| |
Hepatic Steatosis in Hepatitis C: Comparison of Diabetic and Nondiabetic Patients in the Hepatitis C Antiviral Long-Term Treatment Against Cirrhosis Trial (HALT-C Study)
|
| |
| |
Clinical Gastroenterology & Hepatology Feb 2007
Anna S.F. Lok_, James E. Everhart, Raymond T. Chung, Latha Padmanabhan, Joel K. Greenson, Mitchell L. Shiffman#, Gregory T. Everson__, Karen L. Lindsay, Herbert L. Bonkovsky, Adrian M. Di Bisceglie, William M. Lee, Timothy R. Morgan##, Marc G. Ghany___, Chihiro Morishima
"....In summary, hepatic steatosis is a frequent finding among patients with chronic hepatitis C and advanced fibrosis, even those with HCV non-3 genotype. In this cohort with predominantly genotype 1 infection, hepatic steatosis was associated strongly with metabolic factors that contribute to nonalcoholic fatty liver disease. Our study showed that steatosis correlates with increasing stages of fibrosis up to but not including cirrhosis. Among nondiabetic patients, steatosis had a negative impact on SVR to re-treatment with pegylated IFN and ribavirin. These findings provide a rationale to investigate the impact of therapies directed at hepatic steatosis in slowing the rate of progression of liver disease and in improving SVR to antiviral treatment of chronic hepatitis C. Furthermore, the discordant findings between nondiabetic and diabetic patients indicate that these 2 groups should be considered separately when analyzing metabolic factors and liver disease outcomes...."
ABSTRACT
Background & Aims: Hepatic steatosis often is observed in patients with chronic hepatitis C and has been reported to be associated with hepatic fibrosis and impaired treatment response in some studies. Our aim was to determine the prevalence of and risk factors for hepatic steatosis among Hepatitis C Antiviral Long-term Treatment against Cirrhosis patients, and to determine the relationship between steatosis, fibrosis, and sustained virologic response (SVR) to re-treatment with pegylated interferon and ribavirin.
Methods: Baseline data from 1143 Hepatitis C Antiviral Long-term Treatment against Cirrhosis patients, with a mean body mass index of 30, 5% with genotype 3, 38% with cirrhosis, and 24% with diabetes were analyzed.
Results:
Steatosis scores of 0, 1, 2, 3, and 4 were observed in 19%, 42%, 30%, 8%, and 1% of patients, respectively.
High body mass index, triglyceride and alanine aminotransferase levels, and genotype 3 were associated with higher grades of steatosis.
Among nondiabetic patients, steatosis scores of 0-2 but not scores of 3-4 were associated significantly with cirrhosis. For diabetic patients, there was no association between steatosis and cirrhosis.
Similarly, steatosis scores of 2-4 were associated with a lack of SVR among nondiabetic but not among diabetic patients.
Conclusions:
In this cohort with predominantly hepatitis C virus genotype 1 infection, steatosis was associated strongly with metabolic factors that contribute to nonalcoholic fatty liver disease.
Steatosis correlated with increasing stages of fibrosis up to but not including cirrhosis. Steatosis had a negative impact on SVR among nondiabetic but not diabetic patients. The discordant findings between nondiabetic and diabetic patients indicate that these 2 groups should be considered separately when analyzing metabolic factors and liver disease outcomes.
Background
Hepatic steatosis is a commonly observed histologic feature of chronic hepatitis C, being found in 40%-70% of patients.1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14 Steatosis is found more often in patients with hepatitis C virus (HCV) genotype 3: 70%-90% vs 30%-60% in those with genotype 1.3, 8, 11, 15 Several studies have implicated a direct viral effect of genotype 3 on hepatic steatosis,11, 16, 17, 18, 19 whereas risk factors for hepatic steatosis in genotype 1 overlap with those observed for nonalcoholic fatty liver disease.4, 6, 7, 8, 11 However, many of these studies involved small numbers of patients and not all studies examined metabolic risk factors prospectively.
