VOLUME 88 • NUMBER 8 • AUGUST 2021 • www.ccjm.org

Autoimmune brain disease: Antiobesity drug : Think before testing An individualized approach Vertebral body enhancement The obesity paradox, heart failure, and cardiorespiratory fi tness Subungual mucous cysts Antibody-mediated Eruptive seborrheic keratosis autoimmune encephalitis Alcohol use disorder in the elderly PICC insertion in patients with chronic kidney disease: Key considerations

EDITORIAL STAFF ADVERTISING Brian F. Mandell, MD, PhD, Editor in Chief Sima Sherman, Director of Sales and Marketing Pelin Batur, MD, Deputy Editor SHERMAN MEDICAL MARKETING GROUP Craig Nielsen, MD, Deputy Editor 1628 John F. Kennedy Blvd., #2200, Philadelphia, PA 19103 Mary T. Cusick, MS, Executive Editor (610) 529-0322 • [email protected] David A. Huddleston, Managing Editor Allison Siegel, MSSA, Senior Editor SUBSCRIPTIONS Amy Slugg Moore, Manuscript Editor U.S. and possessions: Personal $155; institutional $183; single Ross Papalardo, CMI, Medical Art Director copy/back issue $20 Philip Lammers, Editorial Project Leader Foreign: $200; single copy/back issue $20

PUBLISHING OPERATIONS Institutional (multiple-reader rate) applies to libraries, schools, Peter G. Studer, Executive Publisher hospitals, and federal, commercial, and private institutions and Bruce M. Marich, Production Manager organizations. Individual subscriptions must be in the names of, Iris Trivilino, Production Manager, Special Projects billed to, and paid by individuals. Laurie Weiss, Accountant (Billing) Please make check payable to Cleveland Clinic Journal of and mail to: Cleveland Clinic Education Foundation, P.O. Box 373291, ASSOCIATE EDITORS Cleveland, OH 44193-3291. To purchase a subscription with a Alejandro C. Arroliga, MD credit card, please visit www.ccjm.org. Moises Auron, MD Diana Basali, MD REPRINTS Eden Bernstein, MD (610) 529-0322 • [email protected] Daniel J. Brotman, MD Abhijit Duggal, MD Alejandro DuranCrane, MD PHOTOCOPYING Ruth M. Farrell, MD, MA Authorization to photocopy items for internal or personal use Kathleen Franco, MD is granted by Cleveland Clinic Journal of Medicine (ISSN 0891-1150 Steven M. Gordon, MD [print], ISSN 1939-2869 [online]), published by Cleveland Clinic, Brian Griffi n, MD provided that the appropriate fee is paid directly to Copyright Justin Hanks, DO Clearance Center, 222 Rosewood Drive, Danvers, MA 01923 USA Kristin Highland, MD (978) 750-8400. Prior to photocopying items for educational David L. Keller, MD classroom use, please contact Copyright Clearance Center, Inc., Umesh Khot, MD at the address above. For permission to reprint material, please Mandy C. Leonard, PharmD fax your request with complete information to the Republication Angelo A. Licata, MD, PhD department at CCC, fax (978) 750-4470. For further information Bryce Montane, MD visit CCC online at www.copyright.com. To order bulk reprints, Atul C. Mehta, MD see above. Christian Nasr, MD Robert M. Palmer, MD CHANGE OF ADDRESS Kyle Richardville, MD To report a change of address, send a recent mailing label along David D.K. Rolston, MD with new information to: Gregory Rutecki, MD Bernard J. Silver, MD AMA, Data Verifi cation Unit, 330 N. Wabash Ave., Suite 39300, Tyler Stevens, MD Chicago, IL 60611-5885 • Phone (800) 621-8335 • Fax (312) Theodore Suh, MD, PhD, MHSc 464-4880 • [email protected] Marc Williams, MD Cleveland Clinic Journal of Medicine uses the AMA database of names and addresses. The database includes all US CCJM-UK EDITION and not just AMA members. Only the AMA can update Olaf Wendler, MD, PhD, FRCS, Chief Editor changes of address and other data. Heather Muirhead, MHA, Clinical Institute Education and Training Manager SUBSCRIPTIONS, EDITORIAL, BILLING, AND PRODUCTION 1950 Richmond Rd., TR404, Lyndhurst, OH 44124 • Phone (216) EDITORS EMERITI 444-2661 • Fax (216) 444-9385 • [email protected] • www.ccjm.org John D. Clough, MD Herbert P. Wiedemann, MD James S. Taylor, MD DISCLAIMER Statements and opinions expressed in the Cleveland Clinic Journal of Medicine are those of the authors and not necessarily of Cleveland CLEVELAND CLINIC Tom Mihaljevic, MD Clinic or its Board of Trustees. President and Chief Executive Offi cer Cleveland Clinic Journal of Medicine [ISSN 0891-1150 (print), ISSN 1939-2869 (online)] is published monthly by Cleveland Clinic at CLEVELAND CLINIC EDUCATION INSTITUTE 1950 Richmond Rd., TR404, Lyndhurst, OH 44124. James K. Stoller, MD, MS, Chairman Steven Kawczak, PhD, Senior Director, Professional COPYRIGHT© 2021 THE CLEVELAND CLINIC FOUNDATION. Development and Knowledge Resources ALL RIGHTS RESERVED. PRINTED IN U.S.A.

CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 417 AUGUST 2021

TABLE OF CONTENTS

FROM THE EDITOR ...... Autoimmune brain disease: Think before testing 420 Immune serologic tests are not specifi c for any defi ned clinical diagnosis and thus should not be used to drive therapeutic decisions. Brian F. Mandell, MD, PhD

THE CLINICAL PICTURE ...... Pseudopathologic vertebral body enhancement 424 Contrast-enhanced computed tomography revealed sclerotic enhancement of the T1 vertebral body and the vertebral bodies from C2 to T2. Yi Pan, MD; Peixin Tan, MD; Songxi Xie, MD; Jiewen Chen, MMed; Yi-Long Wu, MD

THE CLINICAL PICTURE ...... Shedding light on subungual digital mucous cysts 426 Transillumination is a quick and easy way to diagnose digital mucous cysts and rule out other subungual masses. Taylor Bullock, BS; Daniel Michalik, MD; Rashmi Unwala, MD; John Anthony, MD

THE CLINICAL PICTURE ...... Eruptive seborrheic keratosis: A perilous clue 428 An abrupt increase in seborrheic keratoses in patients with an underlying malignancy is called the Leser-Trélat sign. Akanksha Kaushik, MD; Soufila KT, MD; Manju Daroach, MD; Divya Aggarwal, MD; Pulkit Rastogi, MD, DM; Muthu Sendhil Kumaran, MD, DNB

1-MINUTE CONSULT ...... Can I place a peripherally inserted central catheter 431 in my patient with chronic kidney disease? Future need for permanent dialysis access is an important consideration. Nora Hilda Hernandez Garcilazo, MD; Mohamed Hassanein, MD; Tushar J. Vachharajani, MD; Evamaria Anvari, MD

CONTINUED ON PAGE 419

Upcoming Features

■ Teriparatide: Labeling changes, long-term dosing ■ Stress testing and noninvasive coronary imaging ■ Ketogenic diets for type 1 diabetes

418 CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 AUGUST 2021

CONTINUED FROM PAGE 418

1-MINUTE CONSULT ...... How do you effectively evaluate the elderly 434 for alcohol use disorder? A brief review of clues to an underrecognized and underdiagnosed problem with physical as well as emotional and social consequences. Moronkeji Fagbemi, MD, FASAM

CURRENT DRUG THERAPY ...... Antiobesity drug therapy: An individualized 440 and comprehensive approach As part of a comprehensive plan, these drugs can provide lasting obesity control and additional health benefi ts, including decreased cardiovascular risk. Yael Mauer, MD, MPH; Marcie Parker, PharmD, BCACP; Sangeeta R. Kashyap, MD

REVIEW ...... The obesity paradox in heart failure: 449 What is the role of cardiorespiratory fi tness? What should these patients be advised about weight management and about cardiorespiratory fi tness, a major factor infl uencing the paradox? Asad Khan, MD; Erik H. Van Iterson, PhD; Luke J. Laffin, MD

REVIEW ...... CME MOC. Antibody-mediated autoimmune encephalitis: 459 A practical approach Symptoms may include subacute, progressive neuropsychiatric symptoms with associated cognitive dysfunction, movement disorders, and autoimmune seizures. Justin R. Abbatemarco, MD; Chen Yan, MD; Amy Kunchok, MBBS; Alexander Rae-Grant, MD, FAAN, FRCPC

DEPARTMENTS ...... CME Calendar 422 CME/MOC Instructions 472

Visit WWW.CCJM.ORG Test your knowledge of clinical topics

CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 419 FROM THE EDITOR

Autoimmune brain disease: Think before testing

As a rheumatologist, I have the opportunity to see many patients in con- sultation for the evaluation of complex and often diffi cult-to-defi ne symp- toms. The buzzword referral diagnosis is frequently “autoimmune.” In 2021, autoim- mune is seemingly the evil humor or miasma of centuries past revisited, with the added distinction that there are laboratory tests that can be ordered that, if positive, seem to provide superfi cial evidence for the validity of this diagnosis. The problem is that many immune serologic tests are not specifi c for any defi ned clinical diagnosis and thus should not be used to drive therapeutic decisions. However, such testing may create signifi cant angst and expectations in patients and their families. Indiscriminate testing often begets additional testing and specialty referrals. Obtain- ing an antinuclear antibody test in a patient with fatigue, brain fog, and diffuse pain as the primary symptoms—but with no diagnostically relevant basic laboratory testing or physical examination abnormalities—is invariably unhelpful to the patient1 and, if posi- tive, is often emotionally and fi nancially costly. Many of the fi nal diagnoses, even if these autoimmune tests are positive, are not autoimmune in nature. This is because many of these serologies have little specifi c- ity (poor positive predictive value). Within this bucket of autoimmune tests are rheu- matoid factor and the antinuclear antibody test and related antibody tests (anti-SSA, anti-RNP, and even to some extent anti-DNA). In a second bucket are tests that detect antibodies pathogenically linked to specifi c clinical syndromes, such as anti-glomerular basement membrane antibody, linked to the glomerular basement membrane subset of Goodpasture syndrome; anti-AQP4, linked to neuromyelitis optica; and anti-acetylcho- line receptor, linked to myasthenia gravis. These tests are of course also susceptible to misinterpretation (false positives) if ordered indiscriminately without the appropriate clinical context. But these antibodies do have a strong pathogenic link to specifi c clini- cal disorders. Since truly pathogenic antibodies have been identifi ed that cause or drive heretofore unexplained complex clinical syndromes, it is no surprise that research has moved to- ward trying to identify additional ones. There is hope in the scientifi c and patient com- munities that such identifi cation can result in new approaches to molecular therapy. The 2012 autobiography by Susannah Cahalan, Brain on Fire: My Month of Mad- ness, and the movie based on it relate the intense story of a young woman ultimately diagnosed with anti-N-methyl-d-aspartate (NMDA) receptor-associated autoimmune encephalitis—a disorder with associated antibody that has a strong pathogenic link to clinical disease. Without knowledge of the antibody to the NMDA receptor, this dis- order might still remain a mystery to diagnose, and a huger challenge to treat. In this issue of the Journal, Abbatemarco et al present a practical discussion of recognizing and managing patients with this and other forms of autoimmune encephalitis. As a nonneurologist, I fi nd these syndromes striking for several reasons. There is a biology that ultimately will be teased apart to explain how these specifi c antibodies sort to specifi c syndromes. It is fairly easy to conceptualize how antibodies binding to surface

doi:10.3949/ccjm.88b.08021

420 CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 MANDELL

receptors can disrupt the function of the cells bearing those receptors and their associ- ated intercellular networks. Perhaps careful clinical characterization of these syndromes will provide insight into further understanding of complex brain networking, specifi - cally, what is the connection between characteristic focal facial seizures and encepha- litis, or encephalomyelitis and sensory neuropathies? But we still have a way to go in understanding how antibodies recognizing intracellular targets are pathogenic, and why there are links between these syndromes and certain malignancies. From the literature, I am not quite sure of the predictive value of these antibodies. For me, a takeaway from the article by Abbatemarco et al is that clinical suspicion for these rare syndromes should come fi rst, followed by appropriate referral and subsequent thoughtful consideration for ordering these antibody panels from both blood and cere- brospinal fl uid.

Brian F. Mandell, MD, PhD Editor in Chief

1. Nashi RA, Shmerling RH. Antinuclear antibody testing for the diagnosis of systemic lupus erythematosus. Med Clin M Am 2021; 105:387–396. doi:10.1016/j.mcna.2020.10.003

CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 421 CME CALENDAR CME MOC

PRIMARY CARE WOMEN’S HEALTH: CARDIOVASCULAR UPDATE 2021 ESSENTIALS AND BEYOND FOR THE PRIMARY CARE PROVIDER September 18 October 28–29 AUGUST Live stream Live stream

HOSPITAL MEDICINE 2021 LESSONS FROM THE THORACIC TUMOR NOVEMBER August 5–6 BOARD: MULTIDISCIPLINARY Live stream DISCUSSIONS, DEBATES, STATE-OF-THE-ART DIAGNOSIS AND TREATMENTS AND TREATMENT OF DEMENTIA UPDATE: A COMPREHENSIVE Sep 18, 2021 November 4 REVIEW FOR THE CLINICIAN Live Stream Live stream August 7–8 Live stream GLOBAL EP UPDATE: September 24 CONTROVERSIES, INNOVATIONS, INTENSIVE REVIEW OF Live stream RESEARCH August 14–15 November 6 Live stream GENETICS EDUCATION SYMPOSIUM – Live stream/virtual GENETICS AND GENOMICS: TRANSTHYRETIN CARDIAC AMYLOIDOSIS APPLICATIONS FOR THE PREVENTION, WASOG/AASOG 2021: IN AFRICAN AMERICANS: DETECTION, AND TREATMENT OF CANCER MULTIDISCIPLINARY MEETING WHAT PHYSICIANS NEED TO KNOW September 30 FOR SARCOIDOSIS AND ILD August 24 Virtual/live stream November 29–December 2 Live stream Hollywood, FL OCTOBER INTENSIVE REVIEW OF DECEMBER AND METABOLISM VIRTUAL UPDATE August 27–29 October 1 Virtual MASTERING THE MANAGEMENT Live stream OF THE AORTIC VALVE December 3–4 SEPTEMBER PRACTICAL MANAGEMENT OF STROKE New York, NY October 1 STATE-OF-THE-ART DIAGNOSIS Live stream RESUSCITATING A DYING MARROW: AND TREATMENT OF DEMENTIA EMERGING CONCEPTS AND TREATMENT September 9, 23, and 30 ADVANCES IN CONGENITAL Live stream ADVANCES IN MYELOID MALIGNANCIES HEART DISEASE SUMMIT December 10 October 1–2 Atlanta, GA DIABETES, OBESITY, Live stream AND CARDIOVASCULAR DISEASE VIRTUAL SUMMIT IMPROVING END-OF-LIFE CARE 2022 September 9–11 IN THE ICU: CHALLENGES Live stream AND OPPORTUNITIES JANUARY October 6 to November 23 THE PRACTICE OF ECHOCARDIOGRAPHY Virtual webcast SHAPING THE MANAGEMENT AT CLEVELAND CLINIC 2021 OF PARKINSON DISEASE: DEBATING September 11 STATE-OF-THE-ART DIAGNOSIS THE MOST CONTROVERSIAL ISSUES Live stream AND TREATMENT OF DEMENTIA AND DISCUSSING THE LATEST October 7,14,21, and 28 BREAKTHROUGHS COMPREHENSIVE LIFELONG Live stream January 22–23 EXPEDITIOUS (CLE) CARE Lake Tahoe, NV OF AORTIC DISEASE WAKE UP TO SLEEP DISORDERS 2021: September 17–18 A CLEVELAND CLINIC SLEEP DISORDERS FEBRUARY Cleveland, OH CENTER UPDATE October 9–10 MULTIDISCIPLINARY APPROACH INTENSIVE REVIEW Live stream TO THE CONTEMPORARY MANAGEMENT OF GASTROENTEROLOGY OF HEART FAILURE AND PRIMARY CARE UPDATE February 25 September 17–20 October 15–16 Cleveland, OH Virtual stream and webcast Live stream

FOR SCHEDULE UPDATES AND TO REGISTER, VISIT: WWW.CCFCME.ORG/LIVE

422 CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 38 MARCHAUGUST 2021 2021

THE CLINICAL PICTURE Yi Pan, MD Peixin Tan, MD Songxi Xie, MD Department of Radiation , Guangdong Department of Radiation Oncology, Guangdong Department of Radiation Oncology, Guangdong Lung Cancer Institute, Guangdong Provincial Lung Cancer Institute, Guangdong Provincial Provincial People’s Hospital, Guangdong Academy People’s Hospital, Guangdong Academy of Medical People’s Hospital, Guangdong Academy of Medical of Medical Sciences, School of Medicine, South China Sciences, School of Medicine, South China University Sciences, School of Medicine, South China University University of Technology, Guangzhou, PR China of Technology, Guangzhou, PR China of Technology, Guangzhou, PR China

Jiewen Chen, MMed Yi-Long Wu, MD Department of Radiation Oncology, Guangdong Guangdong Lung Cancer Institute, Guangdong Provincial People’s Hospital, Guangdong Academy Provincial People’s Hospital, Guangdong Academy of Medical Sciences, School of Medicine, South China of Medical Sciences, School of Medicine, South China University of Technology, Guangzhou, PR China University of Technology, Guangzhou, PR China Pseudopathologic vertebral body enhancement

Figure 1. Contrast-enhanced thoracic computed Figure 2. Contrast-enhanced computed tomography tomography showed sclerotic enhancement of the T1 noted thrombosis in the left brachiocephalic vein (red vertebral body (red arrow). arrow) and contrast fi lling of the paravertebral veins, usually occult on contrast enhancement (green arrow).

45-year-old man presented to the emergency de- the left and right main bronchi. Bronchoscopy-guided A partment with severe dyspnea and cough that had radiofrequency ablation was performed to relieve air- gradually worsened over the past 3 months. His medical way obstruction. Moderately differentiated squamous history was notable for stage IIA esophageal squamous cell carcinoma was confi rmed by endobronchial biopsy cell carcinoma that had been treated with esophagec- study. tomy 15 years ago; and 13 years ago, he had developed The patient’s dyspnea improved after broncho- mediastinal lymph node metastasis and had received scopic therapy. Repeat CT 5 days later showed no chemoradiotherapy. He had been a heavy smoker and evidence of the vertebral body enhancement (Figure drinker but had quit 15 years ago. 3). Esophagoscopy showed an anastomotic stenosis 24 Emergency contrast-enhanced thoracic computed cm from the incisors and smooth mucosa in the esopha- tomography (CT) revealed a tumor mass in the lower gus. Anastomotic stenosis was consistent with the tumor trachea, sclerotic enhancement of the T1 vertebral location in the lower trachea and carina. Esophageal body (Figure 1) and from the C2 to T2 vertebral bod- stenosis and smooth esophageal mucosa demonstrated ies, and thrombosis in the left brachiocephalic vein external tumor compression of the esophagus, which (Figure 2). The patient rapidly developed type 2 respi- did not support the diagnosis of esophageal carcinoma. ratory failure (defi ned as a Pao2 < 8.0 kPa and a Paco2 Based on the patient’s smoking and irradiation his- > 6.0 kPa) and underwent bedside fi beroptic bron- tory and the long interval between the last anticancer choscopy, which revealed an obstructive tumor in the treatment and the appearance of the second tumor, lower trachea that had invaded the carina and both the obstructive tumor was diagnosed as a second pri- mary bronchogenic carcinoma. doi:10.3949/ccjm.88a.20150 The patient underwent disease staging with posi-

424 CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 PAN AND COLLEAGUES

tron emission tomography CT, which showed in- creased 18F-fl uorodeoxyglucose uptake at the back wall of the trachea alone, indicating that the tumor was localized.

■ TREATMENT The patient received 4 cycles of capecitabine and anlo- tinib, followed by 4 cycles of anlotinib. This achieved a partial response. At last follow-up, the patient was alive and without disease progression.

■ PSEUDOPATHOLOGIC VERTEBRAL BODY ENHANCEMENT The differential diagnosis of sclerotic lesions on Figure 3. Repeat computed tomography showed contrast-enhanced CT includes tumor metastasis, the disappearance of the sclerotic vertebral body mastocytosis, sarcoidosis, osteomyelitis, lymphoma, enhancement. Paget disease, and pseudopathologic vertebral body enhancement. For patients with a history of cancer, on CT, were seen. Eight consecutive vertebral bodies the leading cause is metastatic malignancy. were involved, which is uncommon with metastasis. A limited number of case reports showed that However, when the contrast agent was injected into pseudopathologic vertebral body enhancement may the contralateral arm during the second CT, the verte- exist in the presence of obstruction of the superior bra enhancement disappeared (Figure 3). vena cava or brachiocephalic vein. This is uncommon and easily misdiagnosed as sclerotic osseous metastasis Pseudopathologic vertebral body enhancement due in clinical practice.1,2 to brachiocephalic vein narrowing is rare. However, The mechanism of pseudopathologic sclerotic sclerotic bone metastasis based on contrast-enhanced enhancement of vertebral bodies is not well charac- CT should prompt a careful evaluation when narrow- terized. However, an elevated venous pressure that ing or obstruction of the brachiocephalic vein and induced contrast agent refl ux into the intravertebral paravertebral collateral veins is present on imaging. venous plexus was proposed as a main reason.3 Spine magnetic resonance imaging or positron emis- In our patient, narrowing of the left brachiocephal- sion tomography CT should be performed to confi rm ic vein proximal to the superior vena cava resulted in the diagnosis. ■ elevated venous pressure and blood fl ow into the verte- bral venous plexus, causing refl ux of contrast agent.1,4 ■ DISCLOSURES Sclerotic enhancement of the vertebral bodies and The authors report no relevant fi nancial relationships which, in the context of their contribu- vertebral venous collaterals, which are usually occult tions, could be perceived as a potential confl ict of interest.

■ REFERENCES obstruction on computed tomography. Spine J 2015; 15(6):1295– 1301. doi:10.1016/j.spinee.2013.07.440 1. Berritto D, Abboud S, Kosmas C, Riherd D, Robbin M. Vertebral 4. Chen JY, Mamourian AC, Messe SR, Wolf RL. Pseudopathologic body enhancement mimicking sclerotic osseous lesions in the setting brain parenchymal enhancement due to venous refl ux from left- of bilateral brachiocephalic vein thrombosis. Skeletal Radiol 2015; sided injection and brachiocephalic vein narrowing. AJNR Am J 44(2):303–305. doi:10.1007/s00256-014-2037-9 Neuroradiol 2010; 31(1):86–87. doi:10.3174/ajnr.A1688 2. Thomas N, Oliver TB, Sudarshan T. Vanishing bone metastases—a pitfall in the interpretation of contrast enhanced CT in patients with Address: Yi Pan, MD, Department of Radiation Oncology, Guangdong superior vena cava obstruction. Br J Radiol 2011; 84(1005):e176– Lung Cancer Institute, Guangdong Provincial People’s Hospital, Guang- e178. doi:10.1259/bjr/50676625 dong Academy of Medical Sciences, School of Medicine, South 3. Kim YK, Sung YM, Hwang KH, Cho EK, Choi HY. Pseudopathologic China University of Technology, 106 Zhongshan Er Rd, Guangzhou, vertebral body enhancement in the presence of superior vena cava 510080, PR China; [email protected]

CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 425 THE CLINICAL PICTURE Taylor Bullock, BS Daniel Michalik, MD Rashmi Unwala, MD John Anthony, MD Department of , Department of Dermatology, Department of Dermatology, Department of Dermatology, Cleveland Clinic; Cleveland Clinic Lerner Cleveland Clinic, Cleveland, OH Cleveland Clinic, Cleveland, OH Cleveland Clinic, and Clinical Assistant College of Medicine of Case Western Professor, Cleveland Clinic Lerner College Reserve University, Cleveland, OH of Medicine of C ase Western Reserve University, Cleveland, OH

Shedding light on subungual digital mucous cysts

Raised area Longitudinal groove Area of translucency

Transillumination is a quick and easy way to diagnose DMC Linear hemorrhages and rule out other subungual Figure 1. Subungual digital mucous cyst of the thumbnail: (A) longitudinal groove, distal linear hemorrhages, and a raised area at the proximal nail; and (B) a translucent area, masses 3 mm × 5 mm, beneath the nail matrix.

