEG.0000000000001232. [77] Scherer T, Lindtner C, Zielinski E, O’Hare J, Filatova N, Buettner C. Short [98] Borrelli A, Bonelli P, Tuccillo FM, Goldfine ID, Evans JL, Buonaguro FM, et al. term voluntary overfeeding disrupts brain insulin control of adipose tissue Role of gut microbiota and oxidative stress in the progression of non-alco- lipolysis. J Biol Chem 2012;287:33061–33069. doi: 10.1074/jbc.M111. holic fatty liver disease to hepatocarcinoma: Current and innovative ther- 307348. apeutic approaches. Redox Biol 2018;15:467–479. doi: 10.1016/j.redox. [78] Iwen KA, Scherer T, Heni M, Sayk F, Wellnitz T, Machleidt F, et al. Intranasal 2018.01.009. insulin suppresses systemic but not subcutaneous lipolysis in healthy [99] Ampuero J, Gallego-Durán R, Romero-Gómez M. Association of NAFLD with humans. J Clin Endocrinol Metab 2014;99:E246–E251. doi: 10.1210/jc. subclinical atherosclerosis and coronary-artery disease: meta-analysis. Rev 2013-3169. Esp Enferm Dig 2015;107:10–16. [79] Brands M, Verhoeven AJ, Serlie MJ. Role of mitochondrial function in insulin [100] Zhou YY, Zhou XD, Wu SJ, Fan DH, Van Poucke S, Chen YP, et al. Nonalco- resistance. Adv Exp Med Biol 2012;942:215–234. doi: 10.1007/978-94- holic fatty liver disease contributes to subclinical atherosclerosis: A system- – 007-2869-1_9. atic review and meta-analysis. Hepatol Commun 2018;2:376 392. doi: 10. [80] Bugianesi E, Gastaldelli A, Vanni E, Gambino R, Cassader M, Baldi S, et al. 1002/hep4.1155. Insulin resistance in non-diabetic patients with non-alcoholic fatty liver [101] Jaruvongvanich V, Wirunsawanya K, Sanguankeo A, Upala S. Nonalcoholic disease: sites and mechanisms. Diabetologia 2005;48:634–642. doi: 10. fatty liver disease is associated with coronary artery calcification: A system- – 1007/s00125-005-1682-x. atic review and meta-analysis. Dig Liver Dis 2016;48:1410 1417. doi: 10. [81] Koliaki C, Szendroedi J, Kaul K, Jelenik T, Nowotny P, Jankowiak F, et al. 1016/j.dld.2016.09.002. Adaptation of hepatic mitochondrial function in humans with non-alcoholic [102] Fan Y, Wei F, Zhou Y, Zhang H. Association of non-alcoholic fatty liver fatty liver is lost in steatohepatitis. Cell Metab 2015;21:739–746. doi: 10. disease with impaired endothelial function by flow-mediated dilation: A – 1016/j.cmet.2015.04.004. meta-analysis. Hepatol Res 2016;46:E165 E173. doi: 10.1111/hepr. [82] Yeniova AO, Küçükazman M, Ata N, Dal K, Kefeli A, Bas¸yigit S, et al. High- 12554. sensitivity C-reactive protein is a strong predictor of non-alcoholic fatty liver [103] Cai J, Zhang S, Huang W. Association between nonalcoholic fatty liver disease and carotid atherosclerosis: a meta-analysis. Int J Clin Exp Med disease. Hepatogastroenterology 2014;61:422–425. 2015;8:7673–7678. [83] Wood KL, Miller MH, Dillon JF. Systematic review of genetic association [104] Valbusa F, Bonapace S, Grillo C, Scala L, Chiampan A, Rossi A, et al. Non- studies involving histologically confirmed non-alcoholic fatty liver disease. alcoholic fatty liver disease is associated with higher 1-year all-cause reho- BMJ Open Gastroenterol 2015;2:e000019. doi: 10.1136/bmjgast-2014- spitalization rates in patients admitted for acute heart failure. Medicine 000019. (Baltimore) 2016;95:e2760. doi: 10.1097/MD.0000000000002760. [84] Peter A, Kantartzis K, Machicao F, Machann J, Wagner S, Templin S, et al. [105] Granér M, Nyman K, Siren R, Pentikäinen MO, Lundbom J, Hakkarainen A, Visceral obesity modulates the impact of apolipoprotein C3 gene variants on et al. Ectopic fat depots and left ventricular function in nondiabetic men with liver fat content. Int J Obes (Lond) 2012;36:774–782. doi: 10.1038/ijo. nonalcoholic fatty liver disease. Circ Cardiovasc Imaging 2014;8:e001979. 2011.154. doi: 10.1161/CIRCIMAGING.114.001979. [85] Petersen KF, Dufour S, Hariri A, Nelson-Williams C, Foo JN, Zhang XM, et al. [106] Balcioglu AS, Çiçek D, Akinci S, Eldem HO, Bal UA, Okyay K, et al. Arrhyth- Apolipoprotein C3 gene variants in nonalcoholic fatty liver disease. N Engl J mogenic evidence for epicardial adipose tissue: heart rate variability and Med 2010;362:1082–1089. doi: 10.1056/NEJMoa0907295. turbulence are influenced by epicardial fat thickness. Pacing Clin Electro- [86] Zhang RN, Zheng RD, Mi YQ, Zhou D, Shen F, Chen GY, et al. APOC3 physiol 2015;38:99–106. doi: 10.1111/pace.12512. rs2070666 is associated with the hepatic steatosis independently of [107] Targher G, Valbusa F, Bonapace S, Bertolini L, Zenari L, Pichiri I, et al. PNPLA3 rs738409 in Chinese Han patients with nonalcoholic fatty liver dis- Association of nonalcoholic fatty liver disease with QTc interval in patients eases. Dig Dis Sci 2016;61:2284–2293. doi: 10.1007/s10620-016-4120-7. with type 2 diabetes. Nutr Metab Cardiovasc Dis 2014;24:663–669. doi: [87] Mahdessian H, Taxiarchis A, Popov S, Silveira A, Franco-Cereceda A, 10.1016/j.numecd.2014.01.005. Hamsten A, et al. TM6SF2 is a regulator of liver fat metabolism influencing [108] Hung CS, Tseng PH, Tu CH, Chen CC, Liao WC, Lee YC, et al. Nonalcoholic triglyceride secretion and hepatic lipid droplet content. Proc Natl Acad Sci U fatty liver disease is associated with QT prolongation in the general popula- – S A 2014;111:8913 8918. doi: 10.1073/pnas.1323785111. tion. J Am Heart Assoc 2015;4:e001820. doi: 10.1161/JAHA.115.001820. [88] Luukkonen PK, Zhou Y, Hyötyläinen T, Leivonen M, Arola J, Orho-Melander M, [109] Mantovani A, Rigolon R, Pichiri I, Bonapace S, Morani G, Zoppini G, et al. et al. The MBOAT7 variant rs641738 alters hepatic phosphatidylinositols and Nonalcoholic fatty liver disease is associated with an increased risk of heart increases severity of non-alcoholic fatty liver disease in humans. J Hepatol block in hospitalized patients with type 2 diabetes mellitus. PLoS One 2017; – 2016;65:1263 1265. doi: 10.1016/j.jhep.2016.07.045. 12:e0185459. doi: 10.1371/journal.pone.0185459. [89] Mancina RM, Dongiovanni P, Petta S, Pingitore P, Meroni M, Rametta R, et al. [110] Targher G, Mantovani A, Pichiri I, Rigolon R, Dauriz M, Zoppini G, et al. Non- The MBOAT7-TMC4 variant rs641738 increases risk of nonalcoholic fatty alcoholic fatty liver disease is associated with an increased prevalence of liver disease in individuals of European descent. Gastroenterology 2016; atrial fibrillation in hospitalized patients with type 2 diabetes. Clin Sci – 150:1219 1230.e6. doi: 10.1053/j.gastro.2016.01.032. (Lond) 2013;125:301–309. doi: 10.1042/CS20130036. [90] Williams AL, Jacobs SB, Moreno-Macías H, Huerta-Chagoya A, Churchhouse C, [111] Ozveren O, Izgi C, Eroglu E, Simsek MA, Turer A, Kucukdurmaz Z, et al. Márquez-Luna C, et al. Sequence variants in SLC16A11 are a common risk Doppler tissue evaluation of atrial conduction properties in patients with – factor for type 2 diabetes in Mexico. Nature 2014;506:97 101. doi: 10. non-alcoholic fatty-liver disease. Ultrason Imaging 2016;38:225–235. 1038/nature12828. doi: 10.1177/0161734615595015. [91] Rusu V, Hoch E, Mercader JM, Tenen DE, Gymrek M, Hartigan CR, et al.Type2 [112] Targher G, Valbusa F, Bonapace S, Bertolini L, Zenari L, Rodella S, et al. diabetes variants disrupt function of SLC16A11 through two distinct mecha- Non-alcoholic fatty liver disease is associated with an increased incidence of nisms. Cell 2017;170:199–212.e20. doi: 10.1016/j.cell.2017.06.011. atrial fibrillation in patients with type 2 diabetes. PLoS One 2013;8:e57183. [92] Yu J, Marsh S, Hu J, Feng W, Wu C. The pathogenesis of nonalcoholic fatty doi: 10.1371/journal.pone.0057183. liver disease: Interplay between diet, gut microbiota, and genetic back- [113] Targher G, Byrne CD, Lonardo A, Zoppini G, Barbui C. Non-alcoholic fatty ground. Gastroenterol Res Pract 2016;2016:2862173. doi: 10. liver disease and risk of incident cardiovascular disease: A meta-analysis. J 1155/2016/2862173. Hepatol 2016;65:589–600. doi: 10.1016/j.jhep.2016.05.013. [93] Ferolla SM, Armiliato GN, Couto CA, Ferrari TC. The role of intestinal bac- [114] Wu S, Wu F, Ding Y, Hou J, Bi J, Zhang Z. Association of non-alcoholic fatty teria overgrowth in obesity-related nonalcoholic fatty liver disease. liver disease with major adverse cardiovascular events: A systematic review Nutrients 2014;6:5583–5599. doi: 10.3390/nu6125583. and meta-analysis. Sci Rep 2016;6:33386. doi: 10.1038/srep33386. [94] Wigg AJ, Roberts-Thomson IC, Dymock RB, McCarthy PJ, Grose RH, [115] Dulai PS, Singh S, Patel J, Soni M, Prokop LJ, Younossi Z, et al. Increased Cummins AG. The role of small intestinal bacterial overgrowth, intestinal risk of mortality by fibrosis stage in nonalcoholic fatty liver disease: Sys- permeability, endotoxaemia, and tumour necrosis factor alpha in the patho- tematic review and meta-analysis. Hepatology 2017;65:1557–1565. doi: genesis of non-alcoholic steatohepatitis. Gut 2001;48:206–211. doi: 10. 10.1002/hep.29085. 1136/gut.48.2.206. [116] Hagström H, Nasr P, Ekstedt M, Hammar U, Stål P, Askling J, et al. Cardio- [95] Boursier J, Mueller O, Barret M, Machado M, Fizanne L, Araujo-Perez F, et al. vascular risk factors in non-alcoholic fatty liver disease. Liver Int 2019;39: The severity of nonalcoholic fatty liver disease is associated with gut 197–204. doi: 10.1111/liv.13973.

370 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 362–370 Review Article

Role of Granulocyte Colony-stimulating Factor Therapy in Cirrhosis, ‘Inside Any Deep Asking Is the Answering’

Cyriac Abby Philips*1, Philip Augustine2, Rizwan Ahamed2, Sasidharan Rajesh3, Tom George3, Gopakumar C. Valiathan4 and Solomon K. John4

1The Liver Unit and Monarch Liver Lab, Cochin Gastroenterology Group, Ernakulam Medical Centre, Kochi, Kerala, India; 2Department of Gastroenterology, Cochin Gastroenterology Group, Ernakulam Medical Centre, Kochi, Kerala, India; 3Interventional Radiology, Hepatobiliary Division, Cochin Gastroenterology Group, Ernakulam Medical Centre, Kochi, Kerala, India; 4Department of Hepatobiliary and Transplant Surgery, Cochin Gastroenterology Group, Ernakulam Medical Centre, Kochi, Kerala, India

Abstract Introduction

Liver cirrhosis progresses through multiple clinical stages Cirrhosis is the final common pathological pathway of ongoing which culminate in either death or liver transplantation. liver injury that arises due to multiple etiologies of insults that Availability of organs, timely listing and prompt receipt of vary geographically. Even though the etiologies differ, there donor-livers pose difficulties in improving transplant-listed are multiple drivers and progression factors, which culminate and transplant outcomes. In this regard, regenerative in common pathological characteristics that lead to cirrhosis. therapies, particularly with granulocyte colony-stimulating These include hepatic necrosis, degeneration, and replace- factor (GCSF), has become a lucrative option for improving ment of hepatic parenchyma by scar tissue surrounding failed transplant-free survival. However, the literature is confusing regeneration in the form of hepatic nodules that ultimately with regards to patient selection and real outcomes. In this lead to portal hypertension and liver failure. Fibrosis is the exhaustive review, we describe the basics of liver fibrosis and most crucial precursor that drives the central pathological cirrhosis through novel insights from a therapeutic point of process in cirrhosis. view, discuss preclinical studies on GCSF in advanced liver Currently, strategies in the treatment of cirrhosis are disease to improve on clinical utility, shed light on the aimed at management of complications of cirrhosis and pertinent literature of GCSF in advanced cirrhosis, and portal hypertension, and direct treatment strategies for cir- provide astute inputs on growth factor therapy in decom- rhosis, except treatment for known causes for cirrhosis pensated cirrhosis. progression (such as autoimmune hepatitis, and chronic Citation of this article: Philips CA, Augustine P, Ahamed R, viral hepatitis B and C), are lacking. Treatment of cirrhosis Rajesh S, George T, Valiathan GC, et al. Role of granulocyte and restoration or replacement of functional regenerative colony-stimulating factor therapy in cirrhosis, ‘inside any potential can only happen once the molecular mechanisms deep asking is the answering’. J Clin Transl Hepatol that drive progression to cirrhosis are better understood. 2019;7(4):371–383. doi: 10.14218/JCTH.2019.00034. These molecular mechanisms, even though generally understood, lack clarity in the current literature.1 In the current review, we discuss pertinent mechanisms that lead to cirrhosis, and regenerative or restorative therapeutic strategies aimed at prevention and reversal of cirrhosis based on the currently known molecular mechanisms of cirrhosis. We Keywords: GCSF; Growth factor; Portal hypertension; Hepatocellular carcinoma; Fibrosis. also explore the current literature on clinical trials of regener- Abbreviations: ACLF, acute-on-chronic liver failure; C/EBP-a, CCAAT enhancer- ative strategies in cirrhosis and provide a critical appraisal of binding protein-alpha; CTGF, connective tissue growth factor; CTP, Child-Turcotte- the same with particular emphasis on granulocyte-colony- Pugh; ECM, extracellular matrix; EpCAM, epithelial cell adhesion molecule; FGF, stimulating-factor (GCSF)-based treatments. fibroblast growth factor; GCSF, granulocyte-colony-stimulating-factor; GCSF-R, GCSF receptor; HBV, hepatitis B virus; HGF, hepatocyte growth factor; HSCs, hepatic stellate cells; IL, interleukin; JAK, Janus kinase; LSECs, liver sinusoidal Pathogenesis of cirrhosis and complexities associated endothelial cells; MAPK, mitogen-activated protein kinase; MCP, macrophage with ideal therapeutic implications colony-stimulating factor; MELD, model for end-stage liver disease; MMPs, matrix metalloproteinases; MSCs, mesenchymal stem cells; NF-kB, nuclear factor kappa-light-chain-enhancer of activated B cells; PDGF, platelet-derived Chronic liver injury results in progressive accumulation of growth factor; PPAR, peroxisome proliferator-activated receptor; SMT, standard extracellular matrix (ECM) with distortion of hepatic paren- b medical care; STAT, signal transducer and activator of transcription; TGF- , trans- chymal architecture, an event that is spearheaded by myofi- forming growth factor-beta; TIMPs, tissue inhibitors of MMPs; TNF-a, tumor necrosis factor-alpha; VEGF, vascular endothelial growth factor. broblasts that form due to activation of hepatic stellate cells Received: 11 August 2019; Revised: 20 September 2019; Accepted: 5 October (HSCs) and multiple other cell types. This fibril-forming 2019 collagen deposition that replaces the low density, basement *Correspondence to: Cyriac Abby Philips, The Liver Unit and Monarch Liver Lab, membrane-like interstitial matrix, along with the accumula- Cochin Gastroenterology Group, Ernakulam Medical Centre, Automobile Road, Palarivattom, Kochi, 682025, Kerala, India. Tel: +91-484-2907000, Fax: tion of other matrix proteins such as hyaluronan, elastin, +91-484-2907000, E-mail: [email protected] fibronectin and proteoglycans, is central to fibrosis and its

