Orlistat for Overweight Subjects with Nonalcoholic Steatohepatitis

To the Editor:We read with great interest the article by Harrison et al. reporting theeffect of Orlistat, an inhibitor of fat absorption, on the metabolicabnormalities, histopathologic findings, and cytokine levels in subjectswith nonalcoholic steatohepatitis (NASH).1 They showed that whencompared to controls, Orlistat did not enhance weight loss or improveliver enzymes, measures of insulin resistance, and histopathology. Cer-tainly, this study is important because it provides scientific informationon this clinically relevant condition. However, we think that somepoints should be discussed.

First, nonalcoholic fatty liver disease (NAFLD) is the most com-mon liver abnormality in the Western world and is strongly associatedwith the features of the metabolic syndrome (MetS) and insulin resis-tance.2 The pathogenesis of NAFLD is a multiple-hit process resultingfrom hepatic fat deposition that is related to several conditions, includ-ing insulin resistance and central obesity. Additional hits, such as oxi-dative stress or adipocytokines (leptin, adiponectin, tumor necrosisfactor-alpha, interleukin-6, etc.), can further enhance liver damageleading to NASH or fibrosis.3 In addition, a clear relationship wasdemonstrated between the components of the MetS and histopatho-logic findings of NASH.4-6 In light of these data, we think that thedescription of the components of the MetS should be an essential partof the present study that investigates the relationship of adipocyto-kines, inflammation, and fibrosis in NASH.

Second, as shown in Table 1 of the article, most of the study par-ticipants are obese and some of them are even morbidly obese. Obesityis a strong risk factor for diabetes mellitus (DM), prediabetes, hyper-tension, and also dyslipidemia. Although it was stated in the article thatonly four subjects were diabetic at baseline, there is no informationregarding the blood pressure, high density lipoprotein cholesterol andtriglyceride levels, and also the glucose tolerance status of the subjects.It is well known that all components that constitute the MetS are alsorisk factors for DM and also prediabetes, namely, impaired fastingglucose and/or impaired glucose intolerance. We think that some ofthe study participants may still have overt glucose dysregulation orDM without implementation of the glucose tolerance test. In addition,matching the groups for glucose and body mass index may not beenough to make clear comparisons at this point, because DM and evenglucose intolerance is itself a predictor of presence of insulin resistance.Moreover, plasma insulin and adipocytokine levels differ according tothe degree of glucose dysregulation.7,8 The same subject is also true forhypertension and dyslipidemia.9,10

We think all these points make the resultant comparisons and cor-relations questionable. Therefore, we would like to ask the authorswhether they can present some new results by categorizing the subjectswith NASH according to metabolic confounders such as impairedglucose tolerance, dyslipidemia, and blood pressure. This may providethe readers clearer information about the effect of Orlistat on theinflammatory mechanism in NAFLD.

TEOMAN DOGRU1

CEMAL NURI ERCIN1

ILKER TASCI2

SERKAN TAPAN3

ZEKI YESILOVA1

AHMET UYGUN1

1 Department of Gastroenterology2 Department of Internal Medicine and3 Department of BiochemistryGulhane Medical SchoolAnkara, Turkey

References

1. Harrison SA, Fecht W, Brunt EM, Neuschwander-Tetri BA. Orlistat foroverweight subjects with nonalcoholic steatohepatitis: a randomized, pro-spective trial. HEPATOLOGY 2009;49:80-86.

2. Vuppalanchi R, Chalasani N. Nonalcoholic fatty liver disease and nonal-coholic steatohepatitis: Selected practical issues in their evaluation andmanagement. HEPATOLOGY 2009;49:306-317.

3. Kim CH, Younossi ZM. Nonalcoholic fatty liver disease: a manifestationof the metabolic syndrome. Cleve Clin J Med 2008;75:721-728.

4. van der Porten D, Milner KL, Hui J, Hodge A, Trenell MI, Kench JG, etal. Visceral fat: a key mediator of steatohepatitis in metabolic liver disease.HEPATOLOGY 2008;48:449-457.

5. Cheung O, Kapoor A, Puri P, Sistrun S, Luketic VA, Sargeant CC, etal. The impact of fat distribution on the severity of nonalcoholic fattyliver disease and metabolic syndrome. HEPATOLOGY 2007;46:1091-1100.

6. Marchesini G, Bugianesi E, Forlani G, Cerrelli F, Lenzi M, Manini R, et al.Nonalcoholic fatty liver, steatohepatitis, and the metabolic syndrome.HEPATOLOGY 2003;37:917-923.

