Consensus of Gastroesophageal Reflux Disease in Taiwan With

+ MODEL Advances in Digestive Medicine (2015) xx,1e10

Available online at www.sciencedirect.com ScienceDirect

journal homepage: www.aidm-online.com

ORIGINAL ARTICLE Consensus of gastroesophageal reﬂux disease in Taiwan with endoscopy-based approach covered by National Health Insurance Bor-Shyang Sheu a,b, Cheng-Tang Chiu c, Yi-Chia Lee d, Chi-Yang Chang e, Deng-Chyang Wu f, Jyh-Ming Liou d, Ming-Shiang Wu d, Wei-Lun Chang a,b, Chun-Ying Wu g, Jaw-Town Lin h,* a Institute of Clinical Medicine, National Cheng Kung University Hospital, College of Medicine, National Cheng Kung University, Tainan, Taiwan b Department of Internal Medicine, National Cheng Kung University Hospital, College of Medicine, National Cheng Kung University, Tainan, Taiwan c Department of Gastroenterology and Hepatology, Linkou Chang Gung Memorial Hospital, Chang Gung University, College of Medicine, Taoyuan, Taiwan d Department of Internal Medicine, National Taiwan University Hospital, Taipei, Taiwan e Department of Internal Medicine, E-Da Hospital, I-Shou University, Kaoshiung, Taiwan f Department of Internal Medicine, Kaohsiung Medical University, Kaohsiung, Taiwan g Department of Internal Medicine, Taichung Veterans General Hospital, Taichung, Taiwan h School of Medicine, Fu Jen Catholic University, New Taipei City, Taiwan

Received 10 April 2015; accepted 14 May 2015

KEYWORDS Summary Background and aims: Gastroesophageal reflux disease (GERD) is emerging as a clin- Consensus; ical complication in the Orient. The consensus comprises recommendations to GERD control under Gastroesophageal the advantage of endoscopy-based approach covered by the Taiwan National Health Insurance. reflux; Methods: The steering committee defined the consensus scope to cover diagnostic, therapeutic, Level of evidence; unresolved, controversial, or long-term proton pump inhibitor-related issues to GERD. The litera- National health ture review emphasized domestic data, after which the draft statements and statement evidence insurance; levels were defined. Thirty-five experts of GERD in Taiwan formed the expert group to conduct the Proton pump inhibitor; consensus conference by a modified Delphi process to vote anonymously to reach a consensus, Recommendation defined by an agreement of 80% for each statement, and to set the recommendation grade.

* Corresponding author. School of Medicine, Fu Jen Catholic University, Number 510, Zhongzheng Road, Xinzhuang District, New Taipei City 24205, Taiwan. E-mail address: [email protected] (J.-T. Lin). http://dx.doi.org/10.1016/j.aidm.2015.05.002 2351-9797/Copyright ª 2015, The Gastroenterological Society of Taiwan, The Digestive Endoscopy Society of Taiwan and Taiwan Association for the Study of the Liver. Published by Elsevier Taiwan LLC. All rights reserved.

Results: The consensus included 22 statements, including seven on diagnostic approach, seven on therapeutic suggestion, and eight on unresolved, controversial, or long-term proton pump inhibitor-related issues to GERD. The consensus highlighted that the endoscopy approach to GERD can deﬁne the disease spectrum and exclude malignant potential. The questionnaire survey can not only deﬁne GERD, but also monitor treatment response and quality of life. The consensus addressed suggestions for the unresolved issues related to extraesophageal presentation and adverse concerns of GERD after long-term use of proton pump inhibitors. In the endemic area of upper gastrointestinal cancers, Helicobacter pylori eradication is suggested to reduce progression of gastric precancerous lesions, and endoscopic surveillance of Barrett’s esophagus with dysplasia de- serves prospective research. Conclusion: The consensus comprises recommendations for the management of GERD in a high upper gastrointestinal cancer area with a national coverage of endoscopic approach. Copyright ª 2015, The Gastroenterological Society of Taiwan, The Digestive Endoscopy Society of Taiwan and Taiwan Association for the Study of the Liver. Published by Elsevier Taiwan LLC. All rights reserved.

Introduction term issues of GERD for future improvement in such endemic area with upper gastrointestinal cancers. Diagnosis and treatment of gastroesophageal reflux disease (GERD) are becoming more important [1e4]. GERD has not Methods only chronic recurrent bothersome clinical symptoms, but also long-term impacts on the quality of life and perfor- Steering committee set the consensus scope and mance of diseased individuals. In the Western world, due to structure a low prevalence of esophageal cancer and upper gastrointestinal malignancy, empiric antisecretory medication, such as proton pump inhibitors (PPIs), has been widely To establish the expert consensus of GERD in Taiwan, the applied for patients with typical presentations such as acid steering committee was initiated by J.T. Lin, chaired by regurgitation or heart burn sensation at the epigastric or B.S. Sheu along with eight other opinion leaders from the midchest regions to indicate reflux-related disorders [4e8]. Gastroenterological Society of Taiwan (C.T. Chiu, Y.C. Lee, Nevertheless, the current role of esophageal endoscopy G.Y. Chang, D.C. Wu, C.M. Liou, M.S. Wu, W.L. Chang, and remains helpful to confirm the diagnosis of erosive esoph- C.Y. Wu). The steering committee defined the scope ses- agitis and then to validate the degree of esophageal mu- sions of the consensus, conducted a literature search and cosa defects according to the Los Angeles grading review, formulated draft statements, and defined the classification [9e14]. The different disease spectrums, such statement evidence level. as erosive or nonerosive GERD, will have different treatment modalities. Accordingly, even PPI can be empirically Steering committee members to conduct literature started in the Western world, the role of endoscopic-based search and review approach for GERD shall remain with more accurate diagnostic determination to the disease spectrum and to The literature searches included Medline, Embase, the exclude the malignant potentials of upper gastrointestinal Cochrane Central Register of Controlled Trial, and ISI Web tract in the Eastern world with high cancer endemics. of Knowledge, with manual searches of bibliographies of In Taiwan, endoscopy can be covered by the Taiwan key articles and proceedings of abstracts of major gastro- National Health Insurance program, and based on the enterology conferences held over the past 7 years. The key endoscopic findings to define the disease spectrum of words used in the search included gastroesophageal reflux, GERD, PPIs can be allowed for durations ranging from at PPI, Barrett’s esophagus, extraesophageal symptoms, nar- least 4 months up to 12 months. Accordingly, a nationwide row band image, upper gastrointestinal endoscopy, etc. model is required to offer an objective evidence of erosive The members of the steering committee summarized the GERD control. The advantage of the current consensus is findings in the three scope sessions of this consensus: (1) that it offers Taiwanese experts’ experiences with diagnostic approach; (2) therapeutic assessment; and (3) endoscopic-based approach supported by the National unresolved, controversial, or long-term PPI-related issues Health Insurance in a near 95% nationwide coverage. The associated with GERD. Based on the review of the litera- current consensus has provided a strong evidence of its ture, the draft statements of the consensus were estab- validity for GERD management in a nationwide cohort lished by the session leader(s) of each scope session. For setting, including the diagnostic approach and therapeutic each statement, the level of evidence was defined ac- assessment. The consensus also addressed important con- cording to modified grading of the Oxford Centre for cerns to highlight the unresolved, controversial, and long- Evidence-Based Medicine Levels of Evidence (March 2009),