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 371–383 371

Copyright: © 2019 Authors. This article has been published under the terms of Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0), which permits noncommercial unrestricted use, distribution, and reproduction in any medium, provided that the following statement is provided. “This article has been published in Journal of Clinical and Translational Hepatology at DOI: 10.14218/JCTH.2019.00034 and can also be viewed on the Journal’s website at http://www.jcthnet.com”. Philips C.A. et al: Critical review of GCSF use in cirrhosis progression to cirrhosis.2 The ECM has the potential to In clinical and experimental fibrosis models, control or secrete various cytokines, such as transforming growth elimination of the etiological agent responsible for the chronic factor-beta (TGF-b), platelet-derived growth factor (PDGF), inflammation has been shown to promote fibrosis reversal, hepatocyte growth factor (HGF), connective tissue growth due to the complete disappearance of myofibroblasts. Even factor (CTGF), tumor necrosis factor-alpha (TNF-a), fibroblast then, a small subset of myofibroblasts can escape apoptosis growth factor (FGF) and vascular endothelial growth factor during liver fibrosis regression, acquiring a phenotype similar (VEGF), that promote angiogenesis, another critical event to but distinct from quiescent HSCs. These ‘fugitives’ then that leads to exaggerated wound healing in the form of rapidly reactivate into myofibroblasts in response to repeti- fibrous tissue formation. While this injury and ECM deposition tion of fibrogenic stimuli and rapidly contribute to liver is ongoing, remodeling of the ECM to preserve healthy hepatic fibrosis.12,13 Early liver fibrosis, which lacks ECM crosslinking parenchymal structure and function becomes critical in the and marked angiogenesis, has the best potential to revert to maintenance of liver health.3 typical architecture, provided the chronic insult is adequately This balance is maintained by the matrix metalloprotei- controlled. Hence, the initiation, progression and reversal of nases (MMPs; specifically, MMP-1, -2, -8 and -13) and their fibrosis is a highly complex process with multiple interactions inhibitors, the tissue inhibitors of MMPs (TIMPs). When the at the cellular and molecular levels.14,15 injury is chronic and the ECM deposition is overwhelming due Taken together, the pleomorphic action of GCSF through to the persistence of HSC activation and abnormal neo- multiple molecular mechanisms in the liver microenviron- angiogenesis, the activity of TIMPs takes the upper hand. ment could increase fibrosis and liver disease progression This tipping of balance toward prolonged activity of TIMPs apart from its beneficial effects on granulopoiesis. Hence, to (mostly TIMP-1 and -2) results in anti-apoptotic effects on promote liver regeneration or restoration, the therapeutic HSCs that further promoted fibrogenesis.4 intervention(s) must target multiple pathways and not Chronic liver inflammation leads to hepatocyte necrosis/ just one of the ‘central’ pathways. Progression of fibrosis to apoptosis, paracrine stimulation, Kupffer cell activation, reac- cirrhosis happens through multiple ‘central’ pathways — this tive oxidation and cytokine deliberation that activate HSCs is akin to a control headquarters (chronic injury) and multiple that then transform into myofibroblasts with profibrogenic ‘metro-rail-lines’ (molecular mechanisms) and major associ- potential. HSC activation also occurs through lipid peroxide ated stations (central pathways) in a large city, rather than a release, TNF-a and interferon-gamma production via the single central railway station in a town. This complicated aspect activation of nuclear factor kappa-light-chain-enhancer of of liver fibrosis progression and the complexities associated activated B cells (NF-kB)-interferon regulatory factor 3 with treatment of fibrosis/cirrhosis is demonstrated in Fig. 1. pathway and toll-like receptors (through lipopolysaccharide production by gut microbial dysbiosis).5 Stimulated HSCs secrete macrophage colony-stimulating factor (MCP)-1, Pathophysiology associated with GCSF in the context interleukin (IL)-6 and chemokines that act on their respective of chronic liver disease receptors, leading to macrophage activation, neutrophil infiltration, and chemotaxis, which activate mitochondrial oxida- GCSF is a 25 kDa secreted glycoprotein encoded by the CSF3 tion in hepatocytes leading to their apoptosis; furthermore, gene. The central physiological role played by GCSF is in the these become a strong trigger for fibrogenesis, as phagocy- regulation of neutrophil production in health and particularly tosis of damaged hepatocytes by myofibroblasts enhances in emergency responses to infections and bone marrow the fibrogenic activation through NADPH oxidase and the aplasia. In healthy humans, the serum concentrations of Janus kinase (JAK) signal transducer and activator of tran- GCSF are typically undetectable or detectable at deficient scription (STAT) and phosphoinositide 3-kinase/Akt path- levels, which markedly increases in the presence of an ways.6,7 The HSCs and myofibroblasts proliferate and lay infectious stimulus. Most of the tissues in the body secrete down ECM, and further promote hepatic inflammation GCSF after stimulatory effects, such as induction of IL-1, through enhanced TIMP expression, also under the influence lipopolysaccharide and TNF-a produced by the macrophages, of secreted angiotensin II [via mitogen-activated protein endothelial cells, fibroblasts and related mesenchymal cells kinase (MAPK) signal transduction pathways], upregulated (Fig. 2). cannabinoid receptor 1, and circulating adipokine leptin [via IL-17 is a potent upstream extracellular regulator of tissue (JAK)-signal transduction, leading to suppression of peroxi- production of GCSF, especially in the bone marrow. Ligation of some proliferator-activated receptor g (PPAR-g)].8 the extracellular domain of the GCSF receptor (GCSF-R) by Other cells of the liver microenvironment regulating GCSF results in cellular responses due to signals that arise enhancement or reduction of fibrosis involve natural killer from the cytoplasmic domain of the GCSF-R. The GCSF-R cells, T cells, monocytes, liver sinusoidal endothelial cells is expressed by neutrophils and its precursors, such as (LSECs), ductular cells, cholangiocytes, and portal fibro- metamyelocytes, myelocytes, promyelocytes, myeloblasts, blasts.9 The role of monocytes in inflammation and fibrosis myeloid progenitor cells, and primitive hemopoietic stem is important regarding GCSF therapy. Fibrogenesis is pro- cells. The GCSF-R signals through the JAK/STAT pathway moted by the subset of proinflammatory monocytes (CD14 and through Lyn phosphorylation that activates PI3-kinase/ + and CD16+ in humans). Monocytes are a source of circu- Akt pathways, which are pertinent to the progression of liver lating fibrocytes, which differentiate into collagen-producing fibrosis. fibroblasts, which are in turn closely related to the bone GCSF also activates Ras-MAPK through activation of tyro- marrow mesenchymal stem cells (MSCs).10,11 The TGF-b sine kinases, Lyn and Hck. This is of utmost importance secreted by myofibroblasts promotes hepatocyte apoptosis because the Ras/Raf/MEK/ERK signaling pathway has been after activation. TGF-b1 activates Smad2, and Smad3 implicated in the occurrence and development of hepatocel- promote fibrogenesis. Reduction and clearance of activated lular carcinoma in cirrhosis. GCSF has been shown to stim- HSCs are central to fibrosis regression. ulate tumor cell growth and migration in vitro, and to promote

372 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 371–383 Philips C.A. et al: Critical review of GCSF use in cirrhosis

Fig. 1. The ‘Metro-Rail Concept’ to targeted therapy of the pathophysiology of cirrhosis. The development of cirrhosis follows well-coordinated steps that begin with the etiology, leading to sustained chronic inflammation (the control or command center; black bubble) that activates and promotes multiple pathways (rail-pathways, colored lines) that feature prominent mediators of inflammation and fibrosis (central stations, grey bubbles) and parallel assisting pathway intermediaries (secondary stations, small white bubbles) that ultimately lead to the destination (red bubble). Therapies that target only few of the pathway mediators do not tend to improve outcomes as expected; however, targeting the command center (etiology) along with controlling the central-stations (central inflammatory and fibrosis pathways) would impede progression to destination (cirrhosis). Abbreviations: BMDSC, bone marrow-derived stem cells; CCRs, chemokine receptors; CD, cluster of differentiation; CTGF, connective tissue growth factor; ECM, extracellular matrix; HGF, hepatocyte growth factor; HSC, hepatic stellate cell; IL, interleukin; IRF, interferon regulatory factor; JAK/ STAT, Janus kinase/signal transducers and activators of transcription; LPS, lipopolysaccharide; MAP, mitogen activated protein; MCP, monocyte chemoattractant protein; MFB, myofibroblasts; NADPH, reduced form of nicotinamide adenine dinucleotide phosphate; NF-kB, nuclear factor kappa B; PDGF, platelet derived growth factor; PHT, portal hypertension; PI3K/Akt, phosphoinositide-3-kinase-protein kinase B; PPAR, peroxisome proliferator-activated receptors; SMA, smooth muscle antibody; SMADs, homologues of the Drosophila protein, mothers against decapentaplegic (Mad) and the Caenorhabditis elegans protein Sma; TGF, transforming growth factor; TIMP, tissue inhibitors of matrix metalloproteinase; TLR, toll-like receptor; TNF, tumor necrosis factor; VEGF, vascular endothelial growth factor. tumor progression in vivo by autocrine stimulation of tumor by GCSF in hematopoietic stem cells, and hence the utility of cells and paracrine activation of the tumorigenic stroma.16,17 exogenous GSCF to decrease hepatocyte apoptosis cannot be Ordelheide et al.18 demonstrated that GCSF promoted free possible through this mechanism of action.21 Buck et al.22 fatty acid-induced insulin resistance in humans and that showed that, in response to liver injury, activation of riboso- human adipocytes and myotubes treated with GCSF became mal S6 kinase phosphorylation of C/EBPb in activated HSCs is insulin-resistant. Insulin resistance is a major driver of liver critical for the progression of liver fibrosis. Hence, in the pres- fibrogenesis and carcinogenesis. In cirrhosis patients with ence of GCSF, molecular mechanisms of chronic liver injury metabolic syndrome, obesity and insulin resistance, the use that increases fibrosis are possibly upregulated to augment of GCSF could probably augment the disease process.19 disease progression. CCAAT enhancer-binding protein-a (C/EBP-a) regulates adi- GCSF is also produced by a variety of nonhematopoietic pocyte differentiation and induces apoptosis in HSCs in vivo cells, including fibroblasts and endothelial cells, and induces and in vitro.Taoet al.20 showed that in the mouse liver fib- the proliferation and migration of endothelial cells, promotes rosis model, the upregulation of C/EBP-a decreased ECM dep- angiogenesis, and upregulates inflammatory cell infiltration osition, including collagen and hydroxyproline content, and into tissues where GCSF-R is expressed.23 Shojaei et al.24 markers of liver damage were reduced significantly; immuno- identified GCSF as a strong inducer of prokinectin-Bv8 histochemistry showed an increase of apoptosis in HSCs, expression, both in vitro and in vivo, the latter of which pro- while hepatocytes were less affected. motes neovascularization and tumoral progression in gastro- On the other hand, C/EBPb was found to be selectively intestinal malignancies. Hepatic angiogenesis is closely upregulated in granulocytic-macrophage progenitors in the associated with the progression of fibrosis in chronic liver dis- presence of GCSF. However, beneficial C/EBPa is not induced eases and GCSF demonstrably induces endothelial activation

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 371–383 373 Philips C.A. et al: Critical review of GCSF use in cirrhosis

Fig. 2. GCSF-activated molecular pathways associated with fibrosis regression and progression. GCSF is a ‘double-edged sword’ and cannot be considered a true ally in the armamentarium against fibrosis and cirrhosis. Abbreviation: GCSF, granulocyte colony-stimulating factor. and upregulates downstream inflammatory pathways associ- GCSF at 10 mcg/kg/day for 5 days. In portal hypertension, ated with angiogenesis and fibrosis progression.25 splenic congestion and splenomegaly are attributed to portal The belief that GCSF improves synthetic liver function and congestion, elevated portal pressures and closely related to decreases fibrosis stems from the understanding that mobi- increased tissue hyperplasia and fibrosis. The increase in lization of CD34+ bone marrow-derived cells home into the spleen size results in increased splenic blood flow, translating liver microenvironment, transform into hepatic progenitor to increased portal hypertension, which may worsen with cells, and restore lost hepatocyte volume. Subsequently, repeated GCSF use in patients with cirrhosis.31 possible reduction in expression in angiopoietin which leads In the study by Nakamura et al.,32 hepatic arterial infusion to decreased neoangiogenesis ameliorates fibrosis. The of low, mid and high doses of autologous-derived CD34+ cells enhancement of fibrolytic activity of CD133+ cells induced mobilized by GCSF showed mild improvement in serum through GCSF on monocyte and bone marrow activation albumin level, without significant sustained improvement possibly through IL-10 mediated Stat3 regeneration pathway in liver disease severity scores. The spleen size in cirrhosis has been postulated. However, most of these processes patients did increase with GCSF but it was a transient phe- remain clinical hypotheses, lacking proper ‘fate of cell’ tracer nomenon since the GCSF treatment was only for 5 days. Sim- studies.26–28 ilarly, Gaia et al.33 reported significant reversible spleen Several studies have shown that GCSF promoted chromo- enlargement with stable serum liver enzyme levels after somal changes in healthy persons associated with the GCSF administration in patients with alcoholic cirrhosis, and modification of gene expression profile. Even though the Lorenzini et al.34 demonstrated increased splenomegaly with long-term mutagenic implications in healthy persons with stable levels during GCSF administration in patients with viral such changes are insignificant, in cirrhosis patients, in the hepatitis. These studies concentrated on short-term use of presence of an inflammatory microenvironment, exogenous GCSF, and longer use or multiple dosing regimen of GCSF as GCSF-associated genetic expression as well as chromosomal seen with recent studies could be associated with greater aberration are concerning and need further study.29 Healthy chances of splenic enlargement and elevation in portal pres- persons, when injected with GCSF, develop marked neutro- sures. All of these studies lacked the fate of hematopoietic cell philic response within 4 h, mobilize bone marrow-activated investigation and, as such, conclusive evidence of liver cell stem cells after 3 days, that peaks at the fifth day, which is restoration or regeneration could not be ascertained. associated with splenic enlargement over a week’s time. In humans, GCSF could either be beneficial or detrimental, Stroncek et al.30 demonstrated that the spleen length depending on the disease in context. For example, perioper- increased by 20% or more in healthy subjects treated with ative GCSF use was found to reduce postoperative morbidity by