65-year-old man presented with a Clinically, this presentation appeared to A thumbnail deformity that had been pres- be consistent with a subungual digital mu- ent for 3 months. He said he had no pain and cous cyst (DMC) causing nail deformity due no history of trauma to the nail. to compression of the nail matrix. To confi rm Examination of the thumbnail revealed a the diagnosis, a sterile 22-gauge needle was longitudinal groove, distal linear hemorrhages, used to drill through the nail plate into the and a raised area at the proximal end that ap- translucent area. A copious amount of clear, peared to be a subungual nodule affecting the gelatinous material, characteristic of a DMC, underside of the nail plate, and transillumina- was expressed from the puncture site. tion showed a translucent area 3 mm by 5 mm ■ beneath the nail matrix (Figure 1). Examina- DIFFERENTIAL DIAGNOSIS tion of all other digits was unremarkable. Longitudinal grooves in the nail bed can be caused by median nail dystrophy, trauma, doi:10.3949/ccjm.88a.20125 compression of the nail matrix from tumors,

426 CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 BULLOCK AND COLLEAGUES

and physiologic furrows and ridges that are tion of fl uid in DMC lacks a cystic lining, so accentuated in diseases such as lichen planus DMC is not a true cyst. DMC can cause nail and Darier disease. Tumors that can affect deformities by compressing nail matrix cells, the nail matrix include various nail fi broma, most commonly proximal and superfi cial to DMC, pyogenic granuloma, glomus tumor, the proximal nail fold.1 subungual exostosis, squamous cell carcinoma, Subungual DMC is much more diffi cult to and .1 diagnose than its superfi cial counterpart, not The translucency of this patient’s subun- only because it is less accessible, but also be- gual nodule was highly suggestive of a DMC, cause it lacks the characteristic appearance of and the clear, gelatinous material expressed superfi cial DMC and does not cause intermit- from the cyst confi rmed it. In addition, a red tent mucinous discharge.3 Though DMC is lunula, as seen in our patient (Figure 1), is a commonly asymptomatic when left untreated, 2 common fi nding in subungual DMC. subungual DMC most often requires nail avul- ■ sion and to rule out other subungual DIGITAL MUCOUS CYSTS masses, including malignancies.4 DMC, also known as myxoid pseudocyst or Transillumination is a quick and easy way synovial cyst, commonly presents as a superfi - to diagnose DMC and rule out other subungual cial, dome-shaped, shiny, cystic nodule locat- masses, as seen in this case. The fi nding of a ed near the distal interphalangeal joint on the translucent cyst can spare the patient from un- dorsum of the fi ngers. The cyst is commonly dergoing nail avulsion and surgery. When eval- diagnosed clinically based on the appearance uating subungual masses, all clinicians should and the history of intermittent discharge of a be aware of the utility of transillumination in mucoid substance.3 DMC is more common in the diagnosis of subungual DMC. ■ people with osteoarthritis and in women. While superfi cial DMC is common, the ■ DISCLOSURES prevalence of subungual DMC is unknown, The authors report no relevant fi nancial relationships which, in the context as few have been reported. The focal collec- of their contributions, could be perceived as a potential confl ict of interest.

■ REFERENCES ungual tumors. J Hand Surg Am 2012; 37(6):1276–1286. doi:10.1016/j.jhsa.2012.04.001 1. Patterson J. Cutaneous mucinoses. In: Weedon's Skin 4. Jabbour S, Kechichian E, Haber R, Tomb R, Nasr M. Man- . 5th ed. New York: Elsevier; 2021:448. agement of digital mucous cysts: a systematic review and 2. de Berker D, Goettman S, Baran R. Subungual myxoid treatment algorithm. Int J Dermatol 2017; 56(7):701–708. cysts: clinical manifestations and response to therapy. J doi:10.1111/ijd.13583 Am Acad Dermatol 2002; 46(3):394–398. Address: Taylor Bullock, BS, Department of Dermatology, doi:10.1067/mjd.2002.119652 Cleveland Clinic, 9500 Euclid Avenue, Cleveland, OH 44195; 3. Willard KJ, Cappel MA, Kozin SH, Abzug JM. Benign sub- [email protected]

CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 427 THE CLINICAL PICTURE Akanksha Kaushik, MD Soufi la KT, MD Manju Daroach, MD Department of Dermatology, , Department of Dermatology, Venereology, Department of Dermatology, Venereology, and Leprology, Postgraduate Institute of and Leprology, Postgraduate Institute of and Leprology, Postgraduate Institute of and Research, Medical Education and Research, Medical Education and Research, Chandigarh, India Chandigarh, India Chandigarh, India

Divya Aggarwal, MD Pulkit Rastogi, MD, DM Muthu Sendhil Kumaran, MD, DNB Department of Histopathology, Postgraduate Department of Histopathology, Postgraduate Department of Dermatology, Venereology, Institute of Medical Education and Research, Institute of Medical Education and Research, and Leprology, Postgraduate Institute of Chandigarh, India Chandigarh, India Medical Education and Research, Chandigarh, India Eruptive seborrheic keratosis: A perilous clue

An abrupt increase in seborrheic keratoses in patients with an underlying malignancy Figure 1. An erythematous, fi rm, nontender Figure 2. Multiple seborrheic keratoses is called the nodule measuring 3 cm × 3 cm in the right of various sizes arranged in a “Christmas Leser-Trélat midaxillary line. tree” pattern on the patient’s back. sign 51-year-old man presented with a The physical examination revealed an A 1-month history of multiple eruptive erythematous nodule measuring 3 cm by 3 cm seborrheic keratoses on his back and a single in the right midaxillary line (Figure 1). The painless nodule on his chest. He reported oc- nodule was fi rm, mobile, and nontender on casional dry cough and loss of appetite over palpation, and it had a normal temperature. the past 3 months, but he did not seek medi- Also noted were multiple seborrheic keratoses cal care for them. He had no history of fever, of various sizes arranged in a “Christmas tree” weight loss, night sweats, or gastrointestinal pattern on his back (Figure 2). The rest of the complaints. examination was unremarkable. A punch biopsy was taken from the nod- doi:10.3949/ccjm.88a.20124 ule. The histopathology report described an

428 CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 KAUSHIK AND COLLEAGUES

unremarkable epidermis with clusters of pleo- morphic tumor cells in the dermis (Figure 3a) arranged in small glands inciting a desmoplas- tic reaction (Figure 3b). The tumor cells had coarse chromatin and a moderate amount of cytoplasm. Immunostaining results were posi- tive for cytokeratin 7 (Figure 3c). Overall, the features suggested a possible adenocarcinoma. The patient underwent whole-body 18F- fl uorodeoxyglucose (FDG) positron emission tomography with contrast-enhanced comput- ed tomography, which revealed a soft-tissue mass lesion with speckled calcifi cation in the right middle lobe of the lung. The lesion was FDG-avid (ie, with high uptake of FDG), het- erogeneously enhancing, and lobulated. The lesion reached up to the hilum, abutting the mediastinum and encasing the right middle lobe bronchus. There were FDG-avid lymph nodes in the right axillary and supraclavicular regions, a single FDG-avid lesion in the left adrenal gland, and multiple subcutaneous and muscu- Figure 3. A: The epidermis appears relatively unremark- lar deposits distributed in the chest wall, left able. The dermis shows a mild to moderate degree of thigh, right gluteal region, and upper back. perivascular and periadnexal mononuclear infl ammatory Based on those results, we made a diagno- infi ltrate (black arrows) along with a tumor deposit in the sis of metastatic non-small-cell lung cancer, deep dermis (red arrow) (hematoxylin and eosin stain; magnifi cation × 20). B: The tumor is composed of cells ar- adenocarcinoma type, not otherwise speci- ranged in small glands (arrow) inciting a desmoplastic reac- fi ed. Palliative chemotherapy with paclitaxel tion. The tumor cells are moderately pleomorphic and have and carboplatin was started, and the patient coarse chromatin and a moderate amount of cytoplasm received oncology follow-up care. Response (hematoxylin and eosin stain; magnifi cation × 200). C: Im- to chemotherapy could not be ascertained, munostaining shows tumor cells positive for cytokeratin 7 as the patient was lost to follow-up owing to (arrows) (CK 7 immunostain; magnifi cation × 200). COVID-19-related lockdown. The exact pathogenesis of the Leser-Trélat ■ ERUPTIVE SEBORRHEIC KERATOSIS sign is unclear. One hypothesis attributes it to An abrupt increase in the size and number of growth factors released by tumor cells, such seborrheic keratoses in patients with an un- as growth hormone, epidermal growth factor, derlying malignancy is called the Leser-Trélat and transforming growth factor alpha. An- sign. More than 50% of associated malignan- other suggests that extracellular matrix com- cies are adenocarcinomas, especially those ponents, such as glycosaminoglycans released of the stomach, colon, rectum, and breast,1 from stroma of the tumor, become incorporat- although this sign has also been reported in ed in distant normal skin, causing epithelial other malignancies, including lung cancer.2 alteration and eruption of seborrheic kerato- The association of the Leser-Trélat sign with ses.5 malignancy is debatable, with suggestions that The Leser-Trélat sign may be the initial the sign may exist independent of an under- presentation, or it can be detected concur- lying occult malignancy or may be associated rently with or after diagnosing an internal with nonmalignant conditions such as benign malignancy. Eruptive seborrheic keratoses can neoplasms, pregnancy, or human immunodefi - occur anywhere, but the most common sites ciency virus infection.3,4 are the back, chest, and extremities.5

CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 429 ERUPTIVE SEBORRHEIC KERATOSIS

Evaluation of a patient with the Leser-Tré- Trélat sign and thus was biopsied. The results lat sign should begin with a detailed history helped us reach the fi nal diagnosis. and clinical examination. Special investiga- Despite the nonspecifi c nature of the tions should be performed to look for the oc- Leser-Trélat sign, our case exemplifi es the cult primary malignancy. importance of performing a thorough evalu- Our patient was apparently doing well ex- ation in patients presenting with sudden-on- cept for the relatively abrupt appearance of set eruptive seborrheic keratoses. ■ multiple eruptive seborrheic keratoses, which prompted us to investigate further for an oc- ■ DISCLOSURES cult malignancy. The nodule on his chest The authors report no relevant fi nancial relationships which, in the context wall could not be explained by the Leser- of their contributions, could be perceived as a potential confl ict of interest.

■ REFERENCES uretic hormone secretion and Leser-Trélat syndrome as uncommon paraneoplastic manifestations of renal malignancy—a geriatric 1. Sardon C, Dempsey T. The Leser-Trélat sign. Cleve Clin J Med 2017; experience: a case report. J Med Case Reports 2019; 13(1):188. 84(12):918. doi:10.3949/ccjm.84a.17021 doi:10.1186/s13256-019-2122-8 2. Asri H, Soualhi M. The sign of Leser-Trélat: think in the adenocarci- 5. Bölke E, Gerber PA, Peiper M, et al. Leser-Trélat sign presenting in a noma of the lung. Pan Afr Med J 2018; 30:270. patient with ovarian cancer: a case report. J Med Case Reports 2009; doi:10.11604/pamj.2018.30.270.16337 3:8583. doi:10.4076/1752-1947-3-8583 3. Heaphy MR Jr, Millns JL, Schroeter AL. The sign of Leser-Trélat in a case of adenocarcinoma of the lung. J Am Acad Dermatol 2000; 43(2 Address: Dr. M. Sendhil Kumaran, Department of Dermatology, Venere- pt 2):386–390. doi:10.1067/mjd.2000.104967 ology and Leprology, PGIMER, Chandigarh, India; 4. Nyanti L, Samsudin A, Tiong IK. Syndrome of inappropriate antidi- [email protected]

Changed your address? Not receiving your copies? To receive Cleveland Clinic Journal of Medicine, make sure the American Medical Association has your current information. Cleveland Clinic Journal of Medicine uses the AMA database of physician names and addresses to deter- mine its circulation. All physicians are included in the AMA database, not just members of the AMA. Only YOU can update your data with the AMA. ■ If your address has changed, send the new information To contact the American Medical Association: to the AMA. If you send the update by mail, enclose a ■ PHONE 800-621-8335 recent mailing label. Changing your address with the AMA will redirect all of your medically related mailings to the ■ FAX 312-464-4880 new location. ■ E-MAIL [email protected] ■ ■ Be sure the AMA has your current primary specialty and US MAIL type of practice. This information determines who receives Send a recent mailing label along with Cleveland Clinic Journal of Medicine. new information to: ■ If you ever notifi ed the AMA that you did not want to receive mail, you will not receive Cleveland Clinic Journal American Medical Association of Medicine. If you wish to reverse that decision, simply AMA Plaza notify the AMA, and you will again receive all AMA mailings. Data Verification Unit 330 N. Wabash Ave., Suite 39300 ■ Please allow 6 to 8 weeks for changes to take effect. Chicago, IL 60611-5885

430 CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 1-MINUTE CONSULT Nora Hilda Hernandez Garcilazo, MD Mohamed Hassanein, MD Department of , Department of Nephrology and Hypertension, Michigan State University, East Lansing, MI Glickman Urological and Kidney Institute, Cleveland Clinic, Cleveland, OH Tushar J. Vachharajani, MD Evamaria Anvari, MD Department of Nephrology and Hypertension, Department of Nephrology and Hypertension, Glickman Urological and Kidney Institute, Cleveland Glickman Urological and Kidney Institute, Cleveland Clinic, Cleveland, OH; Professor, Cleveland Clinic Clinic, Cleveland, OH; Clinical Assistant Professor, BRIEF ANSWERS Lerner College of Medicine of Case Western Reserve Cleveland Clinic Lerner College of Medicine of Case University, Cleveland, OH Western Reserve University, Cleveland, OH TO SPECIFIC CLINICAL Q: Can I place a peripherally inserted QUESTIONS central catheter in my patient with chronic kidney disease?

obliterate the involved veins and prevent their A 45-year-old man is admitted to the hospital use for future creation of a permanent dialysis ac- for sepsis secondary to osteomyelitis. He has dia- cess.4 Clinically diagnosed thrombosis has been betes mellitus, hypertension, and stage 3 chronic reported to occur in 1% to 4% of patients with kidney disease (CKD), with a glomerular fi ltra- 2 a PICC. However, in a 2000 study using venog- tion rate of 46 mL/min/1.73m . He is treated with raphy to evaluate patency of the vessels, Allen intravenous (IV) antibiotics and improves clini- et al5 reported a much higher incidence, with cally. He will need 6 weeks of IV antibiotics after thrombosis evident in 23.3% of patients after discharge. Should a peripherally inserted central PICC insertion. catheter (PICC) be placed for IV access? Higher rates of thrombosis are associated The decision to place a PICC must be with larger catheter sizes, the use of cephalic individualized for the patient. Current A: veins (due to smaller size compared with basilic guidelines do not provide explicit con- traindications for creating permanent vascular veins), greater number of lumens, placement of access, but the general consensus is that poor multiple catheters, and patient factors including PICC insertion malignancy or history of venous thromboembo- candidates include those with advanced demen- is a strong lism.5,6 Central venous stenosis may also occur, tia, left ventricular ejection fraction less than 7 20%, poor vasculature on imaging, or terminal although it is not as common as thrombosis. independent risk illness (life expectancy < 6–12 months).1 In ad- PICC insertion is also a strong independent factor for failure risk factor for failure of an arteriovenous fi stula, dition, national guidelines and the American of an arterio- Board of Internal Medicine’s Choosing Wisely the preferred method of vascular access for he- 7 8 initiative recommend against PICC placement modialysis. McGill et al, in a 2016 observational venous fi stula in patients expected to need permanent dialysis study, found that PICC insertion before or after access in the future (CKD stages 3–5).2 initiation of hemodialysis was associated with fail- ure to transition to a form of permanent access, ■ PICC PROS: CONVENIENCE, LOW COST with only 24.7% of patients transitioning to a working arteriovenous fi stula, and 11.5% transi- PICCs have become increasingly popular in tioning to a functioning arteriovenous graft. This recent years due to their ease of placement, is very important because the transition from cen- convenience for patients, and cost-effective tral venous access to an arteriovenous fi stula or maintenance. Up to 56% of PICCs are placed graft is associated with better survival and fewer to administer IV antibiotics.3 hospitalizations, due to lower risk of serious infec- tions such as endocarditis and bacteremia.8 ■ PICC CONS: BLOOD VESSEL RISKS In patients such as our 45-year-old man, PICCs are highly associated with phlebitis, an end-stage kidney disease (ESKD) life plan thromboembolism, central vein thrombosis, should be created with input from a nephrolo- and stenosis of the involved vessels, which may gist to determine early access needs and to avoid unnecessary procedures and complica- doi:10.3949/ccjm.88a.20173 tions, while also considering life expectancy

CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 431 PICC LINE

Catheter tip Catheter tip

Right Right basilic vein basilic vein

Figure 1. A peripherally inserted central catheter (left) is inserted at the right basilic vein, through the axillary and subclavian veins and into the superior vena cava. The midline catheter (right) is also inserted at the right basilic vein with the tip just below the axilla.

and kidney replacement alternatives.2 peripheral veins but occupy a smaller ending near the axilla, have also been associated with ■ CASE CONTINUED symptomatic venous thrombosis.4 Catheter The team decides to place a PICC through the location plays an important role, with guide- lines suggesting avoiding cephalic, basilic, right basilic vein (Figure 1), and the infection 2,4 resolves with 6 weeks of IV antibiotics. brachial, and subclavian veins. However, 4 months later, he is readmitted A proposed alternative to a PICC is a small-bore, 4-French or 6-French tunneled in- for acute kidney injury and recurrence of os- ternal jugular catheter (Figure 2). It tends to teomyelitis with a paravertebral abscess. The last longer and is associated with fewer com- abscess is surgically drained, and the infectious plications, decreasing the risk of central ve- disease consult recommends 8 weeks of IV an- nous stenosis.10,11 A 2017 retrospective study tibiotics. After a thorough discussion with the by Bhutani et al10 found lower rates of deep nephrology team and the patient, the decision vein thrombosis in tunneled small-bore cen- is made against placing a PICC. tral venous catheters than with PICCs, which may be explained by the shorter length of the ■ PICC ALTERNATIVES catheter and better catheter-to-vein size ratio. Unfortunately, all methods of IV access can However, whether they produce less damage produce venous damage, either by direct trau- to the peripheral vessels or cause central vein ma at the puncture site or by device contact stenosis has not been fully studied.1 along the walls of the vein.9 It has been hy- It has also been suggested that placement pothesized that the more area within a ves- of internal jugular catheters by a skilled proce- sel that a foreign object occupies, the greater duralist with ultrasonographic and fl uoroscopic the possibility of thrombosis due to increased guidance may result in less venous trauma, stasis and direct-contact damage.3 However, reducing the risk of vessel stenosis compared midline catheters, which are also inserted into with nonguided methods.11

432 CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 HERNANDEZ GARCILAZO AND COLLEAGUES

But even after an arteriovenous fi stula has been successfully created, patients with ESKD Right internal requiring hemodialysis must continue vessel- jugular vein preservation strategies as part of their ESKD Catheter tip management plan.1 If IV antibiotics are need- ed, it may be possible to select an agent that can be administered 3 times a week on dialysis days, using the functioning hemodialysis ac- cess. This will avoid the need for a different catheter, decreasing the risk of central venous stenosis and allowing for the creation of other arteriovenous fi stulas if the current one fails.3 Lifelong vessel-preservation strategies Patients who may progress to ESKD and may require hemodialysis access in the future should be identifi ed early so that they can be provided with timely education regarding ves- sel preservation. This includes patients with stage 3 to stage 5 CKD, patients already on kidney replacement therapy such as hemodi- alysis or peritoneal dialysis, and patients who Figure 2. A small-bore (Hohn) catheter placed have a functional transplanted kidney. Such in the internal jugular vein is an alternative to patients should be encouraged to advocate to a peripherally inserted central catheter. preserve their vessels and work with the treat- ment team in balancing the risks and benefi ts patients with advanced CKD (stages 3–5).12 Pa- of every intervention, including blood draws tients and health professionals are encouraged and use of IV and arterial devices.4,9 Medical to visit the website www.saveyourvein.org to alert bracelets and signs at the bedside of hos- further educate themselves on the importance of pitalized patients with CKD indicating the vein preservation.2 ■ need to restrict needle use is essential in edu- cating and alerting the medical community.4 ■ DISCLOSURES It has also been proposed that a nephrology The authors report no relevant fi nancial relationships which, in the context consult be requested before placing a PICC in of their contributions, could be perceived as a potential confl ict of interest.

■ REFERENCES 8. McGill RL, Ruthazer R, Meyer KB, Miskulin DC, Weiner DE. Peripherally inserted central catheters and hemodialysis outcomes. Clin J Am Soc Nephrol 1. Dumaine C, Kiaii M, Miller L, et al. Vascular access practice patterns in Canada: 2016; 11(8):1434–1440. doi:10.2215/CJN.01980216 a national survey. Can J Kidney Health Dis 2018; 5:2054358118759675. 9. Saad T. Vein preservation and alternative venous access. Endovasc Today doi:10.1177/2054358118759675 2008; 32–36. https://evtoday.com/articles/2008-june/EVT0608_03-php. Ac- 2. Lok CE, Huber TS, Lee T, et al. KDOQI clinical practice guideline for vas- cessed July 14, 2021. cular access: 2019 update. Am J Kidney Dis 2020; 75(4 suppl 2):S1–S164. 10. Bhutani G, El Ters M, Kremers WK, et al. Evaluating safety of tunneled small doi:10.1053/j.ajkd.2019.12.001 bore central venous catheters in chronic kidney disease population: a quality 3. Shingarev R, Allon M. Peripherally inserted central catheters and other intra- improvement initiative. Hemodial Int 2017; 21(2):284–293. vascular devices: how safe are they for hemodialysis patients? Am J Kidney doi:10.1111/hdi.12484 Dis 2012; 60(4):510–513. doi:10.1053/j.ajkd.2012.07.003 11. Sasadeusz KJ, Trerotola SO, Shah H, et al. Tunneled jugular small-bore central 4. Vachharajani TJ, Hassanein M, Liaqat A, Haddad N. Vessel preservation catheters as an alternative to peripherally inserted central catheters for in chronic kidney disease. Adv Chronic Kidney Dis 2020; 27(3):177–182. intermediate-term venous access in patients with hemodialysis and chronic doi:10.1053/j.ackd.2020.03.006 renal insuffi ciency. 1999; 213(1):303–306. 5. Allen AW, Megargell JL, Brown DB, et al. Venous thrombosis associated with doi:10.1148/radiology.213.1.r99se12303 the placement of peripherally inserted central catheters. J Vasc Interv Radiol 12. Williams AW, Dwyer AC, Eddy AA, et al. Critical and honest conversations: 2000; 11(10):1309–1314. doi:10.1016/s1051-0443(07)61307-4 the evidence behind the ‘Choosing Wisely’ campaign recommendations by 6. Chopra V, Flanders SA, Saint S. The problem with peripherally inserted cen- the American Society of Nephrology. Clin J Am Soc Nephrol 2012; 7(10):1664– tral catheters. JAMA 2012; 308(15):1527–1528. doi:10.1001/jama.2012.12704 1672. doi:10.2215/CJN.04970512 7. El Ters M, Schears GJ, Taler SJ, et al. Association between prior peripherally in- serted central catheters and lack of functioning arteriovenous fi stulas: a case- Address: Nora Hilda Hernandez Garcilazo, MD, Department of Internal Medicine, control study in hemodialysis patients. Am J Kidney Dis 2012; 60(4):601–608. Michigan State University, Medical Arts Building, 1322 E. Michigan Ave. Suite 300, doi:10.1053/j.ajkd.2012.05.007 Lansing, MI 48912; [email protected]

CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 433 1-MINUTE CONSULT Moronkeji Fagbemi, MD, FASAM Unit Chief, In-patient Detoxifi cation Unit, Addiction Service, BronxCare Hospital Center, Bronx, NY; Assistant Clinical Professor of Medicine, Icahn School of Medicine at Mount Sinai, New York, NY

BRIEF ANSWERS TO SPECIFIC CLINICAL QUESTIONS Q: How do you effectively evaluate the elderly for alcohol use disorder?