The clinical implications of hepatic steatosis among patients with chronic hepatitis C also are uncertain. Obesity, steatosis, and, more recently, insulin resistance (IR) have been reported to be associated with more advanced hepatic fibrosis.2, 6, 7, 8, 9, 10, 14, 20, 21, 22, 23, 24, 25 However, the association between hepatic steatosis and fibrosis has not been a universal finding, and many studies have not evaluated multiple potential risk factors.3, 4, 12, 26, 27 Furthermore, most studies have not had a sufficiently large number of patients with advanced liver disease to confirm a relationship between hepatic steatosis and cirrhosis, and only a few studies have examined the effect of steatosis on fibrosis progression using paired biopsy specimens.10, 21 Thus, the role of hepatic steatosis in relation to other metabolic factors in fibrosis progression in patients with chronic hepatitis C is still unclear.
Hepatic steatosis also has been reported to be associated with a lower rate of sustained virologic response (SVR) to interferon (IFN) and ribavirin therapy in some but not all studies of treatment-naive patients.3, 11, 28, 29, 30 These mixed findings may be related to varying proportions of patients with steatosis and other more established predictors of response such as HCV genotype in each study.
The Hepatitis C Antiviral Long-term Treatment against Cirrhosis (HALT-C) Trial is a prospective, randomized, controlled study designed to determine whether long-term pegylated IFN therapy can reduce the risk of progression to cirrhosis, decompensated liver disease, and/or hepatocellular carcinoma in patients with chronic hepatitis C who have advanced fibrosis or cirrhosis.31 The study included a large lead-in arm in which 1145 patients who had not previously responded to standard IFN treatment were re-treated with pegylated IFN and ribavirin, the best current therapy, to exclude patients who might achieve virologic response from entering the randomized study.32 Data obtained before treatment and the response to lead-in therapy were analyzed for the current report. The aims of this analysis were as follows: (1) to determine the prevalence of hepatic steatosis among HALT-C patients at entry into the study and to identify the risk factors associated with hepatic steatosis, (2) to determine whether hepatic steatosis is associated with increased hepatic fibrosis independent of other metabolic factors, and (3) to determine whether hepatic steatosis is associated independently with a lower rate of SVR to pegylated IFN and ribavirin therapy in previous nonresponder patients. The prospective assessment of a broad spectrum of metabolic factors among a large cohort of patients permitted the independent evaluation of these factors including diabetes.
Patients and Methods
The HALT-C trial was conducted in 10 clinical centers in the United States. Details of the study design and entry criteria have been reported previously.31 The study protocol was approved by the institutional review board of each participating center, and written consent was obtained from all patients.
Patient Population
Entry criteria included the presence of antibody to hepatitis C virus (anti-HCV) and HCV RNA in serum, failure to respond to the most recent treatment of standard IFN with or without ribavirin, and the presence of bridging fibrosis or cirrhosis (Ishak fibrosis score, 3-6) on a liver biopsy examination performed within 12 months of enrollment. All the entry biopsy examinations were performed at least 2 months after completion of the prior course of therapy; 92% of the biopsy examinations were performed more than 6 months after discontinuation of treatment. Patients with other co-existent liver disorders and those with hepatic decompensation were excluded. Patients with steatosis alone or mild-moderate steatohepatitis were enrolled but those with severe steatohepatitis, defined as the presence of marked steatosis, many Mallory bodies, and extensive zone 3 pericellular fibrosis, were excluded. Patients with poorly controlled diabetes and patients with active alcohol abuse within the past 12 months also were excluded.
Baseline Evaluation
Baseline evaluation included a complete history, physical examination (including height, weight, and waist circumference), review of historical information regarding prior treatment, assessment of lifetime alcohol and cigarette use and physical activity, an abdominal ultrasound, laboratory tests, and a liver biopsy examination.