374 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 371–383 Philips C.A. et al: Critical review of GCSF use in cirrhosis decreasing monocyte and lymphocyte activation. However, in in elegant animal studies. About 5 to 10 ng/mL of HGF is patients with chemotherapy-related lung injury and fibrosis, required for growth promotion of adult rat hepatocytes in GCSF worsened clinical outcomes by exacerbating lung primary culture, and the ideal blood levels required for injury. A similar detrimental outcome was noted in patients human hepatocyte growth promotion in human studies have with rheumatoid arthritis.35 The role of GCSF in cell recruit- not been confirmed. Exogenously administered HGF in animal ment in infected and healthy individuals is different from a models of acute liver failure has shown beneficial effects in potential role in producing tissue injury. These are yet to improving survival, and the role of HGF in clinical practice is to be studied in different stages of advanced liver cirrhosis. predict prognosis in patients with severe liver failure. HGF has However, preclinical studies36,37 that shed light on various pleo- a very short half-life (; 5 m), and the quality of HGF induction morphic actions of GCSF in patients with advanced liver disease, with GCSF use and its continued benefits remain unknown.40 with regards to pro- and anti-inflammation, fibrosis reversal or In another study, Spahr and colleagues41 randomized promotion, and intermediate and long-term portal hypertensive 58 patients with alcoholic hepatitis and underlying cirrhosis outcomes, especially in decompensated cirrhosis, are currently with mean model for end-stage liver disease (MELD) score of lacking and mandatory before further trials on GCSF is 19, early after hospital admission, to standard medical undertaken. therapy (SMT) or combined with GCSF injections and autologous hepatic arterial infusion of bone marrow-derived mono- Critical appraisal of GCSF in decompensated cirrhosis nuclear cells. At the end of 90 days follow-up, two and four patients died in the experimental and SMT groups, respec- One of the strategies to restore liver functionality is cell tively. Adverse events were not significant between groups, therapy, aimed at restoring or regenerating hepatocytes and on follow-up liver biopsy from the baseline, improvement that maintain liver function, an aspect that is physiologically in steatosis was notable but proliferating hepatocyte progen- lacking in cirrhosis. Sources of hepatocytes include normal itor cells decreased in both groups. A weak regenerative stim- liver in which hepatocytes themselves proliferate to restore ulation and resistance to the promotion of regeneration in functionality, liver progenitor cells that differentiate and decompensated alcoholic cirrhosis could have led to the proliferate under specific circumstances, and blood-derived poor responses seen with regenerative therapy in this study. 42 stem cells that infiltrate the liver, transform into hepatocytes Han and colleagues administered GCSF at 5 to 10 mcg/ and proliferate. The ideal source of liver cell function and kg/day for 4 days to patients with decompensated hepatitis B- mass restoration in a chronic liver disease state is currently related liver cirrhosis and compared them to those receiving poorly understood. Most clinical trials on GCSF have been GCSF mobilized peripheral blood mononuclear cells, collected conducted without adequate knowledge on the accurate by leukapheresis followed by infusion through the hepatic beneficial pathway that promotes quantity restoration of artery. In the GCSF group, one patient died from variceal existing healthy hepatocytes or regeneration or formation of bleeding combined with hepatic coma at 3 weeks after new liver cells. GCSF, and two patients from repeated variceal bleeding at 5 Beneficial control over molecular mechanisms involved in months after GCSF. Among the surviving patients, only two HSC activation, amelioration in production and enhanced required no albumin supplementation during the follow-up, degradation of ECM with mitigation in activated myofibro- while ten required continued albumin supplementation 4 blasts form the components of ideal ‘regenerative’ therapy for weeks after GCSF therapy due to repeated ascites removal liver cirrhosis. Such an ideal scenario is currently unavailable and new-onset hepatorenal syndrome in one patient. The dif- with cell therapy-based interventions in cirrhosis. Treatment ferences in serum albumin and Child-Turcotte-Pugh (CTP) or control of etiology responsible for chronic liver inflamma- scores between the two groups were significant. This was tion (i.e. abstinence from alcohol, weight loss in obese indicative of worsening clinical outcomes with GCSF in patients, and antiviral therapy for hepatitis B and C) is the patients with decompensated liver cirrhosis. best currently available intervention that can truly ‘regener- Xing and colleagues43 studied the effects of GCSF in ate/restore’ liver function in cirrhosis. However, in patients patients with hepatitis B virus (HBV)-related cirrhosis. They with advanced cirrhosis, control of etiology may not fully found that CD34+ cells were already higher at baseline establish acceptable liver function, and such patients become in cirrhosis patients compared to healthy controls and that ideal candidates for cell therapies aimed at liver regeneration GCSF use dramatically increased circulating numbers of in the absence of liver transplantation options. At present, the CD34+ bone marrow-derived stem cells. However, such various cell-based therapeutic strategies studied in patients increments in levels of the CD34+ stem cell population in with cirrhosis include infusion of autologous hepatocytes, the systemic circulation did not translate to clinical improve- epithelial cell adhesion molecule (EpCAM) positive fetal liver ments in the treated patients compared to the control group. stem cells, bone marrow-derived differentiated or undiffer- Gaia et al.44 studied the effects of multiple courses of GCSF entiated MSCs, autologous GCSF mobilized cultured CD34+ (3-day GCSF course, 5 mcg/kg every 12 h; administered at bone marrow stem cells, peripheral blood mononuclear cells 3-month intervals for a total of four courses) in patients with from GCSF mobilized peripheral blood, and direct use of GCSF decompensated cirrhosis. CD34+ bone marrow-derived stem to induce bone marrow-derived stem cells in peripheral cells were found to increase in patients receiving GCSF during circulation. Conclusive evidence for improving transplant- the first cycle, without peak level maintenance during subse- free survival or sustained improvement in liver disease quent cycles. Four patients died of progressive liver failure severity scores have not been fully realized with these treat- during the treatment period, in whom peak levels of CD34+ ments38 (Table 1). cells were comparable to those who completed treatment. Spahr et al.39 showed that GCSF mobilized CD34+ cells, The CTP score improved without sustenance, while the increased HGF, and induced proliferation of hepatic progenitor MELD score did not show significant changes from baseline cells within 7 days of administration. HGF has been found to but rather increase beyond 9 months after GCSF use. The ameliorate liver fibrosis and prevent fulminant hepatic failure authors also studied the fate of induced CD34+ stem cells

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 371–383 375 Philips C.A. et al: Critical review of GCSF use in cirrhosis

Table 1. Clinical trials of granulocyte-colony stimulating factor (GCSF) therapy in patients with cirrhosis, outcomes, and critique

Follow up / Outcome / Salient Author / Country / Study type / Year features Comments

Spahr et al.39/ Switzerland / · 7 days follow up · Initial study showing safety of GCSF randomized controlled trial (RCT) / · Clinical outcomes not discussed and mobilization of bone marrow- 2008 · Drug safety demonstrated derived stem cells in patients with GCSF vs. standard of care · Changes in CD34+ cells advanced liver disease demonstrated · Included patients with cirrhosis and · Cytokines and aminopyrine breath alcoholic hepatitis tests similar between treated and control groups Han et al.42/ China / RCT / 2008 · 6 months follow up · GCSF did not improve liver function GCSF vs. autologous peripheral blood · Both groups showed improvement or liver disease severity monocyte cell (PBMC) transplantation in serum albumin and prothrombin · Autologous PBMC transplantation was time a superior modality of treatment in · Liver tests did not show significant advanced liver disease compared to improvement between groups GCSF · Improved Child-Turcotte-Pugh (CTP) score in the PBMC group Spahr et al.41/ Switzerland / RCT / · 3 months follow up · GCSF and stem cell infusions did not 2013 · Clinical outcomes similar between result in expansion of the hepatic GCSF vs. standard of care treated and control groups progenitor cell compartment within · Primary end point of 3-point decrease the liver microenvironment in model for end-stage liver disease · No improvement in liver (MELD) score same with GCSF and function standard medical care · Authors concluded insufficient · Histologically, only steatosis improved regenerative stimulation or resistance to liver regenerative drive in patients with decompensated alcoholic cirrhosis with exogenous therapy Xing et al.43/ China / RCT / 2013 · In hepatitis B virus (HBV)-related · No significant differences in total GCSF vs. standard of care cirrhosis bilirubin, albumin and prothrombin · Proportion of CD34+ cells increased time between the treated and control after GCSF groups · Matrix metalloproteinase level · No significant differences were significantly high before and observed in the cure and after GCSF improvement rates between the · Short-term disease severity not two groups affected by GCSF use Gaia et al.44/ Italy / non randomized, · 12 months follow up · GCSF could be safely administrated control study/ 2013 · 3 day GCSF course (5 mcg/kg every up to four times over a 1-year Multiple courses of GCSF in 12 h), administered at 3-month period in decompensated cirrhotic decompensated cirrhosis intervals for a total of four courses patients · Feasibility and safety explored · CD34+ cells increase · Telomere length was monitored to unsustained peak levels in subse- rule out early cell aging caused quent cycles by GCSF · Four patients died of progressive liver failure (CD34+ cells comparable to those who survived) · CTP score improved without maintenance, MELD score had no significant changes but worsened beyond 9 months · CD184 (repair of liver injury) reduction and loss of C-met (increases fibrosis) noted

(continued)

376 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 371–383 Philips C.A. et al: Critical review of GCSF use in cirrhosis

Table 1. (continued ) Follow up / Outcome / Salient Author / Country / Study type / Year features Comments

Kedarisetty et al.47/ India / RCT / · 12 month follow up · GCSF-only arm not studied 2015 · GCSF (5 mcg/kg/d) for 5 days and · Role of darbepoetin in regeneration GCSF + darbepoetin vs. placebo in then every third day (12 total doses) and amelioration of sepsis events not decompensated cirrhosis + subcutaneous darbepoetin-a studied (40 mcg/week) for 4 weeks · Pre- and post-treatment liver biopsy · Liver disease severity scores, sepsis done in only 5 patients in the treat- events and pre- and post-treatment ment group and 2 patients in the liver biopsies assessed. Survival at control group. Under-powered con- 12 months higher in the GCSF + dar- clusion regarding augmentation of bepoetin group (68.6% vs. 26.9%) hepatic regeneration · CTP scores were reduced by 48.6% in the GCSF group vs. 39.1% in the control group · MELD scores reduced by 40.4% after GCSF use · Need for large-volume paracentesis was significantly reduced and lower proportion of patients developed septic shock after GCSF use · On liver biopsy pre- and post-GCSF, increase in the proportion of CD34+ cells and CD133+ cells noted Prajapati et al.56/ India / RCT / 2017 · 6 months follow up · Acute-on-chronic liver failure GCSF in decompensated cirrhosis · GCSF at 300 mcg subcutaneous twice patients also included daily for 5 days plus standard medical · MELD progression not discussed therapy (SMT) or SMT alone · Specific extrahepatic and liver-related · In the GCSF group, 17 patients died events between groups not discussed and 9 were lost to follow-up · In the control group, 30 patients died and 11 were lost to follow-up · Survival with GCSF was higher (79 vs. 68%) · In the GCSF group, 66% of patients showed improvement or stability in the CTP score at 6 months, while in the control group it was 51% Verma et al.59/ India / RCT / 2018 · 12 months follow up · Increase in CD34+ cells at day 6 was Multiple courses of GCSF with or · Growth hormone (1U subcutaneous expected, does not translate to without growth hormone in per day) with GCSF (5 mcg/kg) sub- improved liver regeneration decompensated cirrhosis cutaneously every 12 hours for 5 · Long-term sustenance in CD34+ cell days, then every 3 months for 3 days levels and linked clinical events not till 12 months studied · GCSF-only arm · More than expected liver stiffness · Standard medical care-only arm improvement not explained with · The primary outcome was transplant- GCSF use; antifibrotic effects of GCSF free survival at 1 year not studied and remain unexplained · Survival significantly higher in GCSF- · Improved liver stiffness measure- treated patients ments not substantiated with liver · CD34+ cells increased at day 6 histology assessment · Significant decrease in clinical scores, · Very low MELD score patients and improvement in nutrition, better those not requiring liver transplant control of ascites, lesser infection listing also included in the study episodes · Striking decrease in liver stiffness after GCSF treatment

(continued)

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 371–383 377 Philips C.A. et al: Critical review of GCSF use in cirrhosis

Table 1. (continued ) Follow up / Outcome / Salient Author / Country / Study type / Year features Comments

Newsome et al.61/ United Kingdom / · 3 months follow up · The first multicenter, open-label, RCT / 2018 · Inclusion MELD scores of 11 to 15 randomized, controlled phase 2 trial Safety and efficacy of GCSF and · Subcutaneous GCSF (lenograstim) on GCSF in cirrhosis haemopoietic stem cell infusions in 15 mcg/kg for 5 days, or treatment · Very rigorous high-quality trial patients with compensated cirrhosis with GCSF for 5 days followed by leu- · Sufficiently powered kapheresis and intravenous infusion · Challenges findings of other similar of three doses of CD133+ hemato- studies poietic stem cells (0$2 3 106 cells · New onset ascites, sepsis and hepatic per kg per infusion) encephalopathy requiring multiple · Co-primary outcomes included hospital admissions after use of improvement in severity of liver GCSF/stem cell infusions compared to disease (change in MELD) at 3 months placebo and the trend of change in MELD score · Adverse events were greater in the over time treatment groups compared to · No improvement in liver dysfunction controls or markers of liver fibrosis occurred after the administration of GCSF or GCSF + stem-cell infusions · GCSF / GCSF + stem infusions worsened liver function and increased patient morbidity and mortality Anand et al.55/ India / RCT / 2019 · 12 months follow up · Lower MELD and lower CTP score GCSF / GCSF + erythropoietin in · GCSF given at a dose of 5 mcg/kg cirrhosis patients had better sur- decompensated cirrhosis subcutaneous at days 1, 2, 3, 4, 5 and vival; this could be true even without then every third day till day 60 (total treatment intervention 22 doses) · Response to treatment in patients · Erythropoietin given subcutaneously with higher grades of liver disease at dose of 500 IU/kg twice a week for severity was poor 2 months (total 16 doses) · Role of erythropoietin alone not · Follow-up until end of 12 months assessed · Combination revealed significant · Need for regenerative therapy in improvement in CTP and MELD scores lower MELD scores debatable compared to GCSF alone · Reduction in mortality better with combination (16.6% vs. 36.7%) · The combination treatment showed decreased acute kidney injury, encephalopathy and refilling of ascites incidence compared to monotherapy · Response poor in grade 3 ascites and better in Child B cirrhosis with MELD <16 Philips et al.62/ India / Real-world · 12 months follow up · Non-randomized, historical controls experience / 2019 · GCSF 10 mcg/kg per day for 5 days, · Included all patients who required GCSF in decompensated cirrhosis followed by 5 mcg/kg/day once every short- and intermediate-term trans- needing liver transplantation in the third day for total 12 doses plant-free survival intermediate term · Per protocol analysis (n = 56) and · Large number of patients, clarity in intention to treat analysis (n = 100) follow-up, and definition and identifi- · 16%, 43% and 75% patients died at cation of events 3, 6, and 12 months respectively · Provided novel data on CTP (>11) and · Sepsis most common cause of death, MELD (>20) cut-off at which GCSF use in 53% patients needs to be avoided · 9% developed hepatocellular carcinoma at the end of follow-up · Patients receiving GCSF had higher mortality at end of 12 months compared to controls (75% vs. 46%)

378 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 371–383 Philips C.A. et al: Critical review of GCSF use in cirrhosis expressing immature CD133 and CD117 markers (considered lower mortality rates in the short- and intermediate-term to home into the liver microenvironment, promoting liver remain debatable in the absence of quality long-term effects regeneration) and found no significant differences between with such experimental treatments. pretreatment and post-treatment levels of the same. Inter- Prajapati et al.56 conducted the largest randomized con- estingly, the authors found that CD184 and C-Met expression trolled study of GCSF therapy, using 253 decompensated cir- decreased significantly after treatment with GCSF in cirrhosis rhosis patients. The authors showed that the cumulative patients. CD184 is involved in repairing liver injury upon survival was significantly higher in GCSF-treated patients triggering MSCs to migrate, transdifferentiate, and fuse with compared to controls (79 vs. 68%) and that significantly hepatocytes, while the loss of c-Met was found to accelerate more patients in the GCSF group had an improvement in the development of liver fibrosis through deregulation of mul- CTP scores at 180 days. Even though those authors stated tiple molecular pathways. These findings demonstrate wor- that the inclusion of only decompensated cirrhosis as one of sening of liver fibrosis in patients with advanced liver the strengths of their study, the table detailing patient char- disease receiving GCSF.45,46 acteristics shows the inclusion of acute-on-chronic liver Kedarisetty et al.,47 utilized GCSF and darbepoetin in failure (ACLF) patients also. ACLF patients have completely patients with decompensated cirrhosis for comparison to different disease mechanisms, progression, clinical outcome placebo. The cumulative probability of survival at 12-months and possibly, distinct response to GCSF therapy in compari- was 68.6% in the GCSF + darbepoetin group and 26.9% in the son to patients with decompensated cirrhosis, and hence the placebo-treated patients. Sepsis events were lower in the study by Prajapati and colleagues56 does not fully realize GCSF + darbepoetin group compared to placebo-treated the potential of GCSF use in decompensated cirrhosis. The patients. It has also been demonstrated that erythropoietin authors do not discuss MELD progression in their study therapy reduced hypotension related to sepsis and promoted among GCSF-treated patients even though baseline values cardioprotective effects in animal models of sepsis, with a for the same are mentioned and the study is restricted to an reduction in acute kidney injury events during endotoxemia.48 intermediate follow up period. In a study by Preheim and colleagues,49 GCSF was admin- Chavez-Tapia et al.57 conducted a systematic review and istered to control and cirrhotic rats before and after induction metanalysis on GCSF use in patients with ACLF. The authors of pneumococcal pneumonia. The authors elegantly demon- included all randomized clinical trials comparing the use of strated that GCSF administered before infection did not any regimen of GCSF against placebo or no intervention; ulti- protect cirrhotic or control rats but did so after infection and mately, two trials involving 102 patients were included. A sig- significantly reduced mortality in control but not cirrhotic rats. nificant reduction in short-term overall mortality was In human trials on GCSF in cirrhotics, the findings are quite observed in patients receiving GCSF compared to controls. the opposite (i.e. GCSF prevented sepsis events).50 The Nonetheless, higher mortality secondary to gastrointestinal mechanism of action of GCSF in amelioration or prevention bleeding was noted in the GCSF-treated patients. The of sepsis in advanced cirrhosis is not yet elucidated. However, authors concluded that the GCSF-treated ACLF patients in the study by Fiuza et al.,51 GCSF was found to improve and had significantly reduced short-term mortality with limited increase recruitment of neutrophils to sites of infection and to evidence. Another metanalysis by Yang et al.50 included enhance neutrophil transendothelial migration in cirrhotic five studies with mostly 3 months follow-up in patients with patients in the absence of neutrophil adhesion, which could ACLF. Those authors concluded that GCSF treatment in be one of the reasons underlying the beneficial action of GCSF patients with advanced liver failure significantly improved in sepsis. liver function, reduced the incidence of sepsis, and prolonged Multiple studies have shed light on the detrimental effects short-term survival. Thus, the short-term survival with GCSF of GCSF in sepsis. GCSF has immunosuppressive effects on use in ACLF has been clearly defined. Long-term benefits, monocytes, macrophages, dendritic cells, and T lymphocytes transplant-free survival, and GCSF use in different etiologies, when exogenously administered, and high levels of GCSF in patients with sepsis and ACLF remain unknown and a have been found to negatively regulate IL-17 production, matter for future study. thereby worsening sepsis.52 Studies have also shown that In patients with ACLF, in contrast to those with decom- high level of GCSF at baseline was associated with poor pensated cirrhosis, immune dysregulation and higher sepsis outcome in sepsis. Stephens et al.53 demonstrated worsening events have been shown to predict poor outcomes. In studies of liver dysfunction with elevations in troponin levels patients on GCSF in ACLF, possible explanations for improved with sepsis treated with GCSF. Segal et al.54 reported that the outcomes include a reduction in sepsis by amelioration of novel existence of a hepatobiliary hybrid progenitor popula- immune dysfunction observed in ACLF, through an increase in tion anatomically restricted to the ductal plate of fetal liver, fraction of circulating and intrahepatic myeloid and plasma- with a transcriptional profile distinct from that of fetal hepa- cytoid dendritic cells. GCSF has also been shown to improve tocytes, mature hepatocytes and mature biliary epithelial survival in patients with ACLF due to reactivation of chronic cells, in human fetal liver using single-cell RNA sequencing. hepatitis B as well as alcoholic hepatitis. However, these This opens up newer horizons on regenerative therapies for studies were mostly from single centers. advanced liver disease. Robust data on the utility of GCSF in ACLF for specific Strong conclusions cannot be made from the data in the etiologies is lacking, the precise mechanism of action promot- Kedarisetty et al.47 study on survival benefits with GCSF due ing clinical benefit remains to be defined, and the dose and to the absence of a GCSF-only treatment arm. Similar find- duration in specific groups of ACLF classified as per severity ings were echoed in the study by Anand et al.55 The authors compared to decompensated cirrhosis is still warranted. showed that addition of erythropoietin to GCSF led to better Larger multicenter double-blind randomized trials in homo- regenerative response than GCSF monotherapy in patients geneous patient groups for validating the role and potential with low MELD score (<16). The need for regenerative cell mechanisms of action of GCSF in ACLF is the next step.58 therapy in patients with low MELD scores and relatively In this regard, the results of a large multicenter, open,