Please cite this article in press as: Sheu B-S, et al., Consensus of gastroesophageal reflux disease in Taiwan with endoscopy-based approach covered by National Health Insurance, Advances in Digestive Medicine (2015), http://dx.doi.org/10.1016/j.aidm.2015.05.002 + MODEL Taiwan expert consensus on GERD 3 as applied in our previous consensus [15]. The draft state- possible reason is related to Helicobacter pylori eradication ments were refined at the steering committee meeting held as a common practice for the treatment of the intragastric in Kaohsiung, Taiwan during June 2014. pathology. A previous meta-analysis showed a significant association between absence of H. pylori infection and GERD symptoms, and a positive association between anti-H. Expert group meeting to achieve agreement of pylori therapy and occurrence of both de novo and statement and recommendation grading rebound/exacerbated GERD [19]. In a community-based screening program in Taiwan, eradication of H. pylori A total of 35 experts (the names are listed in the supple- infection has led to increased esophagitis [20]. Although mentary data), including 10 members in the steering com- the causal relationship between GERD and H. pylori infec- mittee and 25 members who accepted the invitation of the tion may be confounded by other etiologies, an inverse steering committee, comprised the expert group of the association is generally observed. Taiwan GERD consensus. The draft statements from the four session groups were sent to all the experts, together with pertinent literature, prior to the consensus meeting in Kaohsiung in August 2014. Statement I-2: The validated questionnaire is useful During the 2-day consensus meeting, for each draft in the diagnosis of GERD, monitoring of treatment statement from the four scope sessions, the supporting response, and evaluation of quality of life. evidence from the keynote literature summary by the (agreement: 100%, level of evidence: 3b, recom- steering committee was presented serially in the following mendation: A) order: cover diagnostic, therapeutic, unresolved, controversial, long-term PPI-related issues for GERD. Based on a modified Delphi process through two separate iterations, all Several validated questionnaires have been developed participants voted anonymously for the first round of for the assessment of GERD. Clinical applications of the statements and modified the statements through discus- GERD questionnaires may include the diagnosis of GERD, sion. The modified statements were followed by a second assessment of treatment response, and evaluation of GERD round of voting with electronic keypads until a consensus symptoms and their effects on the quality of life. For was reached at the agreement percentage of 80%. If the example, the GERD Questionnaire has been proved to be a agreement was < 80%, the statement was rejected. The useful tool to diagnose GERD with an area under the expert members also discussed the level of evidence sug- receiver operating characteristic curve of 0.7 [21]. A higher gested by the steering committee and then graded the GERD Questionnaire score is significantly correlated with a recommendation level by voting for each statement. The higher risk of erosive esophagitis [22]. The Reflux Disease grade of recommendation ranged from A to D, as was used Questionnaire, which includes both the presence of symp- in our previous consensus in Taiwan [15]. The level of toms and the severity/frequency of GERD, has been found recommendation was defined as the grade with the highest to be useful in monitoring its therapeutic response to PPIs number of votes of the expert group members. The con- [23,24]. The Short-Form 36 questionnaire has been used in ferences were underwritten by unrestricted grants from the the evaluation of health status and quality of life in GERD Gastroenterological Society of Taiwan. Mandatory written patients [25]. disclosures of financial conflicts of interest within the period of 3 years prior to the meetings were obtained from all experts prior to voting. Statement I-3: In Taiwan where upper GI malignancies remain prevalent, upper GI endoscopy is useful to Consensus statements evaluate the diagnosis, set the disease spectrum of GERD, and exclude the possibility of cancer. (agreement: 97%, level of evidence: 3a, recom- Section I: diagnostic approach mendation: A)

Statement I-1: In Taiwan, the prevalence of GERD is In Taiwan, concomitant GERD symptoms are commonly increasing while that of Helicobacter pylori infection observed in patients with upper aerodigestive tract neo- is declining. plasms, and approximately two-thirds of the patients with (agreement: 100%, level of evidence: 2a, recom- fresh hypopharyngeal cancer have erosive esophagitis, mendation: A) active H. pylori infection, or gastric/duodenal ulcers [26]. In patients with gastric cancer, similarly, concomitant GERD/dyspepsia symptoms are very common [27]. There- fore, in this population with prevalent upper GI malig- In Taiwan, both the endoscopic indication and the diagnosis nancies, upper endoscopy is well accepted by of GERD have substantially increased, and the increasing gastroenterologists as the ﬁrst-line diagnostic tool for the trend of GERD is likely related to a higher proportion of low- evaluation of GERD symptoms, categorization of disease grade erosive esophagitis detected by endoscopy [16]. The spectrum of GERD (such as erosive GERD and non-erosive possible reason can be related to the increase in population reﬂux disease (NERD)), and most importantly, exclusion of with obesity and metabolic syndrome [17,18]. Another the possibility of upper GI tract cancer.