7. Dogru T, Sonmez A, Tasci I, Bozoglu E, Yilmaz MI, Genc H, et al. Plasmavisfatin levels in patients with newly diagnosed and untreated type 2 dia-betes mellitus and impaired glucose tolerance. Diabetes Res Clin Pract2007;76:24-29.

8. Erdem G, Dogru T, Tasci I, Sonmez A, Tapan S. Low plasma apelin levelsin newly diagnosed type 2 diabetes mellitus. Exp Clin Endocrinol Diabetes2008;116:289-292.

9. Sung SH, Chuang SY, Sheu WH, Lee WJ, Chou P, Chen CH. Adiponec-tin, but not leptin or high-sensitivity C-reactive protein, is associated withblood pressure independently of general and abdominal adiposity. Hyper-tens Res 2008;31:633-640.

10. Tasci I, Dogru T, Naharci I, Erdem G, Yilmaz MI, Sonmez A, et al. Plasmaapelin is lower in patients with elevated LDL-cholesterol. Exp Clin Endo-crinol Diabetes 2007;115:428-432.

Copyright © 2009 by the American Association for the Study of Liver Diseases.Published online in Wiley InterScience (www.interscience.wiley.com).DOI 10.1002/hep.22822Potential conflict of interest: Nothing to report.

Reply:

We thank Dogru et al. for their thoughtful comments on ourarticle describing the results of a pilot study using orlistat as apotential adjunct to achieve weight loss in calorically restrictedpatients with nonalcoholic steatohepatitis (NASH). We agree thatthe histopathological changes identified as NASH are the result ofcomplex and poorly understood relationships among lipotoxicity,insulin resistance, and dysregulated adipokines and cytokines. Assummarized in their letter, a number of much larger studies havealready examined these relationships. We feel that analyzing sub-groups of a small cohort such as the one we enrolled in this trialwould not contribute to what has already been described. We agreethat some of our patients may have had previously unrecognizeddiabetes, and this seems to be a frequent finding when patients withNASH are studied closely during study enrollment. However, weare not aware of any data to suggest that patients with diabetesrespond differently to weight reduction than those without diabetes

HEPATOLOGY, Vol. 49, No. 6, 2009 CORRESPONDENCE 2127

in terms of their NASH, so this was not viewed as a potentialconfounder. Moreover, we are not aware of any data to suggest thatinhibition of intestinal lipase with orlistat has an anti-inflammatoryeffect independent of its effect of facilitating weight loss, so it isunlikely that any conclusions could be made in this regard using thedata collected in this study.

STEPHEN HARRISON, M.D.1

BRENT NEUSCHWANDER-TETRI, M.D.2

ELIZABETH BRUNT, M.D.3

1Department of Gastroenterology, Brooke Army Medical Center, FortSam Houston, San Antonio, TX

2Department of Internal Medicine, Division of Gastroenterology andHepatology, Saint Louis University School of Medicine, St. Louis, MO

3Department of Anatomic and Molecular Pathology, WashingtonUniversity School of Medicine, St. Louis, MO

Copyright © 2009 by the American Association for the Study of Liver Diseases.Published online in Wiley InterScience (www.interscience.wiley.com).DOI 10.1002/hep.22968Potential conflict of interest: Nothing to report.

Hepatitis C Virus Replication in Patients with Occult Hepatitis C Virus Infection

To the Editor:

We read with interest the review by Welker and Zeuzem1 on occulthepatitis C virus (HCV) infection. In this review, the authors claimthat viral replication has not been proven in occult HCV infection.However, the presence of the antigenomic HCV RNA strand has beendemonstrated in liver and in peripheral blood mononuclear cells (PB-MCs) of patients with occult HCV who are negative for anti-HCV andserum HCV RNA2 (and references in Carreno et al.2). There is nodoubt that detection of the antigenomic strand is an indicator of viralreplication as recognized by Welker and Zeuzem in their review.1

Furthermore, ultracentrifugation of serum samples before perfor-mance of real-time reverse transcription polymerase chain reaction(RT-PCR) detects very low levels of viremia in patients with occultHCV infection, and the density of the circulating HCV particles wasidentical to those found in the serum of patients with chronic hepatitisC.2 In addition, family members of patients with occult HCV infec-tion show serological markers of HCV infection even more frequentlythan those of patients with chronic hepatitis C.3 Other authors havereported that low-level persisting HCV remains infectious.4 Finally,HCV-specific CD4� T cell proliferative responses to nonstructural 3and/or 4 proteins have been detected among patients infected withoccult HCV,2 which may not have occurred unless there was ongoingHCV replication, either within the liver or in immune-privileged sites.So, taking all these data into account, one may convincingly assumethat HCV replication takes place in patients with occult HCV infec-tion who no doubt are potentially infectious.