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 371–383 379 Philips C.A. et al: Critical review of GCSF use in cirrhosis

Fig. 3. The conundrum associated with GCSF and advanced liver disease in animal and human studies. Striking a balance is an unmet need to improve outcomes, which is currently still a matter of bench work before bedside use can be considered. Abbreviation: GCSF, granulocyte colony-stimulating factor. randomized, and controlled trial in patients with ACLF (The in GCSF-treated patients associated with a surprisingly striking Graft Trial) to evaluate efficacy and safety of subcutaneously decrease in liver stiffness measurements. Explanations for administered GCSF is much awaited (ClinicalTrials.gov Iden- such robust fibrosis reversal and improvement in synthetic tifier: NCT02669680). liver function was not supported by strong research data in Verma and colleagues59 studied clinical outcomes in the discussion. Reasons for GCSF related to ‘liver regeneration’ decompensated cirrhosis with multiple courses of GCSF with/ as discussed by the Verma et al.59 remain hypothetical. without growth hormone compared to SMT. Very low MELD The type of bone marrow-derived stem cell utilized has score patients who were exempt from liver transplantation been shown to affect outcomes in cell therapy for liver listing were also included in the study. The mean survival and diseases; for example, puritan mesenchymal cells, derived cumulative probability of transplant-free survival were higher monocyte or macrophage fractions were shown to have

380 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 371–383 Philips C.A. et al: Critical review of GCSF use in cirrhosis better efficacy in amelioration of chronic liver injury in characterized. Much of essential basic science work on GCSF preclinical studies. With GCSF use and an associated in liver fibrosis has not provided in-depth knowledge regard- general increase in bone marrow-derived stem cells, the ing its actions in amelioration of chronic liver injury and much needed ‘fate of cell’ tracer studies to correctly identify fibrosis. Even though clinical trials, mostly from the Indian the probable beneficial pathway associated with this therapy subcontinent, have shown improved outcomes with GCSF remain to be performed.60 use, rigorously designed high-quality trials and real-world Newsome and colleagues61 assessed the safety and effi- evidence have shown the contrary (Fig. 3). Hence, in ques- cacy of GCSF and CD133+ hemopoietic stem-cell infusions in tioning the depth of available data, the answers regarding the patients with compensated liver cirrhosis with MELD scores utility of GCSF in treating patients with cirrhosis remain ranging from 11 to 15.5 and found that liver dysfunction or unclear. fibrosis did not improve when compared to patients receiving To clearly understand the role of GCSF in liver cirrhosis, a standard care, and furthermore patients receiving the treat- systematic approach to the problem is warranted. First, one ment had a higher incidence of adverse liver-related events. must try to define and delineate the pathways affected This study cautioned on the use of growth factor therapy in through use of exogenous GCSF in chronic liver disease patients with compensated cirrhosis and moderately high animal models and in humans. It needs to be clear, with MELD scores. regards to GCSF therapy, if we are attempting to reduce Recently, Philips and colleagues62 published real-world portal hypertension or improve liver failure. With such an experience of GCSF in a large group of patients with cirrhosis attempt, the deleterious effects of GCSF could also be and active decompensations with higher MELD scores. Cirrho- studied. Second, dose-finding studies to clarify outcomes in sis patients with active ascites, jaundice, or both completed chronic liver disease models need to be performed. Third, the GCSF treatment (10 mcg/kg/day for 5 days, followed by roles of GCSF as an anti-inflammatory, antifibrotic or liver cell 5 mcg/kg/day once every third day for total 12 doses). restoration therapy need to be clearly defined, and as such A matched historical control group was used for comparing the dosing that promotes such specific activity needs to be outcomes. Among them, 16%, 43% and 75% of patients identified through thorough quality bench work. Fourth, died at 3, 6 and 12 months respectively, after GCSF treat- short-, intermediate- and long-term use of exogenously ment. Sepsis was the most frequent cause of death (in 53% administered GCSF (at specific doses in the finite period or of patients), followed by progressive liver failure (in 33%). multiple courses over more extended periods) and its effect Notably, a higher number of patients compared to the histor- on fibrogenesis and carcinogenesis has to be demonstrated. ical control group developed hepatocellular carcinoma at the All of these need thorough fate-of-cell tracer studies to end of 12 months. Acute variceal bleeds, overt hepatic ence- identify the true potential and action of GCSF in the liver phalopathy, intensive care unit admissions, and liver disease microenvironment. severity scores were higher after GCSF use at 12 months. Ultimately, once the ideal dose, duration, patient popula- A CTP score of >11 and MELD-sodium score of >20 predicted tion and window of opportunity based on liver disease worse outcomes at all time points and 12 months with GCSF severity has been defined, large multicenter trials following use, respectively. The modified intention to treat analysis this homogenous dosing regimen, with well-defined inclusion demonstrated poor overall survival at 6 months with GCSF criteria and patient to follow up methodology, remain the therapy compared to the historical controls (48% vs. 75%, unmet need. Until then, early and timely liver transplantation p = 0.04). The authors concluded that survival in decompo- remains the most beneficial treatment for patients with sition was shorter than what was expected in the natural decompensated cirrhosis. history of the disease after GCSF use in patients with advanced cirrhosis. Conflict of interest In the study by Kedarisetty et al.,47 a lower proportion of patients developed septic shock during the follow-up period Dr. Cyriac Abby Philips received advisory fees and research compared with controls, and by the end of 1 month after grant support from CiplaÒ,MylanÒ and Samarth Life-SciencesÒ. treatment the mean level of a-fetoprotein was significantly The other authors have no conflict of interests related to this higher in the growth factor group (6.6 ± 3.6 ng/mL) than in publication. the controls (4.7 ± 2.7 ng/mL). The latter was considered to be associated with hepatic regeneration. However, this was 62 contrary to findings described by Philips et al. and the Author contributions occurrence of sepsis as well as liver cancer was found to be 63 higher with GCSF use. Seehofer et al. demonstrated, in an Designed the study and wrote the initial manuscript (CAP, PA, animal model of chronic liver disease, that hepatic regenera- SR), conducted critical revisions to the manuscript (RA, TG, tion was slightly inhibited in the GCSF group. A study on the GCV, SKJ), made illustrations and obtained metadata for the effect of GCSF in liver fibrosis found that it significantly manuscript (CAP, RA, SR, TG). All authors approved the final 64 decreased the survival rate of mice. version of the manuscript.

Conclusions References

The way forward [1] Zhou WC, Zhang QB, Qiao L. Pathogenesis of liver cirrhosis. World J Gastro- enterol 2014;20:7312–7324. doi: 10.3748/wjg.v20.i23.7312. Liver fibrosis and its progression, and the ultimatum of [2] Puche JE, Saiman Y, Friedman SL. Hepatic stellate cells and liver fibrosis. – associated portal hypertension and liver failure, is a highly Compr Physiol 2013;3:1473 1492. doi: 10.1002/cphy.c120035. [3] Elpek GÖ. Cellular and molecular mechanisms in the pathogenesis of liver complex disease mechanism. Furthermore, the mechanisms fibrosis: An update. World J Gastroenterol 2014;20:7260–7276. doi: that define human liver regeneration remain inadequately 10.3748/wjg.v20.i23.7260.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 371–383 381 Philips C.A. et al: Critical review of GCSF use in cirrhosis

[4] Ray K. Liver: Key role for av integrins in myofibroblasts in liver fibrosis. Nat [29] Shaw BE, Confer DL, Hwang W, Pulsipher MA. A review of the genetic and Rev Gastroenterol Hepatol 2014;11:4. doi: 10.1038/nrgastro.2013.227. long-term effects of G-CSF injections in healthy donors: a reassuring lack of [5] Jiang JX, Mikami K, Venugopal S, Li Y, Török NJ. Apoptotic body engulfment evidence for the development of haematological malignancies. Bone Marrow by hepatic stellate cells promotes their survival by the JAK/STAT and Transplant 2015;50:334–340. doi: 10.1038/bmt.2014.278. Akt/NF-kappaB-dependent pathways. J Hepatol 2009;51:139–148. doi: [30] Stroncek D, Shawker T, Follmann D, Leitman SF. G-CSF-induced spleen size 10.1016/j.jhep.2009.03.024. changes in peripheral blood progenitor cell donors. Transfusion 2003;43: [6] Schuppan D, Kim YO. Evolving therapies for liver fibrosis. J Clin Invest 2013; 609–613. doi: 10.1046/j.1537-2995.2003.00384.x. 123:1887–1901. doi: 10.1172/JCI66028. [31] Bolognesi M, Merkel C, Sacerdoti D, Nava V, Gatta A. Role of spleen enlarge- [7] Wasmuth HE, Tacke F, Trautwein C. Chemokines in liver inflammation and ment in cirrhosis with portal hypertension. Dig Liver Dis 2002;34:144–150. fibrosis. Semin Liver Dis 2010;30:215–225. doi: 10.1055/s-0030-1255351. doi: 10.1016/S1590-8658(02)80246-8. [8] Simões E Silva AC, Miranda AS, Rocha NP, Teixeira AL. Renin angiotensin [32] Nakamura T, Torimura T, Iwamoto H, Kurogi J, Inoue H, Hori Y, et al. CD34(+) system in liver diseases: Friend or foe? World J Gastroenterol 2017;23: cell therapy is safe and effective in slowing the decline of hepatic reserve 3396–3406. doi: 10.3748/wjg.v23.i19.3396. function in patients with decompensated liver cirrhosis. J Gastroenterol [9] Song Y, Zhao Y, Wang F, Tao L, Xiao J, Yang C. Autophagy in hepatic fibrosis. Hepatol 2014;29:1830–1838. doi: 10.1111/jgh.12622. Biomed Res Int 2014;2014:436242. doi: 10.1155/2014/436242. [33] Gaia S, Smedile A, Omedè P, Olivero A, Sanavio F, Balzola F, et al. Feasibility [10] Zimmermann HW, Seidler S, Nattermann J, Gassler N, Hellerbrand C, and safety of G-CSF administration to induce bone marrow-derived cells Zernecke A, et al. Functional contribution of elevated circulating and mobilization in patients with end stage liver disease. J Hepatol 2006;45: hepatic non-classical CD14CD16 monocytes to inflammation and human 13–19. doi: 10.1016/j.jhep.2006.02.018. liver fibrosis. PLoS One 2010;5:e11049. doi: 10.1371/journal.pone. [34] Lorenzini S, Isidori A, Catani L, Gramenzi A, Talarico S, Bonifazi F, et al. Stem 0011049. cell mobilization and collection in patients with liver cirrhosis. Aliment Phar- [11] Kisseleva T, Brenner DA. The phenotypic fate and functional role for bone macol Ther 2008;27:932–939. doi: 10.1111/j.1365-2036.2008.03670.x. marrow-derived stem cells in liver fibrosis. J Hepatol 2012;56:965–972. doi: [35] Roberts AW. G-CSF: a key regulator of neutrophil production, but that’s not 10.1016/j.jhep.2011.09.021. all! Growth Factors 2005;23:33–41. doi: 10.1080/08977190500055836. [12] Dooley S, Hamzavi J, Ciuclan L, Godoy P, Ilkavets I, Ehnert S, et al. Hepato- [36] Gross-Weege W, Weiss M, Schneider M, Wenning M, Harms B, Dumon K, cyte-specific Smad7 expression attenuates TGF-beta-mediated fibrogenesis et al. Safety of a low-dosage Filgrastim (rhG-CSF) treatment in non- and protects against liver damage. Gastroenterology 2008;135:642–659. neutropenic surgical intensive care patients with an inflammatory process. doi: 10.1053/j.gastro.2008.04.038. Intensive Care Med 1997;23:16–22. doi: 10.1007/s001340050285. [13] Guimarães EL, Empsen C, Geerts A, van Grunsven LA. Advanced glycation [37] Lanthier N. Haemopoietic stem cell therapy in cirrhosis: the end of the story? end products induce production of reactive oxygen species via the activation Lancet Gastroenterol Hepatol 2018;3:3–5. doi: 10.1016/S2468-1253(17) of NADPH oxidase in murine hepatic stellate cells. J Hepatol 2010;52:389– 30359-X. 397. doi: 10.1016/j.jhep.2009.12.007. [38] Nicolas CT, Hickey RD, Chen HS, Mao SA, Lopera Higuita M, Wang Y, et al. [14] Ellis EL, Mann DA. Clinical evidence for the regression of liver fibrosis. Concise review: Liver regenerative medicine: from hepatocyte transplanta- J Hepatol 2012;56:1171–1180. doi: 10.1016/j.jhep.2011.09.024. tion to bioartificial livers and bioengineered grafts. Stem Cells 2017;35: [15] Jung YK, Yim HJ. Reversal of liver cirrhosis: current evidence and expect- 42–50. doi: 10.1002/stem.2500. ations. Korean J Intern Med 2017;32:213–228. doi: 10.3904/kjim.2016. [39] Spahr L, Lambert JF, Rubbia-Brandt L, Chalandon Y, Frossard JL, Giostra E, 268. et al. Granulocyte-colony stimulating factor induces proliferation of hepatic [16] Delire B, Stärkel P. The Ras/MAPK pathway and hepatocarcinoma: patho- progenitors in alcoholic steatohepatitis: a randomized trial. Hepatology genesis and therapeutic implications. Eur J Clin Invest 2015;45:609–623. 2008;48:221–229. doi: 10.1002/hep.22317. doi: 10.1111/eci.12441. [40] Mizuno S, Nakamura T. Hepatocyte growth factor: a regenerative drug [17] Li L, Zhao GD, Shi Z, Qi LL, Zhou LY, Fu ZX. The Ras/Raf/MEK/ERK signaling for acute hepatitis and liver cirrhosis. Regen Med 2007;2:161–170. doi: pathway and its role in the occurrence and development of HCC. Oncol Lett 10.2217/17460751.2.2.161. 2016;12:3045–3050. doi: 10.3892/ol.2016.5110. [41] Spahr L, Chalandon Y, Terraz S, Kindler V, Rubbia-Brandt L, Frossard JL, et al. [18] Ordelheide AM, Gommer N, Böhm A, Hermann C, Thielker I, Machicao F, et al. Autologous bone marrow mononuclear cell transplantation in patients with Granulocyte colony-stimulating factor (G-CSF): A saturated fatty acid- decompensated alcoholic liver disease: a randomized controlled trial. PLoS induced myokine with insulin-desensitizing properties in humans. Mol One 2013;8:e53719. doi: 10.1371/journal.pone.0053719. Metab 2016;5:305–316. doi: 10.1016/j.molmet.2016.02.001. [42] Han Y, Yan L, Han G, Zhou X, Hong L, Yin Z, et al. Controlled trials in hepatitis B [19] Muzzi A, Leandro G, Rubbia-Brandt L, James R, Keiser O, Malinverni R, et al. virus-related decompensate liver cirrhosis: peripheral blood monocyte trans- Insulin resistance is associated with liver fibrosis in non-diabetic chronic plant versus granulocyte-colony-stimulating factor mobilization therapy. hepatitis C patients. J Hepatol 2005;42:41–46. doi: 10.1016/j.jhep.2004. Cytotherapy 2008;10:390–396. doi: 10.1080/14653240802129901. 09.022. [43] Xing TJ, Xu HT, Xian JC, Shen ML, Li H, Ye J, et al. Mechanism and efficacy of [20] Tao LL, Cheng YY, Ding D, Mei S, Xu JW, Yu J, et al. C/EBP-a ameliorates CCl mobilization of granulocyte colony-stimulating factor in the treatment (4)-induced liver fibrosis in mice through promoting apoptosis of hepatic of chronic hepatic failure. Hepatogastroenterology 2013;60:170–175. stellate cells with little apoptotic effect on hepatocytes in vitro and in vivo. doi: 10.5754/hge12590 Apoptosis 2012;17:492–502. doi: 10.1007/s10495-012-0700-y. [44] Gaia S, Olivero A, Smedile A, Ruella M, Abate ML, Fadda M, et al. Multiple [21] Hirai H, Zhang P, Dayaram T, Hetherington CJ, Mizuno S, Imanishi J, et al. courses of G-CSF in patients with decompensated cirrhosis: consistent mobi- C/EBPbeta is required for ‘emergency’ granulopoiesis. Nat Immunol 2006;7: lization of immature cells expressing hepatocyte markers and exploratory 732–739. doi: 10.1038/ni1354. clinical evaluation. Hepatol Int 2013;7:1075–1083. doi: 10.1007/s12072- [22] Buck M, Chojkier M. A ribosomal S-6 kinase-mediated signal to C/EBP-beta is 013-9473-9. critical for the development of liver fibrosis. PLoS One 2007;2:e1372. doi: [45] Hao NB, Li CZ, Lü MH, Tang B, Wang SM, Wu YY, et al. SDF-1/CXCR4 axis 10.1371/journal.pone.0001372. promotes MSCs to repair liver injury partially through trans-differentiation [23] Gutschalk CM, Herold-Mende CC, Fusenig NE, Mueller MM. Granulocyte and fusion with hepatocytes. Stem Cells Int 2015;2015:960387. doi: 10. colony-stimulating factor and granulocyte-macrophage colony-stimulating 1155/2015/960387. factor promote malignant growth of cells from head and neck squamous [46] Marquardt JU, Seo D, Gómez-Quiroz LE, Uchida K, Gillen MC, Kitade M, et al. cell carcinomas in vivo. Cancer Res 2006;66:8026–8036. doi: 10. Loss of c-Met accelerates development of liver fibrosis in response to CCl(4) 1158/0008-5472.CAN-06-0158. exposure through deregulation of multiple molecular pathways. Biochim [24] Shojaei F, Wu X, Qu X, Kowanetz M, Yu L, Tan M, et al. G-CSF-initiated Biophys Acta 2012;1822:942–951. doi: 10.1016/j.bbadis.2012.02.012. myeloid cell mobilization and angiogenesis mediate tumor refractoriness to [47] Kedarisetty CK, Anand L, Bhardwaj A, Bhadoria AS, Kumar G, Vyas AK, et al. anti-VEGF therapy in mouse models. Proc Natl Acad Sci U S A 2009;106: Combination of granulocyte colony-stimulating factor and erythropoietin 6742–6747. doi: 10.1073/pnas.0902280106. improves outcomes of patients with decompensated cirrhosis. Gastroenter- [25] Elpek GÖ. Angiogenesis and liver fibrosis. World J Hepatol 2015;7:377–391. ology 2015;148:1362–1370.e7. doi: 10.1053/j.gastro.2015.02.054. doi: 10.4254/wjh.v7.i3.377. [48] Chousterman BG, Arnaud M. Is there a role for hematopoietic growth factors [26] Huebert RC, Rakela J. Cellular therapy for liver disease. Mayo Clin Proc 2014; during sepsis? Front Immunol 2018;9:1015. doi: 10.3389/fimmu.2018. 89:414–424. doi: 10.1016/j.mayocp.2013.10.023. 01015. [27] Serefhanoglu S, Goker H, Buyukasik Y, Turgut M, Sayinalp N, Haznedaroglu [49] Preheim LC, Snitily MU, Gentry MJ. Effects of granulocyte colony-stimulating IC, et al. Changes in vascular endothelial growth factor, angiopoietins, and factor in cirrhotic rats with pneumococcal pneumonia. J Infect Dis 1996;174: Tie-2 levels with G-CSF stimulation in healthy donors. Ann Hematol 2009; 225–228. doi: 10.1093/infdis/174.1.225. 88:667–671. doi: 10.1007/s00277-008-0657-7. [50] Yang Q, Yang Y, Shi Y, Lv F, He J, Chen Z. Effects of granulocyte colony- [28] Cong M, Iwaisako K, Jiang C, Kisseleva T. Cell signals influencing hepatic stimulating factor on patients with liver failure: a meta-analysis. J Clin fibrosis. Int J Hepatol 2012;2012:158547. doi: 10.1155/2012/158547. Transl Hepatol 2016;4:90–96. doi: 10.14218/JCTH.2016.00012.