monitoring improves diagnosis by showing weak acidic and nonacidic refluxes [42]. In patients with refractory GERD Statement I-4: PPI-first approach may be considered symptoms, the true diagnosis of GERD can be confirmed by for the treatment-naı¨ve patients with low risk of abnormal acid exposure on 24-hour PH monitoring. By upper GI cancer or patients with relapse GERD symp- contrast, negative results of PH monitoring suggest that toms shortly after treatment. symptoms are unlikely due to acid reflux [43]. (agreement: 100%, level of evidence: 3a, recommendation: A)

Statement I-7: In selected patients for whom invasive PPIs are the most effective drugs to suppress gastric acid diagnostic testing of GERD is not feasible, esophageal secretion and relieve reflux symptoms. Therefore, an radiographic or scintigraphic study could be an empirical short course (usually 7e14 days) of PPIs, the so- option. called PPI test, is useful in the diagnosis of GERD [28,29].In (agreement: 82%, level of evidence: 4, recommen- patients with GERD symptoms, a previous meta-analysis has dation: B) shown pooled sensitivity and specificity of 71% and 41%, respectively [30]. The data indicated that the PPI test was sensitive, but less specific. In patients with noncardiac Traditional diagnoses of GERD, including endoscopy, chest pain, the PPI test showed pooled sensitivity and manometry, and pH studies, may be cumbersome in some specificity of 80% and 74%, respectively, to confirm GERD cases, such as in patients who have undergone total or [31]. A meta-analysis indicated that the optimal duration proximal gastrectomy, those with nasogastric tube or gas- may be 1 week and the optimal cutoff value should be a trostomy, and children. Esophageal radiographic or scinti- decrease of heartburn score by > 75% [32]. graphic studies can be an alternative choice to define the presence of GERD [44,45]. More specifically, radionuclide scintigraphy may provide important diagnostic clues under Statement I-5: Image-enhanced endoscopy with or special conditions, and the reported sensitivities of without magnification may help diagnosis of patients esophageal scintigraphy in the diagnosis of GERD could be with NERD and predict who will respond to PPI. up to 76% [46,47]. (agreement: 88%, level of evidence: 3a, recommendation: B) Section II: therapeutic assessment Image-enhanced endoscopy provides a significant improvement over standard white-light imaging for the diagnosis of nonerosive GERD, with better visualization of Statement II-1: In GERD patients with persistent subtle erosive changes in the esophagus. The sensitivity and symptoms despite PPI therapy, double-dosing or specificity of narrow-band images in differentiation of switching to a different PPI after ensuring the nonerosive reflux diseases and controls were 65% and 83%, compliance may provide symptoms relief. respectively, based on the presence of dilated intra- (agreement: 94%, level of evidence: 1b, recom- papillary capillary loops under endoscopic magnification mendation: B) [33]. The inter- and intraobserver agreement in the grading of endoscopic esophagitis could be improved when image- enhanced endoscopy was used [34e36]. Certain specific narrow band image findings can even correlate with a Persistent heartburn and other upper gastrointestinal positive therapeutic response [37]. symptoms despite PPI therapy have become clinical prob- lems that are increasing in gastroenterology practices. Approximately 25% of PPI recipients will require higher doses than initially prescribed because of insufficient con- Statement I-6: In patients with atypical or refractory trol of GERD symptoms or lack of healing of the esophageal GERD symptoms, esophageal function studies, such as mucosa [48]. A randomized control trial suggested that manometry, pH meter, or impedance-pH monitoring, double dosing of lansoprazole or switching to esomeprazole may be useful in the investigation of etiology. can deal with patients suffering from persistent symptoms (agreement: 100%, level of evidence: 3a, recom- despite lansoprazole therapy [49]. More studies confirming mendation: A) similar effectiveness of double dosing or switching to a different PPI can really be helpful. In patients with atypical or refractory GERD symptoms, it is important to identify those with true GERD and other etiology mimicking GERDs such as atypical presentation of Statement II-2: PPI treatment with on-demand or achalasia and some of esophageal motility disorders continuous therapy can be administered for GERD [38,39]. Additional diagnosis with esophageal manometry patients with symptoms relapse. and 24-hour pH monitoring is required to detect other dis- (agreement: 97%, level of evidence: 1a, recom- ease entities that may manifest like GERD [40,41]. The mendation: A) addition of impedance to conventional 24-hour pH

Several studies showed that there are no differences in symptomatic remission and improvement of quality of life Statement II-6: Barrett’s esophagus remains uncom- between on-demand and continuous therapy with PPIs for mon in Taiwan compared with Western countries. GERD [50e52]. However, continuous therapy has a higher (agreement: 100%, level of evidence: 3a, recom- healing rate of erosive esophagitis [53,54]. However, most mendation: A) of the studies focus on mild GERD cases. Further studies should be undertaken to investigate severe GERD. The prevalence of Barrett’s esophagus in Taiwan was 0.06e2.0%, and it varies with study design [59e62]. Chen et al [16] found that the prevalence of Barrett’s esophagus Statement II-3: The response to PPI therapy for did not change over time, although the prevalence of GERD symptoms relief in patients with non-erosive reflux was increasing. Chang et al [61] used a prospective design disease is generally lower than that in those with with a standardized biopsy protocol and histological erosive esophagitis. confirmation for Barrett’s esophagus to include both (agreement: 97%, level of evidence: 1a, recom- symptomatic and asymptomatic patients from a large mendation: A) ethnic Chinese population. The prevalence (0.85%) of Bar- rett’s esophagus in this study is lower than that in Western A multicenter study showed that the heartburn resolu- countries. The prevalence of Barrett’s esophagus varies tion rates at 4 weeks were higher for patients with erosive from 2.4% to 13.6% in Western countries, according to some esophagitis than for those with nonerosive esophagitis [55]. studies [63e66], using the same design (prospective A systemic review reveals that PPI can provide higher enrollment, endoscopic biopsy from standard four quad- therapeutic gain for heartburn resolution in patients with rants, pathological confirmation, and a large case number). nonerosive reflux disease than in those with erosive esophagitis. Statement II-7: Prague C and M criteria is recommended in the endoscopic description of Barrett’s Statement II-4: PPI therapy for NERD patients may be esophagus. augmented with prokinetics and alginates. (agreement: 94%, level of evidence: 3a, recom- (agreement: 82%, level of evidence: 1b recommen- mendation: B) dation: B) The utility of Prague C and M criteria has been proved by Administration of mosapride citrate in addition to a large prospective study in Taiwan [61]. A multinational omeprazole improved GERD symptoms and gastric emptying endoscopic study showed that the intraclass correlation in PPI-resistant nonerosive GERD patients with delayed coefficient value for the scores of the C value was 0.92 and gastric emptying [56]. Another study illustrated that that for M value was 0.94, indicating an excellent inter- omeprazole combined with sodium alginate provided higher observer agreement [67]. In other words, Prague C and M resolution of heartburn than omeprazole alone in patients criteria have good validity and reliability for description of with nonerosive GERD [57]. Due to the limited case Barrett’s esophagus. However, a long-term follow-up study numbers in the previous two studies, more evidence is should be conducted to verify the relationship between the anticipated in future studies. clinical outcome of Barrett’s esophagus and the severity of Prague C and M values.