Baseline blood tests included complete blood counts, liver panel, basic metabolic panel, prothrombin time/international normalized ratio, fasting glucose level, triglyceride level, insulin level, -fetoprotein level, HCV genotype, quantitative HCV RNA level, thyroid-stimulating hormone level, and tests to exclude other causes of liver disease. Assays for HCV genotype and HCV RNA were performed at a single laboratory (University of Washington, Seattle, WA), as described previously.31 Insulin levels were tested using a radioimmunoassay33 in a single laboratory (Michigan Diabetes Research and Training Center, University of Michigan, Ann Arbor, MI). All other blood tests were performed at the hospital laboratories of the participating clinical centers.
The body mass index (BMI) was calculated using the following formula: weight in kilograms/height in meters2. Patients were considered to be overweight if their BMI was between 25 and 29, and obese if their BMI was 30 or higher. Patients were considered to be diabetic if they had a current diagnosis of diabetes or if their fasting blood glucose level exceeded 126 mg/dL.34 IR was estimated by the homeostasis model assessment (HOMA): fasting insulin (U/mL) ~ fasting glucose (mmol/L)/22.5.35
Recreational physical activity was obtained by self-report from a frequency questionnaire that was adapted from the third National Health and Nutrition Examination Survey, coded by metabolic equivalent (MET) levels, and grouped as minimal (<15 MET/wk), light (15-29 MET/wk), moderate (30-79 MET/wk), or vigorous (80 MET/wk).33, 36, 37 Self-reported nonrecreational activities were categorized as light, moderate, or vigorous.
Interpretation of Liver Histology
Baseline liver biopsy specimens were reviewed in conference by a committee of 11 study hepatic pathologists who were masked to patient data. The modified Histology Activity Index, also known as the Ishak score, was used for grading inflammation (grades, 0-18) and for staging fibrosis (stages, 0-6).38 Fibrosis was evaluated with the Masson trichrome stain, inflammation and steatosis with the H&E stain, and iron with Perls' stain. Hepatic steatosis was graded as follows: 0 (<1%), 1 (1%-5%), 2 (6%-33%), 3 (34%-67%), and 4 (>67%) according to the percentage of hepatocytes with fat. Hepatocellular iron was graded 0 to 4+, with 4+ representing massive iron deposition in all parts of the hepatic lobule, 1+ as a small amount of staining in periportal hepatocytes only, whereas 2+ and 3+ had intermediate amounts of iron staining. All the histologic scores were determined by a consensus of the committee members or in cases of divided opinions by a vote of the majority.
Treatment Protocol and Definition of Sustained Virologic Response
All patients initially were re-treated with peginterferon alfa-2a (Pegasys; Roche Laboratories, Nutley, NJ) 180 mcg/wk and ribavirin (Copegus, Roche) 1000-1200 mg/day for 24 weeks. Patients with undetectable HCV RNA levels at week 20 continued to receive combination therapy until week 48. SVR was defined as undetectable HCV RNA levels in serum at week 72. HCV RNA was tested using the COBAS Amplicor HCV Test, v. 2.0 assay (Roche Molecular Systems, Branchburg, NJ), which has a detection limit of 100 IU/mL.
Statistical Analyses
Baseline data from all patients enrolled into the lead-in phase of the HALT-C trial were analyzed. The χ2 and t tests were used to identify variables that were significantly different between the different grades of steatosis (0-1 vs 2-4), cirrhosis vs no cirrhosis (Ishak fibrosis score 5-6 vs 2-4), and SVR vs non-SVR. The χ2 test and analysis of variance were used to compare the 4 grades of steatosis. Logistic regression with both backward and forward selection methods were performed; variables that were significant in either of the selection methods were introduced into the prediction models. Analyses were performed for the entire cohort, followed by separate analyses of nondiabetic and diabetic patients. Statistically significant cofactors in the full analysis were included in the subanalyses. In the logistic regression, the odds ratios for all numeric variables represent unit increases with the exception of platelets, international normalized ratios, and triglyceride levels, which were rescaled as follows: platelets, unit increases of 50,000/mm3; international normalized ratios, unit increases of 0.1; and triglyceride levels, unit increases of 10 mg/dL.