382 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 371–383 Philips C.A. et al: Critical review of GCSF use in cirrhosis

[51] Fiuza C, Salcedo M, Clemente G, Tellado JM. Granulocyte colony-stimulating [58] Simonetto DA, Shah VH, Kamath PS. Improving survival in ACLF: growing factor improves deficient in vitro neutrophil transendothelial migration evidence for use of G-CSF. Hepatol Int 2017;11:473–475. doi: 10. in patients with advanced liver disease. Clin Diagn Lab Immunol 2002; 1007/s12072-017-9834-x. 9:433–439. doi: 10.1128/cdli.9.2.433-439.2002. [59] Verma N, Kaur A, Sharma R, Bhalla A, Sharma N, De A, et al. Outcomes after [52] Mohammad RA. Use of granulocyte colony-stimulating factor in patients with multiple courses of granulocyte colony-stimulating factor and growth severe sepsis or septic shock. Am J Health Syst Pharm 2010;67:1238–1245. hormone in decompensated cirrhosis: A randomized trial. Hepatology doi: 10.2146/ajhp090325. 2018;68:1559–1573. doi: 10.1002/hep.29763. [53] Stephens DP, Thomas JH, Higgins A, Bailey M, Anstey NM, Currie BJ, et al. [60] Gilchrist ES, Plevris JN. Bone marrow-derived stem cells in liver repair: Randomized, double-blind, placebo-controlled trial of granulocyte colony- 10 years down the line. Liver Transpl 2010;16:118–129. doi: 10. stimulating factor in patients with septic shock. Crit Care Med 2008;36: 1002/lt.21965. 448–454. doi: 10.1097/01.CCM.0B013E318161E480. [61] Newsome PN, Fox R, King AL, Barton D, Than NN, Moore J, et al. Granulocyte [54] Segal JM, Kent D, Wesche DJ, Ng SS, Serra M, Oulès B, et al. colony-stimulating factor and autologous CD133-positive stem-cell therapy Single cell analysis of human foetal liver captures the transcriptional profile in liver cirrhosis (REALISTIC): an open-label, randomised, controlled phase 2 of hepatobiliary hybrid progenitors. Nat Commun 2019;10:3350. doi: trial. Lancet Gastroenterol Hepatol 2018;3:25–36. doi: 10.1016/S2468- 10.1038/s41467-019-11266-x. 1253(17)30326-6. [55] Anand L, Bihari C, Kedarisetty CK, Rooge SB, Kumar D, Shubham S, et al. [62] Philips CA, Augustine P, Rajesh S, Ahamed R, George T, Padsalgi G, et al. Early cirrhosis and a preserved bone marrow niche favour regenerative Granulocyte colony-stimulating factor use in decompensated cirrhosis: lack response to growth factors in decompensated cirrhosis. Liver Int 2019;39: of survival benefit and probable predisposition to hepatocellular carcinoma. 115–126. doi: 10.1111/liv.13923. J Clin Exp Hepatol 2019. doi: 10.1016/j.jceh.2019.05.003. [56] Prajapati R, Arora A, Sharma P, Bansal N, Singla V, Kumar A. Granulocyte [63] Seehofer D, Neumann UP, Schirmeier A, Carter J, Cho SY, Lederer A, et al. colony-stimulating factor improves survival of patients with decompensated Synergistic effect of erythropoietin but not G-CSF in combination with cur- cirrhosis: a randomized-controlled trial. Eur J Gastroenterol Hepatol 2017; cumin on impaired liver regeneration in rats. Langenbecks Arch Surg 2008; 29:448–455. doi: 10.1097/MEG.0000000000000801. 393:325–332. doi: 10.1007/s00423-008-0290-x. [57] Chavez-Tapia NC, Mendiola-Pastrana I, Ornelas-Arroyo VJ, Noreña-Herrera C, [64] Ogiso T, Nagaki M, Takai S, Tsukada Y, Mukai T, Kimura K, et al. Granulocyte Vidaña-Perez D, Delgado-Sanchez G, et al. Granulocyte-colony stimulating colony-stimulating factor impairs liver regeneration in mice through the factor for acute-on-chronic liver failure: systematic review and meta-analysis. up-regulation of interleukin-1beta. J Hepatol 2007;47:816–825. doi: 10. Ann Hepatol 2015;14:631–641. doi: 10.1016/S1665-2681(19)30757-4. 1016/j.jhep.2007.06.017.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 371–383 383 Review Article

Intensive Care Management of Acute Liver Failure: Considerations While Awaiting Liver Transplantation

Anil Seetharam*

Transplant and Advanced Liver Disease Center, University of Arizona College of Medicine, Phoenix, AZ, USA

Abstract review of the European Liver Transplant Registry identified ALF as the primary indication for transplant in only 8% of Acute liver failure is a unique clinical phenomenon charac- transplants performed over a 20-year period.3,4 A key terized by abrupt deterioration in liver function and altered element in identifying ALF is the absence of preexisting liver mentation. The development of high-grade encephalopathy disease (distinguishing it from acute-on-chronic liver disease) and multisystem organ dysfunction herald poor prognosis. and international societies have guided definitions for Etiologic-specific treatments and supportive measures are severity of liver injury.5 routinely employed; however, liver transplantation remains A widely accepted working definition of ALF includes: an the only chance for cure in those who do not spontaneously international normalized ratio (INR) >1.5 and any degree of recover. The utility of artificial and bioartificial assist therapies mental alteration (encephalopathy) in a subject without as supportive care—to allow time for hepatic recovery or as a preexisting cirrhosis and with illness of <26 weeks’ duration.6 bridge to liver transplantation—has been examined but stud- Currently, overall short-term survival (2 years) in those ies have been small, with mixed results. Given the severity of undergoing transplantation approaches 90%.7 Liver trans- derangements, intensive critical care is needed to success- plantation (LT) remains the only definitive treatment for sub- fully bridge patients to transplant, and evaluation of candi- jects who fail to spontaneously recover or respond to dates occurs rapidly in parallel with serial reassessments of etiology-specific medical therapy. Moreover, patients are operative fitness. Psychosocial assessment is often subopti- often managed in an intensive care unit (ICU) setting while mal and relative contraindications to transplant, such as awaiting LT. The approach to the diagnostic evaluation of ALF ventilator-dependence may be overlooked. While often em- has been described elsewhere.8,9 The purpose of this review ployed to guide evaluation, no single prognostic model dis- will be to summarize recent advances in critical care manage- criminates those who will spontaneously recover and those ment, prognostic modeling, and evaluation of the ALF trans- who will require transplant. The purpose of this review will be plant candidate. to summarize approaches in critical care, prognostic modeling, and medical evaluation of the acute liver failure trans- Etiology-specific therapy plant candidate. Citation of this article: Seetharam A. Intensive care man- Etiology-specific treatment for ALF is often administered in an agement of acute liver failure: Considerations while awaiting ICU setting (Table 1).10 Both acetaminophen and non-acet- – liver transplantation. J Clin Transl Hepatol 2019;7(4):384 aminophen drug-induced liver injury, acute viral hepatitis, 391. doi: 10.14218/JCTH.2019.00032. autoimmune hepatitis, Wilson’s Disease, and vascular disorders represent examples for which etiology-specific therapy can be attempted to regain parenchymal function.11 Introduction In instances of known or suspected acetaminophen ingestion, N-acetylcysteine (NAC) can be given as an antidote Acute liver failure (ALF) is an infrequent condition in which orally or intravenously; its efficacy and safety is well-estab- rapid deterioration of liver function results in altered menta- lished.12,13 NAC should be given as early as possible, most tion and coagulopathy in individuals without pre-existing liver commonly via intravenous administration (loading dose of disease.1 Approximately 2,000 cases are thought to occur in 150 mg/kg in 5% dextrose over 15 m; maintenance dose of the USA annually.2 Studies estimating ALF prevalence and 50 mg/kg given over 4 h, followed by 100 mg/kg adminis- incidence in the European Union are lacking; however, tered over 16 h or 6 mg/kg/h) but may have value 48 h or more after ingestion.14 At this time, there is no consensus as Keywords: Acute liver failure; Thromboelastography; Intracranial hypertension; to whether a standard 72-h period is optimal or the further Liver assist therapy; Transplantation. continuation based on clinical course. Medications other than Abbreviations: ALF, acute liver failure; CPP, cerebral perfusion pressure; ICH, acetaminophen may be implicated as the causative agent for intracranial hypertension; ICP, intracranial pressure; ICU, intensive care unit; ALF.15 These are generally thought to represent examples of INR, international normalized ratio; KCC, King’s College Criteria; LT, liver transplantation; MAP, mean arterial pressure; MELD, model for end-stage liver disease; idiosyncratic drug hepatotoxicity. Investigation has sug- NAC, N-acetylcysteine; rFVIIa, recombinant activated factor seven. gested that a course of NAC (dosage outlined above) may Received: 31 July 2019; Revised: 6 October 2019; Accepted: 27 October 2019 abrogate non-acetaminophen-related drug-induced liver *Correspondence to: Anil Seetharam, Banner Transplant and Advanced Liver injury.16 In a recent randomized case-control study of 80 sub- Disease, University of Arizona College of Medicine, 441 N. 12th Street, 2nd Floor, Phoenix, AZ 85006, USA. Tel: +1-602-521-5800; Fax: +1-602-521-5337, jects with non-acetaminophen-induced liver failure, use of E-mail: [email protected] NAC was associated with a significantly lower mortality