Statement II-5: GERD patients with higher BMI will Section III: unresolved, controversial, or long-term have poorer symptoms relief by PPI treatment. care issues (agreement: 97%, level of evidence: 2a, recommendation: A) Extraesophageal syndromes, including chronic cough, asthma, and chronic laryngitis, are present in a substantial There was a strong positive association between body proportion of patients with gastroesophageal reflux disease mass index (BMI) and symptoms of GERD in a large cohort of [68]. However, management of these extraesophageal women [58]. For Los Angeles Grade A or B reflux esopha- symptoms remains debatable in clinical practice [68e73]. gitis, a higher BMI decreases the rate of sustained symp- Whether PPI therapy is beneficial in the management of tomatic response after 8 weeks of esomeprazole therapy, these symptoms has attracted much attention. Whether and increases the need for medication and the failure rate long-term PPI therapy and surveillance programs may of on-demand therapy [13]. In addition, a BMI of > 25 kg/m2 reduce the risk and mortality of esophageal adenocarci- is an independent risk factor to determine the healing of noma, respectively, in patients with Barrett’s esophagus Los Angeles Grade C or D reflux esophagitis by esomepra- are also important clinical issues [74e77]. Whether long- zole [14]. Reducing BMI to > 1.5 kg/m2, especially for those term PPI users have a higher risk of potential adverse ef- with an initial BMI of > 25 kg/m2, could be promising to fects, such as fracture and enteric infection, is also a improve the healing of RE-CD by esomeprazole [14]. controversial issue [78e84]. Another important issue is

Nationwide population-based cohort studies showed that the annual incidence of esophageal adenocarcinoma Statement III-2: In adult asthma patients with typical ranged from 0.15% to 0.19% in males and from 0.05% to reflux symptoms, proton pump inhibitor therapy may 0.08% in females in unselected patients with Barrett’s improve pulmonary function. esophagus [76]. A retrospective cohort study showed that (agreement: 82%, level of evidence: 1a, recom- the proportion of high-grade dysplasia and esophageal mendation: B) adenocarcinoma detected was higher among patients with low-grade dysplasia who underwent endoscopic surveillance every 3 months or less [77]. Considering the low incidence of esophageal cancer and high cost of endoscopy, A meta-analysis including 2524 patients showed that PPI endoscopic surveillance is recommended only for high-risk therapy may improve the morning peak expiratory flow, as patients. Risk factors for progression to esophageal can- compared to placebo, in adult asthma patients with reflux cer include male gender, longer duration of Barrett’s symptoms [69]. However, PPI therapy was not effective in esophagus, higher BMI, higher waist-to-hip ratio, longer asthma patients who did not report reflux symptoms and in length of the Barrett’s esophagus segment, presence of pediatric asthma patients [69e72]. Besides, the clinical hiatal hernia, visible nodularity detected by endoscopy, significance of the magnitude of this improvement remains higher degree of dysplasia, etc. [76,77]. Future prospective debatable [69,70]. Further trials are warranted to identify cohort studies and intervention trials are warranted to the subgroup of patients that might benefit more from PPI suggest the optimal intervals of endoscopic surveillance for therapy. Barrett’s esophagus.

Statement III-3: There is insufﬁcient evidence to support proton pump inhibitor therapy can improve Statement III-6: Although observational studies laryngeal symptoms in suspected GERD-related showed the association of long-term use of proton chronic laryngitis. pump inhibitor with hip and vertebral fracture, the (agreement: 97%, level of evidence: 1a, recom- causal relationship remains uncertain. mendation: B) (agreement: 100%, level of evidence: 2a, recommendation: B)

A meta-analysis failed to show a signiﬁcant symptom reduction by PPI therapy in patients with chronic laryngitis A meta-analysis of four cohort studies and six case- [73]. However, the study scale is limited in usual and deﬁ- econtrol studies, including 223,210 fracture cases, showed nition of chronic laryngitis with diversity among different increased risks of hip and vertebral, but not of wrist/fore- studies in meta-analysis [73,74]. arm, fractures in PPI users [77]. However, PPI use was not