Alcohol use disorder (AUD) is a sig- A: nifi cant problem in the elderly and TABLE 1 one that is often undiagnosed, resulting in in- Presentation of alcohol problems creasing emotional, physical, and social con- in the elderly a sequences. In 2019, Americans over age 65 Anxiety accounted for 16.5% of the population (about 54 million), with a projection to reach 22% of Poor hygiene, urinary or fecal incontinence the total population (81 million) by 2040.1,2 It Malnutrition is estimated that AUD affl icts about 1% to 3% Confusion, memory loss, dementia, or delirium of the elderly, but in treatment settings such as Falls doctors’ offi ces and emergency departments, it may be 10 times more frequent. Studies from Marital problems primary care settings show that alcohol prob- Sleep problems lems exist in 10% to 15% of older adults, 30% Depression or mood swings of hospitalized older adults in general medi- Financial problems A comprehen- cine, and about 50% of those hospitalized in sive patient psychiatric divisions.3 Recent studies show Seizures (new-onset, idiopathic) assessment that although the vast majority of those with Worsening of chronic medical problems (hyper- alcohol use disorder see their doctors regularly tension, diabetes, heart failure) includes for a range of issues, fewer than 1 in 10 ever a Note: Crime, antisocial, and substance-seeking illegal behav- 4 an in-depth gets treatment for drinking. iors are not common in this group. history, physical ■ THE PROBLEM OF UNDERDIAGNOSIS examination, ing medical problems with the possibility of In older adults, AUD has an atypical presenta- underlying substance use such as AUD and and other tion (Table 1), which contributes to the un- instead attribute the problems to aging. testing derdiagnosis in this population. Missed diag- 5 Fourth, clinicians may have a therapeu- nosis has many reasons. tic nihilism about alcohol use in the elderly: First, there is a general lack of awareness “What is the point in intervening? It is okay and sensitization about AUD among physi- for him to use, and besides, at this age, he de- cians, with a stereotypical view of AUD as serves a break.” a phenomenon of young and middle-aged And fi nally, the widespread use of prescrip- adults. tion medications among the elderly can hinder Second, clinicians may also be hesitant the detection of AUD. This is further compli- or embarrassed to screen for AUD in a senior cated by survey data showing that only about citizen, and this can be compounded by the 47% of primary care physicians ask about maxi- unwillingness of older adults to acknowledge mum alcohol intake, and only 13% use a for- alcohol problems, due to stigma. mal screening tool for alcohol problems.6 Third, clinicians may fail to link coexist- Older adults are likely to experience more doi:10.3949/ccjm.88a.20123 problems with relatively small amounts of al-

434 CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 FAGBEMI

cohol use due to changes in the pharmacoki- tial to identifying those who need treatment. netics and pharmacodynamics of alcohol, re- This should be explored at every opportunity, sulting in higher blood-alcohol levels. These such as visits for changes in health status or are related to decreases in body water and medications and after major life events such as body mass with age, hence a smaller volume of retirement or the loss of a spouse, in addition distribution, increased sensitivity, and slower to routine health visits. metabolism. Furthermore, many older adults The National Institute on Alcohol Abuse tend to have multiple comorbidities such as and Alcoholism (NIAAA) and the Substance hypertension, other cardiovascular problems, Abuse and Mental Health Service Adminis- and diabetes that can worsen with chronic al- tration (SAMHSA) recommend that men cohol use, as well as having drug-drug interac- and women age 65 and older consume no tions from multiple medications.7,8 more than 1 standard drink per day or 7 stan- dard drinks per week,12,13 unlike the guidelines ■ UNDERSTANDING THE TYPES OF AUD for adults younger than age 65, which are as IN OLDER ADULTS follows: for women, no more than 1 standard drink per day or 7 standard drinks per week; Older individuals who meet AUD criteria for men, no more than 2 standard drinks per can be divided into 2 groups: those who de- day and not more than 14 standard drinks per veloped AUD before age 60, and those who week.12–15 The NIAAA defi nes “risky use” as developed it after age 60. Patients with earlier exceeding the recommended limits of 4 drinks onset of AUD account for about two-thirds of per day (56 g/d based on the US standard of 14 the elderly AUD population and have a more g of ethanol per drink) or 14 drinks per week severe course of illness. They also tend to be (196 g/d) for healthy adult men ages 21 to 64, predominately male and have more alcohol- or 3 drinks per day or 7 drinks per week (42 related medical and psychiatric comorbidities, g/d or 98 g/week) for all adult women of any including being less well-adjusted and having age and men age 65 or older.16 In addition, more antisocial traits. Those with later onset older men should not consume more than 4 of AUD tend to have a milder clinical picture standard drinks on any day. (A standard drink Older adults and few medical problems, possibly because of containing 12 to 14 g of ethanol is equivalent the shorter exposure to alcohol. These patients can experience to a 12-oz can of beer, a 5-oz glass of wine, or also tend to be women, are more affl uent, and about 1.5 oz of 80-proof liquor.) more problems are likely to have begun alcohol misuse after a Screening questions should be asked in a stressful event such as the loss of a spouse, job, with small confi dential setting and in a nonthreatening, or home, or retirement. Hence, it is necessary nonjudgmental manner. Note that a patient amounts to explore these areas when taking the history. may have cognitive impairment that inter- of alcohol Other risk factors for development of later feres with the ability to provide complete and onset of AUD, apart from personal and fam- accurate responses during the medical history, ily history of AUD, include onset of chronic as well as to self-monitor alcohol intake and pain, predisposition to affective or anxiety understand feedback from healthcare provid- disorders, and decreased alcohol metabolism 10 7,9,10 ers. This may necessitate involving family with age. members or friends after discussing permission with the patient. ■ SCREENING: ASK ABOUT DRINKING Formal screening tools for identifying al- The United States Preventive Services Task cohol misuse should be used at every oppor- Force recommends screening adults age 18 tunity as they are brief, easy to recall, and and older for alcohol misuse and providing highly sensitive and specifi c. They can be those engaged in risky or hazardous drinking self-administered by the patient while waiting (a pattern of drinking that increases the risk of in the doctor’s offi ce and reviewed during the physical or psychological problems) with brief face-to-face encounter. Commonly used tools behavioral counseling interventions to reduce include: alcohol misuse.11 Given the high use of medi- • CAGE (cut-annoyed-guilty-eye), a simple cal services by the elderly, clinicians are essen- 4-item yes-or-no questionnaire

CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 435 ALCOHOL USE DISORDER

TABLE 2 Short Michigan Alcoholism Screening Test—Geriatric Version (SMAST-G) Please answer yes or no to the following quesƟ ons: 1. When talking with others, do you ever underestimate how much you drink? 2. After a few drinks, have you sometimes not eaten or been able to skip a meal because you didn’t feel hungry? 3. Does having a few drinks help decrease your shakiness or tremors? 4. Does alcohol sometimes make it hard for you to remember parts of the day or night? 5. Do you usually take a drink to relax or calm your nerves? 6. Do you drink to take your mind off your problems? 7. Have you ever increased your drinking after experiencing a loss in your life? 8. Has a doctor or nurse ever said they were worried or concerned about your drinking? 9. Have you ever made rules to manage your drinking? 10. When you feel lonely, does having a drink help?

Extra question (asked, but not calculated in the fi nal score): Do you drink alcohol and take mood or mind-altering drugs, including prescription tranquilizers, prescription sleeping pills, prescription pain pills, or any illicit drugs?

Scoring: 1 point for each “yes” answer, and total the responses. A score of 2 or more points indicates an alcohol problem, and a brief intervention should be conducted.

Adapted from reference 17.

• AUDIT (Alcohol Use Disorder Identi- defi ned as mild for the presence of 2 to 3 symp- Hypertension, fi cation Test), with good sensitivity and toms, moderate for 4 to 5 symptoms, or severe cardiovascular specifi city, or its shorter version AUDIT-C for 6 or more symptoms. disease, and questionnaire, which is good at identifying Assessment and diagnosis provide data those with hazardous drinking for a comprehensive and effective treatment diabetes can • MAST-G (Michigan Alcohol Screening plan, a process called SBIRT (screening, brief worsen with Test–Geriatric Version) and its shorter intervention, and referral to treatment). Plans version (SMAST-G), shown in Table 2.17 range from simple brief intervention in the chronic alcohol MAST-G and SMAST-G were designed clinic to admission to a medically managed use for older patients who drink less and are bet- facility. After discharge, older patients with ter than the CAGE tool at identifying older AUD need to participate in alcohol treatment adults with AUD.18 Using more than 1 screen- programs with a focus specifi c to the elderly ing tool can provide more data on alcohol that incorporates both pharmacologic treat- quantity consumed, alcohol use patterns, and ment and nonpharmacologic treatments, and alcohol-related consequences, which may to participate in support groups such as Alco- help identify AUD in older patients across holics Anonymous. Other treatment options a wide range of demographic characteristics. are cognitive behavior therapy and motiva- SAMHSA has published a compilation of tional interviewing techniques and admission these instruments.19 to a therapeutic community.

■ DIAGNOSIS ■ MEDICAL HISTORY, PHYSICAL EXAMINATION, TESTS According to the American Psychiatric Asso- ciation, the diagnosis of AUD requires a pa- A comprehensive patient assessment includes tient to meet 2 of 11 criteria during the same an in-depth medical history, physical exami- 12-month period.20 The severity of AUD is nation, and other testing (Table 3). History-

436 CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 FAGBEMI

taking should inquire about falls, sleep problems, physiologic dependence and with- TABLE 3 drawals, level of cognitive function, medical Assessment of alcohol use disorder and psychiatric comorbidities, medication in the elderly a history with potential of drug-to-drug and Previous diagnosis of alcohol-use disorder drug-to-alcohol interactions, and surgical his- tory such as bariatric surgery (especially Roux- Drinking patterns, physiologic dependence, withdrawals en-Y gastric bypass), as this can increase the Presence of intoxication risk of AUD postoperatively. A psychosocial Neuropsychiatric comorbidities or manifestations including suicidal evaluation should include anxiety, presence of ideation chronic pain, level of social activities, family dynamics (ie, level of interaction with family Medical comorbidities and complications including chronic pain; current medications and friends), fi nances, housing, legal issues, and diet. The evaluation should also include Psychosocial evaluation including housing, dietary issues, fi nances, questions about depression and suicide, as well legal issues, sociability, family as diagnostic tools such as the Patient Health Prior treatment, including pharmacotherapy: success, failure, relapse, Questionnaire 9,21 if warranted. Also impor- participation in support groups tant is assessing the patient’s level of readiness Patient’s level of motivation to change and motivation to change. Motivational tech- a Assessment should include screening, a thorough medical and psychosocial history, niques may be useful for patients exhibiting physical examination, and appropriate laboratory tests. less willingness to change. The physical examination should be thor- ough, assessing for intoxication and alcohol transferrin test, if available, suggest recent al- breath. Consider using noninvasive tools such cohol abuse, particularly when corroborated as a breathalyzer or obtaining urine or blood with elevated levels of other liver-associated samples to check for alcohol levels with ethyl enzymes. Other uses of this test include long- glucuronide or ethyl sulfate testing. Also as- term monitoring for early detection of relapse sess for alcohol withdrawal using a withdrawal drinking during medical treatment, enabling Older adults scale such as the Clinical Institute Withdraw- early intervention. 22 with AUD al Assessment for Alcohol. The physical ex- ■ amination should also assess for comorbidities A WORD ON TREATMENT benefi t and complications of AUD such as hyperten- Even though an elaborate and in-depth evalu- from programs sion, cardiomyopathy, neuropathy, myopathy, ation of treatments for AUD in older adults is that offer and alcoholic liver disease including cirrhosis. beyond the scope of this article, several studies Other testing should include the Mini- have documented that older adults with AUD age-appropriate Mental State Examination and electrocardiog- seem to do best in programs that offer age- care raphy to identify arteriosclerosis, arrhythmias, appropriate care, including individual, group, and cardiomegaly. Also, include laboratory and family therapy, and in self-help group tests such as gamma-glutamyl transpeptidase, meetings such as Alcoholics Anonymous with hepatic transaminases such aspartate ami- providers who are knowledgeable about aging- notransferase and alanine aminotransferase related issues. For older patients, these pro- (a 2:1 ratio pattern suggests chronic alcohol grams can result in higher attendance at group use), basic blood chemistry panel, and com- meetings and a greater likelihood of complet- plete blood cell count to check for elevated ing treatment than in younger patients.25 mean corpuscular volume. These tests, though Pharmacotherapy for short-term control of essential in the overall scheme, are by them- alcohol withdrawal includes benzodiazepines selves poor screening tools for AUD in older such as lorazepam and treatment of medical patients and thus should be used in conjunc- and psychiatric comorbidities. Treatment of tion with results from the medical history, acute withdrawal in patients with multiple physical examination, and formal screening severe comorbidities or AUD complications tools.23,24 should be done in a medically supervised Elevated levels on a carbohydrate-defi cient setting. Pharmacotherapy should follow the

CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 437 ALCOHOL USE DISORDER

principle of “start low, go slow” and should been found to reduce alcohol consumption be closely monitored for early detection of and increase abstinence rates.27 It is renally ex- adverse effects such as cognitive impairment, creted unmetabolized; thus, before prescribing hypotension, and falls.5 it, the patient’s renal function should be deter- mined, as renal dysfunction is not uncommon Naltrexone in the elderly. Acamprosate is most effective Studies have shown that pharmacotherapy at maintaining abstinence in patients who are such as naltrexone, which is thought to block not currently drinking alcohol.27,28 the endogenous opioid system contribution to alcohol priming effects, can be an important Disulfi ram ally in preventing relapse when used in con- The evidence is inconsistent on the effi cacy of junction with behavioral interventions and disulfi ram, the only alcohol-sensitizing medi- treatments.26 The oral dose approved by the cation FDA-approved to treat AUD in the US Food and Drug Administration (FDA) is elderly. It decreases alcohol consumption by 50 mg daily, but it can be initiated at 25 mg/ irreversibly binding with the enzyme aldehyde day with a 25-mg/day increase as often as dehydrogenase, hence causing a disulfi ram- weekly, to a maximum of 150 mg/day, using ethanol reaction. Daily dosage ranges from desire-to-drink and other patient symptoms as 250 mg to 500 mg. a measure of relative risk of relapse to heavy Factors that make disulfi ram a less suit- drinking. able choice for many older patients with AUD Compared with placebo, naltrexone has include problems with adherence and serious been found to be well tolerated, with no sig- adverse effects (drowsiness, optic neuritis, nifi cant differences in frequency of self-report- peripheral neuropathy, hepatotoxicity). It is ed adverse effects or in liver enzyme values. contraindicated in patients with history of sei- But treatment has resulted in reduced craving zure, psychosis, or cerebrovascular accident, for alcohol, as well as in fewer drinking days and in those not willing to achieve complete 5,28 and relapse events.26 abstinence. Long-acting naltrexone, available as a ■ 380-mg dose given every 4 weeks as a deep THE BOTTOM LINE intramuscular gluteal injection, is an alterna- AUD is a signifi cant problem in the elderly, tive, especially if patients do not adhere to the and as this segment of the population contin- daily oral regimen. Avoid long-acting naltrex- ues to grow, we can expect to see more elderly one in patients taking opioids and in those patients with AUD. Fortunately, studies have with elevated levels of liver enzymes. shown that elderly patients with AUD have very good outcomes when it is diagnosed and Acamprosate treatment is initiated, especially with age- Another medication useful in AUD reha- specifi c care and programming. It behooves bilitation is acamprosate, an amino acid de- clinicians to be knowledgeable about the pre- rivative that fosters gamma-aminobutyric acid sentations, screening, and assessment of AUD neurotransmission. It seems to interact with in the elderly, with the goal of timely inter- glutamate at the N-methyl-d-aspartate recep- ventions and referral to appropriate care. ■ tor, although its exact mechanism is unclear. It is an alternative and has an FDA-approved ■ DISCLOSURES dosing of 1,998 mg/day, usually administered The authors report no relevant fi nancial relationships which, in the context as two 333-mg capsules 3 times a day. It has of their contributions, could be perceived as a potential confl ict of interest.

438 CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 FAGBEMI

■ REFERENCES 16. National Institute on Alcohol Abuse and Alcoholism. Helping patients who drink too much: a clinician’s guide. Updated 2005 1. United States Census Bureau. Quick facts. https://www.census.gov/ edition. https://www.niaaa.nih.gov/guide. Accessed July 20, 2021. quickfacts/fact/table/US/PST045219. Accessed July 20, 2021. 17. Blow FC, Brower KJ, Schulenberg JE, et al. The Michigan Alcoholism 2. US Department of Health and Human Services, Administration for Screening Test - Geriatric Version (MAST-G): A new elderly-specifi c Community Living. 2020 Profi le of older Americans. https://acl.gov/ screening instrument. Alcoholism: Clinical and Experimental Re- aging-and-disability-in-america/data-and-research/profi le-older- search 1992; 16:372. americans. Accessed July 20, 2021. 18. Moore AA, Seeman T, Morgenstern H, Beck JC, Reuben DB. Are 3. Caputo F, Vignoli T, Leggio L, Addolorato G, Zoli G, Bernardi M. there differences between older persons who screen positive on the Alcohol use disorders in the elderly: a brief overview from epide- CAGE questionnaire and the Short Michigan Alcoholism Screening miology to treatment options. Exp Gerontol 2012; 47(6):411–416. Test-Geriatric Version? J Am Geriatr Soc 2002; 50(5):858–862. doi:10.1016/j.exger.2012.03.019 doi:10.1046/j.1532-5415.2002.50211.x 4. Dryden J. Alcohol problems severely undertreated. Washington 19. Substance Abuse and Mental Health Services Administration (SAM- University School of Medicine in St. Louis. May 17, 2021. https:// HSA). A guide to substance abuse services for primary care clinicians. medicine.wustl.edu/news/alcohol-problems-severely-undertreated/. Treatment Improvement Protocol (TIP) series, No. 24. Published Accessed July 20, 2021. 1997. https://www.ncbi.nlm.nih.gov/books/NBK64829/#A45987. 5. Lal R, Pattanayak RD. Alcohol use among the elderly: issues and Accessed July 20, 2021. considerations. J Geriatr Ment Health 2017; 4:4–10. 20. Diagnostic and Statistical Manual of Mental Disorders, 5th ed. doi:10.4103/jgmh.jgmh_34_16 Arlington, VA: American Psychiatric Association, 2013. 6. Friedmann PD, McCullough D, Chin MH, Saitz R. Screening and 21. American Psychological Association. Patient health questionnaire intervention for alcohol problems. A national survey of primary care (PHQ-9). https://www.apa.org/depression-guideline/patient-health- physicians and psychiatrists. J Gen Intern Med 2000; 15(2):84–91. doi:10.1046/j.1525-1497.2000.03379.x questionnaire.pdf. Accessed July 20, 2021. 7. Offsay J. Treatment of alcohol-related problems in the elderly. Ann 22. Sullivan JT, Sykora K, Schneiderman J, Naranjo CA, Sellers EM. Long Term Care 2007; 15:39–44. Assessment of alcohol withdrawal: the revised clinical institute 8. National Institute on Alcohol Abuse and Alcoholism. Alcohol and withdrawal assessment for alcohol scale (CIWA-Ar). Br J Addict 1989; aging. April 1998. https://pubs.niaaa.nih.gov/publications/aa40.htm. 84(11):1353-7. doi:10.1111/j.1360-0443.1989.tb00737.x Accessed July 12, 2021. 23. Di Bari M, Silvestrini G, Chiarlone M, et al. Features of excessive 9. Liberto JG, Oslin DW. Early versus late onset of alcohol- alcohol drinking in older adults distinctively captured by behavioral ism in the elderly. Int J Addict 1995; 30(13–14):1799–1818. and biological screening instruments. An epidemiological study. J Clin doi:10.3109/10826089509071056 Epidemiol 2002; 55(1):41–47. doi:10.1016/s0895-4356(01)00408-5 10. Trevisan LA. Elderly alcohol use disorders: epidemiology, screening, 24. Hietala J, Koivisto H, Anttila P, Niemelä O. Comparison of the com- and assessment issues. Psychiatric Times 2014; 31(5). bined marker GGT-CDT and the conventional laboratory markers of 11. US Preventive Services Task Force (USPSTF). Unhealthy alcohol use alcohol abuse in heavy drinkers, moderate drinkers and abstainers. in adolescents and adults: screening and behavioral counseling Alcohol Alcohol 2006; 41(5):528–533. doi:10.1093/alcalc/agl050 interventions. Updated November 13, 2018. https://www.uspreven- 25. Sattar SP, Petty F, Burke WJ. Diagnosis and treatment of alcohol de- tiveservicestaskforce.org/uspstf/recommendation/unhealthy-alcohol- pendence in older alcoholics. Clin Geriatr Med 2003; 19(4):743–761. use-in-adolescents-and-adults-screening-and-behavioral-counseling- doi:10.1016/s0749-0690(03)00045-4 interventions. Accessed July 20, 2021. 26. Oslin D, Liberto JG, O’Brien J, Krois S, Norbeck J. Naltrexone as an 12. National Institute on Alcohol Abuse and Alcoholism. Drinking levels adjunctive treatment for older patients with alcohol dependence. defi ned. https://www.niaaa.nih.gov/alcohol-health/overview-alco- Am J Geriatr 1997; 5(4):324–332. hol-consumption/moderate-binge-drinking. Accessed July 20, 2021. doi:10.1097/00019442-199700540-00007 13. Substance Abuse and Mental Health Services Administration (SAM- 27. Winslow BT, Onysko M, Hebert M. Medications for alcohol use HSA). SAMHSA—Get connected: linking older adults with resources disorder. Am Fam Physician 2016; 93(6):457–465. pmid:26977830 on medication, alcohol, and mental health. 2019 edition. https:// 28. Substance Abuse and Mental Health Services Administration (SAM- store.samhsa.gov/sites/default/fi les/d7/priv/sma03-3824_2.pdf. Ac- HSA). Treating substance use disorder in older adults. Treatment cessed July 20, 2021. Improvement Protocol (TIP) series, No. 26. Updated 2020. https:// 14. Dufour M, Fuller RK. Alcohol in the elderly. Annu Rev Med 1995; store.samhsa.gov/sites/default/fi les/SAMHSA_Digital_Download/ 46:123–132. doi:10.1146/annurev.med.46.1.123 PEP20-02-01-011%20PDF%20508c.pdf. Accessed July 20, 2021. 15. Blow FC, Barry KL. Use and misuse of alcohol among older women. National Institute on Alcohol Abuse and Alcoholism. June 2003. Address: Moronkeji Fagbemi, MD, FASAM, 1285 Life Recovery Center, https://pubs.niaaa.nih.gov/publications/arh26-4/308-315.htm Ac- BronxCare Hospital, 1285 Fulton Avenue, Bronx, NY 10456; cessed July 20, 2021. [email protected]

CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 439 CURRENT DRUG THERAPY Yael Mauer, MD, MPH Marcie Parker, PharmD, BCACP Sangeeta R. Kashyap, MD Department of Internal Medicine and , Ambulatory Care Clinical Specialist, Department Department of Endocrinology, Diabetes, and Cleveland Clinic, Cleveland, OH; Assistant of , Cleveland Clinic Beachwood Family Metabolism, Cleveland Clinic, Cleveland, OH; Professor of Medicine, Cleveland Clinic Lerner Health and Surgery Center, Beachwood, OH Professor, Cleveland Clinic Lerner College of College of Medicine of Case Western Reserve Medicine of Case Western Reserve University, University, Cleveland, OH Cleveland, OH

Antiobesity drug therapy: An individualized and comprehensive approach

ABSTRACT ntiobesity medications are signifi cantly A underprescribed. Only 2% of US adults Obesity affects 42% of US adults and has a devastat- eligible for obesity pharmacotherapy receive ing impact on health. Although many patients initially it.1 Contributing to this underutilization are lose weight with diet and exercise, long-term weight inadequate training of prescribers and stigma- loss is diffi cult to achieve. Pharmacotherapy, as part of a tization of obesity as resulting from a perceived comprehensive plan, can help patients lose weight and lack of willpower on the part of the patient.2,3 avoid regaining it. Choosing an antiobesity drug regimen Familiarizing clinicians with the pathophysiol- should be an individualized, shared decision-making pro- ogy of obesity and helping them provide an in- cess that accounts for patient preferences, comorbidities, dividualized and comprehensive plan for their and out-of-pocket costs. We review antiobesity drugs and patients is the purpose of this review. propose an individualized and comprehensive approach to obesity management. ■ OBESITY RATES ARE HIGH AND GETTING HIGHER KEY POINTS Obesity (body mass index [BMI] ≥ 30 kg/m2) Antiobesity medications as part of a comprehensive plan affects 42.4% of US adults, or more than 107 can help patients achieve lasting obesity control and million people, and 9.2% of US adults have provide independent health benefi ts, including decreased severe obesity (BMI ≥ 40 kg/m2).2,4 These rates cardiovascular risk. have increased dramatically since 1999, when 30.5% of Americans were obese and 4.7% were severely obese,2 and they continue to in- All patients suffering from obesity should be counseled to crease, so that it is now estimated that 51% of adopt a healthful low-calorie diet, exercise regularly, get US adults will be obese by 2030.5 adequate sleep, and manage stress. The prevalence of obesity is similar in both sexes, but women are more likely to be severe- Antiobesity management plans should include lifestyle ly obese (11.5% vs 6.9%).4 Young, middle- counseling, discontinuation of obesity-inducing medica- aged, and older adults all have similar obesity tions when possible, and, when indicated, weight loss rates. The obesity rate in non-Hispanic Black surgery. people is 49.6%, which is higher than in His- panic people (44.8%), non-Hispanic White people (42.2%), and non-Hispanic Asian peo- ple (17.4%). Rates of severe obesity are 13.8% in Black people, 9.3% in non-Hispanic White people, 7.9% in Hispanic people, and 2% in doi:10.3949/ccjm.88a.20080 non-Hispanic Asian people.4

440 CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 MAUER AND COLLEAGUES

Obesity rates decrease with increasing in- lar physical activity have failed to achieve a come: 39% obesity at 130% or less of the fed- healthy weight.7 Choosing the right drug regi- eral poverty level vs 31.2% in those at higher men involves an individualized, shared deci- than 350% of the federal poverty level.6 Rates sion-making process that accounts for patient also decrease with college education: 27.8% in preferences, comorbidities, and out-of-pocket college graduates vs 40% in nongraduates.6 costs (Table 1, Table 2).

■ ADVERSE EFFECTS ON HEALTH Tolerability Adverse effects frequently limit the use of Obesity is a chronic disease characterized by a anti obesity medications. long-term positive energy balance resulting in 3 Liraglutide and semaglutide, glucagon- excess adiposity. Weight in relation to height like peptide 1 (GLP-1) receptor agonists that (ie, the BMI) and waist circumference are sur- carry a US Food and Drug Administration rogates for measuring adiposity. (FDA) indication for long-term treatment of Obesity leads to structural abnormalities obesity, are generally well tolerated. (Semaglu- such as venous stasis and hepatic steatosis, tide was approved in June 2021.) Gastrointes- physiologic derangements such as insulin re- tinal effects are common and include nausea, sistance and infl ammation, and functional im- abdominal cramping, and diarrhea. Liraglu- pairments such as gastroesophageal refl ux dis- tide requires daily injections, and semaglutide ease, urinary incontinence, osteoarthritis, and requires weekly injections, and although some disability. Obesity increases the risk of devel- patients fi nd injections undesirable, they rare- oping more than 200 other chronic diseases ly discontinue this therapy because of its mode and thus is associated with signifi cant rates of of administration. Semaglutide is available in morbidity and death.3 an oral formulation, but this is not approved for the treatment of obesity. ■ DRUG THERAPY CAN AID WEIGHT LOSS Phentermine-topiramate, another FDA- Obesity is infl uenced by genetics, epigenetics, approved drug, is also well tolerated. The socioeconomic status, access to food and exer- combination can achieve higher weight loss By 2030, cise, psychological health, iatrogenic factors, at lower doses of each medication, thus lessen- and behavior, and thus it is challenging to ing the chance of dose-dependent adverse ef- 51% of US adults treat. Furthermore, biological responses to loss fects. Irritability, insomnia, neuropathy, prob- may be obese of adipose tissue trigger a decrease in metabol- lems with memory, and changes in taste are ic rate and increase in energy effi ciency while the adverse effects most frequently reported. inducing hunger and decreasing satiety.7 Patients often fi nd soda drinks less appealing, Drug therapy helps patients adhere to life- which can help their weight management. style changes and overcome these responses. When patients experience severe topira- Typically, monotherapy leads to a loss of 3% mate-associated effects such as neuropathy, or to 8% of total body weight from baseline, and phentermine-associated effects such as irrita- combination therapy can result in even great- bility, one can consider prescribing the other er loss.6,7 Moreover, some antiobesity drugs medication by itself. Generally, phentermine can provide independent health benefi ts, such is very well tolerated, and side effects wane as decreases in blood pressure, harmful lipid with continued use. However, the clinician levels, waist circumference, insulin resistance, may recommend as-needed use for patients nonalcoholic fatty liver disease, risk of major who experience signifi cant adverse effects. For cardiovascular events, and progression of dia- example, if a patient tends to eat excessively 1,5,8–11 betic kidney disease. during the weekends, phentermine can be taken just on weekends. ■ CHOOSING THE RIGHT DRUG Phentermine carries a theoretical poten- Pharmacotherapy is indicated in patients with tial for abuse and addiction, but this concern a BMI of 30 kg/m2 or higher, or 27 kg/m2 or is not supported in the literature. Topiramate higher with obesity-associated complications is generally less well tolerated, with frequent in whom a healthy low-calorie diet and regu- neurologic and psychiatric adverse effects.

CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 441 ANTIOBESITY DRUG THERAPY

TABLE 1 Drugs approved for long-term treatment of obesity Average wholesale Drug and class Ideal candidates Adverse effects Contraindicationsa price b Liraglutide Patients with coronary Constipation, diarrhea, nau- Family or personal history of Liraglutide Semaglutide artery disease, prediabetes, sea, headache, fatigue, and medullary thyroid cancer or $1,619 (GLP-1 receptor and diabetes injection site reactions. Serious multiple endocrine neoplasia Semaglutide agonist) but rare: increased heart rate, type 2. Use with caution in renal impairment, pancreatitis, patients with severe chronic $1,022 and suicidal ideation. Potential kidney disease and history of risk of thyroid C-cell tumor. pancreatitis.

Semaglutide is associated with an increased incidence of diabetic retinopathy complica- tions, probably attributable to rapid correction of hyperglyce- mia in patients with diabetes. Naltrexone- Patients with depression, Glaucoma, hepatotoxicity, Uncontrolled hypertension, $365 bupropion those interested in smok- increase in heart rate and seizures, anorexia, bulimia, (opioid receptor ing cessation, and those blood pressure, headache, drug or alcohol withdrawal, or antagonist and with food addiction and nausea, constipation, vomit- chronic opioid use. DNRI) strong cravings ing, dry mouth. Serious but rare: suicidal ideation and a lower seizure threshold. Orlistat Patients who do not want Headaches, fl atulence, cramp- None, but not recommended $108 (lipase inhibitor) to take a systemic drug, or ing, fecal incontinence, oily for patients with malabsorp- patients who eat a moder- spotting, decreased absorption tion (eg, after gastric bypass ate- or high-fat diet of medications and fat-soluble surgery). vitamins. Gastric disturbances can be reduced by taking with psyllium. Serious but infre- quent: liver injury, cholelithia- sis, nephrolithiasis. Phentermine- Patients with chronic Increased heart rate, dizziness, Uncontrolled anxiety or $239 topiramate ER c migraines neuropathy, insomnia, anxiety, depression, cardiovascular (sympatho- depression, cognitive impair- disease, uncontrolled hyper- mimetic amine ment, and dry mouth. Serious tension, hyperthyroidism, and GABA recep- but rare: suicidal ideation, glaucoma, and history of tor modulator) acidosis, hypokalemia, rise in se- substance dependence. rum creatinine, myopia, or glau- coma. Minimal risk of seizures with rapid discontinuation.

a All antiobesity medications are contraindicated in pregnancy. Because of potential teratogenicity of many antiobesity drugs, a pregnancy test should be done before prescribing, and women should be counseled on effective birth control. b Lexicomp average wholesale price for 30-day supply of maximum doses as of May 2021. c A controlled substance. DNRI = dopamine-norepinephrine reuptake inhibitor; ER = extended release; GABA = gamma aminobutyric acid; GLP-1 = glucagon-like protein-1

442 CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 MAUER AND COLLEAGUES

TABLE 2 Other drugs used for treating obesity Average wholesale Drug and class Ideal candidates Side effects Contraindicationsa price b Lisdexam- Patients with attention Insomnia, irritability, anxiety, dry None, but we recommend $402 fetamine c defi cit hyperactivity mouth, increased heart rate and against using in patients (amphetamine disorder blood pressure. Controlled sub- with cardiovascular prodrug) stance with theoretical potential disease. for abuse and dependence.

Phentermine d N/A Headaches, increased blood pres- Uncontrolled anxiety and $21.30 (sympathomimetic sure and heart rate, irritability, hypertension, cardiovas- amine) insomnia, constipation, diarrhea, cular disease, hyperpara- impotence, dizziness. Controlled thyroidism, glaucoma, and substance with theoretical poten- history of drug depen- tial for abuse and dependence. dence. Serious but rare: pulmonary hyper- tension, valvular disease. Topiramate e Patients with chronic Insomnia, xerostomia, constipa- Hyperthyroidism, $10 (GABA receptor migraines tion, paresthesias, dizziness, glaucoma. modulator) anxiety, depression, drowsiness, language and memory impair- ments. Very rare: seizures with rapid discontinuation. Metformin e,f Patients with diabetes Diarrhea, nausea, abdominal pain. Severe chronic kidney $5 (biguanide) and prediabetes Serious but rare: lactic acidosis. disease.

Pramlintide e,f Patients with type 1 or Hypoglycemia, headaches, nausea, Gastroparesis and hypo- $694 (amylin analogue) type 2 diabetes vomiting. glycemic unawareness. SGLT-2 Patients with type 2 dia- Genitourinary infections, hypovole- Severe chronic kidney $600–$700 inhibitors e,f betes, hypertension, heart mia, increased low-density lipopro- disease and ketogenic diet failure, cardiovascular tein cholesterol, and hyperkalemia. (concern for euglycemic disease, diabetic kidney Serious but rare: diabetic ketoaci- ketoacidosis). disease dosis, bone fractures, amputations, Fournier gangrene.

a All antiobesity medications are contraindicated in pregnancy except for metformin in patients with diabetes. Because of the potential teratogenic effect of many antiobesity medications, a pregnancy test should be obtained before prescribing, and women should be counseled on effective birth control. b Lexicomp average wholesale price for 30-day supply of maximum doses as of May 2021. c Approved for treatment of binge-eating disorder. d Approved for short-term use; however, it is often prescribed long-term in US states where this regulation is not strictly enforced. e Off-label use. f Occasionally prescribed for patients who do not have diabetes or another US Food and Drug Administration indication for their use; however, this is not our practice. GABA = gamma aminobutyric acid; SGLT-2 = sodium-glucose cotransporter-2

Naltrexone-bupropion, another FDA-ap- bance. A stool softener or coprescription with proved option, is usually well tolerated, with metformin can be considered to promote more constipation, headaches, irritability, anxiety, regular bowel movements. and insomnia being commonly reported. These Lisdexamfetamine, FDA-approved for can be ameliorated by reducing doses, eg, skip- treating binge-eating disorder but not obe- ping the afternoon dose to reduce sleep distur- sity per se, has very infrequent side effects. It

CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 443 ANTIOBESITY DRUG THERAPY

has potential for abuse and dependence, but dividual components are available generically whether this actually occurs in patients taking and can be prescribed separately, in combina- it for binge-eating disorder is not well estab- tion or individually (average wholesale price lished. $21.30 for phentermine 18.75 mg plus topira- Metformin, a diabetes drug frequently used mate 100 mg). off-label to treat obesity, is also well tolerated. Despite the listed prices, both phenter- Its most common adverse effects include ab- mine and topiramate can be bought very inex- dominal pain, nausea, and diarrhea. These are pensively at certain retailers at less than $10 lessened when using extended-release formula- per month. Phentermine-topiramate can be tions and when taken with meals. Dose reduc- used alternatively with phentermine alone to tion is often required to manage adverse effects. reduce overall costs. Orlistat, FDA-approved for treating obe- Phentermine is FDA-approved for short- sity, is poorly tolerated, with abdominal pain, term use only—3 months of therapy with 6 nausea, bloating, fl atulence, and diarrhea be- months between courses. This is not strictly ing very common and bothersome effects. enforced in all US states. Phentermine is the Gastric disturbances can be reduced when this most commonly prescribed antiobesity drug in drug is taken with psyllium. the United States.5 Orlistat, average wholesale price $108, ■ COST has no generic formulation and often is not covered by insurance. It is available over the Antiobesity drugs are not frequently covered by counter at a lower dose. insurance, and their cost limits patients’ choices. Lisdexamfetamine, average wholesale Most expensive price $402, has no generic alternative but is Liraglutide and semaglutide. Liraglutide, av- often covered by insurance when prescribed erage wholesale price (30-day supply of maxi- for binge-eating disorder. mum dose) $1,619, and semaglutide (average Most affordable wholesale price $1,022) are the most expen- Obesity is Metformin, average wholesale price $5, is an sive choices, have no generic alternatives, and economical option available in generic form multifactorial are often not covered by insurance. Lower and covered by insurance. Phentermine and and thus doses are more affordable. These drugs are usu- topiramate are also affordable at less than $10 ally covered by insurance when prescribed for is diffi cult per month, and bupropion is another afford- diabetes, but when prescribed for this indica- able option. to treat tion, the doses are lower and thus there is less Very frequently, manufacturers offer cus- with lifestyle effect on weight loss. tomer-assistance programs, but these programs Depending on insurance coverage, other often exclude patients with Medicaid and modifi cations GLP-1 receptor agonists may be more suitable Medicare insurance. alone for patients with diabetes and obesity. Intermediate cost ■ EFFICACY IN WEIGHT LOSS Naltrexone-bupropion, average wholesale price Semaglutide is the most effective agent. Phenter- $365, has no generic option and is not usually mine-topiramate, naltrexone-bupropion, liraglu- covered by insurance. However, its individual tide, lisdexamfetamine, and pramlintide are also components are available as generic formula- associated with a high percentage of total body tions and can be prescribed separately at much weight loss. Orlistat, topiramate alone, metfor- lower cost (average wholesale price $50 for nal- min, and sodium-glucose cotransporter-2 (SGLT- trexone 25 mg plus bupropion XL 300 mg). Bu- 2) inhibitors are associated with a more modest propion can also be prescribed alone for obesity weight loss. Further, orlistat is effective only in management and is usually covered by insur- those who consume a moderate- or high-fat diet. ance (average wholesale price $16.50). Table 3 lists percentages of total body weight loss Phentermine-topiramate, average whole- observed in randomized controlled trials of the sale price $239, has no generic formulation various drugs.10–29 Of note, not all available dos- and is usually not covered by insurance. Its in- ages of the medications today were included in

444 CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 MAUER AND COLLEAGUES

TABLE 3 Weight loss in randomized, double-blind, placebo-controlled trials % TBWL vs Medication Patients Dose placebo Liraglutide10 Diabetes, body mass index (BMI) ≥ 27 kg/m2 1.8 mg 2.7 3.0 mg 4.0 Liraglutide11 Prediabetes, BMI ≥ 30 or ≥ 27 with hypertension or dyslipidemia 3.0 mg 5.4 Liraglutide12 No diabetes, BMI ≥ 30 or ≥ 27 with hypertension or dyslipidemia, 3.0 mg 6.0 who lost ≥ 5% total body weight with a low-calorie diet Naltrexone-bupropion13 Diabetes, BMI ≥ 27 32/360 mg 3.2 Naltrexone-bupropion14 No diabetes, BMI ≥ 30 or ≥ 27 with hypertension or dyslipidemia 16/360 mg 3.7 32/360 mg 4.8 Naltrexone-bupropion15 No diabetes, BMI ≥ 30 or ≥ 27 with hypertension or dyslipidemia 32/360 mg 5.2 Naltrexone-bupropion16 No diabetes, BMI ≥ 30 or ≥ 27 with hypertension or dyslipidemia 32/360 mg 4.2 Orlistat17 BMI 30–43 120 mg TID 3.0 Orlistat18 BMI 30–44 120 mg TID 3.7 Orlistat19 Type 2 diabetes, clinically stable on oral sulfonylureas, BMI 28–40 120 mg TID 1.9 Phentermine-topiramate ER20 No diabetes, BMI ≥ 35, blood pressure ≤ 140/90 mm Hg 3.75/23 mg 3.5 15/92 mg 9.3 Phentermine-topiramate ER21 BMI 27–45 with at least 2 of the following: hypertension, dyslipid- 7.5/46 mg 8.0 emia, diabetes, prediabetes, abdominal obesity 15/92 mg 10.8 Phentermine-topiramate ER22 BMI 27–45 with at least 2 of the following: hypertension, dyslipid- 7.5/46 mg 7.1 emia, diabetes, prediabetes, abdominal obesity 15/92 mg 8.5 Phentermine-topiramate ER23 BMI 30–45 7.5/46 mg 6.8 15/92 mg 7.5 Lisdexamfetamine24 Adults with binge eating disorder, BMI 25–45 30 mg 3.3 50 mg 5.2 70 mg 5.3 Phentermine23 BMI 30–45 7.5 mg 3.7 15 mg 4.4 Topiramate23 BMI 30–45 46 mg 3.4 92 mg 4.7 Metformin25 No diabetes, BMI ≥ 24 (≥ 22 in Asian Americans), elevated fasting 850 mg BID 2.3 glucose or impaired glucose tolerance Pramlintide26 No diabetes, BMI 30–50 120 μg TID 5.6 360 μg TID 6.8 Canaglifl ozin27 No diabetes, BMI 27–50 50 mg 0.9 100 mg 1.6 300 mg 1.4 Semaglutide10 No diabetes, BMI ≥ 30, or ≥ 27 with at least 1 obesity-associated 2.4 mg/week 12.4 comorbidity Semaglutide11 Diabetes, BMI ≥ 27 2.4 mg/week 6.2

BID = twice a day; ER = extended release; TID = three times a day; TBWL = total body weight loss

CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 445 ANTIOBESITY DRUG THERAPY

the trials, and all studies included lifestyle modi- glutide and semaglutide), SGLT-2 inhibitors, fi cations in addition to pharmacotherapy. and pramlintide are sometimes prescribed for weight loss in patients without diabetes or an- ■ IN PATIENTS WITH DIABETES other FDA-approved indication. This is not Weight loss can help patients improve glyce- our practice. mic control, and certain diabetes drugs have Orlistat and naltrexone-bupropion can the added benefi t of helping patients lose also be considered in this group since they have been shown to improve insulin sensitivity. weight. GLP-1 receptor agonists and SGLT-2 ■ IN PATIENTS WITH CARDIOVASCULAR inhibitors are great choices for patients with DISEASE high BMI and diabetes, as they lower hemo- globin A1C and carry a low risk of hypogly- Several antiobesity drugs have shown a posi- cemia. Some agents—liraglutide, semaglutide, tive impact on cardiometabolic risk factors dulaglutide, empaglifl ozin, canaglifl ozin, and such as blood pressure, waist circumference, dapaglifl ozin—offer cardiovascular benefi ts in insulin sensitivity, and lipid profi le. These these already high-risk patients. Use is limited drugs include liraglutide, semaglutide, nal- by high copays. Further, most of the GLP-1 trexone-bupropion, and orlistat. Naltrexone- drugs are injectable and have frequent adverse bupropion is contraindicated in patients with gastric effects. Semaglutide is an ideal choice uncontrolled hypertension but is a suitable because it has been shown to produce more choice for this patient group once blood pres- weight loss than others in this category, re- sure control is established. quires only weekly injection, and is the only Phentermine-topiramate has also been GLP-1 receptor agonist available in oral for- shown to improve cardiometabolic markers, mulation. but due to its stimulant effect on the heart, SGLT-2 inhibitors are usually well toler- it should be avoided in patients with known ated, with frequency of urination being the or suspected coronary artery disease, to avoid Multiple agents most commonly reported adverse effect. Uri- infarction. Lisdexamfetamine should similarly nary tract and vaginal infections are also com- be avoided in these patients. In patients at or combination mon in these patients, who are already at risk high risk for cardiovascular disease, such as agents, off-label for infection. Canaglifl ozin should likely be those with multiple metabolic comorbidities, smoking history, and family history of coro- and on-label, avoided in patients with known or suspected peripheral artery disease, as it is associated nary artery disease, a thorough history and examination should be performed before pre- are sometimes with a modest yet higher risk of amputations. scribing stimulants. required Some studies have shown canaglifl ozin to be for clinically associated with a higher incidence of frac- ■ ONCE A DRUG IS CHOSEN, tures, making other SGLT-2 inhibitors poten- FOLLOW-UP IS ESSENTIAL signifi cant tially more suitable for those with osteoporosis weight loss or frequent falls. Once a medication is chosen, patients should Metformin is a more affordable option be evaluated for weight loss and adverse ef- that promotes weight loss, lowers hemoglobin fects at least monthly for the fi rst 3 months, A1C, and has low potential for hypoglycemia. then at least every 3 months. If the medica- However, metformin has not been shown to tion is effective (≥ 5% total body weight loss provide additional cardiovascular benefi ts. at 3 months), safe, and tolerable, it should be 7,30 Metformin should be avoided in patients with continued indefi nitely. Multiple or com- severe chronic kidney disease (glomerular fi l- bination agents, off-label and on-label, are tration rate < 30 mL/min/1.73 m2). sometimes required for clinically signifi cant Pramlintide is FDA-approved for type 1 weight loss. and type 2 diabetes, but its use is limited by ■ multiple daily injections and a tendency to BEYOND DRUG THERAPY cause hypoglycemia. All patients suffering from obesity should be GLP-1 receptor analogues (other than lira- counseled to adopt a healthful low-calorie

446 CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 MAUER AND COLLEAGUES

diet, exercise regularly, get adequate sleep, and pharmacotherapy.32 Weight loss surgery also manage stress. achieves superior glycemic control and reduc- The clinician should also eliminate or re- tion of cardiovascular risk factors in patients place the patient’s current obesity-inducing with diabetes and obesity.30 However, despite medications with more favorable ones when- its effectiveness and safety, only 0.5% of eli- ever appropriate. For example, the clinician gible patients receive this treatment.33 The may consider bupropion instead of paroxetine, reasons behind this underutilization are not amitriptyline, nortriptyline, or mirtazapine in fully understood but likely include an overes- patients with depression or anxiety. Second- timation of the surgical risks and an underes- generation antihistamines can be prescribed timation of the potential benefi ts by primary instead of fi rst-generation ones. providers, who in turn fail to make an initial Beta-blocker use should be limited to those recommendation.33 with an indication (such as an arrhythmia). Another potential barrier is a lack of When beta-blockers are indicated, carvedilol and knowledge by primary providers regarding bar- nebivolol are associated with less weight gain.1,7 iatric surgery options and qualifying patient For contraception, oral contraceptive pills characteristics. Lastly, underutilization may and intrauterine devices should be considered also be due to variable coverage for bariatric over medroxyprogesterone acetate.1 surgery across insurances. In treating patients with diabetes, met- formin, pramlintide, SGLT-2 inhibitors, and ■ TAKE-HOME POINTS GLP-1 receptor agonists should be considered, • Obesity, a chronic disease with devastating as they promote weight loss. Thiazolidinedio- health consequences, is exceedingly prevalent nes, sulfonylureas, and insulin can cause pa- and affects certain groups disproportionately, tients to gain weight, and thus should be lim- including women, non-Hispanic Blacks, and ited to those with specifi c indications (such as people with lower education and income. insulin for type 1 diabetes mellitus), those un- • The prevalence of obesity has increased in the able to tolerate or afford preferred medications, past 20 years, and it is expected to worsen. Weight loss or those whose blood glucose remains uncon- • Obesity is multifactorial and thus diffi cult trolled. Basal insulin is more favorable than to treat with lifestyle modifi cations alone. surgery can 1,7 preprandial or biphasic insulin. Analogue in- • Pharmacotherapy, as part of a comprehen- 31 result in sulins are recommended over human insulin. sive plan, can help patients achieve mean- When treating patients with chronic in- ingful and lasting obesity control. It can superior fl ammatory diseases, disease-modifying anti- also provide independent health benefi ts, glycemic rheumatics and biologics are preferred over including decreased cardiovascular risk. control and steroids in patients who tolerate them and are • Despite its benefi ts, antiobesity drug thera- able to afford them. py is signifi cantly underutilized. reduction of Bariatric weight loss surgery is indicated • Choosing a drug for weight loss should be cardiovascular in patients who have not achieved a healthy an individualized, shared decision-making weight though lifestyle changes and who meet process that accounts for patients’ prefer- risk factors 1,30 one of the following criteria : ences, comorbidities, and out-of-pocket 2 in patients • BMI 40 kg/m or greater expenses. 2 with diabetes • BMI 35 kg/m or greater, with obesity-asso- • A comprehensive antiobesity management ciated complications plan should also include lifestyle counsel- and obesity 2 • BMI 30 kg/m or greater in patients with ing, discontinuation of obesity-inducing diabetes. medications when possible, and weight A substantial body of evidence has demon- loss surgery, when indicated. ■ strated that weight loss surgery is more effec- tive in promoting long-term weight loss and ■ DISCLOSURES in improving associated comorbid conditions The authors report no relevant fi nancial relationships which, in the context of than intensive lifestyle modifi cations and their contributions, could be perceived as a potential confl ict of interest.

CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 447 ANTIOBESITY DRUG THERAPY