384 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 384–391

Table 1. Etiology-specific therapies for ALF. Directed therapy of ALF when this may be deleterious, and transplant evaluation should not etiology is known may increase transplant-free survival. be delayed while awaiting a response.25 ’ Etiology Therapya Wilson s disease, an inherited metabolic disorder in copper transport, is a relatively infrequent cause of ALF.26 This fulmi- Acetaminophen and NAC: loading dose is 150 nant presentation is considered uniformly fatal without trans- non-acetaminophen mg/kg in 5% dextrose over plantation. Treatment to acutely lower serum copper and to drug-induced liver 15 m; maintenance dose is limit further hemolysis can include albumin dialysis, continu- injury 50 mg/kg given over 4 h ous hemofiltration, plasmapheresis, or plasma exchange;6 followed by 100 mg/kg although, such copper lowering measures can be employed, administered over 16 h or 6 recovery is very rare without LT.27 mg/kg/h Vascular disorders of the liver may also be associated with Hepatitis B Antiviral therapy with ALF. Budd-Chiari syndrome, characterized by acute hepatic nucleos(t)ide analogue: outflow obstruction, may precipitate severe acute liver injury entecavir or tenofovir requiring consideration of hepatic vein recanalization or shunting; preexisting hypercoagulable state must be inves- – Herpes (herpes simplex Acyclovir 5 10 mg/kg tigated and mandates treatment with systemic anticoagula- virus) intravenous every 8 h for at tion.28 Acute ischemic injury to the liver may cause ALF, least 7 days typically in patients with preexisting cardiac dysfunction ren- Autoimmune hepatitis Prednisone 40–60 mg PO dering hepatic tissue sensitive to minor reductions in systolic daily blood pressure—treatment in these instances generally Wilson’s disease Albumin dialysis, focuses on restoration of normal hemodynamics. continuous hemofiltration, Mushroom poisoning (most commonly associated with plasmapheresis, or plasma Aminita phalloides) may cause ALF. Typically, gastrointestinal exchange symptoms including profuse vomiting and diarrhea occur after mushroom ingestion. Gastric lavage, supportive meas- Budd-Chiari syndrome Attempts at hepatic vein ures, as well as various agents with antidotal properties (ben- recanalization with zylpenicllin, NAC, silymarin) can be considered.29 transjugular or direct Acute fatty liver of pregnancy is characterized by extensive portosystemic shunt; steatosis in the third trimester of pregnancy. Development, in systemic anticoagulation part, is related to inheritance patterns in mother and fetus, Mushroom Poisoning Gastric lavage, activated particularly of long-chain 3-hydroxylacyl-CoA dehydrogenase (Aminita phalloides) charcoal (also referred to as LCHAD) deficiency, which is linked to fetal ingestion NAC; silymarin fatty oxidation defects.30 Prompt delivery of the fetus is the Acute fatty liver of Delivery of the fetus preferred management and often reverses liver injury. pregnancy a Coagulopathy Etiology-directed therapy is administered concomitantly with intensive care support and medical evaluation for transplant. Abbreviations: ALF, acute liver failure; NAC, N-acetylcysteine. In ALF, prothrombin time and its derivative INR are elevated and are considered predictors of increased bleeding risk— though magnitude of effect is undefined. Previous investiga- (28% in NAC group vs. 53% control, p = 0.02) and shorter tion has found a concomitant and proportional reduction in length of stay.17 plasma levels of both procoagulants and natural anticoagu- Acute viral hepatitis acquisition with hepatitis A, B, D or E lant proteins, in conjunction with a significant elevation in may progress to ALF.7 In instances of serologically identified plasma levels of factors-VIII (referred to as FVIII) and Von hepatitis A and E, management is largely supportive. Nucleos Willebrand factor, culminating in an overall efficient, albeit (t)ide analogue therapy can be attempted in acute hepatitis B reduced, thrombin generation capacity in comparison with 31 or D; however, reversal of severe injury with early agents healthy individuals. Global hemostasis as assessed with (e.g. lamivudine) has rarely been achieved.18 Improved effi- thromboelastography may be normal by several compensa- cacy has been reported with new-generation therapy (ente- tory mechanisms, even in patients with markedly elevated INR.32 In the absence of overt bleeding, measures to cavir and tenofovir); a recent investigation found higher correct the INR with plasma are not recommended and may short-term efficacy with tenofovir in cases of acute-on- potentially obscure trends marking prognosis.6 chronic liver failure with suspected hepatitis B virus reactiva- Vitamin K deficiency has been reported in patients with tion.19 Herpes virus infection is an uncommon cause of ALF ALF33 and vitamin K (10 mg subcutaneously) can be admin- but has been reported in both immunocompromised and 20,21 istered (Table 2). Treatment of clotting factor deficiency is immunocompetent subjects. Empiric treatment should generally reserved for clinically significant bleeding or in prep- – be considered utilizing acyclovir (5 10 mg/kg every 8 h for aration for a high-risk invasive procedure (e.g., intracranial at least 7 days). Cytomegalovirus, varicella-zoster, Epstein- pressure (ICP) monitor placement). Fresh frozen plasma infu- Barr, and adenovirus have all been reported as viral causes of sion alone infrequently corrects a severely elevated INR and ALF, albeit rare, particularly in pediatric or immunocompro- carries risk of volume overload and transfusion-related lung – mised populations.22 24 injury.34 Use of recombinant activated factor seven (referred Autoimmune hepatitis may present with ALF.6 Initiation of to as rFVIIa) may be considered, as administration in combi- steroid therapy may be considered for some patients (predni- nation with frozen fresh plasma provides temporary correc- sone starting at 40–60 mg/day). However, in some patients tion of coagulopathy without volume overload.35,36 Important

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 384–391 385 Seetharam A.: Liver transplantation for acute liver failure barriers to the rFVIIa utilization include availability, cost, and Independent of vascular resistance, relative adrenal insuf- reports of thromboembolic phenomenon.37 There is no con- ficiency has been commonly observed in over half of patients sensus regarding the prophylactic administration of platelets presenting with ALF.41 Although a mortality benefit with in thrombocytopenic patients with ALF and this is generally hydrocortisone usage in ALF subjects has not been demon- reserved for cases of overt bleeding or prior to invasive strated, utility as an adjunctive measure to reduce systemic 42 procedures.6 vasopressor requirement has been demonstrated. Use of corticosteroids in the context of ALF must be individualized, taking into account etiology of liver injury and risk of Circulatory dysfunction/renal failure infection. Substantial renal dysfunction (either functional or as a ALF, similar to cirrhosis or sepsis, is characterized by low result of acute tubular necrosis) may occur in greater than 8 systemic vascular resistance. Diminished tone may compro- 50% of patients with ALF. Acute kidney injury more com- mise peripheral tissue oxygenation and promote multisystem monly occurs when acetaminophen is the etiology of injury and in the elderly.43 Renal failure contributes to morbidity organ failure. Maintenance of adequate mean arterial pres- and mortality and is associated with a poorer prognosis.44 sure (referred to as MAP) is of particular importance in ALF When support is required, continuous renal replacement subjects with increased ICP to maintain cerebral perfusion therapy is preferred over conventional hemodialysis, as the pressure (CPP) [CPP equates to MAP minus ICP]. Intravas- former has been shown in randomized trials to result in cular volume depletion is common at ALF presentation and improved stability in cardiovascular and intracranial parame- resuscitation is often required. Hypotensive patients with ALF ters compared to the latter.45 Avoidance of nephrotoxic should be resuscitated with normal saline and changed to agents, including intravenous contrast agents, is standard of half-normal saline containing 75 mEq/L sodium bicarbonate if care.6 acidotic.38 Volume expanding solutions should contain dextrose to prevent hypoglycemia if present.39 Subjects remaining hypotensive despite volume resusci- Encephalopathy/ICP management tation should be cultured for infection and considered for vasopressor support to maintain a MAP of at least 75 mmHg Cerebral edema and intracranial hypertension (ICH) repre- 38 or a CPP of 60–80 mmHg. While no studies have defined the sent the most serious complications of ALF.46 Although patho- optimal vasopressor, norepinephrine is known to augment genesis is only partly understood, there is evidence that both peripheral organ perfusion while minimizing tachycardia and systemic and local inflammation and circulating neurotoxins, 40 preserving splanchnic/hepatic blood flow. Vasopressin and/ including ammonia, play roles.47 Ammonia infusion causes or analogues may potentiate effects of norepinephrine and brain edema in animal models, and an arterial ammonia allow a decrease in its infusion rate, mitigating intense vaso- level >200 mg/dL in humans is strongly associated with cer- constriction in peripheral tissues that may lead to bowel and/ ebral herniation.48,49 Encephalopathy can be precipitated by or limb ischemia.6 infection or may occur as a result of low systemic blood

Table 2. Organ-specific supportive measures for ALF in the intensive care unit. Supportive measures to address multisystem organ dysfunction occurring in ALF generally requires intensive critical care.

Organ system Derangement Supportive measures

Hematologic Concomitant and proportional reduction in both Vitamin K 10 mg intravenous x 1 considered in those procoagulants and anticoagulants with nutritional deficiency Significant elevation in plasma levels of factors-VIII Fresh frozen plasma, platelets, and rFVIII reserved (FVIII) and Von Willebrand factor for active bleeding or invasive procedure Reduced thrombin generation capacity Cardiovascular Low systemic vascular resistance Hypotensive patients resuscitated with normal Diminished tone compromising peripheral tissue saline and changed to half-normal saline containing oxygenation 75 mEq/L sodium bicarbonate if acidotic Volume expanding solutions with dextrose to prevent hypoglycemia Vasopressor (norepinephrine) support to maintain a mean arterial pressure >75 mmHg or a cerebral perfusion pressure of 60–80 mmHg Renal Prerenal kidney injury from diminished effective Continuous renal replacement therapy preferred circulating volume, acute tubular necrosis, or over conventional hemodialysis reduced function Neurologic Systemic and local inflammation and circulating Consideration of intracranial monitor placementa neurotoxins, including ammonia that promote Head of bed elevation cerebral edema and intracranial hypertension Avoid endotracheal suction Mannitol a Intracranial pressure monitoring placement is variable and based on center expertise. Abbreviation: ALF, acute liver failure.

386 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 384–391 Seetharam A.: Liver transplantation for acute liver failure pressure and vasodilatation.50 Incidence of cerebral edema allow time for hepatic recovery or as a bridge to LT.65 Owing to and ICH in ALF is directly related to severity of hepatic ence- the relative rarity of the condition, clinical heterogeneity and phalopathy, with episodes of ICH occurring between 65–75% complexity in administration, studies have been small with in those with the highest grade.51 mixed results. In the FULMAR trial examining use of an As patients progress to higher grades of encephalopathy, extracorporeal device in patients with ALF, there was no intubation and mechanical ventilation are mandatory for improvement in overall survival in the cohort but there was airway protection. Propofol is routinely chosen for sedation a trend of potential benefit in patients with acetaminophen as because it may reduce cerebral blood flow.52 Frequent neuro- the etiology of ALF.66 Therapeutic plasma exchange has logical evaluation for signs of ICH (e.g., sluggish pupillary hypothesized applicability in ALF through removal of inhibi- reflexes and posturing) should be conducted. To reduce the tors of hepatic regeneration. In a multicenter randomized incidence of ICH, patients should have the head elevated at controlled trial, three 5-l sessions of therapeutic plasma 30 degrees and stimulation (including endotracheal suction- exchange significantly improved survival, although the sur- ing) and pain should be minimized.6,53 vival benefit was inferior to that seen with transplantation.67 Invasive and noninvasive monitoring strategies have been Other forms of cellular therapy (including hepatocyte trans- utilized to assess progression of cerebral edema.54 Use of ICP plantation) have showed potential as future treatments for monitoring devices is controversial but employed by many ALF but are not ready for routine clinical practice.68–70 ICUs. The principal concern is for intracranial hemorrhage Recent society guidelines suggest consideration of available with ICP placement, particularly in ALF subjects with support systems in patients with ALF with an expected poor advanced encephalopathy and active infection. Experience prognosis without transplantation but who have clear medical is limited with noninvasive strategies, such as transcranial or psychiatric contraindications to surgery.3 Doppler, jugular bulb oximetry, and pupillometry.55 If ICP monitoring is performed, both ICP and CPP are followed with Prognostic models a goal ICP below 20–25 mmHg and CPP maintained above – 56 50 60 mmHg. A number of different systems are in current worldwide use to Based on the role of arterial ammonia in pathogenesis, assess prognosis in patients with ALF (Table 3).71–73 All investigation has focused on the use of lactulose administra- systems utilize admission laboratory measures (coagulop- tion in ALF. In a study of ALF patients who received lactulose athy of highest weight) and recognize the development of compared to a matched group of patients who did not, a small encephalopathy and advanced age as markers of poor prog- increase in survival time was seen in those receiving lactu- nosis. One of the oldest and most utilized tools are the King’s lose, but there was no difference in the severity of encephal- College Criteria (KCC), the original iteration now used for 57 opathy or outcome. One concern regarding the use of 3 decades, being the first to recognize the importance of eti- lactulose in this setting is the precipitation of bowel gas with ology of ALF/acetaminophen toxicity carrying higher chance excessive usage, thereby complicating subsequent transplant of recovery than others.74 With over 3 decades of utilization, 11 surgery. the KCC is noted to have high specificity (relatively few cases Osmotic agents, such as mannitol, can be effective in fulfilling criteria would be ‘unnecessarily transplanted’); 58 decreasing cerebral edema. Mannitol was found to transi- however, it suffers from relatively low sensitivity (significant ently correct episodes of elevated ICP in ALF patients and also number of cases not fulfilling criteria would progress and die 59 to improve survival. Administration of intravenous mannitol without earlier identification and consideration of LT). The – (in a bolus dose of 0.5 1.0 g/kg) is recommended as first-line latest model from the King’s College is a dynamic outcome therapy of ICH in patients with ALF but repeated doses may prediction model and has been validated for use with ALF 6 trigger serum hyperosmolality. In patients with ALF and from acetaminophen toxicity.75 The model uses a number of severe hepatic encephalopathy, a controlled trial of the pro- clinical variables sequentially assessed from admission to phylactic induction of hypernatremia with hypertonic saline 72 h and has demonstrated good discrimination between sur- – (to a serum sodium 145 155 mEq/L) suggested a lower inci- vivors and non-survivors. 60 dence of ICH. Utilization of hypertonic saline as treatment Recently, the United States Acute Liver Failure Group for established ICH has not been studied. Barbiturate agents prospectively enrolled over 1900 subjects with ALF, (thiopental or pentobarbital) may also be considered to managed with and without transplantation.76 Investigators decrease ICP that has been nonresponsive to other meas- aimed to develop a model for ALF (all etiologies included) to 61 ures. Significant systemic hypotension may limit use and predict transplant-free survival at 21 days. Clinical demo- may necessitate vasopressor administration to maintain graphics and laboratory values were collected at enrollment adequate mean arterial pressure. In addition, barbiturate and recorded serially up to 1 week. Variables of prognostic clearance is markedly reduced in patients with ALF, poten- value adopted in the predictive model included admission tially confounding neurological assessments. Hypothermia coma grade, ALF etiology and vasopressor requirement, and 62 may prevent or control ICH in patients with ALF. Pilot admission INR and bilirubin values. The model correctly pre- – 8 studies suggest potential use (core temperature of 33 34 C) dicted outcome of illness in 66.3% of subjects, slightly out- 63 as a bridge to LT. Induced hypothermia has not been com- performing KCC and the model for end-stage liver disease pared to normothermic conditions in a controlled trial and (referred to as MELD) score. Performance appeared best in there is theoretical concern about a negative effect on patients with non-acetaminophen etiology and high-grade 64 hepatic regeneration. encephalopathy. A number of other scoring systems have been proposed to Advanced systems identify candidates most at risk for death and need for LT. Non-liver specific indices, such as the sequential organ failure Over the years, several artificial and bioartificial liver assist assessment which is widely used to quantify severity of therapies have been tested for utility as supportive care to multiorgan failure in other forms of critical illness, have

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 384–391 387 Seetharam A.: Liver transplantation for acute liver failure

Table 3. Prognostic models for ALF. Evaluation of various prognostic scoring systems in ALF cohorts.

Prognostic ALF Subjects Subjects died or Study model etiology studied, n transplanted Sensitivity Specificity AUROC

McPhail KCC All 2153 Unknown 0.55 0.79 0.76 et al.79 McPhail MELD All 2153 Unknown 0.74 0.67 0.78 et al.79 Koch et al.76 ALFSG All 1974 987 (50%) Not Not 0.84 reported reported Cholongitas SOFA APAP 125 58 (46%) 0.67 0.80 0.79 et al.78 only

Abbreviations: ALF, acute liver failure; ALFSG, Acute Liver Failure Study Group; APAP, acetaminophen; AUROC, area under the receiver operating curve; KCC, King’s College Criteria; MELD, model for end-stage liver disease score; SOFA, sequential organ failure assessment score. been utilized with comparable performance to KCC in pre- A number of shortcomings may be present despite thor- diction of non-survival; although, their organ “non-specific- ough evaluation. Living donor donation for ALF (particularly in ity” compromises applicability in determining benefit with LT areas of the world where access to cadaveric grafts may be and use is limited.77,78 The MELD score, widely used for liver low) raises ethical concern of coercion for potentially compat- prioritization/allocation in chronic liver disease, has been ible family members. In addition, given the acuity of illness investigated in ALF with similar performance to that of the and associated multisystemic organ dysfunction, determining KCC.79 An emerging theme in ALF prognostication is the futility or when the patient is too sick to transplant may be a need for individualized, dynamic assessments as opposed to clinical challenge, as widely accepted relative contraindica- historically static ones at presentation. An investigation utiliz- tions to LT (bacteremia or ventilatory-dependent respiratory ing the ALF early dynamic model followed clinical variables failure) may be overlooked if responding to critical care over 3 days and outperformed KCC and MELD in predicting measures. outcome in an observational study of 380 subjects.80 Adult ALF subjects who meet program criteria for listing in Recently, investigation has focused on the use of bio- the USA are granted status 1A, placing the patient at the markers in prediction of outcome with ALF. Circulating blood highest priority on the waiting list. In order to be listed as levels of caspase-cleaved and uncleaved cytokeratin K18 status 1A, the ALF candidate must be at least 18 years-old, (referred to as CK18), an apoptosis cell death marker, has with life expectancy of less than 7 days. Additionally, he or she shown promise.81 Subjects with spontaneous recovery of ALF must be in the ICU and meet at least 1 of the following have been demonstrated to have higher level of caspase acti- criteria: 1) ventilator dependence; 2) requiring renal replace- vation than in subjects who required transplant or died.82 Cir- ment therapy; 3) INR greater than 2.0.88 The mean time culating levels in concert with standard blood measures of between listing and LT is 2 days in the USA.89 coagulation and renal function demonstrated superior sensi- Currently, ALF accounts for approximately 8% of all liver tivity and specificity to KCC in predicting ALF outcome; transplants, as per data from the Scientific the Registry of although, further studies are needed.83 HLA-DR monocyte Transplant Recipients and the European Liver Transplant expression has been identified as a potential biomarker of Registry, with 1-year survival rates of 84% in the USA and ALF severity and outcome in acetaminophen-related ALF.84 79% in Europe, respectively.4,90 Outcomes are inferior when compared with patients receiving a transplant due to chronic liver disease; though survival rates even out beyond 1 year.91 Transplant evaluation and listing In a series of over 1400 patients from the United Network for Organ Sharing database of LT for ALF, body mass index While advances in critical care have improved survival with greater than 30, serum creatinine greater than 2 mg/dL, ALF, LT remains the standard of care for those with progres- 85 recipient age greater than 50 years, and a history of life sive injury. Given the severity of injury, multisystem organ support were independent factors of poor post-transplant dysfunction, and encephalopathy, the usual protocol for can- outcomes, with a survival of less than 50% in recipients didate evaluation is accelerated. Psychosocial assessment of with all factors pretransplant.92 An analysis of the European the potential recipient may be challenging, as ALF patients Liver Transplant Registry identified male gender, older age of with advanced stages of hepatic encephalopathy may be the donor (>60 years-old), older age of the recipient unable to respond to direct questioning. Candidate interview (>50 years-old), incompatible graft, and reduced graft size is often suboptimal, with collateral information often garnered as predictors of poor prognosis.4 from historical chart review and interview of available friends and family.86 Assessment of critical issues, such as adherence and social support, is often speculative. In certain cases of Conclusions ALF, individual centers may overlook traditional contraindications to LT, such as alcohol abuse, drug use or suicide ALF represents abrupt deterioration in hepatic function from attempt; as such, the decision to deny or support transplant diverse etiologies. Associated multisystem organ dysfunction listing should be done involving activation of a multidiscipli- and dense encephalopathy with chance for progression to nary transplant team with thorough documentation of brainstem herniation mandates prompt recognition and decision-making.87 transfer to a liver transplant center. While etiologic-specific