Please cite this article in press as: Sheu B-S, et al., Consensus of gastroesophageal reflux disease in Taiwan with endoscopy-based approach covered by National Health Insurance, Advances in Digestive Medicine (2015), http://dx.doi.org/10.1016/j.aidm.2015.05.002 + MODEL Taiwan expert consensus on GERD 7 associated with osteoporosis and low bone mineral density Acknowledgments [78,79,84]. Therefore, the causal relationship of the association detected by observational studies remains uncer- This work was supported by Taiwan expert groups, including tain because the possibility of confounding by indication contributing authors in the steering committee and the could not be excluded. Therefore, well-designed random- following experts: Dr Chia-Long Lee, Hsin-Chu Cathay ized trials are warranted to clarify the causal relationships General Hospital, Hsin-Chu; Drs Chen-Ming Hsu and Ming- of the association. Yao Su, Chang Gung Memorial Hospital, Linko; Drs Ping- Huei Tseng and Mei-Jyh Chen, National Taiwan University Hospital, Taipei; Drs Ching-Liang Lu, Chih-Yen Chen, and Statement III-7: Although observational studies Jiing-Chyuan Luo, Taipei Veterans General Hospital, Taipei; showed the association of long-term use of proton Dr Chien-Lin Chen, Buddhist Tzu Chi General Hospital, pump inhibitor with Clostridium difficile associated Hualien; Dr Yone-Han Mah, Lotung Saint Mary’s Hospital, diarrhea, the causal relationship remains uncertain. Lotung; Dr Ming-Jen Sheu, Chi Mei Medical Center, Tainan; (agreement: 94%, level of evidence: 2a, recom- Drs Chao-Hung Kuo and Jeng-Yih Wu, Kaohsiung Medical mendation: B) University Hospital, Kaohsiung; Drs Yao-Chun Hsu and Wen- Lun Wang, E-Da Hospital, Kaohsiung; Dr Chun-Chao Chang, A meta-analysis of six cohort studies and 17 case- Taipei Medical University Hospital, Taipei; Drs Chi-Sen econtrol studies, including close to 300,000 patients, Chang, Han-Chung Lien, and Teng-Yu Lee, Taichung Veter- showed increased risks of Clostridium difficile-associated ans General Hospital, Taichung, Dr Chun-Che Lin, Chung diarrhea among patients on long-term PPI therapy [79,80]. Shan Medical University Hospital, Taichung; Drs Ping-I Hsu Similarly, randomized trials are warranted to confirm or and Sung-Shuo Kao, Kaohsiung Veterans General Hospital, refute the causal relationships of the association. Kaohsiung; Drs Seng-Kee Chuah and Keng-Liang Wu, Kaoh- siung Chang Gung Memorial Hospital, Kaohsiung; and Dr Hsiu-Chi Cheng, National Cheng Kung University, Tainan, Taiwan. This work was funded by the Gastroenterological Statement III-8: Eradication of Helicobacter pylori Society of Taiwan. improves gastritis and prevents progression of gastric atrophy and intestinal metaplasia and is recommended in patients on long-term proton pump inhibitor therapy. References (agreement: 94%, level of evidence: 1a, recommendation: A) [1] Hunt RH. Importance of pH control in the management of GERD. Arch Intern Med 1999;159:649e57. [2] Malfertheiner P, Megraud F, O’Morain C, Bazzoli F, El-Omar E, Earlier observational studies showed an increased risk of Graham D, et al. Current concepts in the management of atrophic gastritis in patients with reflux esophagitis and Helicobacter pylori infectiondthe Maastricht III Consensus Helicobacter pylori infection that were treated with PPI report. Gut 2007;56:772e81. therapy [82]. The subsequent four randomized control trials [3] Yang HB, Sheu BS, Wang ST, Cheng HC, Chang WL, Chen WY. that included 546 patients further showed that gastric H. pylori eradication prevents the progression of gastric in- inflammation could be reduced by H. pylori eradication testinal metaplasia in reflux esophagitis patients using long- [3,82e84]. Longer follow-up periods are warranted to term esomeprazole. Am J Gastroenterol 2009;104:1642e9. assess whether the risk of atrophic gastritis or intestinal [4] Spechler SJ. Epidemiology and natural history of gastro- metaplasia in long-term PPI users can really be controlled esophageal reflux disease. Digestion 1992;51(Suppl. 4): 3se7s. by H. pylori eradication. [5] Richardson P, Hawkey CJ, Stack WA. Proton pump inhibitors. Pharmacology and rationale for use in gastrointestinal dis- e Dissemination strategies and legal issues orders. Drugs 1998;56:307 35. [6] Chiba N, De Gara CJ, Wilkinson JM, Hunt RH. Speed of healing and symptom relief in grade II to IV gastroesophageal reflux These statements are based on the best available evidence disease: a meta-analysis. Gastroenterology 1997;112: to pursue better quality of care and will be updated every 5 1798e810. years. They are not suitable for deciding the standard of [7] Holloway RH, Dent J, Narielvala F, Mackinnon AM. Relation care in specific cases. This consensus statement will be between oesophageal acid exposure and healing of oeso- disseminated through the following: (1) presentations given phagitis with omeprazole in patients with severe reflux e at the annual society meeting of Taiwan Digestive Week in esophagitis. Gut 1996;38:649 54. 2014; (2) release of the copies of these statements in [8] Tytgat GN. Long-term therapy for refluxive esophagitis. N Engl J Med 1995;333:1148e50. electronic and paper format to national societies/associa- [9] Dent J, Brun J, Fendrick A. An evidence-based appraisal of tions of gastroenterologists for their iterations; and (3) reflux disease managementdthe Genval workshop report. release on the website of our society link. Gut 1999;44:S1e16. [10] Talley NJ, Lauritsen K, Tunturi-Hihnala H, Lind T, Moum B, Bang C, et al. Esomeprazole 20 mg maintains symptom con- Conflicts of interest trol in endoscopy-negative gastro-oesophageal reflux disease: a controlled trial of ‘on-demand’ therapy for 6 months. All authors declare no conflicts of interest. Aliment Pharmacol Ther 2001;15:347e54.