■ REFERENCES modifi cation: the COR-BMOD trial. Obesity (Silver Spring, MD) 2011; 19(1):110–120. doi:10.1038/oby. 2010.147 1. Velazquez A, Apovian CM. Updates on obesity pharmacotherapy. 19. Davidson MH, Hauptman J, DiGirolamo M, et al. Weight control Ann N Y Acad Sci 2018; 1411(1):106–119. doi:10.1111/nyas.13542 and risk factor reduction in obese subjects treated for 2 years with 2. United States Census Bureau. Quick Facts United States. Accessed orlistat: a randomized controlled trial. JAMA 1999; 281(3):235–242. March 25, 2021. www.census.gov/quickfacts/fact/table/US/PST045219 doi:10.1001/jama. 281.3.235 3. Jastreboff AM, Kotz CM, Kahan S, Kelly AS, Heymsfi eld SB. Obesity 20. Hauptman J, Lucas C, Boldrin MN, Collins H, Segal KR. Orlistat in the as a disease: the Obesity Society 2018 position statement. Obesity long-term treatment of obesity in primary care settings. Arch Fam 2019; 27(1):7–9. doi:10.1002/oby.22378 Med 2000; 9(2):160–167. doi:10.1001/archfami.9.2.160 4. Hales CM, Carroll MD, Fryar CD, Ogden CL. Prevalence of obesity 21. Hollander PA, Elbein SC, Hirsch IB, et al. Role of orlistat in the and severe obesity among adults: United States, 2017–2018. NCHS treatment of obese patients with type 2 diabetes. A 1-year random- Data Brief 2020 Feb;(360):1–8. https://www.cdc.gov/nchs/data/data- ized double-blind study. Diabetes Care 1998; 21(8):1288–1294. briefs/db360-h.pdf doi:10.2337/diacare.21.8.1288 5. Bersoux S, Byun TH, Chaliki SS, Poole KG. Pharmacotherapy for 22. Allison DB, Gadde KM, Garvey WT, et al. Controlled-release phen- obesity: what you need to know. Cleve Clin J Med 2017; 84(12): termine/topiramate in severely obese adults: a randomized con- 951–958. doi:10.3949/ccjm.84a.1609 trolled trial (EQUIP). Obesity (Silver Spring, MD) 2012; 20(2):330–342. 6. Ogden CL, Fakhouri TH, Carroll MD, et al. Prevalence of obesity doi:10.1038/oby.2011.330 among adults, by household income and education — United States, 23. Gadde KM, Allison DB, Ryan DH, et al. Effects of low-dose, 2011–2014. MMWR Morb Mortal Wkly Rep 2017; 66(50):1369–1373. controlled-release, phentermine plus topiramate combination on doi:10.15585/mmwr.mm6650a1 weight and associated comorbidities in overweight and obese adults 7. Apovian CM, Aronne LJ, Bessenen DH, et al, Endocrine Society. (CONQUER): a randomised, placebo-controlled, phase 3 trial. Lancet Pharmacological management of obesity: an Endocrine Society clini- 2011; 377(9774):1341–1352. doi:10.1016/S0140-6736(11)60205-5 cal practice guideline. J Clin Endocrinol Metab 2015; 100(2):342–362. Erratum in: Lancet 2011; 377(9776):1494. doi:10.1210/jc.2014-3415 24. Garvey WT, Ryan DH, Look M, et al. Two-year sustained weight loss 8. Srivastava G, Apovian CM. Current pharmacotherapy for obesity. and metabolic benefi ts with controlled-release phentermine/topi- Nature Reviews Endocrinol 2018; 14(1):12–24. doi:10.1038/nrendo.2017.122 ramate in obese and overweight adults (SEQUEL): a randomized, 9. Crawford AR, Alamuddin N, Amaro A. Cardiometabolic effects of placebo-controlled, phase 3 extension study. Am J Clin Nutr 2012; anti-obesity pharmacotherapy. Curr Atheroscler Rep 2018; 20(4):18. 95(2):297–308. doi:10.3945/ajcn.111.024927 doi:10.1007/s11883-018-0719-9 25. Aronne LJ, Wadden TA, Peterson C, Winslow D, Odeh S, Gadde 10. Wilding JPH, Batterham RL, Calanna S. Once-weekly semaglutide in KM. Evaluation of phentermine and topiramate versus phenter- adults with overweight or obesity. N Engl J Med 2021; 384(11):989. mine/topiramate extended-release in obese adults. Obesity 2013; doi: 10.1056/NEJMoa2032183 21(11):2163–2171. doi:10.1002/oby.20584 11. Davies M, Færch L, Jeppesen OK. Semaglutide 2·4 mg once a week 26. McElroy SL, Hudson JI, Mitchell JE, et al. Effi cacy and safety of in adults with overweight or obesity, and type 2 diabetes (STEP 2): lisdexamfetamine for treatment of adults with moderate to severe a randomised, double-blind, double-dummy, placebo-controlled, binge-eating disorder: a randomized clinical trial. JAMA Psychiatry phase 3 trial. Lancet 2021;397(10278): 971-984. 2015; 72(3):235–246. doi:10.1001/jamapsychiatry.2014.2162 doi: 10.1016/S0140-6736(21)00213-0 27. Diabetes Prevention Program Research Group 1. Long-term safety, 12. Davies MJ, Bergenstal R, Bode B, et al. Effi cacy of liraglutide tolerability, and weight loss associated with metformin in the for weight loss among patients with type 2 diabetes: the SCALE Diabetes Prevention Program Outcomes Study. Diabetes Care 2012; diabetes randomized clinical trial. JAMA 2015; 314(7):687–699. 35(4):731–737. doi:10.2337/dc11-1299 doi:10.1001/jama.2015.9676 28. Smith SR, Aronne LJ, Burns CM, et al. Sustained weight loss fol- 13. Pi-Sunyer X, Astrump A, Fujioka K, et al; SCALE Obesity and Predia- lowing 12-month pramlintide treatment as an adjunct to lifestyle betes NN8022-1839 Study Group. A randomized, controlled trial of intervention in obesity. Diabetes Care 2008; 31(9):1816–1823. 3.0 mg of liraglutide in weight management. N Engl J Med 2015; doi:10.2337/dc08-0029 373(1):11–22. doi:10.1056/NEJMoa141189 29. Bays HE, Weinstein R, Law G, Canovatchel W. Canaglifl ozin: effects 14. Wadden TA, Hollander P, Klein S, et al. Weight maintenance and ad- in overweight and obese subjects without diabetes mellitus. Obesity ditional weight loss with liraglutide after low-calorie-diet-induced (Silver Spring, MD) 2014; 22(4):1042–1049. doi:10.1002/oby.20663 weight loss: the SCALE Maintenance randomized study. Int J Obes 30. American Diabetes Association. 8. Obesity management for the (Lond) 2013; 37(11):1443–1451. doi:10.1038/ijo.2013.120 treatment of type 2 diabetes: standards of medical care in diabe- 15. Hollander P, Gupta AK, Plodkowski R, et al; COR-Diabetes Study tes—2020. Diabetes Care 2020; 43(suppl 1):S89–S97. Group. Effects of naltrexone sustained-release/bupropion sus- doi:10.2337/dc20-S008 tained-release combination therapy on body weight and glycemic 31. American Diabetes Association. 9. Pharmacologic approaches to parameters in overweight and obese patients with type 2 diabetes. glycemic treatment: standards of medical care in diabetes—2020. Diabetes Care 2013; 36(12):4022–4029. doi:10.2337/dc13-0234. Diabetes Care 2020; 43(suppl 1): S98–S110. doi:10.2337/dc20-S009 Erratum in: Diabetes Care 2014; 37(2):587. 32. Sudlow AC, le Roux CW, Pournaras DJ. Review of advances in anti- 16. Greenway FL, Fujioka K, Plodkowski RA, et al. Effect of naltrexone obesity pharmacotherapy: implications for a multimodal treatment plus bupropion on weight loss in overweight and obese adults approach with metabolic surgery. Obes Surg 2019; 29(12):4095– (COR-I): a multicenter, randomised, double-blind, placebo-con- 4104. doi:10.1007/s11695-019-04206-7 trolled, phase 3 trial. Lancet 2010; 376(9741):595–605. 33. Campos GM, Khoraki J, Browning MG, Pessoa BM, Mazzini doi:10.1016/S0140-6736(10)60888-4 GS, Wolfe L. Changes in utilization of bariatric surgery in the 17. Apovian CM, Aronne L, Rubino D, et al; COR-II Study Group. A United States from 1993 to 2016. Ann Surg 2020; 271(2):201–209. randomized, phase 3 trial of naltrexone SR/bupropion SR on weight doi:10.1097/SLA.0000000000003554 and obesity-related risk factors (COR-II). Obesity (Silver Spring, MD) 2013; 21(5):935–943. doi:10.1002/oby.20309 Address: Yael Mauer, MD, MPH, Cleveland Clinic Beachwood Family 18. Wadden TA, Foreyt JP, Foster GD, et al. Weight loss with naltrexone Health and Surgery Center, 26900 Cedar Road, Beachwood, OH 44122; SR/bupropion SR combination therapy as an adjunct to behavior [email protected]

448 CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 REVIEW Asad Khan, MD Erik H. Van Iterson, PhD Luke J. Laffi n, MD Department of Internal Medicine, Cleveland Director, Cardiac Rehabilitation, Co-director, Center for Blood Pressure Clinic, Cleveland, OH; Clinical Instructor, Section of Preventive Cardiology Disorders, Section of Preventive Cardiology Cleveland Clinic Lerner College of Medicine and Rehabilitation, Heart, Vascular, and Rehabilitation, Heart, Vascular, of Case Western Reserve University, and Thoracic Institute, Cleveland Clinic, and Thoracic Institute, Cleveland Clinic, Cleveland, OH Cleveland, OH Cleveland, OH

The obesity paradox in heart failure: What is the role of cardiorespiratory fi tness?

ABSTRACT besity is a well-established and impor- O tant predictor of morbidity and mortality The obesity paradox describes a survival benefi t for in patients with cardiovascular (CV) disease higher body mass index in patients with heart failure. and other conditions, including chronic kid- But other factors like cardiorespiratory fi tness may play a ney disease and chronic obstructive pulmo- role in heart failure development, severity, and survival. nary disease. Yet some studies report obesity Although more research is needed to better understand is associated with lower mortality in patients the relationships between body mass index and fi tness with heart failure—a fi nding known as the in patients with heart failure, evidence indicates that obesity paradox. recommending weight loss and an exercise program is Though not fully understood, several pos- appropriate for most patients. sible reasons for the obesity paradox have been proposed (Table 1).1–9 KEY POINTS Understanding the obesity paradox has im- Obesity increases the risk of developing heart failure portant clinical implications given the high regardless of fi tness level, but better fi tness attenuates prevalence of obesity in patients with heart the risk. failure (42% of those with preserved ejec- tion fraction [HFpEF] and 36% of those with reduced ejection fraction [HFrEF]).10 What Weight appears to be only part of the obesity paradox should patients be advised about weight man- story. Evidence indicates that cardiorespiratory fi tness is a agement? What should patients be advised major factor infl uencing the paradox. about cardiorespiratory fi tness, a major factor infl uencing the paradox? Fitness modifi es the obesity paradox in patients with This review summarizes current under- heart failure and reduced ejection fraction, with the para- standing of the roles of cardiorespiratory fi t- dox remaining strongest in patients who are less fi t. ness and body mass index (BMI) in patients with heart failure and its development. It also Although more research is needed on risk reduction for discusses how to advise patients about fi tness and body mass in light of the obesity paradox. heart failure, evidence indicates that intentional weight loss and increased fi tness are advisable for select pa- ■ BENEFIT OF FITNESS tients. IN CARDIOVASCULAR DISEASE The effect of cardiorespiratory fi tness on CV outcomes is an active area of clinical research. The standard for measuring cardiorespiratory fi t- doi:10.3949/ccjm.88a.20098 ness is cardiopulmonary exercise testing, using

CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 449 OBESITY PARADOX IN HEART FAILURE

TABLE 1 and risk factors for cardiovascular disease de- velopment, with follow-up for about 10 years. Select theoretical mechanisms Cardiorespiratory fi tness was determined us- of the obesity paradox ing maximal treadmill exercise testing with age-based metabolic equivalent (MET) val- Greater metabolic reserves ues for fi tness levels. Participants also were Less cardiac cachexia stratifi ed by BMI using standard thresholds for Increased concentration of tumor necrosis factor normal weight (18.5–24.9 kg/m2), overweight receptors (25.0–29.9 kg/m2), and obesity (> 30 kg/m2).14 Earlier presentation owing to greater functional Results showed that cardiovascular disease impairment mortality increased with increasing BMI levels.14 Expectedly, the lowest risk for cardiovascular Attenuated response to renin-angiotensin-aldos- terone system disease mortality was a combination of normal weight and high fi tness. However, the relative Higher blood pressure leading to greater use of risk of cardiovascular disease mortality in the cardioprotective medications obese high-fi tness cohort was half that in the low- Adapted from reference 6. fi tness normal-weight cohort, suggesting that fi t- ness is a more important predictor of cardiovascu- an incremental treadmill or upright cycle proto- lar disease mortality than body weight. The effect col. Numerous studies have found associations of low cardiorespiratory fi tness on cardiovascular between poor CV disease outcomes and low disease mortality was also higher than the pres- 11,12 peak exercise oxygen uptake (peak Vo2). ence of diabetes, dyslipidemia, hypertension, or current smoking across all BMI levels.14 Low fi tness predicts poor outcomes In 1996, Blair et al13 were among the fi rst to ■ HEART FAILURE DEVELOPMENT: quantify the effects of cardiorespiratory fi tness CARDIORESPIRATORY FITNESS AND BMI on cardiovascular disease outcomes. After fol- lowing 25,341 men and 7,080 women in a pre- The mechanisms related to obesity that con- Low fi tness tribute to the development of HFpEF and ventive medicine clinic for about 9 years, they is a strong HFrEF include hemodynamic alterations that found that low fi tness was independently associ- may predispose the patient to changes in car- independent ated with increased all-cause mortality in both diac morphology and ventricular function.10 predictor of men (relative risk [RR] 1.52, 95% confi dence interval [CI] 1.28–1.82) and women (RR 2.10, Possible mechanisms cardiovascular 95% CI 1.36–3.21). Low fi tness was associated The mechanisms related to low cardio- disease with statistically signifi cant increased cardio- respiratory fi tness that contribute to the de- mortality vascular disease mortality risk in men (RR 1.70, velopment of heart failure are not well un- 95% CI 1.28–2.25), although the difference was derstood. Low cardiorespiratory fi tness may not statistically signifi cant in women. In both indirectly affect development of cardiovascu- sexes, low fi tness was a more signifi cant prog- lar risk factors (ie, reduced cardiorespiratory nostic factor than other traditional cardiac risk fi tness is associated with a low level of physical factors. Interestingly, elevated BMI (> 27 kg/m2) activity),15 which may accelerate the develop- was not found to be signifi cantly associated with ment of heart failure risk factors including dia- increased mortality in either sex. betes, hypertension, and coronary artery dis- ease. Alternatively, cardiovascular symptoms Fitness may be more important than weight such as angina or dyspnea on exertion may A 1999 prospective observational study by limit habitual physical activity, in turn lead- Wei et al14 also found that low cardiorespirato- ing to reduced cardiorespiratory fi tness. ry fi tness is a strong independent predictor of Even in the absence of traditional cardio- cardiovascular disease mortality in the general vascular disease risk factors, studies demon- population, and perhaps more so than BMI. strate that sedentary aging leads to increased The study assessed nearly 26,000 men for car- stiffness of the left ventricular myocardium, a diorespiratory fi tness, cardiovascular disease, potential substrate for heart failure.16 Higher

450 CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 KHAN AND COLLEAGUES

TABLE 2 Studies assessing BMI and cardiorespiratory fi tness: Effect on heart failure development

Study N Designa End point Main fi ndings Pandey et al19 19,485 Patients stratifi ed by Long-term risk Higher midlife BMI was signifi cantly associated with Cooper Center BMI and peak METs of hospitaliza- greater risk of hospitalization for HF in older age. This Longitudinal into quintiles tion for HF association was attenuated after adjusting for cardio- Study respiratory fi tness. Kenchaiah 21,094 Patients stratifi ed by New onset HF Compared with lean participants, overweight and et al20 BMI and vigorous obese participants had increased HF risk. Vigorous Physicians’ physical activity physical activity conferred decreased HF risk. No Health Study interaction was found between BMI, vigorous physical activity, and HF risk. Hu et al21 59,178 Patients stratifi ed by New onset HF Higher BMI, waist circumference, or waist-to-hip ratio physical activity and was associated with increased HF incidence in men indicators of adipos- and women. The protective effect of physical activity ity (eg, BMI, waist on HF risk was consistent in participants at all levels circumference, waist- of BMI. to-hip ratio) Kokkinos et al22 20,254 Patients stratifi ed by New onset HF Increased cardiorespiratory fi tness was associated BMI and cardio- with progressively lower HF risk regardless of BMI. respiratory fi tness in After adjusting for fi tness, BMI was not a signifi cant quartiles predictor of HF risk.

Pandey et al23 5,109 Patients stratifi ed by New onset HF High cardiorespiratory fi tness was associated with Look AHEAD (with BMI and cardio- lower risk of developing HFpEF. Sustained long-term trial DM) respiratory fi tness into improvement in fi tness was associated with lower risk tertiles of HF after 4 years.

a All studies are retrospective.

AHEAD = Action for Health in Diabetes; BMI = body mass index; DM = diabetes mellitus; HF = heart failure; HFpEF = heart failure with preserved ejection fraction; METs = metabolic equivalents levels of physical activity are associated with nearly 20,000 participants by standard BMI benefi cial effects on cardiovascular measures, thresholds and cardiorespiratory fi tness levels including improved early diastolic fi lling time (low, moderate, and high as determined by and favorable cardiac remodeling.17 In addi- calculated METs achieved with treadmill ex- tion, an animal study showed a direct and fa- ercise testing). A higher BMI during midlife vorable effect of exercise training on cardiac was associated with a signifi cantly greater structure and function, leading to a delayed risk of heart failure hospitalization in older 18 onset of heart failure. patients (age 65 and older), even after ad- Study comparing fi tness and BMI justing for other established heart failure risk The combined impact of cardiorespiratory fi t- factors. When adjusted for cardiorespiratory ness and BMI on heart failure development is fi tness, this association was attenuated, such gaining increasing attention, and many stud- that cardiorespiratory fi tness accounted for ies have been conducted (Table 2).19–23 47% of the heart failure risk associated with Data from the Cooper Center Longitudi- BMI. Furthermore, the BMI-associated risk of nal study19 indicated that cardiorespiratory fi t- hospitalization for heart failure was more pro- ness may be at least as important as BMI for nounced in participants who had low fi tness developing heart failure. The study stratifi ed or were moderately fi t.

CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 451 OBESITY PARADOX IN HEART FAILURE

A subgroup of about 9,000 participants style modifi cation did not lower the risk of heart underwent repeat measurements of cardiore- failure more than diabetes support and educa- spiratory fi tness and BMI at a median follow- tion groups (HR 0.96, 95% CI 0.75–1.23). up of 4.2 years. Increased cardiorespiratory fi t- However, a pooled multivariate analysis ness, but not BMI, was signifi cantly associated found a statistically signifi cant, graded, inverse with decreased risk of heart failure hospitaliza- association between baseline cardiorespiratory tion in older patients (hazard ratio [HR] 0.91, fi tness and heart failure incidence in partici- 95% CI 0.84–0.98 per 1 MET increase).19 pants who were moderately or highly fi t. Inter- Data from the Physicians’ Health Study estingly, this association was only observed for showed that participation in self-reported vig- heart failure with preserved but not reduced orous activity (defi ned as “working up a sweat”) ejection fraction. Also, the association of BMI 1 to 3 times a month conferred a 26% decrease with heart failure was not stastistically signifi - in new-onset heart failure development.20 In cant after adjusting for baseline cardiorespira- contrast, a 1-kg/m2 increase in BMI increased tory fi tness and traditional risk factors. In a the risk of heart failure by 13%. Adjusting for subset of patients who underwent repeat as- vigorous physical activity did not alter the risk sessment of cardiorespiratory fi tness and BMI of heart failure associated with elevated BMI. at 1 and 4 years, there was a statistically sig- Hu et al21 studied the relationship between nifi cant association between improved fi tness physical activity, heart failure risk, and indi- and lower risk of overall heart failure at 4 years cators of adiposity (ie, BMI, waist circumfer- (HR 0.86, 95% CI 0.79–0.94).23 ence, and waist-to-hip ratio) in nearly 60,000 More information needed on women Finnish participants who were free of heart and type of heart failure failure at enrollment. During a mean follow- Other than in a study by Hu et al,21 which in- up of 18.4 years, the risk of developing heart cluded comparable numbers of men (1,921) failure directly increased with BMI and other and women (1,693), women are vastly under- measures of adiposity for men and women. represented in the studies. The Physicians’ Moderate or high levels of physical activity 20 Cardio- Health Study consisted entirely of men, and were associated with a reduced risk of heart in the Cooper Center Longitudinal Study,19 respiratory failure in both sexes at all levels of BMI and women accounted for less than 10% of par- fi tness may be waist-to-hip ratio. ticipants in the overweight category and less In a study published in 2019, Kokkinos et than 11% in the obese category. Given the at least as al22 stratifi ed 20,000 US men by standard BMI known differences between men and women, important thresholds and cardiorespiratory fi tness. Fitness especially body fat distribution, more studies thresholds were based on quartiles following as BMI for that include women are essential. age and sex-specifi c MET adjustments. After a Another criticism is that only the Look developing mean follow-up of 13.4 years, they found that AHEAD trial23 determined the risk for spe- heart failure heart failure risk increased progressively with cifi c heart failure phenotypes (ie, HFrEF vs decreasing fi tness in each BMI category. Al- HFpEF). In most studies, the primary outcome though age, BMI, and cardiorespiratory fi tness was defi ned by a combination of International were strong independent predictors of heart Classifi cation of Diseases codes, limiting over- failure risk, the association between BMI and all interpretation. heart failure risk was no longer statistically sig- nifi cant after adjusting for fi tness. Each increase Bottom line of 1 MET was associated with a 16% lower risk Despite limitations, these studies, taken as a of heart failure (HR 0.84; 95% CI 0.83–0.86; P whole, have two important implications for < .001). heart failure prevention: • BMI and cardiorespiratory fi tness both af- Clues from patients with diabetes fect heart failure development, but fi tness A recent post hoc analysis of the Look AHEAD is likely the more signifi cant factor (Action for Health in Diabetes) trial23 also ex- • Increased fi tness is associated with a re- amined the impact of fi tness and BMI on heart duced risk of heart failure hospitalization failure development. It found that intensive life- as one ages.

452 CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 KHAN AND COLLEAGUES

TABLE 3 Studies assessing BMI and cardiorespiratory fi tness: Effect on heart failure prognosis Study Na Average LVEF of Designb End point Main fi ndings target groups Lavie et al24 2,066 High fi t = 30.1% Patients stratifi ed Overall In patients with low cardiorespiratory fi tness, Low fi t = 26.0% by BMI and peak mortality BMI ≥ 30 kg/m² was a significant predictor of

VO2 better survival. No obesity paradox seen at the high fi tness level. Clark et al7 1,675 High fi t = 23.4% Patients stratifi ed Death, urgent BMI of obesity class was associated with a Low fi t = 23.2% by BMI and peak status 1A signifi cantly lower risk of death, urgent trans-

VO2 heart trans- plant, or device placement than with normal plant, or VAD BMI in the group with low peak VO2. In the placement. high peak VO2 group, no difference was seen for BMI and survival. 2 Piepoli et 4,623 BMI (kg/m ) Patients stratifi ed All-cause Higher BMI and peak VO2 were signifi cant al27 MECKI < 25 = 31% by BMI and peak mortality and positive predictors of longer survival. When

Score 25 to 30 = 33% VO2 CV death patients in a BMI category were matched Research > 30 to ≤ 35 = 33% according to age, sex, LVEF, and peak VO2, the Group > 35 = 33% protective role of BMI disappeared. McAuley 774 High fi t = 41% Patients stratifi ed Overall Signifi cant positive association between BMI et al26 FIT Low fi t = 40% by BMI and peak mortality category and survival for exercise capacity Project METs < 4 METs, but not ≥ 4 METs. a All patients had established heart failure. b All studies were retrospective.

BMI = body mass index; CV = cardiovascular; FIT = Henry Ford Exercise Testing; LVEF = left ventricular ejection fraction; MECKI = Metabolic Exercise test data com-

bined with Cardiac and Kidney Indexes; METs = metabolic equivalents; VAD = ventricular assist device; VO2 = exercise oxygen uptake

■ HEART FAILURE PROGNOSIS: ing. After a mean follow-up of 10 years, the CARDIORESPIRATORY FITNESS AND BMI authors concluded that the higher the BMI, Studies have been conducted in patients with the lower the mortality in those with a low heart failure to determine the impacts of fi t- level of fi tness, but not in those with a high ness and BMI, and whether fi tness affects the level of fi tness. Thus, exercise capacity should obesity paradox (Table 3).7,24–27 be considered when stratifying risk. Clark et al7 also found that higher fi tness HFrEF: Higher fi tness may negate levels likely mitigate the obesity paradox in pa- the obesity paradox tients with heart failure. They assessed almost In the MECKI Score Research Group study,27 2,000 patients referred for heart transplant 4,623 patients with HFrEF underwent maxi- evaluation. Participants were stratifi ed by BMI mum cardiopulmonary exercise testing at en- and fi tness, as determined by cardiopulmonary rollment and were followed for a median of 3 exercise testing. After 2 years of follow-up, a years. The population was divided according 2 high BMI (≥ 30 kg/m ) was a signifi cant pre- to BMI and peak Vo . On univariate analy- dictor of improved survival in the low-fi tness 2 sis, groups with higher BMI and peak Vo2 had group but not in the high-fi tness group. lower mortality. However, when groups were The Henry Ford Exercise Testing (FIT) matched for age, sex, left ventricular ejection Project26 followed nearly 800 participants with 2 fraction (LVEF), and predicted peak Vo2, the heart failure and a BMI of at least 18.5 kg/m . protective role of BMI disappeared. Participants were grouped into standard BMI categories and then stratifi ed by fi tness (< 4 Fitness: An obesity paradox modifi er or ≥ 4 METs) based on treadmill stress test- The above studies support an obesity paradox-

CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 453 OBESITY PARADOX IN HEART FAILURE

cardiorespiratory fi tness dichotomy in estab- underlying fi tness as an obesity-paradox modi- lished heart failure: obesity is predominantly fi er in these patients are similar to those with protective in patients with low fi tness but not reduced ejection fraction, but that is not well in highly fi t patients. Hence, high fi tness can be established. It is unclear if benefi cial interven- thought of as a modifi er of the obesity paradox. tions in one group are relevant to the other. A strength of the data is the wide range in the mean age of each low-fi tness obese cohort Obesity defi nitions vary by study 8,27 Most of the above studies defi ned obesity (50.8–63 years), indicating that the protec- 2 tive effect of obesity is not limited to younger broadly as a BMI greater than 30 kg/m , limit- patients. Studies also have included a range ing the generalizability of conclusions. Only 27 of mean LVEF (23.6%–40%),2,7,27 suggesting the MECKI Score study subdivided patients that cardiorespiratory fi tness is likely an obe- based on obesity classes. Certain BMI thresh- sity paradox modifi er in patients with reduced olds may exist for which protective effects of LVEF (< 40%), mid-range LVEF (40%–50%), obesity become deleterious. and preserved LVEF (> 50%). ■ IMPACT OF WEIGHT LOSS HFpEF: Does the obesity paradox hold? A meta-analysis by Mahajan et al4 found that The obesity paradox is not as consistently re- weight loss induced by bariatric surgery re- ported for heart failure patients with preserved sulted in signifi cantly improved measures of ejection fraction as it is for those with reduced cardiac function and morphology (diastolic ejection fraction. Aerobic exercise capacity function, left ventricular mass index, and left has been examined in patients with preserved atrial size). However, clinical outcomes (eg, ejection fraction in relation to indices of obe- heart failure incidence) were not assessed. sity and adiposity.28,29 In those trials, BMI Furthermore, patients did not have a diagno- predicted lower exercise capacity but did not sis of heart failure at baseline, so the effect of correlate with cardiac-specifi c functional and bariatric surgery in established heart failure prognostic parameters, including measures of was uncertain. left ventricular function. Vigorous Other studies have not found improved A retrospective analysis of the Treatment cardiac function with weight loss. Kitzman activity 1 to 3 of Preserved Cardiac Function Heart Failure et al34 found that left ventricular mass and With an Aldosterone Antagonist (TOPCAT) times a month relative wall thickness decreased after diet-in- trial30 indicated that the obesity paradox may duced weight loss, but resting cardiac function conferred not hold for HFpEF. It found that a higher did not improve. baseline level of physical activity was associ- a 26% decrease A Swedish registry study with nearly 40,000 ated with lower risk of adverse cardiovascular in new-onset participants without heart failure at baseline events through the duration of the trial (me- evaluated the effects of weight loss from either heart failure dian follow-up 2.4 years), independent of BMI intensive lifestyle intervention or bariatric sur- development and other risk factors. 35 It is possible that the apparent lack of an gery. Baseline weight and BMI did not differ obesity paradox in HFpEF may be because between the cohorts. Surgery led to 18.8 kg more obesity itself is a risk factor for HFpEF. Also, weight loss than lifestyle interventions at 1-year patients with HFpEF and obesity are more follow-up and 22.6 kg more at 2 years. After a likely to have other cardiovascular risk factors median follow-up of 4.1 years, surgery was asso- such as hypertension, diabetes, and obstruc- ciated with lower heart failure incidence than tive sleep apnea that may attenuate any pro- lifestyle modifi cation (4.1% vs 7.6% per 10,000 tective effect of obesity.31 person-years; HR 0.54, 95% CI 0.26–0.81). A 10-kg weight loss from both cohorts combined What about heart failure with mid-range resulted in decreased heart failure incidence ejection fraction? (HR 0.77, 95% CI 0.60–0.97). Heart failure with mid-range ejection fraction Bariatric surgery may also help mitigate (LVEF 40%–50%) is a more recently charac- established heart failure. In a population- terized group that is not well defi ned or under- based study,36 524 patients with heart failure stood.32,33 It is possible that the mechanisms were followed after bariatric surgery, with a