Results
Baseline Characteristics of the Patients
A total of 1143 patients were included in this analysis, 989 (87%) of the biopsy specimens had more than 6 portal triads. Two patients were excluded because the biopsy specimen quality was not adequate for grading steatosis. The population included a broad age range (range, 19-80 y) and large numbers of black (n = 174) and Hispanic (n = 95) minorities and women (n = 318) (Table 1). The vast majority (89%) was infected with HCV genotype 1; only 5% were infected with genotype 3. A total of 214 (19%) patients admitted they were still drinking at the time of entry into the study. Of these, 70% were having 1-2 drinks a week, 20% were having 3-6 drinks a week, and 11% were having 1 or more drinks/day. Baseline biopsy specimens revealed cirrhosis (Ishak fibrosis score, 5-6) in 38% of patients.
Most patients were overweight (40%) or obese (42%). Insulin resistance and hypertriglyceridemia were correspondingly common. A total of 271 (24%) patients met criteria for diabetes, 191 (70%) were known to have diabetes before entering the HALT-C trial, 52 (19%) patients were receiving insulin, and 92 (34%) were receiving oral hypoglycemics (only 4 patients were taking thiazolidinediones). Patients with diabetes were more likely to be older, black, have a poorer metabolic profile, and more severe fibrosis and steatosis (Table 1). Physical activity level did not differ between patients with or without diabetes.
Prevalence of Hepatic Steatosis and Factors Associated With Hepatic Steatosis
Fat was present in more than 5% of hepatocytes in 451 (39%) patients; 19%, 42%, 30%, 8%, and 1% patients had hepatic steatosis scores of 0, 1, 2, 3, and 4, respectively. Because of the small number of patients with a steatosis score of 4, patients with scores of 3 and 4 were combined in subsequent analyses. Diabetic patients were more likely to have high steatosis scores. The percent of nondiabetic and diabetic patients with steatosis scores of 0, 1, 2, 3-4 were 21% vs 13%, 43% vs 39%, 29% vs 35%, and 8% vs 13%, respectively (P < .0001) (Figure 1). Among diabetic patients, blacks had less steatosis than patients of other races; the percentages of blacks, whites, Hispanics, and others with steatosis scores greater than 2 were 34%, 53%, 54%, and 50%, respectively (P = .03). Among nondiabetic patients, steatosis was more common among Hispanics (52%) than among non-Hispanic blacks, whites, and other ethnic groups (34%, 36%, and 27%, respectively) (P = .02), although the number of Hispanic patients was small. There was no association between steatosis and drinking at enrollment or average daily alcohol consumption during the 6 months before enrollment. The percentage of patients with and without alcohol consumption during the 6 months before enrollment with steatosis scores of 0, 1, 2, and 3-4 were 20% vs 15%, 41% vs 44%, 30% vs 31%, and 9% vs 10%, respectively (P = .48).
Increasing grades of hepatic steatosis were correlated significantly with BMI, waist circumference, blood pressure, glucose levels, insulin levels, triglyceride levels, IR (HOMA), and diabetes (Table 2). Although very few patients were infected with HCV genotype 3, there was a striking association between genotype 3 and higher grades of steatosis. Genotype 3 was present in 1%, 4%, 5%, and 12% of patients with steatosis scores of 0, 1, 2, and 3-4, respectively (P = .0007). Increasing grades of steatosis were associated with higher levels of aspartate aminotransferase (AST) and alanine aminotransferase (ALT), and alkaline phosphatase, and higher scores of hepatic inflammation (P = .0036) and fibrosis (P = .0001) but lower grades of hepatocellular iron (P = .0046). Hepatic steatosis decreased with increasing age and HCV RNA level, regardless of HCV genotype. Among the diabetic patients, increasing grades of steatosis were associated with increasing values of glycosylated hemoglobin (HbA1c) (P = .035) but only modestly with diabetes treatment (P = .082) (Table 2).