[11] Chen WY, Chang WL, Tsai YC, Cheng HC, Lu CC, Sheu BS. [27] Liou JM, Lin JT, Wang HP, Huang SP, Lee YC, Shun CT, et al. Double-dosed pantoprazole accelerates the sustained symp- The optimal age threshold for screening upper endoscopy for tomatic response in overweight and obese patients with uninvestigated dyspepsia in Taiwan, an area with a higher reflux esophagitis in Los Angeles grades A and B. Am J Gas- prevalence of gastric cancer in young adults. Gastrointest troenterol 2010;105:1046e52. Endosc 2005;61:819e25. [12] Lundell LR, Dent J, Bennett JR, Blum AL, Armstrong D, [28] Fass R, Ofman JJ, Gralnek IM, Johnson C, Camargo E, Galmiche JP, et al. Endoscopic assessment of oesophagitis: Sampliner RE, et al. Clinical and economic assessment of the clinical and functional correlates and further validation of omeprazole test in patients with symptoms suggestive of the Los Angeles classification. Gut 1999;45:172e80. gastroesophageal reflux disease. Arch Intern Med 1999;159: [13] Sheu BS, Cheng HC, Chang WL, Chen WY, Kao AW. The impact 2161e8. of body mass index on the application of on-demand therapy [29] Bate CM, Riley SA, Chapman RW, Durnin AT, Taylor MD. for Los Angeles grades A and B reflux esophagitis. Am J Evaluation of omeprazole as a cost-effective diagnostic test Gastroenterol 2007;102:2387e94. for gastro-oesophageal reflux disease. Aliment Pharmacol [14] Sheu BS, Chang WL, Cheng HC, Kao AW, Lu CC. Body mass Ther 1999;13:59e66. index can determine the healing of reflux esophagitis with [30] Numans ME, Lau J, de Wit NJ, Bonis PA. Short-term treat- Los Angeles Grades C and D by esomeprazole. Am J Gastro- ment with proton-pump inhibitors as a test for gastroenterol 2008;103:2209e14. esophageal reflux disease: a meta-analysis of diagnostic test [15] Sheu BS, Wu CY, Wu MS, Chiu CT, Lin CC, Hsu PI, et al. characteristics. Ann Intern Med 2004;140:518e27. Consensus on control of risky nonvariceal upper gastroin- [31] Cremonini F, Wise J, Moayyedi P, Talley NJ. Diagnostic and testinal bleeding in Taiwan with National Health Insurance. therapeutic use of proton pump inhibitors in non-cardiac Biomed Res Int 2014;2014:563707. chest pain: a metaanalysis. Am J Gastroenterol 2005;100: [16] Chen MJ, Lee YC, Chiu HM, Wu MS, Wang HP, Lin JT. Time 1226e32. trends of endoscopic and pathological diagnoses related to [32] Pace F, Pace M. The proton pump inhibitor test and the gastroesophageal reflux disease in a Chinese population: diagnosis of gastroesophageal reflux disease. Expert Rev eight years single institution experience. Dis Esophagus 2010; Gastroenterol Hepatol 2010;4:423e7. 23:201e7. [33] Sharma P, Wani S, Bansal A, Hall S, Puli S, Mathur S, et al. A [17] Lee YC, Yen AM, Tai JJ, Chang SH, Lin JT, Chiu HM, et al. The feasibility trial of narrow band imaging endoscopy in patients effect of metabolic risk factors on the natural course of with gastroesophageal reflux disease. Gastroenterology gastro-oesophageal reflux disease. Gut 2009;58:174e81. 2007;133:454e64. quiz 674. [18] Hsu CS, Wang PC, Chen JH, Su WC, Tseng TC, Chen HD, et al. [34] Lee YC, Lin JT, Chiu HM, Liao WC, Chen CC, Tu CH, et al. Increasing insulin resistance is associated with increased Intraobserver and interobserver consistency for grading severity and prevalence of gastro-oesophageal reflux dis- esophagitis with narrow-band imaging. Gastrointest Endosc ease. Aliment Pharmacol Ther 2011;34:994e1004. 2007;66:230e6. [19] Cremonini F, Di Caro S, Delgado-Aros S, Sepulveda A, [35] Miyasaka M, Hirakawa M, Nakamura K, Tanaka F, Mimori K, Gasbarrini G, Gasbarrini A, et al. Meta-analysis: the rela- Mori M, et al. The endoscopic diagnosis of nonerosive reflux tionship between Helicobacter pylori infection and gastro- disease using flexible spectral imaging color enhancement oesophageal reflux disease. Aliment Pharmacol Ther 2003; image: a feasibility trial. Dis Esophagus 2011;24:395e400. 18:279e89. [36] Tseng PH, Chen CC, Chiu HM, Liao WC, Wu MS, Lin JT, et al. [20] Lee YC, Chen TH, Chiu HM, Shun CT, Chiang H, Liu TY, et al. Performance of narrow band imaging and magnification The benefit of mass eradication of Helicobacter pylori endoscopy in the prediction of therapeutic response in pa- infection: a community-based study of gastric cancer pre- tients with gastroesophageal reflux disease. J Clin Gastro- vention. Gut 2013;62:676e82. enterol 2011;45:501e6. [21] Jonasson C, Wernersson B, Hoff DA, Hatlebakk JG. Validation [37] Lynch CR, Wani S, Rastogi A, Keighley J, Mathur S, Higbee A, of the GerdQ questionnaire for the diagnosis of gastro- et al. Effect of acid-suppressive therapy on narrow band oesophageal reflux disease. Aliment Pharmacol Ther 2013; imaging findings in gastroesophageal reflux disease: a pilot 37:564e72. study. Dis Esophagus 2013;26:124e9. [22] Bai Y, Du Y, Zou D, Jin Z, Zhan X, Li ZS, et al. Gastroesoph- [38] Pandolfino JE, Kwiatek MA, Nealis T, Bulsiewicz W, Post J, ageal Reflux Disease Questionnaire (GerdQ) in real-world Kahrilas PJ, et al. Achalasia: a new clinically relevant clas- practice: a national multicenter survey on 8065 patients. J sification by high-resolution manometry. Gastroenterology Gastroenterol Hepatol 2013;28:626e31. 2008;135:1526e33. [23] Lee YC, Lin JT, Wang HP, Chiu HM, Wu MS. Influence of cy- [39] Spechler SJ, Castell DO. Classification of oesophageal tochrome P450 2C19 genetic polymorphism and dosage of motility abnormalities. Gut 2001;49:145e51. rabeprazole on accuracy of proton-pump inhibitor testing in [40] Chen CL, Hsu PI. Current advances in the diagnosis and Chinese patients with gastroesophageal reflux disease. J treatment of nonerosive reflux disease. Gastroenterol Res Gastroenterol Hepatol 2007;22:1286e92. Pract 2013;2013:653989. [24] Tseng PH, Lee YC, Chiu HM, Wang HP, Lin JT, Wu MS. A [41] Kahrilas PJ, Sifrim D. High-resolution manometry and comparative study of proton-pump inhibitor tests for Chinese impedance-pH/manometry: valuable tools in clinical and reflux patients in relation to the CYP2C19 genotypes. J Clin investigational esophagology. Gastroenterology 2008;135: Gastroenterol 2009;43:920e5. 756e69. [25] Lee SW, Chang CM, Chang CS, Kao AW, Chou MC. Comparison [42] Sifrim D, Dupont L, Blondeau K, Zhang X, Tack J, Janssens J. of presentation and impact on quality of life of gastro- Weakly acidic reflux in patients with chronic unexplained esophageal reflux disease between young and old adults in a cough during 24 hour pressure, pH, and impedance moni- Chinese population. World J Gastroenterol 2011;17:4614e8. toring. Gut 2005;54:449e54. [26] Wang CP, Tseng PH, Chen TC, Lou PJ, Yang TL, Hu YL. [43] Connor J, Richter J. Increasing yield also increases false Transnasal esophagogastroduodenoscopy for evaluation of positives and best serves to exclude GERD. Am J Gastro- upper gastrointestinal non-neoplastic disorders in patients enterol 2006;101:460e3. with fresh hypopharyngeal cancer. Laryngoscope 2013;123: [44] Akbunar AT, Alper E, Nak SG, Konuk N, Erkal B, Tamgac F. A 975e9. simple method to increase the diagnostic efficiency of