388 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 384–391 Seetharam A.: Liver transplantation for acute liver failure treatments and a number of systemic supportive measures stage non-acetaminophen acute liver failure. Gastroenterology 2009;137: can be employed, LT remains the only chance for cure in those 856-864, 864.e1. doi: 10.1053/j.gastro.2009.06.006. [17] Nabi T, Nabi S, Rafiq N, Shah A. Role of N-acetylcysteine treatment in non- who do not spontaneously recover. While often employed to acetaminophen-induced acute liver failure: A prospective study. Saudi J Gas- guide evaluation, no single prognostic model discriminates troenterol 2017;23:169–175. doi: 10.4103/1319-3767.207711. those who will spontaneously recover and those who will [18] Terrault NA, Lok ASF, McMahon BJ, Chang KM, Hwang JP, Jonas MM, et al. require transplant. Given the severity of derangements, Update on prevention, diagnosis, and treatment of chronic hepatitis B: AASLD 2018 hepatitis B guidance. Hepatology 2018;67:1560–1599. doi: intensive critical care is needed to bridge ALF patients to 10.1002/hep.29800. transplant; furthermore, evaluation of potential candidates [19] Wan YM, Li YH, Xu ZY, Wu HM, Xu Y, Wu XN, et al. Tenofovir Versus Entecavir must occur rapidly and in the context of serial reassessment for the Treatment of Acute-on-Chronic Liver Failure due to Reactivation of of fitness for surgery. Chronic Hepatitis B With Genotypes B and C. J Clin Gastroenterol 2019;53: e171–e177. doi: 10.1097/MCG.0000000000001038. [20] Levitsky J, Duddempudi AT, Lakeman FD, Whitley RJ, Luby JP, Lee WM, et al. Conflict of interest Detection and diagnosis of herpes simplex virus infection in adults with acute liver failure. Liver Transpl 2008;14:1498–1504. doi: 10.1002/lt.21567. [21] Little L, Rule J, Peng L, Gottfried M, Lee WM. Herpes simplex virus-associated The author has no conflict of interest related to this acute liver failure often goes unrecognized. Hepatology 2019;69:917–919. publication. doi: 10.1002/hep.30236. [22] Gupta E, Ballani N, Kumar M, Sarin SK. Role of non-hepatotropic viruses in acute sporadic viral hepatitis and acute-on-chronic liver failure in adults. Author contributions Indian J Gastroenterol 2015;34:448–452. doi: 10.1007/s12664-015- 0613-0. [23] Brewer EC, Hunter L. Acute liver failure due to disseminated varicella zoster Contributed to data review and writing of manuscript (AS). infection. Case Reports Hepatol 2018;2018:1269340. doi: 10. 1155/2018/1269340. [24] Núñez-Ramos R, Montoro S, Bellusci M, Del Fresno-Valencia MR, Germán- References Díaz M, Urruzuno P, et al. Acute liver failure: Outcome and value of pediatric end-stage liver disease score in pediatric cases. Pediatr Emerg Care 2016. [1] Mendizabal M, Silva MO. Liver transplantation in acute liver failure: A chal- doi: 10.1097/PEC.0000000000000884. lenging scenario. World J Gastroenterol 2016;22:1523–1531. doi: 10. [25] Rahim MN, Liberal R, Miquel R, Heaton ND, Heneghan MA. Acute severe 3748/wjg.v22.i4.1523. autoimmune hepatitis: Corticosteroids or liver transplantation? Liver [2] Lee WM, Squires RH Jr, Nyberg SL, Doo E, Hoofnagle JH. Acute liver failure: Transpl 2019;25:946–959. doi: 10.1002/lt.25451. Summary of a workshop. Hepatology 2008;47:1401–1415. doi: 10. [26] Roberts EA, Schilsky ML. Diagnosis and treatment of Wilson disease: an 1002/hep.22177. update. Hepatology 2008;47:2089–2111. doi: 10.1002/hep.22261. [3] Wendon J, Cordoba J, Dhawan A, Larsen FS, Manns M, Samuel D, et al. EASL [27] Korman JD, Volenberg I, Balko J, Webster J, Schiodt FV, Squires RH Jr, et al. Clinical Practical Guidelines on the management of acute (fulminant) liver Screening for Wilson disease in acute liver failure: a comparison of currently failure. J Hepatol 2017;66:1047–1081. doi: 10.1016/j.jhep.2016.12.003. available diagnostic tests. Hepatology 2008;48:1167–1174. doi: 10. [4] Germani G, Theocharidou E, Adam R, Karam V, Wendon J, O’Grady J, et al. 1002/hep.22446. Liver transplantation for acute liver failure in Europe: outcomes over 20 years [28] Eapen CE, Velissaris D, Heydtmann M, Gunson B, Olliff S, Elias E. Favourable from the ELTR database. J Hepatol 2012;57:288–296. doi: 10.1016/j.jhep. medium term outcome following hepatic vein recanalisation and/or trans- 2012.03.017. jugular intrahepatic portosystemic shunt for Budd Chiari syndrome. Gut [5] Sarin SK, Choudhury A, Sharma MK, Maiwall R, Al Mahtab M, Rahman S, et al. 2006;55:878–884. doi: 10.1136/gut.2005.071423. Acute-on-chronic liver failure: consensus recommendations of the Asian [29] Ye Y, Liu Z. Management of Amanita phalloides poisoning: A literature review Pacific association for the study of the liver (APASL): an update. Hepatol and update. J Crit Care 2018;46:17–22. doi: 10.1016/j.jcrc.2018.03.028. Int 2019;13:353–390. doi: 10.1007/s12072-019-09946-3. [30] Liu J, Ghaziani TT, Wolf JL. Acute fatty liver disease of pregnancy: Updates in [6] Lee WM, Stravitz RT, Larson AM. Introduction to the revised American Asso- pathogenesis, diagnosis, and management. Am J Gastroenterol 2017;112: ciation for the Study of Liver Diseases Position Paper on acute liver failure 838–846. doi: 10.1038/ajg.2017.54. 2011. Hepatology 2012;55:965–967. doi: 10.1002/hep.25551. [31] Agarwal B, Wright G, Gatt A, Riddell A, Vemala V, Mallett S, et al. Evaluation [7] Fontana RJ, Ellerbe C, Durkalski VE, Rangnekar A, Reddy RK, Stravitz T, et al. of coagulation abnormalities in acute liver failure. J Hepatol 2012;57:780– Two-year outcomes in initial survivors with acute liver failure: results from a 786. doi: 10.1016/j.jhep.2012.06.020. prospective, multicentre study. Liver Int 2015;35:370–380. doi: 10. [32] Stravitz RT, Lisman T, Luketic VA, Sterling RK, Puri P, Fuchs M, et al. Minimal 1111/liv.12632. effects of acute liver injury/acute liver failure on hemostasis as assessed by [8] Wang DW, Yin YM, Yao YM. Advances in the management of acute liver thromboelastography. J Hepatol 2012;56:129–136. doi: 10.1016/j.jhep. failure. World J Gastroenterol 2013;19:7069–7077. doi: 10.3748/wjg.v19. 2011.04.020. i41.7069. [33] Pereira SP, Rowbotham D, Fitt S, Shearer MJ, Wendon J, Williams R. Pharma- [9] Squires JE, McKiernan P, Squires RH. Acute liver failure: An update. Clin Liver cokinetics and efficacy of oral versus intravenous mixed-micellar phylloqui- Dis 2018;22:773–805. doi: 10.1016/j.cld.2018.06.009. none (vitamin K1) in severe acute liver disease. J Hepatol 2005;42:365–370. [10] Bernal W, Auzinger G, Dhawan A, Wendon J. Acute liver failure. Lancet 2010; doi: 10.1016/j.jhep.2004.11.030. 376:190–201. doi: 10.1016/S0140-6736(10)60274-7. [34] West KL, Adamson C, Hoffman M. Prophylactic correction of the international [11] Bernal W, Wendon J. Acute liver failure. N Engl J Med 2013;369:2525–2534. normalized ratio in neurosurgery: a brief review of a brief literature. J Neuro- doi: 10.1056/NEJMra1208937. surg 2011;114:9–18. doi: 10.3171/2010.7.JNS091857. [12] Smilkstein MJ, Knapp GL, Kulig KW, Rumack BH. Efficacy of oral N-acetylcys- [35] Shami VM, Caldwell SH, Hespenheide EE, Arseneau KO, Bickston SJ, Macik teine in the treatment of acetaminophen overdose. Analysis of the national BG. Recombinant activated factor VII for coagulopathy in fulminant hepatic multicenter study (1976 to 1985). N Engl J Med 1988;319:1557–1562. doi: failure compared with conventional therapy. Liver Transpl 2003;9:138–143. 10.1056/NEJM198812153192401. doi: 10.1053/jlts.2003.50017. [13] Keays R, Harrison PM, Wendon JA, Forbes A, Gove C, Alexander GJ, et al. [36] Le TV, Rumbak MJ, Liu SS, Alsina AE, van Loveren H, Agazzi S. Insertion of Intravenous acetylcysteine in paracetamol induced fulminant hepatic failure: intracranial pressure monitors in fulminant hepatic failure patients: early a prospective controlled trial. BMJ 1991;303:1026–1029. doi: 10.1136/bmj. experience using recombinant factor VII. Neurosurgery 2010;66:455–458. 303.6809.1026. doi: 10.1227/01.NEU.0000365517.52586.A2. [14] Harrison PM, Keays R, Bray GP, Alexander GJ, Williams R. Improved outcome [37] Pavese P, Bonadona A, Beaubien J, Labrecque P, Pernod G, Letoublon C, et al. of paracetamol-induced fulminant hepatic failure by late administration of FVIIa corrects the coagulopathy of fulminant hepatic failure but may be acetylcysteine. Lancet 1990;335:1572–1573. doi: 10.1016/0140-6736 associated with thrombosis: a report of four cases. Can J Anaesth 2005; (90)91388-q. 52:26–29. doi: 10.1007/BF03018576. [15] Rockey DC, Seeff LB, Rochon J, Freston J, Chalasani N, Bonacini M, et al. [38] Stravitz RT, Kramer DJ. Management of acute liver failure. Nat Rev Gastro- Causality assessment in drug-induced liver injury using a structured expert enterol Hepatol 2009;6:542–553. doi: 10.1038/nrgastro.2009.127. opinion process: comparison to the Roussel-Uclaf causality assessment [39] Etogo-Asse FE, Vincent RP, Hughes SA, Auzinger G, Le Roux CW, Wendon J, method. Hepatology 2010;51:2117–2126. doi: 10.1002/hep.23577. et al. High density lipoprotein in patients with liver failure; relation to sepsis, [16] Lee WM, Hynan LS, Rossaro L, Fontana RJ, Stravitz RT, Larson AM, et al. adrenal function and outcome of illness. Liver Int 2012;32:128–136. doi: 10. Intravenous N-acetylcysteine improves transplant-free survival in early 1111/j.1478-3231.2011.02657.x.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 384–391 389 Seetharam A.: Liver transplantation for acute liver failure