For multivariate logistic regression analysis, all variables that were found to be statistically significant (P < .05) were evaluated, and steatosis grades were condensed into 2 groups: 0-1 (<5% hepatocytes with fat) and 2-4. BMI, triglyceride level, ALT level, and genotype 3 were associated with higher grades of hepatic steatosis; whereas HCV RNA level and hepatic iron grade were associated with less hepatic steatosis (Table 3). The association between diabetes and steatosis was not significant in the multivariate analysis, with an odds ratio of 1.28, and a 95% confidence interval of 0.95-1.72 (P = .11). A significant association between hepatic steatosis and BMI, triglyceride level, and ALT level was present in nondiabetic and diabetic patients. HOMA IR was associated with steatosis in univariate analysis (Table 2), but not in multivariate analysis among either the entire cohort or the nondiabetic or diabetic subgroups.
Correlation Between Hepatic Steatosis and Hepatic Fibrosis
Although there was a significant correlation between hepatic steatosis and hepatic fibrosis, the relation was not linear. Ishak fibrosis and hepatic steatosis scores increased in parallel for patients with no to moderate steatosis (scores, 0-2), but the mean fibrosis score and the percentage of patients with cirrhosis were lower among patients with severe steatosis (scores, 3-4) (Table 2, Figure 2). The percentage of black and white patients with cirrhosis was similar for both diabetic (38% vs 46%; P = .29) and nondiabetic patients (31% vs 35%; P = .50).
Because of the nonlinear relation between steatosis and fibrosis, individual grades of steatosis were analyzed in the logistic regression analysis for cirrhosis. Truncal obesity (waist circumference) and mild to moderate steatosis (scores, 1-2) were significant predictors of cirrhosis (Table 4). When diabetes was forced into the model there was a trend for an association with cirrhosis with an odds ratio of 1.38 (95% confidence interval, 0.98-1.95; P = .068). Other variables significantly associated with cirrhosis included low platelet count, high AST/ALT ratio, high alkaline phosphatase level, increased international normalized ratio, and high inflammatory score. Among nondiabetic patients, mild-moderate steatosis (scores, 1-2) was associated significantly with cirrhosis but severe steatosis (scores, 3-4) was not (Table 4). For diabetic patients, there was no association between steatosis and cirrhosis, regardless of severity (Table 4). Of the metabolic factors examined, including BMI, diabetes, truncal obesity, and HOMA IR, only truncal obesity was associated significantly with cirrhosis among nondiabetic patients; none of the metabolic factors was associated with cirrhosis among diabetic patients (Table 4).
Correlation Between Hepatic Steatosis and Sustained Virologic Response
A total of 180 (15.7%) patients achieved SVR. Logistic regression analysis identified prior treatment with the combination of interferon and ribavirin, high HCV RNA level, genotype 1, high fibrosis score, high AST/ALT ratio, hypoalbuminemia, and severe steatosis (grades 3-4) associated with lack of SVR (Table 5). Of interest, African American ethnicity was associated with a lower rate of SVR in univariate but not in multivariate analyses. Moderate (grade 2) and severe steatosis (grades 3-4) were associated with a lack of SVR among nondiabetic patients (P = .02 and .01, respectively) (Table 5), but steatosis regardless of grade was not associated with SVR among diabetic patients. Other metabolic factors including BMI (P = .78), HOMA IR (P = .59), and diabetes (P = .96) were not associated with SVR. Among diabetic patients, there was no relation between SVR and diabetes treatment, the rates of SVR were 12%, 13%, and 13% for patients on diet control, oral hypoglycemics, and insulin, respectively. Patients who admitted to drinking at enrollment had a similar rate of SVR as those who had abstained: 19% vs 15%.