454 CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 KHAN AND COLLEAGUES

composite of emergency department visits or fect of lean mass on cardiorespiratory fi tness. hospitalizations for heart failure exacerbation Lean mass is used as a surrogate for skeletal as the primary outcome measure. In the 13 to muscle mass, which is independently associ- 24 months after surgery, heart failure exacer- ated with cardiorespiratory fi tness, possibly via bations were signifi cantly reduced (odds ratio endothelial and mitochondrial dysfunction 0.57, 95% CI 0.39–0.82). There were 184 and respiratory muscle abnormalities.28,38 In a heart failure events (43% systolic and 57% 2017 review, reduced lean mass contributed to diastolic). No information on body weight re- impaired cardiorespiratory fi tness, independent duction was reported, so it is unclear if more of cardiac function.39 BMI reductions occur weight loss correlated with fewer events. with loss of lean mass, which may partially ac- In contrast, a study by Zamora et al37 count for the obesity paradox in heart failure.40 of 1,000 patients with ambulatory chronic Osman et al41 prospectively studied 225 HFrEF were followed for 3 years to determine consecutive ambulatory patients with chronic the impact of signifi cant weight loss (defi ned systolic heart failure who were referred for car- as more than 5% of body weight over 1 year) diopulmonary exercise testing. They found that on the mortality rate. Mortality was higher in adjusting peak Vo2 to lean mass provided great- patients who lost signifi cant weight (27.6%) er prognostic strength than adjusting by body than in patients without signifi cant weight weight, particularly in people with obesity. loss (15.3%). Among obese patients, signifi - The pattern of regional tissue deposition, cant weight loss was associated with a higher especially increased proportions of intra-abdom- risk of all-cause death (adjusted HR 2.38, 95% inal fat, may play a key role in exercise intoler- CI 1.31–4.32) than in nonobese patients (ad- ance in patients with HFpEF. Haykowsky et al42 justed HR 1.83, 95% CI 1.16–2.89). found that patients with HFpEF had higher ra- tios of intermuscular fat to skeletal muscle mass Does unintentional weight loss explain the obesity paradox? than healthy controls, and this was signifi cantly related to reduced peak Vo . This evidence sug- Intentional vs unintentional weight loss likely 2 gests that body composition indices such as lean explains the different heart failure outcomes mass play an important role in cardiorespiratory In patients with following weight loss, particularly in patients fi tness regardless of BMI. with HFrEF. established When evaluating candidates for inten- Drug-induced weight loss: heart failure, tional weight loss via bariatric surgery or life- The evidence is unclear obesity is style modifi cations, medical clearance for par- There is little evidence to demonstrate the ticipation requires a certain level of baseline safety and effi cacy of pharmacologic weight predominantly functional status. However, unintentional loss in patients with heart failure. A recent protective in weight loss in patients with advanced HFrEF post hoc analysis of the Functional Impact of may be the result of sarcopenia and cardiac GLP-1 (glucagon-like peptide-1) for Heart patients with cachexia, leading to poor baseline metabolic Failure Treatment (FIGHT) trial43 found that low fi tness reserves and adverse clinical outcomes. Thus, in patients with reduced ejection fraction, the obesity paradox may simply refl ect the there was a treatment-related 4.1-lb weight severity of heart failure, with lower BMI oc- loss for liraglutide vs placebo (95% CI −7.94 curring in end-stage heart failure and obesity, to −0.25; P < .04), but no effect was found in indicating a better baseline metabolic reserve. worsening heart failure, making the clinical Body composition is also important implications unclear. More research is needed Patients with HFpEF and obesity also have sar- to determine whether pharmacologic weight copenia and adipose infi ltration of muscle,31 in- loss is an effective strategy to improve clinical dicating a highly infl amed and catabolic state. outcomes in this patient population. This highlights one of the limitations of using ■ BMI as a surrogate of adiposity, and it demon- WHAT TO ADVISE PATIENTS? strates the need to further describe body compo- Studies support 2 major themes: sition when evaluating heart failure outcomes. • Obesity and low cardiorespiratory fi tness are More attention is being focused on the ef- risk factors for the development of heart failure

CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 455 OBESITY PARADOX IN HEART FAILURE

TABLE 4 The obesity paradox: What we know and what we don’t Setting Established study fi ndings Current limitations Research questions Patients BMI appears to be protective Different obesity classes have Is cardiorespiratory fi tness an obesity with heart predominantly in patients with low not been specifi cally evaluated. paradox modifi er in specifi c classes of failure fi tness. obesity? No separate evaluation of patients with either preserved Is cardiorespiratory fi tness an obesity or mid-range ejection fraction; paradox modifi er in HFpEF and HFmrEF? they are largely grouped with reduced ejection fraction. Heart Improving cardiorespiratory fi tness No differentiation between What type of physical activity leads to the failure may be more important for risk types or duration of physical lowest risk of heart failure development? prevention reduction than lowering BMI. activity. How do BMI and cardiorespiratory fi tness In patients with established dia- Limited specifi city of type of (and interventions) affect development of betes, improved fi tness may de- heart failure as end point (ie, different types of heart failure? crease the risk of developing HFpEF. HFpEF, HFmrEF, or HFrEF). Are fi ndings relevant for women? Increasing BMI and specifi c measures Women underrepresented. of adiposity correlate with increased risk of developing heart failure. Even small amounts of physical activity decrease risk of developing heart failure. Physical activity appears to have a dose-dependent effect on heart failure risk, with the lowest risk as- sociated with highest frequency of physical activity.

Weight Either surgical or lifestyle-based Lack of clinical outcomes data How does medical vs surgical weight loss loss weight loss may reduce morbidity after intentional weight loss affect heart failure morbidity and mortal- from heart failure. for patients with heart failure ity rates, particularly with newer medical and obesity. for obesity? Unintentional weight loss indicates acute illness and contributes to Limited data on specifi c exer- How does supervised exercise for patients poor metabolic reserve, leading to cise training programs in heart with heart failure and obesity affect fi t- worse outcomes. failure outcomes or prevention. ness, weight loss, and outcomes? BMI = body mass index; HFmrEF = heart failure with mid-range ejection fraction; HFpEF = heart failure with preserved ejection fraction; HFrEF = heart failure with reduced ejection fraction

• Obesity in people with low fi tness is pro- heart failure.44,45 A 2018 position paper from tective for those with established heart the Heart Failure Association of the European failure. Society of Cardiology advocates cardiopulmo- How can clinicians use this knowledge to nary exercise testing only for assessing the risk 46 advise patients regarding weight loss and exer- of heart failure. cise training? The answer is unclear. The most Bottom line: Advise to increase fi tness recent American and European heart failure and consider weight loss guidelines give only limited guidance on obe- Although large-scale clinical trials are needed sity management in patients with established to better assess and defi ne the risks and ben-

456 CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 KHAN AND COLLEAGUES

efi ts of weight loss in patients with heart fail- tion fraction, with each responding differently ure, particularly in those with reduced ejec- to therapeutic interventions. Grouping all pa- tion fraction, recommending moderate weight tients with heart failure together in analyses loss may be appropriate. Lifestyle interven- may blur results. Current literature has consis- tions aimed at weight loss and improving car- tent fi ndings in reduced ejection fraction, but diorespiratory fi tness—such as with a phase 2 dedicated analyses of preserved and mid-range (outpatient) cardiac rehabilitation program— ejection fraction are needed. should be considered, as studies suggest they Similarly, it is likely that there are certain reduce heart failure risk by improving fi tness BMI thresholds where the protective effects of in patients with obesity and heart failure. And obesity become deleterious, but different obe- no data suggest harm. sity classes are commonly considered together For heart failure prevention, weight loss through dietary and lifestyle changes can be in studies. Future research should examine if fi t- recommended, given that evidence shows a ness modifi es the obesity paradox in heart failure lower BMI predicts reduced risk of heart fail- when assessing individuals with class II (BMI 2 2 ure development. In patients with established 35–39.9 kg/m ) or class III (> 40 kg/m ) obesity. heart failure and reduced ejection fraction, it Other major gaps in evidence include the appears that intentional weight loss through specifi c weight reduction interventions that lifestyle modifi cation or bariatric surgery may result in better heart failure outcomes in pa- be benefi cial,35 although unintentional weight tients with obesity. Metabolic surgery has loss appears to be detrimental.36 Thus, when been studied the most. How do pharmaco- advising weight loss to obese patients with logical therapies compare? How do supervised heart failure, it is important to consider the exercise programs (particularly cardiac reha- individual’s clinical profi le. bilitation) impact risk in patients with estab- lished heart failure? Which is more important, ■ FUTURE DIRECTIONS FOR RESEARCH weight loss or increased cardiorespiratory fi t- Understanding the overlapping impact of obe- ness? Future studies should assess relative risk sity and cardiorespiratory fi tness in heart fail- reduction of specifi c exercise training com- ure is important to identify gaps in evidence bined with metabolic surgery or pharmaco- and assess future research directions (Table 4). therapy-induced weight loss in patients with A high priority for future studies is to bet- heart failure. ■ ter evaluate the impact of obesity on different heart failure phenotypes. Distinct pathophysi- ■ DISCLOSURES ologic differences exist between heart failure The authors report no relevant fi nancial relationships which, in the context with reduced, mid-range, or preserved ejec- of their contributions, could be perceived as a potential confl ict of interest. ■ REFERENCES of obesity and the obesity paradox on prevalence and prognosis in heart failure. JACC Heart Fail 2013; 1(2):93–102. doi:10.1016/j.jchf.2013.01.006 1. Elagizi A, Kachur S, Lavie CJ, et al. An overview and update on obesity and 7. Clark AL, Fonarow GC, Horwich TB. Impact of cardiorespiratory fi tness on the obesity paradox in cardiovascular diseases. Prog Cardiovasc Dis 2018; the obesity paradox in patients with systolic heart failure. Am J Cardiol 61(2):142–150. doi:10.1016/j.pcad.2018.07.003 2015; 115(2):209–213. doi:10.1016/j.amjcard.2014.10.023 2. Shah R, Gayat E, Januzzi JL Jr, et al; GREAT (Global Research on Acute 8. Van Iterson EH, Kim CH, Uithoven K, Olson TP. Obesity and hemoglobin Conditions Team Network. Body mass index and mortality in acutely de- content impact peak oxygen uptake in human heart failure. Eur J Prev compensated heart failure across the world: a global obesity paradox. J Am Cardiol 2018; 25(18):1937–1946. doi:10.1177/2047487318802695 Coll Cardiol 2014; 63(8):778–785. doi:10.1016/j.jacc.2013.09.072 9. Poirier P, Giles TD, Bray GA, et al. Obesity and cardiovascular disease: 3. Zhang J, Begley A, Jackson R, et al. Body mass index and all-cause mortality pathophysiology, evaluation, and effect of weight loss: an update of the in heart failure patients with normal and reduced ventricular ejection frac- 1997 American Heart Association Scientifi c Statement on Obesity and Heart tion: a dose-response meta-analysis. Clin Res Cardiol 2019; 108(2):119–132. Disease from the Obesity Committee of the Council on Nutrition, Physical doi:10.1007/s00392-018-1302-7 Activity, and Metabolism. Circulation 2006; 113(6):898–918. 4. Mahajan R, Stokes M, Elliott A, et al. Complex interaction of obesity, inten- doi:10.1161/CIRCULATIONAHA.106.171016 tional weight loss and heart failure: a systematic review and meta-analysis. 10. Alpert MA, Lavie CJ, Agrawal H, Aggarwal KB, Kumar SA. Obesity and Heart 2020; 106(1):58–68. doi:10.1136/heartjnl-2019-314770 heart failure: epidemiology, pathophysiology, clinical manifestations, and 5. Powell-Wiley TM, Ngwa J, Kebede S, et al. Impact of body mass index on management. Transl Res 2014; 164(4):345–356. doi:10.1016/j.trsl.2014.04.010 heart failure by race/ethnicity from the Get With the Guidelines-Heart 11. Kodama S, Saito K, Tanaka S, et al. Cardiorespiratory fi tness as a quantita- Failure (GWTG-HF) registry. JACC Heart Fail 2018; 6(3):233–242. tive predictor of all-cause mortality and cardiovascular events in healthy doi:10.1016/j.jchf.2017.11.011 men and women: a meta-analysis. JAMA 2009; 301(19):2024–2035. 6. Lavie CJ, Alpert MA, Arena R, Mehra MR, Milani RV, Ventura HO. Impact doi:10.1001/jama.2009.681

CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 457 OBESITY PARADOX IN HEART FAILURE

12. Kaminsky LA, Arena R, Ellingsen Ø, et al. Cardiorespiratory fi tness and 31. Pandey A, Patel KV, Vaduganathan M, et al. Physical Activity, fi tness, and cardiovascular disease—the past, present, and future. Prog Cardiovasc Dis obesity in heart failure with preserved ejection fraction. JACC Heart Fail 2019; 62(2):86–93. doi:10.1016/j.pcad.2019.01.002 2018; 6(12):975–982. doi:10.1016/j.jchf.2018.09.006 13. Blair SN, Kampert JB, Kohl HW 3rd, et al. Infl uences of cardiorespiratory fi t- 32. Webb J, Draper J, Fovargue L, et al. Is heart failure with mid range ejection ness and other precursors on cardiovascular disease and all-cause mortality fraction (HFmrEF) a distinct clinical entity or an overlap group? Int J Cardiol in men and women. JAMA 1996; 276(3):205–210. pmid: 866756 Heart Vasc 2018; 21:1–6. doi:10.1016/j.ijcha.2018.06.001 14. Wei M, Kampert JB, Barlow CE, et al. Relationship between low cardiorespi- 33. Branca L, Sbolli M, Metra M, Fudim M. Heart failure with mid-range ejec- ratory fi tness and mortality in normal-weight, overweight, and obese men. tion fraction: pro and cons of the new classifi cation of Heart Failure by Eu- JAMA 1999; 282(16):1547–1553. doi:10.1001/jama.282.16.1547 ropean Society of Cardiology guidelines. ESC Heart Fail 2020; 7(2):381–399. 15. Liu K, Daviglus ML, Loria CM, et al. Healthy lifestyle through young adult- doi:10.1002/ehf2.12586 hood and the presence of low cardiovascular disease risk profi le in middle 34. Kitzman DW, Brubaker P, Morgan T, et al. Effect of caloric restriction or age: the Coronary Artery Risk Development in (Young) Adults (CARDIA) aerobic exercise training on peak oxygen consumption and quality of life in study. Circulation 2012; 125(8):996–1004. obese older patients with heart failure with preserved ejection fraction: a doi:10.1161/CIRCULATIONAHA.111.060681 randomized clinical trial. JAMA 2016; 315(1):36–46. 16. Arbab-Zadeh A, Dijk E, Prasad A, et al. Effect of aging and physical activ- doi:10.1001/jama.2015.17346 ity on left ventricular compliance. Circulation 2004; 110(13):1799–1805. 35. Sundstrom J, Bruze G, Ottosson J, Marcus C, Naslund I, Neovius M. doi:10.1161/01.CIR.0000142863.71285.74 Weight loss and heart failure: a nationwide study of gastric bypass surgery 17. Brinker SK, Pandey A, Ayers CR, et al. Association of cardiorespira- versus intensive lifestyle treatment. Circulation 2017; 135(17):1577–1585. tory fi tness with left ventricular remodeling and diastolic function: the doi:10.1161/CIRCULATIONAHA.116.025629 Cooper Center Longitudinal Study. JACC Heart Fail 2014; 2(3):238–246. 36. Shimada YJ, Tsugawa Y, Brown DFM, Hasegawa K. Bariatric surgery and doi:10.1016/j.jchf.2014.01.004 emergency department visits and hospitalizations for heart failure exac- 18. Miyachi M, Yazawa H, Furukawa M, et al. Exercise training alters left erbation: population-based, self-controlled series. J Am Coll Cardiol 2016; ventricular geometry and attenuates heart failure in dahl salt-sensitive 67(8):895–903. doi:10.1016/j.jacc.2015.12.016 hypertensive rats. Hypertension 2009; 53(4):701–707. 37. Zamora E, Diez-Lopez C, Lupon J, et al. Weight loss in obese patients with doi:10.1161/HYPERTENSIONAHA.108.127290 heart failure. J Am Heart Assoc 2016; 5(3):e002468. 19. Pandey A, Cornwell WK, 3rd, Willis B, et al. Body mass index and cardiore- doi:10.1161/JAHA.115.002468 spiratory fi tness in mid-life and risk of heart failure hospitalization in older 38. Kitzman DW, Nicklas B, Kraus WE, et al. Skeletal muscle abnormalities age: fi ndings from the Cooper Center Longitudinal Study. JACC Heart Fail and exercise intolerance in older patients with heart failure and preserved 2017; 5(5):367–374. doi:10.1016/j.jchf.2016.12.021 ejection fraction. Am J Physiol Heart Circ Physiol 2014; 306(9):H1364–1370. 20. Kenchaiah S, Sesso HD, Gaziano JM. Body mass index and vigorous physi- doi:10.1152/ajpheart.00004.2014 cal activity and the risk of heart failure among men. Circulation 2009; 39. Carbone S, Popovic D, Lavie CJ, Arena R. Obesity, body composition and 119(1):44–52. doi:10.1161/CIRCULATIONAHA.108.807289 cardiorespiratory fi tness in heart failure with preserved ejection fraction. 21. Hu G, Jousilahti P, Antikainen R, Katzmarzyk PT, Tuomilehto J. Joint effects Future Cardiol 2017; 13(5):451–463. doi:10.2217/fca-2017-0023 of physical activity, body mass index, waist circumference, and waist-to- 40. Ventura HO, Carbone S, Lavie CJ. Muscling up to improve heart failure hip ratio on the risk of heart failure. Circulation 2010; 121(2):237–244. prognosis. Eur J Heart Fail 2018; 20(11):1588–1590. doi:10.1002/ejhf.1314 doi:10.1161/CIRCULATIONAHA.109.887893 41. Osman AF, Mehra MR, Lavie CJ, Nunez E, Milani RV. The incremental 22. Kokkinos P, Faselis C, Franklin B, et al. Cardiorespiratory fi tness, body mass prognostic importance of body fat adjusted peak oxygen consumption in index and heart failure incidence. Eur J Heart Fail 2019; 21(4):436–444. chronic heart failure. J Am Coll Cardiol 2000; 36(7):2126–2131. doi:10.1002/ejhf.1433 doi:10.1016/s0735-1097(00)00985-2 23. Pandey A, Patel KV, Bahnson JL, et al; Look AHEAD Research Group. 42. Haykowsky MJ, Kouba EJ, Brubaker PH, Nicklas BJ, Eggebeen J, Kitzman Association of intensive lifestyle intervention, fi tness, and body mass DW. Skeletal muscle composition and its relation to exercise intolerance index with risk of heart failure in overweight or obese adults with type 2 in older patients with heart failure and preserved ejection fraction. Am J diabetes mellitus: an analysis from the Look AHEAD trial. Circulation 2020; Cardiol 2014; 113(7):1211–1216. doi:10.1016/j.amjcard.2013.12.031 141(16):1295–1306. doi:10.1161/CIRCULATIONAHA.119.044865 43. Sharma A, Ambrosy AP, DeVore AD, et al. Liraglutide and weight loss 24. Lavie CJ, Cahalin LP, Chase P, et al. Impact of cardiorespiratory fi tness on among patients with advanced heart failure and a reduced ejection frac- the obesity paradox in patients with heart failure. Mayo Clin Proc 2013; tion: insights from the FIGHT trial. ESC Heart Fail 2018; 5(6):1035–1043. 88(3):251–258. doi:10.1016/j.mayocp.2012.11.020 doi:10.1002/ehf2.12334 25. Mancini DM, Eisen H, Kussmaul W, Mull R, Edmunds LH, Jr., Wilson JR. 44. Writing Committee Members; Yancy CW, Jessup M, Bozkurt B, et al; Value of peak exercise oxygen consumption for optimal timing of cardiac American College of Cardiology Foundation/American Heart Association transplantation in ambulatory patients with heart failure. Circulation 1991; Task Force on Practice Guidelines. 2013 ACCF/AHA guideline for the man- 83(3):778–786. doi:10.1161/01.cir.83.3.778 agement of heart failure: a report of the American College of Cardiology 26. McAuley PA, Keteyian SJ, Brawner CA, et al. Exercise capacity and the obe- Foundation/American Heart Association Task Force on practice guidelines. sity paradox in heart failure: the FIT (Henry Ford Exercise Testing) Project. Circulation 2013; 128(16):e240–e327. doi:10.1161/CIR.0b013e31829e8776 Mayo Clin Proc 2018; 93(6):701–708. doi:10.1016/j.mayocp.2018.01.026 45. Ponikowski P, Voors AA, Anker SD, et al; Authors/Task Force Mem- 27. Piepoli MF, Corra U, Veglia F, et al; MECKI Score Research Group. Exercise bers; Document Reviewers. 2016 ESC Guidelines for the diagnosis and tolerance can explain the obesity paradox in patients with systolic heart treatment of acute and chronic heart failure: The Task Force for the failure: data from the MECKI Score Research Group. Eur J Heart Fail 2016; diagnosis and treatment of acute and chronic heart failure of the European 18(5):545–553. doi:10.1002/ejhf.534 Society of Cardiology (ESC). Developed with the special contribution of the 28. Carbone S, Billingsley HE, Rodriguez-Miguelez P, et al. Lean mass abnor- Heart Failure Association (HFA) of the ESC. Eur J Heart Fail 2016; 18(8): malities in heart failure: the role of sarcopenia, sarcopenic obesity, and 891–975. doi:10.1002/ejhf.592 cachexia. Curr Probl Cardiol 2020; 45(11):100417. 46. Corra U, Agostoni PG, Anker SD, et al. Role of cardiopulmonary exercise doi:10.1016/j.cpcardiol.2019.03.006 testing in clinical stratifi cation in heart failure. A position paper from 29. Lavie CJ, Ozemek C, Carbone S, Katzmarzyk PT, Blair SN. Sedentary the Committee on Exercise Physiology and Training of the Heart Failure behavior, exercise, and cardiovascular health. Circ Res 2019; 124(5):799–815. Association of the European Society of Cardiology. Eur J Heart Fail 2018; doi:10.1161/CIRCRESAHA.118.312669 20(1):3–15. doi:10.1002/ejhf.979 30. Hegde SM, Claggett B, Shah AM, et al. Physical activity and prognosis in the TOPCAT trial (Treatment of Preserved Cardiac Function Heart Failure With Address: Luke J. Laffi n, MD, Section of Preventive Cardiology and Rehabilita- an Aldosterone Antagonist). Circulation 2017; 136(11):982–992. tion, JB-1, Cleveland Clinic, 9500 Euclid Avenue, Cleveland, OH 44195; doi:10.1161/CIRCULATIONAHA.117.028002 laffi [email protected]

458 CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 REVIEW CME MOC Justin R. Abbatemarco, MD Chen Yan, MD Amy Kunchok, MBBS Alexander Rae-Grant, MD, FAAN, FRCPC Mellen Center for Multiple Sclerosis, Mellen Center for Multiple Sclerosis, Mellen Center for Multiple Sclerosis, Mellen Center for Multiple Sclerosis, Cleveland Clinic, Cleveland Clinic, Cleveland, OH Cleveland Clinic, Cleveland, OH Cleveland Clinic, Cleveland, OH Cleveland, OH

Antibody-mediated autoimmune encephalitis: A practical approach

ABSTRACT he spectrum and understanding of antibody-mediated autoimmune enceph- Antibody-mediated autoimmune encephalitis (AE) is a T alitis (AE)—an umbrella term for a group of heterogeneous group of infl ammatory central nervous noninfectious, infl ammatory central nervous system disorders. Symptoms typically include subacute, system diseases—have expanded dramati- progressive neuropsychiatric symptoms with associ- cally over the past few years. Familiarity with ated cognitive dysfunction, movement disorders, and AE syndromes ensures prompt diagnosis and autoimmune seizures. The diagnosis should be based on treatment. Practitioners need to stay abreast objective neurologic dysfunction in combination with of developments in this fi eld as the breadth autoantibody testing. Treatment with immunothera- of immune-mediated disorders of the nervous pies requires both short-term and long-term strategies system continues to evolve. depending on the specifi c syndrome and potential for In this paper, we will focus on the clini- relapse. In this paper, we review key features of AE, focus- cal features of common central nervous system ing on syndromes involving cell surface and synaptic cell surface and synaptic antibody syndromes in proteins, and share a practical approach to the diagnosis adults and on the emerging evidence in this area and management, including common pitfalls associated that has led to rapid changes in management with nonspecifi c antibody fi ndings. and treatment over the past decade. We will also briefl y comment on antibody-negative AE. KEY POINTS Antibody-related syndromes such as in- tracellular neuronal antibody-associated en- AE is an umbrella term for a group of infl ammatory cen- cephalitis and encephalitis occurring with tral nervous system disorders associated with neuronal demyelinating disorders are less commonly en- autoantibodies or other biomarkers of central nervous countered in clinical practice and are beyond system autoimmunity. the scope of this article.