[40] Eefsen M, Dethloff T, Frederiksen HJ, Hauerberg J, Hansen BA, Larsen FS. liver failure. Transplantation 2003;75:2034–2039. doi: 10.1097/01.TP. Comparison of terlipressin and noradrenalin on cerebral perfusion, intracra- 0000066240.42113.FF. nial pressure and cerebral extracellular concentrations of lactate and pyru- [64] Munoz SJ. Hypothermia may impair hepatic regeneration in acute liver vate in patients with acute liver failure in need of inotropic support. J Hepatol failure. Gastroenterology 2005;128:1143–1144. doi: 10.1053/j.gastro. 2007;47:381–386. doi: 10.1016/j.jhep.2007.04.015. 2005.02.048. [41] Harry R, Auzinger G, Wendon J. The clinical importance of adrenal insuffi- [65] García Martínez JJ, Bendjelid K. Artificial liver support systems: what is new ciency in acute hepatic dysfunction. Hepatology 2002;36:395–402. doi: 10. over the last decade? Ann Intensive Care 2018;8:109. doi: 10. 1053/jhep.2002.34514. 1186/s13613-018-0453-z. [42] Harry R, Auzinger G, Wendon J. The effects of supraphysiological doses of [66] Saliba F, Camus C, Durand F, Mathurin P, Letierce A, Delafosse B, et al. corticosteroids in hypotensive liver failure. Liver Int 2003;23:71–77. doi: 10. Albumin dialysis with a noncell artificial liver support device in patients 1034/j.1600-0676.2003.00813.x. with acute liver failure: a randomized, controlled trial. Ann Intern Med [43] Leithead JA, Ferguson JW, Bates CM, Davidson JS, Lee A, Bathgate AJ, et al. 2013;159:522–531. doi: 10.7326/0003-4819-159-8-201310150-00005. The systemic inflammatory response syndrome is predictive of renal dys- [67] Larsen FS, Schmidt LE, Bernsmeier C, Rasmussen A, Isoniemi H, Patel VC, function in patients with non-paracetamol-induced acute liver failure. Gut et al. High-volume plasma exchange in patients with acute liver failure: An 2009;58:443–449. doi: 10.1136/gut.2008.154120. open randomised controlled trial. J Hepatol 2016;64:69–78. doi: 10.1016/j. [44] Tujios SR, Hynan LS, Vazquez MA, Larson AM, Seremba E, Sanders CM, et al. jhep.2015.08.018. Risk factors and outcomes of acute kidney injury in patients with acute liver [68] Atta HM. Gene therapy for liver regeneration: experimental studies and pros- – failure. Clin Gastroenterol Hepatol 2015;13:352–359. doi: 10.1016/j.cgh. pects for clinical trials. World J Gastroenterol 2010;16:4019 4030. doi: 10. 2014.07.011. 3748/wjg.v16.i32.4019. [45] Davenport A, Will EJ, Davidson AM. Improved cardiovascular stability during [69] Ibars EP, Cortes M, Tolosa L, Gómez-Lechón MJ, López S, Castell JV, et al. continuous modes of renal replacement therapy in critically ill patients with Hepatocyte transplantation program: Lessons learned and future strategies. – acute hepatic and renal failure. Crit Care Med 1993;21:328–338. doi: 10. World J Gastroenterol 2016;22:874 886. doi: 10.3748/wjg.v22.i2.874. 1097/00003246-199303000-00007. [70] Iansante V, Mitry RR, Filippi C, Fitzpatrick E, Dhawan A. Human hepatocyte [46] Vaquero J, Chung C, Cahill ME, Blei AT. Pathogenesis of hepatic encephalop- transplantation for liver disease: current status and future perspectives. – athy in acute liver failure. Semin Liver Dis 2003;23:259–269. doi: 10. Pediatr Res 2018;83:232 240. doi: 10.1038/pr.2017.284. 1055/s-2003-42644. [71] Simpson KJ, Bates CM, Henderson NC, Wigmore SJ, Garden OJ, Lee A, et al. [47] Scott TR, Kronsten VT, Hughes RD, Shawcross DL. Pathophysiology of cere- The utilization of liver transplantation in the management of acute liver failure: comparison between acetaminophen and non-acetaminophen etiol- bral oedema in acute liver failure. World J Gastroenterol 2013;19:9240– ogies. Liver Transpl 2009;15:600–609. doi: 10.1002/lt.21681. 9255. doi: 10.3748/wjg.v19.i48.9240. [72] Oketani M, Ido A, Nakayama N, Takikawa Y, Naiki T, Yamagishi Y, et al. Eti- [48] Reynolds AS, Brush B, Schiano TD, Reilly KJ, Dangayach NS. Neurological ology and prognosis of fulminant hepatitis and late-onset hepatic failure in Monitoring in Acute Liver Failure. Hepatology 2019;70:1830–1835. doi: 10. Japan: Summary of the annual nationwide survey between 2004 and 2009. 1002/hep.30760. Hepatol Res 2013;43:97–105. doi: 10.1111/j.1872-034X.2012.01105.x. [49] Clemmesen JO, Larsen FS, Kondrup J, Hansen BA, Ott P. Cerebral herniation [73] Ichai P, Legeai C, Francoz C, Boudjema K, Boillot O, Ducerf C, et al. Patients in patients with acute liver failure is correlated with arterial ammonia con- with acute liver failure listed for superurgent liver transplantation in France: centration. Hepatology 1999;29:648–653. doi: 10.1002/hep.510290309. reevaluation of the Clichy-Villejuif criteria. Liver Transpl 2015;21:512–523. [50] Kitzberger R, Funk GC, Holzinger U, Miehsler W, Kramer L, Kaider A, et al. doi: 10.1002/lt.24092. Severity of organ failure is an independent predictor of intracranial hyper- [74] Yoon E, Babar A, Choudhary M, Kutner M, Pyrsopoulos N. Acetaminophen- tension in acute liver failure. Clin Gastroenterol Hepatol 2009;7:1000–1006. induced hepatotoxicity: A comprehensive update. J Clin Transl Hepatol 2016; doi: 10.1016/j.cgh.2009.05.019. 4:131–142. doi: 10.14218/JCTH.2015.00052. [51] Kok B, Karvellas CJ. Management of cerebral edema in acute liver failure. [75] Bernal W, Wang Y, Maggs J, Willars C, Sizer E, Auzinger G, et al. Development Semin Respir Crit Care Med 2017;38:821–829. doi: 10.1055/s-0037- and validation of a dynamic outcome prediction model for paracetamol- 1608772. induced acute liver failure: a cohort study. Lancet Gastroenterol Hepatol [52] Wijdicks EF, Nyberg SL. Propofol to control intracranial pressure in fulminant 2016;1:217–225. doi: 10.1016/S2468-1253(16)30007-3. hepatic failure. Transplant Proc 2002;34:1220–1222. doi: 10.1016/s0041- [76] Koch DG, Tillman H, Durkalski V, Lee WM, Reuben A. Development of a model 1345(02)02804-x. to predict transplant-free survival of patients with acute liver failure. Clin [53] Lele AV, Wilson D, Chalise P, Nazzaro J, Krishnamoorthy V, Vavilala MS. Dif- Gastroenterol Hepatol 2016;14:1199–1206.e2. doi: 10.1016/j.cgh.2016. ferences in blood pressure by measurement technique in neurocritically ill 03.046. patients: A technological assessment. J Clin Neurosci 2018;47:97–102. doi: [77] Craig DG, Zafar S, Reid TW, Martin KG, Davidson JS, Hayes PC, et al. The 10.1016/j.jocn.2017.10.079. sequential organ failure assessment (SOFA) score is an effective triage [54] Rajaram P, Subramanian R. Management of acute liver failure in the intensive marker following staggered paracetamol (acetaminophen) overdose. – care unit setting. Clin Liver Dis 2018;22:403 408. doi: 10.1016/j.cld.2018. Aliment Pharmacol Ther 2012;35:1408–1415. doi: 10.1111/j.1365-2036. 01.013. 2012.05102.x. [55] Kandiah PA, Olson JC, Subramanian RM. Emerging strategies for the treat- [78] Cholongitas E, Theocharidou E, Vasianopoulou P, Betrosian A, Shaw S, Patch D, ment of patients with acute hepatic failure. Curr Opin Crit Care 2016;22: et al. Comparison of the sequential organ failure assessment score with the – 142 151. doi: 10.1097/MCC.0000000000000291. King’s College Hospital criteria and the model for end-stage liver disease score [56] Rajajee V, Fontana RJ, Courey AJ, Patil PG. Protocol based invasive intra- for the prognosis of acetaminophen-induced acute liver failure. Liver Transpl cranial pressure monitoring in acute liver failure: feasibility, safety and 2012;18:405–412. doi: 10.1002/lt.23370. impact on management. Crit Care 2017;21:178. doi: 10.1186/s13054- [79] McPhail MJ, Farne H, Senvar N, Wendon JA, Bernal W. Ability of king’s college 017-1762-6. criteria and model for end-stage liver disease scores to predict mortality of “ ” [57] Butterworth RF. The concept of the inflamed brain in acute liver failure: patients with acute liver failure: A meta-analysis. Clin Gastroenterol Hepatol mechanisms and new therapeutic opportunities. Metab Brain Dis 2016;31: 2016;14:516–525.e5. doi: 10.1016/j.cgh.2015.10.007. 1283–1287. doi: 10.1007/s11011-015-9747-0. [80] Kumar R, Shalimar, Sharma H, Goyal R, Kumar A, Khanal S, et al. Prospective [58] Bunchorntavakul C, Reddy KR. Acute liver failure. Clin Liver Dis 2017;21: derivation and validation of early dynamic model for predicting outcome in 769–792. doi: 10.1016/j.cld.2017.06.002. patients with acute liver failure. Gut 2012;61:1068–1075. doi: 10. [59] Canalese J, Gimson AE, Davis C, Mellon PJ, Davis M, Williams R. Controlled 1136/gutjnl-2011-301762. trial of dexamethasone and mannitol for the cerebral oedema of fulminant [81] Eguchi A, Wree A, Feldstein AE. Biomarkers of liver cell death. J Hepatol hepatic failure. Gut 1982;23:625–629. doi: 10.1136/gut.23.7.625. 2014;60:1063–1074. doi: 10.1016/j.jhep.2013.12.026. [60] Murphy N, Auzinger G, Bernel W, Wendon J. The effect of hypertonic sodium [82] Rutherford A, King LY, Hynan LS, Vedvyas C, Lin W, Lee WM, et al. Develop- chloride on intracranial pressure in patients with acute liver failure. Hepatol- ment of an accurate index for predicting outcomes of patients with acute liver ogy 2004;39:464–470. doi: 10.1002/hep.20056. failure. Gastroenterology 2012;143:1237–1243. doi: 10.1053/j.gastro. [61] Mpabanzi L, Jalan R. Neurological complications of acute liver failure: path- 2012.07.113. ophysiological basis of current management and emerging therapies. Neuro- [83] Bechmann LP, Jochum C, Kocabayoglu P, Sowa JP, Kassalik M, Gieseler RK, chem Int 2012;60:736–742. doi: 10.1016/j.neuint.2011.10.014. et al. Cytokeratin 18-based modification of the MELD score improves predic- [62] Karvellas CJ, Todd Stravitz R, Battenhouse H, Lee WM, Schilsky ML. Thera- tion of spontaneous survival after acute liver injury. J Hepatol 2010;53:639– peutic hypothermia in acute liver failure: a multicenter retrospective cohort 647. doi: 10.1016/j.jhep.2010.04.029. analysis. Liver Transpl 2015;21:4–12. doi: 10.1002/lt.24021. [84] Antoniades CG, Berry PA, Davies ET, Hussain M, Bernal W, Vergani D, et al. [63] Jalan R, Olde Damink SW, Deutz NE, Davies NA, Garden OJ, Madhavan KK, Reduced monocyte HLA-DR expression: a novel biomarker of disease et al. Moderate hypothermia prevents cerebral hyperemia and increase in severity and outcome in acetaminophen-induced acute liver failure. Hepatology intracranial pressure in patients undergoing liver transplantation for acute 2006;44:34–43. doi: 10.1002/hep.21240

390 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 384–391 Seetharam A.: Liver transplantation for acute liver failure

[85] O’Grady J. Liver transplantation for acute liver failure. Best Pract Res Clin than 200 patients over a 17-year period. Ann Surg 2003;237:666–675. Gastroenterol 2012;26:27–33. doi: 10.1016/j.bpg.2012.01.012. doi: 10.1097/01.SLA.0000064365.54197.9E. [86] Samuel D, Saliba F, Ichai P. Changing outcomes in acute liver failure: Can we [90] Freeman RB Jr, Steffick DE, Guidinger MK, Farmer DG, Berg CL, Merion RM. transplant only the ones who really need it? Liver Transpl 2015;21 Suppl 1: Liver and intestine transplantation in the United States, 1997–2006. Am J S36–S38. doi: 10.1002/lt.24310. Transplant 2008;8:958–976. doi: 10.1111/j.1600-6143.2008.02174.x. [87] Carrion AF, Martin P. Non-intensive care unit management of acute liver [91] O’Grady J. Timing and benefit of liver transplantation in acute liver failure. J failure. Clin Liver Dis 2018;22:389–401. doi: 10.1016/j.cld.2018.01.009. Hepatol 2014;60:663–670. doi: 10.1016/j.jhep.2013.10.024. [88] Organ Procurement and Transplantation Network (OPTN) Policies. Available [92] Barshes NR, Lee TC, Balkrishnan R, Karpen SJ, Carter BA, Goss JA. Risk from: https://optn.transplant.hrsa.gov/media/1200/optn_policies.pdf. stratification of adult patients undergoing orthotopic liver transplantation [89] Farmer DG, Anselmo DM, Ghobrial RM, Yersiz H, McDiarmid SV, Cao C, et al. for fulminant hepatic failure. Transplantation 2006;81:195–201. doi: 10. Liver transplantation for fulminant hepatic failure: experience with more 1097/01.tp.0000188149.90975.63.

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 384–391 391 Reviewer Acknowledgement

2019 Reviewer Acknowledgement

Editorial Office of Journal of Clinical and Translational Hepatology

We thank the following reviewers for their contribution and support in 2019.

Mai Ali Abd el Meguid Peter Buch Xingrong Du Egypt United States United States Mohamed Abdallah Al Hajj Chalermrat Bunchorntavakul Atanu Kumar Dutta Abdallah Thailand India United States Ozge Telci Caklili Mohamed A. El-Guindi Sherief Abd-Elsalam Turkey Egypt Egypt Wenqing (Wendy) Cao Gülsüm Ozlem Elpek Anant Agarwalla United States Turkey United States Vincenzo Cardinale Saleh Elwir Avin Aggarwal Italy United States United States Guido Carpino Yu-Chen Fan Kawtar Alkhalloufi Italy China United States Ryan Chadha Jian-Gao Fan United States China Ahmed M. Alkhunaizi Saudi Arabia Naichaya Chamroonkul Myriam Farah Thailand Canada Khaled Amer Egypt Jason Pik Eu Chang Artin Galoosian Singapore United States Masahiro Arai Japan Phunchai Charatcharoenwitthaya Yanhang Gao Thailand China Leon Averbukh United States Limin Chen Antonio Giorgio Canada Italy Timothy A. Bach United States Yongpeng Chen Amit Goel China India Gyorgy Baffy United States José Artur Chies George Boon-Bee Goh Brazil Singapore Adriana de Los Anteles Barrios Guatemala Ashok Choudhury Humberto C Gonzalez India United States Roopjeet Bath United States Alessandro Csermely Kazuhiko Hayashi Italy Japan Ines Bilic-Curcic Jessica Dahmus Noha Helal CROATIA United States Egypt Thomas Graham Bird Victor David W Jing Jing United Kingdom United States China John W Birk Giovan Giuseppe Di Costanzo sandeep kumar karn United States Italy China Iliana B Doycheva Ahyoung Kim q DOI: 10.14218/JCTH.2019.000RA. United States United States

392 Journal of Clinical and Translational Hepatology 2019 vol. 7 | 392–393

John Koskinas Jie Peng Hsin-Yun Sun Greece China Taiwan Karen Krok Cyriac Abby Philips Sridhar Sundaram United States India India Pramod Kumar Madhumita Premkumar Ashraf Tabll India India Egypt Ewa Kurys-Denis Xingshun Qi Tawesak Tanwandee Poland China Thailand Chaoshuang Lin Bo Qin Georgios Tsoulfas China China Greece Cheng-Hai Liu Atoosa Rabiee Lucija Virovic-Jukic China United States Croatia Xiaocong Liu Chitta Ranjan Benjamin l Woolbright United States India United States Shousheng Liu Sahaj Rathi QIULIXIE China Canada China Joyce A. Lloyd Aathira Ravindranath Chencheng Xie United States India United States Lungen Lu Ciro Romano Yongning Xin China Italy China Mojtaba mafi Farzin Roohvand Weijie Xu Iran (Islamic Republic of) Iran (Islamic Republic of) China Sunil K Malonia Mehdi Sadat Po-Chih Yang United States Iran (Islamic Republic of) Taiwan Alessandro Mantovani Ghada Salum Eric M Yoshida Italy Egypt Canada Qing Mao Keisaku Sato Yuexin Zhang China United States China Rosa Maria Martin-Mateos Moinak Sen Sarma Rong Zhang Spain India United States Yong-Ping Mu Ke-Qing Shi Qingzhan Zhang China China United States Seyedehsan Navabi Hao Shi Ze-Hua Zhao United States United States China Paul Naylor Coleman Smith Ming-Hua Zheng United States United States China Bolin Niu Bin Song Yueyong Zhu United States China China James Park Hang Sun Julie Zhu United States China Canada

Journal of Clinical and Translational Hepatology 2019 vol. 7 | 392–393 393 Editorial Board

Avin Aggarwal Peng Hu Ahmed Mesalam David Victor Las Vegas, USA Chongqing, China Cairo, Egypt New York, USA Gianfranco D. Alpini Jing Hua Albert D. Min Genshu Wang Temple, USA Shanghai, China New York, USA Guangzhou, China Moustafa El Awady Yan Huang Paul Naylor Leyi Wang Cairo, Egypt Changsha, China Detroit, USA Urbana, USA Sina Aziz Yuehua Huang Calvin Q. Pan Yongning Xin Karachi, Pakistan Guangzhou, China New York, USA Qingdao, China Gyorgy Baffy Trana Hussaini James S. Park Ming Yan Boston, USA Vancouver, Canada New York, USA Jinan, China Mahmoud Mohamed Bahgat Hartmut Jaeschke María Teresa Pérez-Gracia Dongliang Yang Cairo, Egypt Kansas City, USA València, Spain Wuhan, China Fernando Bessone Wasim Jafri Cyriac Abby Philips Li Yang Rosario, Argentina Karachi, Parkistan Kochi, India Cincinnati, USA Kalyan Ram Bhamidimarri Tatsuo Kanda Nikolaos T. Pyrsopoulos Tian Yang Miami, USA Tokyo, Japan Newark, USA Xi'an, China Peter Buch Lindsey Kennedy Alok Ranjan Eric M. Yoshida Farmington, USA Temple, USA Bethesda, USA Vancouver, Canada Chalermrat Bunchorntavakul John Koskinas Sammy Saab Samar Samir Youssef Bangkok, Thailand Athens, Greece Los Angeles, USA Cairo, Egypt Phunchai Charatcharoenwitthaya Anastasios Koulaouzidis Behnam Saberi Yufeng Yuan Bangkok, Thailand Edinburgh, USA Baltimore, USA Wuhan, China Enqiang Chen Anand V. Kulkarni Moinak Sen Sarma Maysaa El Sayed Zaki Chengdu, China Hyderabad, India Lucknow, India Cairo, Egypt Po-Hung Chen Ashish Kumar Keqing Shi Yuanyuan Zhang Baltimore, USA New Delhi, India Wenzhou, China Chengdu, China Ashok Kumar Choudhury Manoj Kumar Gamal Shiha Xinxin Zhang New Delhi, India New Delhi, India Mansoura, Egypt Shanghai, China Jianqiang Ding Jun Li Surajit Sinha Shicheng Zheng Foshan, China Hangzhou, China Bethesda, USA Chengdu, China Jiangao Fan Bing Liu Coleman Smith Huiping Zhou Shanghai, China Guangzhou, China Washington, USA Richmond, USA Eduardo Fernández-Martínez Xiangming Lao Robert Smolic Hong Zhou Hidalgo, Mexico Guangzhou, China Osijek, Croatia Nanjing, China Heather L Francis Chenghai Liu Martina Smolic Julie Zhu Bryan, USA Shanghai, China Osijek, Croatia Halifax, Canada Catherine Frenette Manqing Liu Jonathan G. Stine La Jolla, USA Wuhan, China Charlottesville, USA George Boon-Bee Goh Fengmin Lu Giovanni Targher Singapore Beijing, China Verona, Italy Yanhang Gao Mingqin Lu Rolf Teschke Changchun, China Wenzhou, China Frankfurt, Germany Ahmet Gurakar Mazyar Malakouti Claudio Tiribelli Baltimore, USA San Antonio, USA Trieste, Italy Steven-Huy Bui Han Alessandro Mantovani Vladimir Maximovich Tsyrkunov Los Angeles, USA Verona, Italy Grodno, Belarus Ying Han Qing Mao Zhengkun Tu Xi’an, China Chongqing, China Changchun, China Kazuhiko Hayashi Matthew McMillin Nagoya, Japan Austin, USA Song-Bing He Fanyin Meng Suzhou, China Temple, USA JCTH