Discussion
In this study, we sought to determine the prevalence of hepatic steatosis among HALT-C patients at entry, and the relation between hepatic steatosis and hepatic fibrosis and the response to antiviral therapy. The HALT-C study offers the advantage of a comprehensive analysis of a large cohort of HCV-infected patients, careful attention to histologic evaluation, and an opportunity to examine multiple risk factors simultaneously. Information on risk factors was collected prospectively by trained study coordinators using standardized questionnaires. All the liver biopsy specimens were scored for steatosis, iron deposition, inflammation, and fibrosis by a panel of liver pathologists based on consensus or majority vote. Because of the large number of patients with cirrhosis (n = 429) and diabetes (n = 271), an accurate assessment of correlations between steatosis and cirrhosis among nondiabetic and diabetic patients was possible in this cohort.
Despite a very high proportion of patients with genotype 1 and a low (5%) representation of HCV genotype 3, we found a high prevalence of hepatic steatosis among patients enrolled in the HALT-C trial, with 39% having more than 5% hepatocytes with fat (steatosis scores, >2). The prevalence of hepatic steatosis in the HALT-C cohort is similar to other reports on patients with genotype 1 infection.3, 8, 11, 15
All of the metabolic factors traditionally associated with hepatic steatosis including obesity, truncal obesity, IR, diabetes, and hypertriglyceridemia were associated significantly with the degree of steatosis on baseline biopsy specimens. Our findings are in accord with other studies, indicating that risk factors for hepatic steatosis in patients with genotype 1 infection are similar to those observed in nonalcoholic fatty liver disease.4, 6, 7, 8, 11 Furthermore, we showed that diabetic patients with hepatitis C had more marked steatosis than nondiabetic patients, and among diabetic patients there was a correlation between glycemic control as reflected by HbA1c and steatosis score. The lack of an association between HOMA-IR and steatosis was unexpected and may be related to the high proportion of patients with cirrhosis. Patients with cirrhosis may have high HOMA-IR owing to impaired carbohydrate metabolism in the liver and/or peripheral hyperinsulinemia and not metabolic syndrome. In this study, we found that HOMA-IR was significantly higher in nondiabetic patients with cirrhosis compared with those without cirrhosis: 10.4 vs 9.5 (P < .0001). In addition, steatosis regresses as liver disease progresses to cirrhosis. Similar to other studies, we found a strong association between steatosis and HCV genotype 3 despite the small number of patients with this genotype. A small (19%) percentage of our patients admitted to drinking at enrollment, but these patients did not have higher steatosis scores, possibly because of the low amount of alcohol consumption (only 11% had >1 drink/day).
We showed an association between hepatic steatosis and fibrosis but the relation was not linear, with Ishak fibrosis and hepatic steatosis scores increasing in parallel for patients with mild to moderate steatosis (scores, 0-2), but the fibrosis score decreased in patients with severe steatosis (scores, 3-4). These findings support other reports that steatosis increases the risk of fibrosis,6, 7, 8, 14, 21, 22, 24 but the relationship ceases once cirrhosis has developed.4 The basis for the apparent regression of steatosis in cirrhosis is unclear but could be related to decreased access of lipoproteins or fat-storing signals as a result of portosystemic shunting and/or sinusoidal capillarization.14, 39, 40 Similar observations have been reported in patients with cirrhosis as a result of fatty liver and may account for the inability to confirm fatty liver as a definitive cause among patients presenting with cryptogenic cirrhosis.14, 41, 42 Our findings suggest that steatosis may have greater use as a predictive tool for fibrosis progression in longitudinal studies of patients with mild fibrosis on initial biopsy examination.