gastroesophageal scintigraphy using the knee-chest position. Barrett’s esophagus in an ethnic Chinese population. Am J Clin Nucl Med 1999;24:842e4. Gastroenterol 2009;104:13e20. [45] Asakura Y, Imai Y, Ota S, Fujiwara K, Miyamae T. Usefulness [62] Kuo CJ, Lin CH, Liu NJ, Wu RC, Tang JH, Cheng CL. Frequency of gastroesophageal reflux scintigraphy using the kneeechest and risk factors for Barrett’s esophagus in Taiwanese pa- position for the diagnosis of gastroesophageal reflux disease. tients: a prospective study in a tertiary referral center. Dig Ann Nucl Med 2005;19:291e6. Dis Sci 2010;55:1337e43. [46] Hsu CP, Chen CY, Hsieh YH, Hsia JY, Shai SE, Kao CH. [63] Csendes A, Smok G, Burdiles P, Korn O, Gradiz M, Rojas J, Esophageal reflux after total or proximal gastrectomy in et al. Prevalence of intestinal metaplasia according to the patients with adenocarcinoma of the gastric cardia. Am J length of the specialized columnar epithelium lining the Gastroenterol 1997;92:1347e50. distal esophagus in patients with gastroesophageal reflux. Dis [47] Lee TH, Shiun YC. Changes in gastroesophageal reflux in Esophagus 2003;16:24e8. patients with nasogastric tube followed by percutaneous [64] Rex DK, Cummings OW, Shaw M, Cumings MD, Wong RK, endoscopic gastrostomy. J Formos Med Assoc 2011;110: Vasudeva RS, et al. Screening for Barrett’s esophagus in co- 115e9. lonoscopy patients with and without heartburn. Gastroen- [48] Carlsson R, Galmiche JP, Dent J, Lundell L, Frison L. Prog- terology 2003;125:1670e7. nostic factors influencing relapse of oesophagitis during [65] Veldhuyzen van Zanten SJ, Thomson AB, Barkun AN, maintenance therapy with anti-secretory drugs: a meta- Armstrong D, Chiba N, White RJ, et al. The prevalence of analysis of long-term omeprazole trials. Aliment Pharmacol Barrett’s oesophagus in a cohort of 1040 Canadian primary Ther 1997;11:473e82. care patients with uninvestigated dyspepsia undergoing [49] Fass R, Sontag SJ, Traxler B, Sostek M. Treatment of patients prompt endoscopy. Aliment Pharmacol Ther 2006;23:595e9. with persistent heartburn symptoms: a double-blind, ran- [66] Lee YC, Cook MB, Bhatia S, Chow WH, El-Omar EM, Goto H, domized trial. Clin Gastroenterol Hepatol 2006;4:50e6. et al. Asian Barrett’s Consortium. Interobserver reliability in [50] Kaspari S, Kupcinskas L, Heinze H, Bergho¨fer P. Pantoprazole the endoscopic diagnosis and grading of Barrett’s esophagus: 20 mg on demand is effective in the long-term management an Asian multinational study. Endoscopy 2010;42:699e704. of patients with mild gastro-oesophageal reflux disease. Eur J [67] Vakil N, van Zanten SV, Kahrilas P, Dent J, Jones R, Global Gastroenterol Hepatol 2005;17:935e41. Consensus Group. The Montreal definition and classification [51] Scholten T, Dekkers CP, Schu¨tze K, Ko¨rner T, Bohuschke M. of gastroesophageal reflux disease: a global evidence-based Gatz G On-demand therapy with pantoprazole 20 mg as consensus. Am J Gastroenterol 2006;101:1900e20. effective long-term management of reflux disease in patients [68] Chang AB, Lasserson TJ, Gaffney J, Connor FL, Garske LA. with mild GERD: the ORION trial. Digestion 2005;72:76e85. Gastro-oesophageal reflux treatment for prolonged non- [52] Bour B, Staub JL, Chousterman M, Labayle D, Nalet B, Nouel O, specific cough in children and adults. Cochrane Database et al. Long-term treatment of gastro-oesophageal reflux dis- Syst Rev 2011;1. CD004823. ease patients with frequent symptomatic relapses using [69] Chan WW, Chiou E, Obstein KL, Tignor AS, Whitlock TL. The rabeprazole: on-demand treatment compared with contin- efficacy of proton pump inhibitors for the treatment of asthma uous treatment. Aliment Pharmacol Ther 2005;21:805e12. in adults: a meta-analysis. Arch Intern Med 2011;171:620e9. [53] Sjo¨stedt S, Befrits R, Sylvan A, Harthon C, Jo¨rgensen L, [70] Mastronarde JG, Anthonisen NR, Castro M, Holbrook JT, Carling L, et al. Daily treatment with esomeprazole is supe- Leone FT, Teague WG, et al. Efficacy of esomeprazole for rior to that taken on-demand for maintenance of healed treatment of poorly controlled asthma. N Engl J Med 2009; erosive oesophagitis. Aliment Pharmacol Ther 2005;22: 360:1487e99. 183e91. [71] Holbrook JT, Wise RA, Gold BD, Blake K, Brown ED, Castro M, [54] Orlando RC, Monyak JT, Silberg DG. Predictors of heartburn et al. Lansoprazole for children with poorly controlled resolution and erosive esophagitis in patients with GERD. asthma: a randomized controlled trial. JAMA 2012;307: Curr Med Res Opin 2009;25:2091e102. 373e81. [55] Dean BB, Gano Jr AD, Knight K, Ofman JJ, Fass R. Effec- [72] Qadeer MA, Phillips CO, Lopez AR, Steward DL, Noordzij JP, tiveness of proton pump inhibitors in nonerosive reflux dis- Wo JM, et al. Proton pump inhibitor therapy for suspected ease. Clin Gastroenterol Hepatol 2004;2:656e64. GERD-related chronic laryngitis: a meta-analysis of random- [56] Futagami S, Iwakiri K, Shindo T, Kawagoe T, Horie A, ized controlled trials. Am J Gastroenterol 2006;101:2646e54. Shimpuku M, et al. The prokinetic effect of mosapride citrate [73] Singh S, Garg SK, Singh PP, Iyer PG, El-Serag HB. Acid-sup- combined with omeprazole therapy improves clinical symp- pressive medications and risk of oesophageal adenocarci- toms and gastric emptying in PPI-resistant NERD patients noma in patients with Barrett’s oesophagus: a systematic with delayed gastric emptying. J Gastroenterol 2010;45: review and meta-analysis. Gut 2014;63:1229e37. 413e21. [74] de Jonge PJ, van Blankenstein M, Grady WM, Kuipers EJ. [57] Manabe N, Haruma K, Ito M, Takahashi N, Takasugi H, Barrett’s oesophagus: epidemiology, cancer risk and impli- Wada Y, et al. Efficacy of adding sodium alginate to omep- cations for management. Gut 2014;63:191e202. razole in patients with nonerosive reflux disease: a ran- [75] Ramus JR, Gatenby PA, Caygill CP, Winslet MC, Watson A. domized clinical trial. Dis Esophagus 2012;25:373e80. Surveillance of Barrett’s columnar-lined oesophagus in the [58] Jacobson BC, Somers SC, Fuchs CS, Kelly CP, Camargo Jr CA. UK: endoscopic intervals and frequency of detection of Body-mass index and symptoms of gastroesophageal reflux in dysplasia. Eur J Gastroenterol Hepatol 2009;21:636e41. women. N Engl J Med 2006;354:2340e8. [76] Ngamruengphong S, Leontiadis GI, Radhi S, Dentino A, [59] Yeh C, Hsu CT, Ho AS, Sampliner RE, Fass R. Erosive esoph- Nugent K. Proton pump inhibitors and risk of fracture: a agitis and Barrett’s esophagus in Taiwan: a higher frequency systematic review and meta-analysis of observational than expected. Dig Dis Sci 1997;42:702e6. studies. Am J Gastroenterol 2011;106:1209e18. [60] Tseng PH, Lee YC, Chiu HM, Huang SP, Liao WC, Chen CC, [77] Targownik LE, Leslie WD, Davison KS, Goltzman D, Jamal SA, et al. Prevalence and clinical characteristics of Barrett’s Kreiger N, et al. The relationship between proton pump in- esophagus in a Chinese general population. J Clin Gastro- hibitor use and longitudinal change in bone mineral density: enterol 2008;42:1074e9. a population-based study [corrected] from the Canadian [61] Chang CY, Lee YC, Lee CT, Tu CH, Hwang JC, Chiang H, et al. Multicentre Osteoporosis Study (CaMos). Am J Gastroenterol The application of Prague C and M criteria in the diagnosis of 2012;107:1361e9.