However, Welker and Zeuzem1 review two studies that unconvinc-ingly failed to detect occult HCV infection in cryptogenic hepatitis,5,6

because of several concerns with the methodology employed. Thus,one study5 used whole blood to assess occult HCV infection, but it wasdemonstrated in patients with occult hepatitis C that the sensitivity ofHCV RNA detection substantially decreased when using whole bloodinstead of PBMCs.2 Another study6 included a very small and hetero-geneous population, and these authors did not perform detection ofHCV RNA either in liver or in serum by PCR after ultracentrifuga-tion. Thus, their results do not exclude the existence of occult HCVinfection. On the other hand, an anti-HCV core antibody assay hasbeen developed recently, which adds another diagnostic tool because ithas been proposed as a potential new marker of the serologically silentform of occult HCV infection.7

In contrast, other studies have confirmed the existence of occultHCV infection by detection of HCV RNA in liver.8,9 Furthermore,contrary to what is concluded by Welker and Zeuzem, HCV proteinshave been detected in liver10 and in lymph nodes11 of patients who arenegative for anti-HCV and serum HCV RNA.

In summary, considering the data available, there is cumulativeevidence strongly supporting the idea that HCV replication takes place

in patients with occult HCV infection. Therefore, in order to diagnoseoccult HCV infection among these patients, it is mandatory to per-form appropriate molecular and immunoserological tests in serum,PBMCs, and liver biopsies.

VICENTE CARRENO

JAVIER BARTOLOME

INMACULADA CASTILLO

JUAN ANTONIO QUIROGA

Fundacion para el Estudio de las Hepatitis ViralesMadrid, Spain

References1. Welker MW, Zeuzem S. Occult hepatitis C: how convincing are the cur-

rent data? HEPATOLOGY 2009;49:665-675.2. Carreno V, Bartolome J, Castillo I, Quiroga JA. Occult hepatitis B virus

and hepatitis C virus infections. Rev Med Virol 2008;18:139-157.3. Castillo I, Bartolome J, Quiroga JA, Barril G, Carreno V. Hepatitis C virus

infection in the family setting of patients with occult hepatitis C. J MedVirol. In press.

4. MacParland SA, Pham TN, Guy CS, Michalak TI. Hepatitis C viruspersisting after clinically apparent sustained virological response to antivi-ral therapy retains infectivity in vitro. HEPATOLOGY 2009; doi:10.1002/hep.22802.

5. Tagiuri NI, McClure P, Ryder SD, Irving WL. Occult hepatitis C infec-tion in patients with histological hepatitis and negative serum PCR forHCV [Abstract]. J Hepatol 2006;44(S2):S157.

6. Halfon P, Bourliere M, Ouzan D, Sene D, Saadoun D, Khiri H, et al.Occult hepatitis C virus infection revisited with ultrasensitive real-timePCR assay. J Clin Microbiol 2008;46:2106-2108.

7. Quiroga JA, Castillo I, Llorente S, Bartolome J, Barril G, Carreno V.Identification of serologically silent occult hepatitis C virus infection bydetecting immunoglobulin G antibody to a dominant HCV core peptideepitope. J Hepatol 2009;50:256-263.

8. Esaki T, Suzuki N, Yokoyama K, Iwata K, Irie M, Anan A, et al. Hepato-cellular carcinoma in a patient with liver cirrhosis associated with negativeserum HCV test but positive liver tissue HCV-RNA. Internal Med 2004;43:279-282.

9. Comar M, Dal Molin G, D’Agaro P, Croce SL, Tribelli C, Campillo C.HBV, HCV and TTV detection by in situ polymerase chain reaction couldreveal occult infection in hepatocellular carcinoma: comparison with bloodmarkers. J Clin Pathol 2006;59:526-529.

10. Verslype C, Nevens F, Sinelli N, Clarysse C, Pirenne J, Depla E, et al.Hepatic immunohistochemical staining with a monoclonal antibodyagainst HCV-E2 to evaluate antiviral therapy and reinfection of liver graftsin hepatitis C viral infection. J Hepatol 2003;38:208-214.

2128 CORRESPONDENCE HEPATOLOGY, June 2009