Multivariate analysis identified markers of portal hypertension, impaired hepatic synthetic function, and increased hepatic inflammation as significant predictors of cirrhosis. Among diabetic patients, neither steatosis nor any of the metabolic factors analyzed was associated with cirrhosis, whereas among nondiabetic patients mild-moderate steatosis and truncal obesity were associated significantly with cirrhosis. The lack of association between steatosis and metabolic factors and cirrhosis among diabetic patients may be related to the high prevalence of obesity, IR, and hypertriglyceridemia in these patients. By contrast, nondiabetic patients had a wider range of BMI and metabolic function, enabling us to evaluate patients with a normal vs abnormal metabolic profile.
In an earlier report of the first 604 patients in the HALT-C trial, factors associated with an SVR included previous treatment with IFN monotherapy, infection with genotypes 2 or 3, a lower AST/ALT ratio, and absence of cirrhosis.32 The current analysis included all patients enrolled in the lead-in phase of the HALT-C trial (n = 1110), and took into account hepatic steatosis and metabolic factors. In accordance with other studies on treatment-naive patients,3, 11, 28 we found that hepatic steatosis was an independent predictor of a lack of SVR during re-treatment of nonresponders. However, the effect of steatosis on SVR was observed only among the nondiabetic patients. Contrary to other reports29, 30 we did not find an association between other metabolic factors such as obesity, diabetes, and IR and SVR. This may be related to the strong relationship among obesity, insulin resistance, and hepatic steatosis. Alternatively, our study of nonresponders with advanced liver disease may have been selected for patients who are mostly obese (mean BMI, 29.7) with IR (mean HOMA IR, 15.6).
Our findings of an association between hepatic steatosis and fibrosis and lack of SVR among the nondiabetics are similar to a recent report on nondiabetic European patients with genotype 1 HCV infection, although the patients in the latter study had a lower BMI and no prior IFN therapy.43
In summary, hepatic steatosis is a frequent finding among patients with chronic hepatitis C and advanced fibrosis, even those with HCV non-3 genotype. In this cohort with predominantly genotype 1 infection, hepatic steatosis was associated strongly with metabolic factors that contribute to nonalcoholic fatty liver disease. Our study showed that steatosis correlates with increasing stages of fibrosis up to but not including cirrhosis. Among nondiabetic patients, steatosis had a negative impact on SVR to re-treatment with pegylated IFN and ribavirin. These findings provide a rationale to investigate the impact of therapies directed at hepatic steatosis in slowing the rate of progression of liver disease and in improving SVR to antiviral treatment of chronic hepatitis C. Furthermore, the discordant findings between nondiabetic and diabetic patients indicate that these 2 groups should be considered separately when analyzing metabolic factors and liver disease outcomes.
_ Division of Gastroenterology, University of Michigan Medical Center, Ann Arbor, Michigan
Department of Pathology, University of Michigan Medical Center, Ann Arbor, Michigan
Department of Health and Human Services, Division of Digestive Diseases and Nutrition, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, Maryland
___ Liver Diseases Branch, Division of Digestive Diseases and Nutrition, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, Maryland
Gastrointestinal Unit (Medical Services), Massachusetts General Hospital and the Department of Medicine, Harvard Medical School, Boston, Massachusetts
New England Research Institutes, Watertown, Massachusetts
# Hepatology Section, Virginia Commonwealth University Health System, Richmond, Virginia
__ Section of Hepatology, Division of Gastroenterology and Hepatology, University of Colorado School of Medicine, Denver, Colorado
Division of Gastrointestinal and Liver Diseases, Keck School of Medicine, University of Southern California, Los Angeles, California
Liver-Biliary-Pancreatic Center and Departments of Medicine and Molecular & Structural Biology, University of Connecticut Health Center, Farmington, Connecticut
Division of Gastroenterology and Hepatology, St. Louis University School of Medicine, St. Louis, Missouri
Division of Digestive and Liver Diseases, University of Texas Southwestern Medical Center, Dallas, Texas
## Division of Gastroenterology, University of California\Irvine, Irvine, California
Department of Laboratory Medicine, University of Washington, Seattle, Washington
|
|
| |
| |
|
|
|