Common clinical presentations include progressive neuro- ■ GENERAL FEATURES cognitive symptoms with concomitant movement disor- OF AUTOIMMUNE ENCEPHALITIS ders, seizures, and autonomic dysfunction that worsens Potential associations of AE encompass para- over weeks to months. neoplastic, parainfectious triggers along with adverse events related to various immunother- 1–3 Objective clinical fi ndings are needed to make the diagnosis apies. of AE, including changes on magnetic resonance imaging, Onset of AE is usually subacute over weeks electroencephalography, and cerebrospinal fl uid analysis. to months, with progressive neurocognitive symptoms including encephalopathy, cogni- tive dysfunction, neuropsychiatric symptoms, and seizures. Other features may include brain- stem syndromes, dysautonomia, and move- doi:10.3949/ccjm.88a.20122 ment disorders.

CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 459 AUTOIMMUNE ENCEPHALITIS

The prevalence and incidence are increas- ing on the specifi c antibody or antibodies.5 ing as testing becomes more widely available. A recent study in Olmsted County, MN, showed ■ WHEN SHOULD I CONSIDER a prevalence of AE of 13.7 per 100,000, similar THE DIAGNOSIS? 4 to that of infectious encephalitis. AE may be considered in patients presenting ■ with subacute onset (over 1 to 3 months) of HOW IS AUTOIMMUNE ENCEPHALITIS cognitive or memory defi cits, alterations in CLASSIFIED? consciousness, seizures, movement disorders, Over the past few decades, there has been rap- or other neuropsychiatric symptoms.8 Accom- id growth in the discovery of antibody-associ- panying neurologic or systemic symptoms sug- ated neurologic diseases. Autoantibodies that gestive of a specifi c antibody-mediated syn- target neuronal antigens can cause a diverse drome can increase clinical suspicion for AE. set of neurologic disorders. This has signifi - Examples include the following: cantly raised awareness of the wide spectrum • Dystonia-chorea for anti-N-methyl-d-as- of disease presentations that may have an un- partate (anti-NMDA) receptor encepha- derlying autoimmune component. litis Antibodies associated with AE are com- • Hyperekplexia (exaggerated startle refl ex) monly divided into 2 groups depending on the for anti-glycine receptor (Gly-R) antibody location of the antigen. The traditional “well- syndrome defi ned” syndromes (eg, anti-Hu or ANNA- • Faciobrachial dystonic seizure (focal or 1, anti-Ri or ANNA-2) target intracellular lateralized coordinated contractions of an neuronal antigens.5 And more recently, a new arm and the face) for anti-leucine-rich group of neuronal cell surface/synaptic pro- glioma-inactivated protein 1 (LGI1) en- teins has been described in association with cephalitis AE. This distinction is important for diag- • Peripheral nerve hyper excitability (dif- nosis and prognosis. Intracellular antibody- fuse involuntary motor-unit activity due A 2018 study mediated syndromes appear to be driven pri- to hyperexcitability of the motor nerve or marily by a CD8+ T-cell cytotoxic response its terminal)9 for anti-contactin-associated found the and usually have a poorer prognosis, with a protein-like 2 (Caspr2) syndrome prevalence of limited response to immunotherapy (Table • Weight loss accompanying gastrointestinal autoimmune 1). In contrast, cell surface/synaptic antibod- symptoms for anti-dipeptidyl-peptidase- ies appear to be directly pathogenic and are like protein 6 (DPPX) encephalitis.5,10 encephalitis more responsive to multimodal immunothera- While certain autoantibody disorders have was similar to pies (Table 2).1,5 Detailed discussions of the a specifi c phenotype, a number of patients pathophysiology of AE have been published with AE do not present with classic “limbic that of infec- by Bradshaw and Linnoila6 and by McKeon.7 encephalitis” and can present with a range of tious encepha- Though the terms paraneoplastic syndrome central nervous system and peripheral nervous litis and AE are sometimes used interchangeably, system involvement.11 Table 3 provides a list not all AE syndromes are paraneoplastic. of potential clinical and radiographic findings Paraneoplastic syndromes are defi ned as neu- suggestive of AE. rologic syndromes occurring in the setting of It is also important to recognize that AE cancer, sometimes preceding the diagnosis of can be antibody-negative. Antibody-negative neoplasm by months or years.1 Paraneoplas- AE may occur due to limitations of currently tic disorders are usually related to intracellu- available testing, especially as novel autoanti- lar neuronal antibodies. Cell surface/synaptic bodies are being discovered. However, objec- antibody-mediated syndromes, however, may tive clinical and radiologic criteria exist to aid also be associated with cancer, though they are diagnosis.8 usually classifi ed as phenotypes of low to mod- Despite the challenges, the diagnosis of erate risk, as these disorders can occur with or AE should be driven by the patient’s clinical without cancer. The strength of association presentation and diagnostic evaluation. This with an underlying neoplasm varies depend- approach includes a detailed clinical history

460 CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 ABBATEMARCO

TABLE 1 Autoantibody biomarkers of autoimmune encephalitis: Intracellular autoantibodies

PERIPHERAL ANTIBODY CENTRAL NERVOUS NERVOUS SYSTEM ASSOCIATED TARGET SYSTEM FEATURES FEATURES OTHER MALIGNANCY High-risk paraneoplastic autoantibodies ANNA-1 (Hu) Limbic encephalitis Sensory neuropathy Gastrointestinal SCLC Encephalomyelitis dysmotility Rare: neuroblastoma Cerebellar ataxia ANNA-2 (Ri) Encephalomyelitis Jaw dystonia SCLC Cerebellar ataxia Laryngospasm Breast carcinoma Rhombencephalitis ANNA-3 Limbic encephalitis Sensory and sensori- SCLC Encephalomyelitis motor neuropathies Cerebellar degeneration Amphiphysin Stiff-person spectrum SCLC disorder Breast or ovarian carcinoma CRMP-5 Limbic encephalitis Polyradiculo- SCLC Cerebellar ataxia neuropathy Thymoma carcinoma Chorea Myelopathy Cranial neuropathies (optic neuritis) Negative GAD65 Stiff-person spectrum Rare disorder antibody Limbic encephalitis testing does Cerebellar ataxia GFAP Meningoencephalitis Ovarian teratoma not rule out Myelitis Adenocarcinomas autoimmune Optic neuritis of various sites encephalitis PCA-1 (Yo) Cerebellar ataxia Breast or ovarian carcinoma PCA-2 Limbic encephalitis Polyneuropathy SCLC Cerebellar ataxia PCA-Tr (DNER) Limbic encephalitis Hodgkin lymphoma Cerebellar ataxia Ma 1 and Diencephalitis Ma1: Common, diverse Ma 2 (Ta) Limbic encephalitis Ma2: Testicular semi- Brain stem encephalitis noma Cerebellar degeneration

ANNA = antineuronal nuclear antibody; CRMP-5 = collapsin response mediator protein 5; DNER = delta/notch-like epidermal growth factor-related receptor; GAD = glutamic acid decarboxylase; GFAP = glial fi brillary acidic protein; PCA = Purkinje cell cytoplasmic antibody; SCLC = small-cell lung cancer

CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 461 AUTOIMMUNE ENCEPHALITIS

TABLE 2 Autoantibody biomarkers of autoimmune encephalitis: Cell-surface and synaptic antibodies

PERIPHERAL CENTRAL NERVOUS NERVOUS SYSTEM OTHER ASSOCIATED ANTIBODY SYSTEM FEATURES FEATURES FEATURES MALIGNANCY AMPAR Limbic encephalitis SCLC Breast carcinoma Thymoma Caspr2 Limbic encephalitis Peripheral nerve Rare, but thymoma hyperexcitability carcinoma reported DPPX Encephalopathy GI dysmotility Rare, but lymphoma Myelopathy Sleep disorder reported D2R Parkinsonism Encephalitis GABA A receptor Encephalitis Thymoma Status epilepticus GABA B receptor Limbic encephalitis SCLC Status epilepticus Opsoclonus myoclonus GQ1b Bickerstaff brain stem Guillain-Barré-like encephalitis illness IgLON5 Sleep disorder Dysphagia Rare Dementia Respiratory Ruling out an failure infectious LGI1 Limbic encephalitis Thymoma Faciobrachial dystonic process is seizures important, NMDA-R Limbic encephalitis Ovarian teratoma given that im- Status epilepticus Movement disorders munotherapies Psychosis can worsen Catatonia infection mGluR1 Cerebellar ataxia Dysgeusia Hodgkin lymphoma mGluR5 Limbic encephalitis Hodgkin lymphoma Glycine receptor Stiff-person spectrum Rare disorder

AMPAR = 2-amino-3-(5-methyl-3-oxo-1,2-oxazol-4-yl) propanoic acid receptor; Caspr2 = contactin-associated protein-like 2; D2R = dopamine 2 receptor; DPPX = dipeptidyl-peptidase-like protein 6; GABA = gamma-aminobutyric acid; GI = gastrointestinal; LGI1 = leucine-rich glioma-inactivated 1; mGluR = metabotropic glutamate receptor; NMDA-R = anti-N-methyl-D-aspartate receptor; SCLC = small cell lung cancer

including a personal or family history of au- with AE may not exhibit all the disease char- toimmunity, identifying infectious risk factors acteristics discussed below. For example, nor- (ie, exposure and travel history) while exclud- mal fi ndings on magnetic resonance imaging ing other conditions in the differential diag- (MRI) and electroencephalography (EEG) are nosis.8 It should also be noted that a patient not uncommon in anti-LGI1 encephalitis.12

462 CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 ABBATEMARCO

TABLE 3 Clinical, diagnostic, and radiographic clues to autoimmune encephalitis FINDINGS COMMENTS Subacute clinical course 1–3 months of symptoms Viral-like prodrome Fever, malaise, headache, gastrointestinal symptoms, etc Neurocognitive defi cits Agitation, apathy, catatonia, delusions, irritability, mania, psychosis, and paranoia Neurologic examination abnormalities Ataxia, brain stem abnormalities, myoclonus, tremor, or myelopathy New-onset focal seizure disorder or status epilepticus Often not responsive to antiepileptic medications New focal electroencephalogram abnormalities Focal epileptic or slow-wave activity particularly arising from the temporal lobes Subacute movement disorder Dyskinesias, dystonia, or choreoathetosis Subacute sleep disturbance Central sleep apnea, central neurogenic hypoventilation, or narco- lepsy Subacute autonomic dysfunction Hyperhidrosis, tachyarrhythmias, labile blood pressure, central hypoventilation, gastrointestinal dysmotility, urinary dysfunction Brain MRI abnormalities Bilateral T2-weighted FLAIR hyperintensities in the medial aspect of the temporal lobes, although multifocal changes involving the gray and white matter are also possible Infl ammatory cerebrospinal fl uid Mild to moderate pleocytosis (white blood cell count 5–100/μL ) Previous or current oncologic disorder or risk factors Increased risk of a paraneoplastic disorder for malignancy such as smoking FLAIR = fl uid-attenuated inversion recovery; MRI = magnetic resonance imaging

■ WHAT ARE THE COMMON CELL-SURFACE/ weeks of the illness (Figure 1). Soon after, SYNAPTIC ANTIBODY SYNDROMES subacute psychiatric symptoms develop in- IN AUTOIMMUNE ENCEPHALITIS? cluding anxiety, personality changes, halluci- nations, paranoid ideas, and frank psychosis. Anti-NMDA receptor encephalitis It is not unusual for patients to alternate be- Anti-NMDA receptor encephalitis was ini- tween hyperactive and catatonic states. Con- tially characterized in 12 women showing a comitant movement disorders such as dyski- characteristic progression of psychiatric symp- nesia (typically orofacial or limb), dystonia, toms ranging from subtle behavioral changes, and choreoathetosis are common. From 60% such as irritability, to frank psychosis. These to 75% of adult patients have been reported symptoms were followed by movement dis- to experience behavioral problems or move- orders, autonomic dysfunction, hypoventila- ment disorders during the fi rst month of the tion, seizures, and coma.13 Anti-NMDA is disorder.15 one of the most frequently identifi ed neuronal Autonomic dysfunction. As the disease autoantibodies in AE.1 Anti-NMDA receptor progresses, autonomic dysfunction becomes encephalitis frequently affects young adults, more prominent and commonly necessitates with a strong female predominance (4:1), monitoring in an intensive care unit. though it has been described in all ages in- Potential complications include tachy- cluding children and the elderly, with variable arrhythmias, hypotension, and central hypo- phenotypes.14 The median age is 20. ventilation requiring mechanical ventilation. Viral-like prodrome. Some patients ex- Approximately 80% of patients require ICU perience a viral-like prodrome that includes admission.15 headache or fever during the initial 1 to 2 Seizures can occur at any time and are

CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 463 AUTOIMMUNE ENCEPHALITIS

NMDA receptor encephalitis: Typical clinical course

Typical clinical and laboratory fi ndings Tier 1 treatments: Tier 2 treatments: IVMP: 1,000 mg x 5 days, IV rituximab: 1,000 mg on days 1 1. MRI: 30% mesial hippocampal 3. EEG: partial/generalized epilepsy IVIg: 0.4 g/kg/day x 5 days FLAIR/T2 hyperintensities +/- extreme delta brush (30%) and 15 or 4 weekly treatments Plasmapheresis: 5 sessions of 375 mg/m2 2. CSF: NMDA receptor antibodies 4. Cancer screening: 50% of fe- over 7–10 days IV cyclophosphamide: 3-6 with lymphocytic pleocytosis male patients will have ovarian monthly cycles of 750 mg/m2 teratoma

Baseline • Viral prodrome

50% to 55% have • Psychiatric manifestations: anxiety, clinical response to 45%–50% are paranoia, or social withdrawal tier 1 treatments non-responders to • Severe cognitive dysfunction tier 1 treatments and require tier 2 medications • Movement disorders: orofacial dyski- nesia, dystonia, or choreoathetosis • Epilepsy

Disease course • Disorder of consciousness • Severe autonomic dysfunction: hyper- thermia, fl uctuations of blood pressure, tachycardia • Central hypoventilation • Increasing seizure frequency Worsening clinical Weeks severity Figure 1. Typical clinical course associated with anti-NMDA receptor encephalitis.

CSF = cerebrospinal fl uid; EEG = electroencephalography; FLAIR = fl uid-attenuated inversion recovery; IV = intravenous; IVIg = intravenous immunoglobulin; IVMP = intravenous methylprednisolone; MRI = magnetic resonance imaging; NMDA = anti-N-methyl-D-aspartate receptor

Based on information in reference 15.

often not responsive to antiepileptic medica- tein, and intrathecal antibody production. tions alone. In a case series of 75 patients with Patterns on EEG can vary and often sim- anti-NMDA receptor encephalitis, almost ply demonstrate slowing.11 Approximately 80% suffered from generalized tonic-clonic one-third of patients develop extreme delta seizures, and 74% had focal seizures without brush; this is described as slowing with delta impaired awareness.16 There has been some activity 1–3 Hz and superimposed bursts of evidence to suggest antiepileptic drugs with rhythmic activity (beta activity 20–30 Hz) on sodium channel-blocking properties (eg, car- these slow waves and may be a poor prognostic bamazepine, oxcarbazepine, lacosamide, phe- sign.18 nytoin) may be more effective, though seizure MRI fi ndings. MRI is usually normal but freedom is usually achieved only when paired may show subtle mesial hippocampal fl uid-at- with immunotherapies.16,17 tenuated inversion recovery (FLAIR) hyper- NMDA receptor immunoglobulin G intensities. Titulaer et al reported that 76% of (IgG) testing should always be done with ce- patients had a CSF pleocytosis, one-third of rebrospinal fl uid (CSF), as serum testing is less the cohort had an abnormal brain MRI, and reliable (100% sensitivity for CSF vs 85% for 90% had changes on EEG including slowing.15 serum).15 CSF analysis usually reveals a mod- Teratomas. Anti-NMDA receptor en- erate lymphocytic pleocytosis, elevated pro- cephalitis may be associated with ovarian

464 CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 ABBATEMARCO

LGI1 encephalitis: Typical clinical course

Typical clinical and laboratory fi ndings Tier 1 treatments: 1. Serum: LGI1 autoantibodies along with hyponatremia 3. CSF: usually normal without evidence of intra- IVMP: 1,000 mg x 5 days, 2. MRI: 30%–40% mesial temporal FLAIR/T2 hyper- thecal antibody production IVIg: 0.4 g/kg/day x 5 days intensities; later in the disease course, patients can 4. EEG: epileptic discharges and focal slowing; Plasmapheresis: 5 sessions have mesial temporal sclerosis usually, FBDS does not have EEG correlation over 7–10 days

Baseline • Paroxysmal dizzy spells > 90% clinical • FBDS response to tier 1 treatments • New-onset epilepsy: autonomic, motor or gelastic seizures 25%–35% experience a relapse primarily with • Neuropsychiatric symptoms: cognitive dysfunction apathy or disinhibition and/or epileptic events • Cognitive dysfunction: amnesia or visual-spatial decline

Disease course • Autonomic dysfunction: bradycardia, hypothermia, neuropathic pain • Increasing seizure frequency: generalized-tonic seizures

Worsening clinical Weeks severity

Figure 2. Typical clinical course associated with anti-LGI1 encephalitis.

CSF = cerebrospinal fl uid; EEG = electroencephalography; FBDS = faciobrachial dystonic seizures; FLAIR = fl uid-attenuated inversion recovery; IVIg = intravenous immunoglobulin; IVMP = intravenous methylprednisolone; LGI1 = leucine-rich glioma-inactivated 1; MRI = magnetic resonance imaging or extraovarian teratomas in 30% to 40% of emerged in the literature.20,21 patients, particularly in young women. It re- Immunotherapy. Anti- NMDA receptor quires prompt surgical treatment.15 encephalitis responds to immunotherapy, but Infection. Herpes simplex virus infection the response can be slow. In the largest study of the central nervous system can also trigger to date of patients with anti- NMDA recep- 19 the production of NMDA receptor IgG. In a tor encephalitis, 81% had signifi cant recovery large prospective study, 27% of patients with at 24 months, but only 53% had clinical im- herpes simplex encephalitis developed anti- provement within 4 weeks of diagnosis.15 NMDA receptor antibodies within 16 weeks of From 10% to 25% of patients have a clini- completing their acyclovir treatment.19 None cal relapse, though symptoms are less severe of the patients had NMDA receptor IgG at the 1,22 index admission. Interestingly, 3 patients (6%) than in the initial presentation. developed CSF NMDA receptor IgG without Maintenance immunotherapy can be used clinical correlation, but the antibody produc- to optimize acute treatment response while tion was not detectable at 1-year follow-up.19 preventing relapses.15 Commonly used agents There have been reports of pediatric anti- include oral corticosteroids, intravenous im- NMDA receptor encephalitis developing after munoglobulin (IVIg), and steroid-sparing Japanese encephalitis infection, but no other agents such as mycophenolate mofetil, azathi- clear postinfectious or vaccination pattern has oprine, rituximab, and cyclophosphamide.23

CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 465 AUTOIMMUNE ENCEPHALITIS

Anti-LGI1 encephalitis ■ WHAT TESTING SHOULD I CONSIDER Anti-LGI1 encephalitis was fi rst characterized FOR AE DIAGNOSIS? 24 in 2010. In contrast to anti-NMDA receptor When AE is highly suspected (Table 3), ini- encephalitis, it typically occurs in men over tial testing should include an antibody panel. age 60, presenting with subacute cognitive Both serum and CSF should be tested for anti- dysfunction and behavioral changes. bodies since CSF is more sensitive and specifi c About 50% of patients also develop facio- for certain antibodies such as NMDA receptor brachial dystonic seizures characterized by fo- IgG, GAD65 IgG, and GFAP IgG, whereas cal or lateralized rapid coordinated movements serum is more sensitive for other antibodies of an arm or the face that may occur hundreds such as LGI1 IgG and Caspr2 IgG. of times a day. These dystonic seizures are very Antibody panels are preferred over specifi c specifi c for anti-LGI1 encephalitis but are not antibody tests, given the often overlapping universally present. Most patients will have clinical syndromes and the possibility of mul- normal surface activity on EEG even if the tiple positive antibodies. faciobrachial seizure events are captured dur- Common testing sites include Mayo Clinic ing the recording. The exact reason for this is or Associated Regional and University Pa- unclear, but it may refl ect subcortical seizure thologists (ARUP) laboratory. Antibody pan- origin, as some of these patients have contra- els include: lateral basal ganglia lesions on MRI.25 • Mayo ENS-2 panel in serum (www. Other associated seizure subtypes include mayocliniclabs.com/test-catalog/Over- subtle autonomic focal seizures and generalized view/92116) tonic-clonic seizures (Figure 2).12 Autonomic • Mayo ENC-2 panel in CSF (www. hyperactivity has also been reported. This in- mayocliniclabs.com/test-catalog/Over- cludes hyperhidrosis, tachycardia, blood pres- view/92117) sure lability, and urinary dysfunction.26 • ARUP Autoimmune Encephalitis Extend- The initial workup for anti-LGI1 enceph- ed Panel in serum (https://ltd.aruplab.com/ Viral infections alitis includes MRI, EEG, and CSF analysis Tests/Pub/3001431). but is usually unrevealing aside from a mild usually cause a Encephalitis has a broad differential hyponatremia on basic metabolic testing. diagnosis more profound CSF-specifi c oligoclonal bands may be seen, The differential diagnosis for encephalitis CSF pleocytosis but CSF can also be non-infl ammatory. Initial is very broad and includes infections, toxic- brain MRI can occasionally demonstrate T2- 27 metabolic encephalopathy, mitochondrial FLAIR hyperintensity of the hippocampus. disorders, nutritional defi ciencies, vascular Unlike in anti-NMDA receptor encephalitis, disorders, malignancy, and demyelinating dis- CSF is less sensitive than serum for the detec- orders. Clinicians should be particularly con- tion of LGI1 antibodies (serum 100% vs CSF cerned about ruling out an infectious process 27,28 about 88%). given the immunotherapies utilized to treat Treatment response. This syndrome char- AE. Infections to consider include herpes sim- acteristically responds briskly to tier 1 treat- plex virus encephalitis, human herpesvirus 6, ments (corticosteroids, intravenous immuno- human immunodefi ciency virus, fungal infec- globulin, plasmapheresis), but 20% to 30% of tion (eg, cryptococcal), mycobacterial infec- patients may experience a relapse necessitat- tion, Whipple disease, and neurosyphilis. ing long-term immunosuppression.22 In one In general, viral infections usually cause a small randomized control trial, IVIg treat- more profound CSF pleocytosis (a white blood ment was associated with a signifi cant reduc- cell count of 50–100/μL). Bacterial or myco- tion in seizure burden and can be considered a bacterial infections can have a lower of CSF steroid-sparing agent.29 glucose concentration, whereas AE usually Malignancy. From 5% to 15% of patients has normal glucose levels.31 have an underlying malignancy, most com- Overall, careful examination usually re- monly thymoma.30 Patients may have residual veals subtle neurologic defi cits that should cognitive defi cits despite initial recovery. prompt further evaluation for AE. Diagnos-

466 CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 ABBATEMARCO

tic red fl ags include newly occurring epileptic Magnetic resonance imaging seizures, movement disorders, and neurocog- MRI of the brain with and without gado- nitive symptoms, especially in the setting of linium contrast is appropriate. Consider MRI or CSF abnormalities. completing an epilepsy protocol, which may It should be mentioned that the prevalence vary between institutions but usually consists of primary psychiatric disorders is much higher of standard T1- and T2-weighted images, than the prevalence of AE. For example, the FLAIR, gradient-echo (T2) sequences, and overall prevalence of schizophrenia, with inci- diffusion-weighted sequences. Imaging se- dence peaking in young adults, is estimated to quences should also include contiguous, thin be 2.7 to 8.3 per 1,000,32 whereas the overall slices to ensure that hippocampal and tempo- prevalence of AE is 13.7 per 100,000.4 Thus, ral lobe lesions can be identifi ed.34 in a young patient with new mood disorder, Consider spinal cord MRI if neurologic a primary psychiatric diagnosis remains more abnormalities suggest concomitant myelitis. likely than AE. Symptoms could encompass motor, sensory, While patients with a preexisting psychi- and autonomic (bladder, bowel, sexual) dys- atric disorder can develop a concomitant au- function that localizes to the spinal cord. toimmune condition, it should also be men- An example could be a well-defi ned truncal tioned that a specifi c isolated psychiatric AE sensory level below which the sensation is al- 35 phenotype has not emerged in the literature tered. 21 despite extensive investigations. Electroencephalographic monitoring Continuous monitoring with EEG can clarify ■ A COMPREHENSIVE EVALUATION the cause of motor symptoms, identify sugges- FOR AUTOIMMUNE ENCEPHALITIS tive patterns such as extreme delta brush, and A comprehensive evaluation for suspected characterize seizure burden. AE includes a range of laboratory studies and Positron emission tomography imaging. Brain fl uorodeoxyglucose-positron emission Paraneoplastic Serum testing tomog raphy with computed tomography Serum testing with an AE antibody panel to (FDG-PET/CT) may illustrate either hypo- syndromes human immunodefi ciency virus, thyroid-stim- metabolism that may correlate with impair- sometimes ment of neuronal activity even in the absence ulating hormone, vitamin defi ciency (B1, B12, 36 precede E, folic acid). of structural disturbance or hypermetabolism that could correlate with increased glucose a cancer Evaluation for other disease markers metabolism caused by synaptic dysfunction or diagnosis Patients with autoimmune encephalitis are at ongoing seizure activity.37 Overall, PET may increased risk of having a second autoimmune be a sensitive marker for AE but is limited by by months disorder. Choice of any specifi c testing should its higher cost, lack of diagnostic specifi city or years be based on clinical suspicion.7,33 and clinical availability.36 Body FDG-PET/CT can be done to screen Cerebrospinal fl uid testing for malignancy.38 CSF studies include an AE antibody panel, routine studies, CSF IgG index, CSF-specifi c ■ WHICH AUTOANTIBODY FINDINGS oligoclonal bands, and comprehensive infec- SHOULD BE INTERPRETED CAUTIOUSLY? tious evaluations with specifi c attention to viral agents (ie, herpes simplex virus type 1, The development of broad antibody panels varicella-zoster virus, West Nile virus, John has led to some unintended consequences, Cunningham virus, and human herpesvirus particularly as these panels may include dis- 6). orders of the central nervous system and the peripheral nervous system.39 In addition, the Urine toxicology screen specifi city and sensitivity can vary for the dif- A urine toxicology test should include screen- ferent autoantibodies and the association with ing for marijuana and cocaine. AE. Therefore, interpretation of autoantibody

CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 467 AUTOIMMUNE ENCEPHALITIS

testing should be combined with a compre- portant for improved patient outcomes. hensive clinical evaluation as outlined above An important caveat for all clinicians in the section “A comprehensive evaluation is that although response to corticosteroid for autoimmune encephalitis.”8 therapy is a typical feature of AE, it does not Caution is particularly needed when in- confi rm the diagnosis of AE. Disorders such terpreting tests for autoantibodies that have as lymphoma and vasogenic edema associated low specifi city for AE. For example, anti- with a brain tumor can also respond dramati- voltage-gated potassium channel (VGKC) cally to steroids. Additionally, some AE pa- antibodies were initially detected in periph- tients do not respond to corticosteroids but eral nerve hyperexcitability disorders such require prolonged treatment with other im- as neuromyotonia and Morvan syndrome. munotherapies.8 Further laboratory evaluation has elucidated Initiation and escalation of treatment de- that LGI1 and Caspr2 are the usual targets of pend on the pretest probability of AE along these complex antibodies and not the VGKC with the clinical severity. If there is a high pre- itself. VGKC antibodies alone are not specif- test probability of a known syndrome, waiting ic for AE and can be seen in 5% of healthy for the results of antibody testing should not controls.40 In VGKC-positive patients with- delay treatment. For example, a case of refrac- out LGI1 and Caspr2 antibodies, additional tory autoimmune-mediated status epilepticus testing is not usually warranted, as there is no in the intensive care unit requires prompt and clear evidence that VGKC titers are indica- aggressive treatment. Caution should be used tive of an autoimmune disorder.30 when the diagnosis is less certain.8 Hashimoto encephalopathy, also known First-line treatments as steroid-responsive encephalopathy associ- First-line treatment for AE involves cortico- ated with autoimmune thyroiditis, has been steroids combined with either IVIg or plasma- historically linked to elevated serum levels 22,44 41 pheresis : of thyroid peroxidase antibodies. In the era • Methylprednisolone 1 g daily for 5 days, Anti-NMDA of more neuronal-specifi c antibodies, thyroid followed by oral prednisone 1 mg/kg (max- peroxidase antibodies have been found to have imum dose 60–80 mg daily, with a pro- receptor limited diagnostic value. An extensive evalu- longed taper over 3–6 months) encephalitis ation to exclude other causes of AE should be • IVIg 400 mg/kg/day for 5 days pursued in cases with elevated thyroid peroxi- is associated • Plasmapheresis, with 5 exchanges over 7 to dase antibody titers, given the unclear clinical 8 days. with ovarian signifi cance of these antibodies in neurologic disorders.42,43 Second-line treatments or extraovarian Second-line treatments can be given as mono- teratomas ■ WHAT IS THE INITIAL TREATMENT therapy or in combination for refractory dis- FOR AUTOIMMUNE ENCEPHALITIS? ease activity 1 to 2 weeks from completion of fi rst-line treatment22: Current treatment guidelines for AE are based • Rituximab 1,000 mg IV, with a repeat dose on a combination of expert opinion, case se- in 2 weeks, or 375 mg/m2 weekly IV infu- ries, and case reports, and evidence from high- sion for 4 weeks quality multicenter randomized trials is lack- • Cyclophosphamide 750 mg/m2 IV month- ing.29 But existing evidence suggests that early ly for 3 to 6 months. initiation of therapy and prompt escalation to second-line immunotherapy may lead to im- Maintenance treatments45 proved clinical outcomes.15,22 • Rituximab, repeat every 6 months, same Still, unanswered questions include the dosing schedule as second-line therapy time frame associated with a response to the • IVIg 0.4 g/kg every 2 to 4 weeks fi rst-line immunotherapy and the optimal du- • Mycophenolate mofetil 500 to 3,000 mg/ ration of sustained immunotherapy. A com- day prehensive evaluation for malignancy is also • Azathioprine 1 to 3 mg/kg/day vital as early detection and treatment are im- • Cyclophosphamide 1 to 2 mg/kg/day orally.

468 CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 ABBATEMARCO

■ HOW DO I MONITOR RESPONSE All treatment decisions must be made in TO TREATMENT IN AE? coordination with the treating oncologist.6 There are no validated biological markers to Additionally, patients may have permanent assess treatment response in AE. Although neurologic defi cits with relatively little hope some studies have suggested that early reduc- for recovery. tion in titers can correlate with better clini- The type and frequency of screening for cal outcomes, antibodies can remain positive malignancy depends on the specifi c antibody 50 even in patients with good outcomes.46 Fur- syndrome. If a patient is diagnosed with an ther, the change in antibody titers does not autoantibody commonly associated with para- consistently correlate with risk of relapse.47 neoplastic syndromes, frequent monitoring is Therefore, serum or CSF antibody titers in recommended. In those cases, we typically or- isolation are not reliable for monitoring treat- der surveillance testing every 6 to 12 months ment response in AE. for at least 3 to 5 years. If the patient has an Imaging, when abnormal, can be repeated antibody with a low likelihood of an underly- to look for improvements over time. Unfor- ing neoplasm, we consider reevaluating once a 50 tunately, even after appropriate treatment is year for 3 years after the index diagnosis. initiated for AE, brain MRI may show irre- Based on the 2011 European Federation of versible changes such as generalized or focal the Neurological Societies guidelines, we will atrophy on follow-up. utilize either whole-body FDG-PET or CT of Monitoring of the treatment response is the chest, abdomen, and pelvis as a screening therefore primarily based on the clinical ex- measure for certain AE patients, with clinical amination. We urge clinicians to use objective consideration of the known cancer associa- measures to determine the true effi cacy of a tions, cancer risk factors, or family history of given treatment. For example, clinicians can cancer.50 use the Scale for the Assessment and Rating In particular, the nature of the antibody of Ataxia (SARA),48 Symbol Digit Modalities determines the risk and type of an underly- Test, or the Montreal Cognitive Assessment ing malignancy and therefore the investiga- Anti-NMDA (MoCA)49 for reliable scores to measure the tion. For example, anti-Yo autoantibody has success of a trial and the utility of long-term a greater than 90% association with breast or receptor treatments. Formal neuropsychological test- ovarian malignancy, so an extensive evalua- encephalitis ing is also a valuable tool to document the tion should be pursued to identify the underly- extent of cognitive damage and to evaluate ing neoplasm.50 responds immunotherapy response. In addition, sex-specifi c tests such as pelvic to immuno- ultrasonography, mammography, and testicu- therapies, ■ WHAT ONCOLOGIC EVALUATION lar ultrasonography should be considered if IS APPROPRIATE FOR PATIENTS WITH AE? the initial evaluation is unrevealing. Further, but response all patients should complete routine age- and Paraneoplastic AE can occur in association can be slow with an underlying malignancy. In particular, sex-appropriate screening measures including small-cell lung cancer, non-small-cell lung screening for breast, colorectal, cervical, and cancer, and neuroblastoma trigger the produc- prostate cancer, along with lung cancer when tion of paraneoplastic autoantibodies.50 The applicable, based on US Preventive Services neurologic sequelae can occur prior to detec- Task Force recommendations. tion of the cancer allowing for early discovery ■ TAKE-HOME MESSAGE and oncologic treatment. Though treating the cancer is the main Our understanding of AE has expanded dra- goal in these patients, they may still require matically over the past few years. Familiarity short-term or long-term immunotherapy for with different AE syndromes will ensure prompt the paraneoplastic disorder. Immunotherapies diagnosis and treatment. In cases that are less that may have dual benefi t to treat the cancer clear, a sound diagnostic approach anchored and autoimmune disorder, such as cyclophos- on objective clinical or radiographic fi ndings is phamide, can be considered. important for optimizing outcomes. Clinicians

CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 469 AUTOIMMUNE ENCEPHALITIS

need to stay abreast of developments in this fi eld ■ DISCLOSURES as the breadth of immune-mediated disorders of The authors report no relevant fi nancial relationships which, in the context the nervous system continues to evolve. ■ of their contributions, could be perceived as a potential confl ict of interest.

■ REFERENCES plex encephalitis: a prospective observational study and retrospec- tive analysis. Lancet Neurol 2018; 17(9):760–772. 1. Dalmau J, Graus F. Antibody-mediated encephalitis. N Engl J Med doi:10.1016/S1474-4422(18)30244-8 2018; 378(9):840–851. doi:10.1056/NEJMra1708712 20. Ma J, Zhang T, Jiang L. Japanese encephalitis can trigger anti- 2. Zekeridou A, Lennon VA. Neurologic autoimmunity in the era of N-methyl-D-aspartate receptor encephalitis. J Neurol 2017; checkpoint inhibitor cancer immunotherapy. Mayo Clin Proc 2019; 264(6):1127–1131. doi:10.1007/s00415-017-8501-4 94(9):1865–1878. doi:10.1016/j.mayocp.2019.02.003 21. Dalmau J, Armangué T, Planagumà J, et al. An update on anti- 3. Kunchok A, Aksamit AJ Jr, Davis JM 3rd, et al. Association be- NMDA receptor encephalitis for neurologists and psychiatrists: tween tumor necrosis factor inhibitor exposure and infl ammatory mechanisms and models. Lancet Neurol 2019; 18(11):1045–1057. central nervous system events. JAMA Neurol 2020; 77(8):937–946. doi:10.1016/S1474-4422(19)30244-3 doi:10.1001/jamaneurol.2020.1162 22. Nosadini M, Mohammad SS, Ramanathan S, Brilot F, Dale RC. Im- 4. Dubey D, Pittock SJ, Kelly CR, et al. Autoimmune encephalitis epi- mune therapy in autoimmune encephalitis: a systematic review. demiology and a comparison to infectious encephalitis. Ann Neurol Expert Rev Neurother 2015; 15(12):1391–1419. 2018; 83(1):166–177. doi:10.1002/ana.25131 doi:10.1586/14737175.2015.1115720 5. Lancaster E. The diagnosis and treatment of autoimmune encephali- 23. López-Chiriboga AS, Flanagan EP. Diagnostic and therapeutic ap- tis. J Clin Neurol 2016; 12(1):1–13. doi:10.3988/jcn.2016.12.1.1 proach to autoimmune neurologic disorders. Semin Neurol 2018; 6. Bradshaw MJ, Linnoila JJ. An overview of autoimmune and 38(3):392–402. doi:10.1055/s-0038-1660819 paraneoplastic encephalitides. Semin Neurol 2018; 38(3):330–343. 24. Irani SR, Alexander S, Waters P, et al. Antibodies to Kv1 potassium doi:10.1055/s-0038-1660821 channel-complex proteins leucine-rich, glioma inactivated 1 protein 7. McKeon A. Autoimmune encephalopathies and dementias. Con- and contactin-associated protein-2 in limbic encephalitis, Morvan’s tinuum (Minneap Minn) 2016; 22(2 Dementia):538–558. syndrome and acquired neuromyotonia. Brain 2010; 133(9):2734– doi:10.1212/CON.0000000000000299 2748. doi:10.1093/brain/awq213 8. Graus F, Titulaer MJ, Balu R, et al. A clinical approach to diagnosis 25. Flanagan EP, Kotsenas AL, Britton JW, et al. Basal ganglia T1 hy- of autoimmune encephalitis. Lancet Neurol 2016; 15(4):391–404. perintensity in LGI1-autoantibody faciobrachial dystonic seizures. doi:10.1016/S1474-4422(15)00401-9 Neurol Neuroimmunol Neuroinfl amm 2015; 2(6):e161. 9. Auger RG. AAEM minimonograph #44: diseases associated with doi:10.1212/NXI.0000000000000161 excess motor unit activity. Muscle Nerve 1994; 17(11):1250–1263. 26. Vernino S. Autoimmune autonomic disorders. Continuum (Minneap doi:10.1002/mus.880171103 Minn) 2020; 26(1):44–57. doi:10.1212/CON.0000000000000812 10. Dutra LA, Abrantes F, Toso FF, Pedroso JL, Barsottini OGP, Hoft- 27. van Sonderen A, Thijs RD, Coenders EC, et al. Anti-LGI1 encepha- berger R. Autoimmune encephalitis: a review of diagnosis and treat- litis: clinical syndrome and long-term follow-up. Neurology 2016; ment. Arq Neuropsiquiatr 2018; 76(1):41–49. 87(14):1449–1456. doi:10.1212/WNL.0000000000003173 doi:10.1590/0004-282X20170176 28. Gadoth A, Pittock SJ, Dubey D, et al. Expanded phenotypes and out- 11. Ariño H, Armangué T, Petit-Pedrol M, et al. Anti-LGI1-associated comes among 256 LGI1/CASPR2-IgG-positive patients. Ann Neurol cognitive impairment: presentation and long-term outcome. Neu- 2017; 82(1):79–92. doi:10.1002/ana.24979 rology 2016; 87(8):759–765. doi:10.1212/WNL.0000000000003009 29. Dubey D, Britton J, McKeon A, et al. Randomized placebo-con- 12. Bastiaansen AEM, van Sonderen A, Titulaer MJ. Autoimmune trolled trial of intravenous immunoglobulin in autoimmune LGI1/ encephalitis with anti-leucine-rich glioma-inactivated 1 or anti- CASPR2 epilepsy. Ann Neurol 2020; 87(2):313–323. contactin-associated protein-like 2 antibodies (formerly called doi:10.1002/ana.25655 voltage-gated potassium channel-complex antibodies). Curr Opin 30. van Sonderen A, Schreurs MW, de Bruijn MA, et al. The relevance of Neurol 2017; 30(3):302–309. doi:10.1097/WCO.0000000000000444 VGKC positivity in the absence of LGI1 and Caspr2 antibodies. Neu- 13. Dalmau J, Tüzün E, Wu HY, et al. Paraneoplastic anti-N-methyl-D- rology 2016; 86(18):1692–1699. doi:10.1212/WNL.0000000000002637 aspartate receptor encephalitis associated with ovarian teratoma. 31. Armangue T, Leypoldt F, Dalmau J. Autoimmune encephalitis as dif- Ann Neurol 2007; 61(1):25–36. doi:10.1002/ana.21050 ferential diagnosis of infectious encephalitis. Curr Opin Neurol 2014; 14. Guasp M, Dalmau J. Encephalitis associated with antibodies 27(3):361–368. doi:10.1097/WCO.0000000000000087 against the NMDA receptor. Med Clin (Barc) 2018; 151(2):71–79. 32. Messias EL, Chen CY, Eaton WW. Epidemiology of schizophre- doi:10.1016/j.medcli.2017.10.015 nia: review of fi ndings and myths. Psychiatr Clin North Am 2007; 15. Titulaer MJ, McCracken L, Gabilondo I, et al. Treatment and prog- 30(3):323–338. doi:10.1016/j.psc.2007.04.007 nostic factors for long-term outcome in patients with anti-NMDA 33. Galli JR, Austin SD, Greenlee JE, Clardy SL. Stiff person syndrome receptor encephalitis: an observational cohort study. Lancet Neurol with Anti-GAD65 antibodies within the national veterans af- 2013; 12(2):157–165. doi:10.1016/S1474-4422(12)70310-1 fairs health administration. Muscle Nerve 2018; 58(6):801–804. 16. de Bruijn MAAM, van Sonderen A, van Coevorden-Hameete MH, et doi:10.1002/mus.26338 al. Evaluation of seizure treatment in anti-LGI1, anti-NMDAR, and 34. Bernasconi A, Cendes F, Theodore WH, et al. Recommendations for anti-GABABR encephalitis. Neurology 2019; 92(19):e2185–e2196. the use of structural magnetic resonance imaging in the care of doi:10.1212/WNL.0000000000007475 patients with epilepsy: a consensus report from the International 17. Feyissa AM, López Chiriboga AS, Britton JW. Antiepileptic drug League Against Epilepsy Neuroimaging Task Force. Epilepsia 2019; therapy in patients with autoimmune epilepsy. Neurol Neuroimmu- 60(6):1054–1068. doi:10.1111/epi.15612 nol Neuroinfl amm 2017; 4(4):e353. 35. Frohman EM, Wingerchuk DM. Clinical practice. Transverse myelitis. doi:10.1212/NXI.0000000000000353 N Engl J Med 2010; 363(6):564–572. doi:10.1056/NEJMcp1001112 18. Schmitt SE, Pargeon K, Frechette ES, Hirsch LJ, Dalmau J, Friedman 36. Probasco JC, Solnes L, Nalluri A, et al. Abnormal brain metabolism D. Extreme delta brush: a unique EEG pattern in adults with anti- on FDG-PET/CT is a common early fi nding in autoimmune en- NMDA receptor encephalitis. Neurology 2012; 79(11):1094–1100. cephalitis. Neurol Neuroimmunol Neuroinfl amm 2017; 4(4):e352. doi:10.1212/WNL.0b013e3182698cd8 doi:10.1212/NXI.0000000000000352 19. Armangue T, Spatola M, Vlagea A, et al. Frequency, symptoms, risk 37. Leypoldt F, Buchert R, Kleiter I, et al. Fluorodeoxyglucose positron factors, and outcomes of autoimmune encephalitis after herpes sim- emission tomography in anti-N-methyl-D-aspartate receptor en-

470 CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 ABBATEMARCO

cephalitis: distinct pattern of disease. J Neurol Neurosurg Psychiatry 45. Shin YW, Lee ST, Park KI, et al. Treatment strategies for auto- 2012; 83(7):681–686. doi:10.1136/jnnp-2011-301969 immune encephalitis. Ther Adv Neurol Disord 2018; 11:1–19. 38. McKeon A, Apiwattanakul M, Lachance DH, et al. Positron emission doi:10.1177/1756285617722347 tomography-computed tomography in paraneoplastic neuro- 46. Dalmau J, Gleichman AJ, Hughes EG, et al. Anti-NMDA-receptor logic disorders: systematic analysis and review. Arch Neurol 2010; encephalitis: case series and analysis of the effects of antibodies. 67(3):322–329. doi:10.1001/archneurol.2009.336 Lancet Neurol 2008; 7(12):1091–1098. 39. Ebright MJ, Li SH, Reynolds E, et al. Unintended consequences of doi:10.1016/S1474-4422(08)70224-2 Mayo paraneoplastic evaluations. Neurology 2018; 91(22):e2057– 47. Irani SR, Bera K, Waters P, et al. N-methyl-D-aspartate antibody en- e2066. doi:10.1212/WNL.0000000000006577 cephalitis: temporal progression of clinical and paraclinical observa- 40. Rossi M, Mead S, Collinge J, Rudge P, Vincent A. Neuronal antibod- tions in a predominantly non-paraneoplastic disorder of both sexes. ies in patients with suspected or confi rmed sporadic Creutzfeldt- Brain 2010; 133(pt 6):1655–1667. doi:10.1093/brain/awq113 Jakob disease. J Neurol Neurosurg Psychiatry 2015; 86(6):692–694. 48. Schmitz-Hübsch T, du Montcel ST, Baliko L, et al. Scale for the as- doi:10.1136/jnnp-2014-308695 sessment and rating of ataxia: development of a new clinical scale. 41. Schiess N, Pardo CA. Hashimoto’s encephalopathy. Ann N Y Acad Sci Neurology 2006; 66(11):1717–1720. 2008; 1142:254–265. doi:10.1196/annals.1444.018 doi:10.1212/01.wnl.0000219042.60538.92 42. Mattozzi S, Sabater L, Escudero D, et al. Hashimoto encephalopathy 49. Hébert J, Day GS, Steriade C, Wennberg RA, Tang-Wai DF. Long- in the 21st century. Neurology 2020; 94(2):e217–e224. term cognitive outcomes in patients with autoimmune encephalitis. doi:10.1212/WNL.0000000000008785 Can J Neurol Sci 2018; 45(5):540–544. doi:10.1017/cjn.2018.33 43. Valencia-Sanchez C, Pittock SJ, Mead-Harvey C, et al. Brain dysfunc- 50. Titulaer MJ, Soffi etti R, Dalmau J, et al. Screening for tumours in tion and thyroid antibodies: autoimmune diagnosis and misdiagno- paraneoplastic syndromes: report of an EFNS task force. Eur J Neurol sis. Brain Commun 2021. doi:10.1093/braincomms/fcaa233. 2011; 18(1):19–e3. doi:10.1111/j.1468-1331.20 44. de Montmollin E, Demeret S, Brulé N, et al. Anti-N-methyl-D-aspar- tate receptor encephalitis in adult patients requiring intensive care. Address: Justin R. Abbatemarco, MD, Mellen Center for Multiple Scle- Am J Respir Crit Care Med 2017; 195(4):491–499. rosis, U10, Cleveland Clinic, 9500 Euclid Avenue, Cleveland, OH 44195; doi:10.1164/rccm.201603-0507OC [email protected]

CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021 471 CONTINUING MEDICAL EDUCATION CME MOC

How to earn AMA PRA Category 1 Credit™ August 2021 CME/MOC activities and ABA, ABIM, ABP, ABPath, ABS MOC points Estimated time to complete each activity: up to 1 hour AMA/PRA Category 1 Credit™ Antibody-mediated autoimmune encephalitis: To read articles as CME activities and claim credit, go to www. A practical approach ccjm.org, click on the “CME/MOC” menu, and then “Articles.” Find the articles that you want to read as CME activities and Release date: August 1, 2021 click on the appropriate links. After reading an article, click on Expiration date: July 31, 2022 the link to complete the activity. You will be asked to log in to your MyCME account (or to create an account). Upon logging in, select “CME,” complete the activity evaluation, and print your certifi cate. Call 216-444-2661 or e-mail [email protected] with questions. Maintenance of Certifi cation (MOC) Points All Cleveland Clinic Journal of Medicine CME activities are now eligible for MOC points. Physicians may claim MOC points in addition to CME credit. Follow the instructions for completing and claiming credit for CME activities. When you log into your MyCME account, select “CME & MOC” and enter your ABIM identifi cation number and your date of birth. The system will store this information after you enter it the fi rst time. Complete the quiz and evaluation and print your CME certifi cate.

FINANCIAL DISCLOSURES: In accordance with the Standards for Com- AUTHOR AND STAFF DISCLOSURES: Authors’ potential confl icts of interest mercial Support issued by the Accreditation Council for Continuing are disclosed within their articles. Cleveland Clinic Journal of Medicine’s staff Medical Education (ACCME), the Cleveland Clinic Foundation Center disclose the following fi nancial relationships that may be relevant to their edi- for Continuing Education requires resolution of all faculty confl icts of torial roles: Dr. Brian F. Mandell (Editor in Chief) reports teaching and speaking interest to ensure CME activities are free of commercial bias. for Genentech; and consulting for Horizon Pharma. Dr. Kristin Highland (As- sociate Editor) has disclosed fi nancial interests (consulting, research, teaching, and speaking) with Actelion Pharmaceuticals, Bayer Healthcare, Boehringer Ingelheim, Eiger Biopharmaceuticals, Gententech, Gossamer Bio, Lilly, Reata Pharmaceuticals, United Therapeutics, and Viela Bio. Dr. Christian Nasr (Associ- ate Editor) reports service on advisory committees or review panels for Exelixis, Horizon Pharma, Neurogastrx, and Nevro Corp.; and consulting for Siemens. DISCLAIMER: The information in these educational activities is provided for general medical education purposes only and is not meant to substitute for the independent medical judgment of a physician relative to diagnostic and treatment options of a specifi c patient’s medical condition. The viewpoints expressed in these CME activities are those of the authors. They do not represent an endorsement by The Cleveland Clinic Foundation. In no event will The Cleveland Clinic Foundation be liable for any decision made or action taken in reliance upon the information provided through these CME activities.

CME ACCREDITATION: Participants will earn MOC points equivalent to the amount of CME credits claimed for the The Cleveland Clinic Foundation Center for Continuing Education is accredited by the activity. Accreditation Council for Continuing Medical Education (ACCME) to provide continuing It is the CME activity provider’s responsibility to submit participant completion information medical education for physicians. to ACCME for the purpose of granting ABA, ABIM, ABPath and ABP credit. Credit will be The Cleveland Clinic Foundation Center for Continuing Education designates this journal- reported within 30 days of claiming credit. based CME activity for a maximum of 1.0 AMA PRA Category 1 Credit™. Physicians should ABS: It is the participant’s responsibility to self-report their participation per current board claim only the credit commensurate with the extent of their participation in the activity. policy. Participants claiming CME credit from this activity may submit the credit hours to the Please Note: To receive MOC you must select the MOC option during the online credit American Osteopathic Association for Category 2 credit. claiming process and complete the required steps. MOC/CC PART II ACCREDITATION: Successful completion of this CME activity, which includes participation in the evaluation component, enables the participant to earn up to: • American Board of (ABA) MOC: 1.0 Lifelong Learning MOC points in the ABA MOCA 2.0® Maintenance of Certifi cation in Anesthesiology Program®. • American Board of Internal Medicine (ABIM) MOC: 1.0 Medical Knowl- edge MOC points in the ABIM MOC Assessment Recognition Program. • American Board of Pathology (ABPath) CC: 1.0 Lifelong Learning credits in the ABPath Continuing Certifi cation Program. • American Board of (ABP) MOC: 1.0 Lifelong Learning & Self- Assessment MOC points in the ABP Maintenance of Certifi cation Program. • American Board of Surgery (ABS) CC: 1.0 Accredited CME & Self-Assess- ment credits toward ABS Continuous Certifi cation Program.

472 CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 88 • NUMBER 8 AUGUST 2021