[78] Janarthanan S, Ditah I, Adler DG, Ehrinpreis MN. Clostridium randomized follow-up study. Am J Gastroenterol 2001;96: difficile-associated diarrhoea and proton pump inhibitor 2882e6. therapy: a meta-analysis. Am J Gastroenterol 2012;107: [82] Kuipers EJ, Nelis GF, Klinkenberg-Knol EC, Snel P, Goldfain D, 1001e10. Kolkman JJ, et al. Cure of Helicobacter pylori infection in [79] Kwok CS, Arthur AK, Anibueze CI, Singh S, Cavallazzi R, patients with reflux oesophagitis treated with long term Loke YK. Risk of Clostridium difficile infection with acid omeprazole reverses gastritis without exacerbation of reflux suppressing drugs and antibiotics: meta-analysis. Am J Gas- disease: results of a randomised controlled trial. Gut 2004; troenterol 2012;107:1011e9. 53:12e20. [80] Kuipers EJ, Lundell L, Klinkenberg-Knol EC, Havu N, [83] Schenk BE, Kuipers EJ, Nelis GF, Bloemena E, Thijs JC, Snel P, Festen HP, Liedman B, et al. Atrophic gastritis and Heli- et al. Effect of Helicobacter pylori eradication on chronic cobacter pylori infection in patients with reflux esophagitis gastritis during omeprazole therapy. Gut 2000;46:615e21. treated with omeprazole or fundoplication. N Engl J Med [84] Targownik LE, Lix LM, Leung S, Leslie WD. Proton-pump in- 1996;334:1018e22. hibitor use is not associated with osteoporosis or accelerated [81] van Grieken NC, Meijer GA, Weiss MM, Bloemena E, bone mineral density loss. Gastroenterology 2010;138: Lindeman J, Baak JP, et al. Quantitative assessment of 896e904. gastric corpus atrophy in subjects using omeprazole: a