| |
Hepatic steatosis and antiretroviral drug use among adults coinfected with HIV and hepatitis C virus
|
| |
| |
AIDS: Volume 19(6) 8 April 2005 p 585-592
Sulkowski, Mark Sa; Mehta, Shruti Hb; Torbenson, Michaela; Afdhal, Nezam Hc; Mirel, Lisab; Moore, Richard Da; Thomas, David La,b
From the aDivision of Infectious Diseases Johns Hopkins School of Medicine, Baltimore, Maryland
bDepartment of Epidemiology, Johns Hopkins School of Public Health, Baltimore, Maryland
cDepartment of Medicine, Beth Israel Deaconess Medical Center, Boston, Massachussetts, USA.
Abstract
Objective: To ascertain the prevalence and severity of hepatic steatosis among patients coinfected with HIV and hepatitis C virus (HCV) who had been taking antiretroviral therapy (ART); to investigate if steatosis is associated with more advanced liver disease, and to identify factors that might contribute to the process.
Methods: Steatosis was assessed among a randomly selected subset of HIV-HCV-coinfected patients who had received at least 2 years of ART in a cohort study at the Johns Hopkins University HIV clinic. Liver histology was evaluated by a single pathologist. The primary outcome measure was the hepatic steatosis grade, which was classified on a 5 point scale: 0, none; 1, steatosis involving < 5% of hepatocytes; 2, 5-29%; 3, 30-60%; 4 > 60%.
Results: Liver histology was assessed in 112 patients, 74% of whom were taking ART at the time of biopsy. The median cumulative exposure to nucleoside reverse transcriptase inhibitors and protease inhibitors was 5.8 and 3.7 years, respectively. No steatosis was detected in 60%; grades 2-4 steatosis was recognized in 18%. In multivariate analysis, steatosis was independently associated with Caucasian race, weight > 86 kg, hyperglycemia, and stavudine use. Patients with steatosis also were more likely to have greater hepatic fibrosis (P = 0.02) and necroinflammatory activity (P = 0.005).
Conclusions: Steatosis was observed in 40% of HIV-HCV-coinfected patients with extensive ART exposure and was associated with more severe HCV-related liver disease. Metabolic abnormalities (excess weight and hyperglycemia) and stavudine use were modifiable risk factors for steatosis in this population.
AUTHOR DISCUSSION
By examining liver tissue from a random sample of ART-experienced, HIV-HCV-coinfected patients, this study provides an important, and possibly the first, estimate of the overall prevalence and determinants of hepatic steatosis in this population. The findings are an important contribution to recent research. In individuals with HIV infection, ART has been strongly linked with a spectrum of metabolic abnormities including diabetes mellitus and hyperlipidemia [15,23,24]. Concurrently, studies conducted in individuals without HIV infection have shown that chronic HCV infection is linked to diabetes mellitus (and other manifestations of insulin resistance), as well as to hepatic steatosis, and that steatosis appears to accelerate progression to cirrhosis and reduce the response to interferon-alfa treatment [17,25-27]. Therefore, several findings are particularly noteworthy.
The majority (60%) of the ART-experienced, HIV-HCV-coinfected individuals in our study had no histological evidence of hepatic steatosis despite prolonged exposure to ART. In addition, the prevalence of steatosis among individuals currently taking ART was similar to that among individuals not taking ART at the time of the biopsy. Hepatic steatosis has previously been observed in liver biopsies from HIV-infected individuals taken before ART was available [6]. While comparable data among HIV-HCV-coinfected individuals are not available, this finding is consistent with previous estimates of the prevalence of hepatic steatosis in HCV-monoinfected individuals referred for medical evaluation [28,29]. For example, Patton et al. [27] observed no hepatic steatosis in 52% of 574 liver biopsies from HCV-infected individuals. Similarly, no hepatic steatosis was noted in 45% of 1428 HCV-infected individuals enrolled in a HCV treatment trial [30] and 53% of 290 French patients with chronic HCV infection [28].
There are a number of potential explanations for a lower than expected prevalence of steatohepatitis in this study. There was a relatively low prevalence of other factors that contribute to steatohepatitis, such as high body mass, HCV genotype 3 infection, and Caucasian race. In addition, too little is known about the dynamics of ART-related steatohepatitis to know how long it takes to occur or how long it persists once ART is stopped. Nonetheless, taken together, these data suggest that the prevalence of hepatic steatosis in HIV-HCV-coinfected patients is not markedly increased by ART in this setting.
Nonetheless, hepatic steatosis was observed in 40% of HIV-HCV-coinfected patients, and nearly 18% had evidence of significant steatosis ( 5% of hepatocytes affected). Since individuals in this study with hepatic steatosis were also more likely to have bridging fibrosis or cirrhosis, it is important to consider whether the association is causal and, more importantly, modifiable. Although it is possible that this relationship reflects another linked exposure (such as alcohol use), no such confounding was detected (steatosis was not associated with past and/or current alcohol use in our population). Furthermore, these data are consistent with those derived from studies of HIV-negative patients, in whom hepatic steatosis was been associated with more advanced fibrosis [20,30].
Higher body weight and hyperglycemia were associated with hepatic steatosis in this investigation, and insulin resistance is common to both [26]. In obese individuals, weight loss has been shown to reduce insulin resistance and reduce hyperglycemia [31]. Therefore, it is tempting to speculate that this and other measures aimed at restoring insulin sensitivity might reduce hepatic steatosis and, in turn, delay progression of hepatic fibrosis.
Nearly 90% of patients with hepatic steatosis had been exposed to a PI and stavudine, and nearly 50% of individuals exposed to these agents had steatosis. After adjusting for potential confounders, exposure to stavudine was associated with a more than 5-fold higher risk of hepatic steatosis. Stavudine has been closely linked with hepatic steatosis (and to lactic acidosis) in other studies, presumably through inhibition of mitochondrial polymerase gamma and depletion of mitochondrial DNA [15,32,33]. Recently, among asymptomatic patients, Walker et al. [32] observed that exposure to stavudine, but not other NRTI, was associated with depletion of hepatic mitochondrial DNA. These data, and the parallel in vitro studies showing greater mitochondrial inhibition with stavudine compared with other NRTI, suggest that stavudine should be avoided in HIV-HCV-coinfected patients with hepatic steatosis [34].
Our data also suggest that PI use also may contribute to hepatic steatosis, particularly in individuals taking stavudine. The PI drugs have been associated with inhibition of insulin-mediated glucose uptake, leading to decrease insulin sensitivity and, in some patients, to the development of diabetes mellitus [24]. In addition, with the exception of atazanavir, PI drugs have been associated with substantial increases in plasma lipids, through induction of hepatic fatty acid and sterol biosynthesis in liver tissues [24,35]. Interestingly, in ritonavir-treated mice, Riddle et al. [35] demonstrated induction of hepatic fatty acid biosynthesis, rapidly leading to accumulation of triglycerides within hepatocytes (i.e., hepatic steatosis). However, the present study was limited by a relatively small number of PI-naive subjects and the lack of data on fasting serum insulin and lipid levels and lipodystrophy.
Since we were interested in the effects of ART on liver disease, we restricted our random sample to patients who had greater than 2 years of exposure to ART, a measure that may limit the relevance of these data to individuals with less or no ART exposure. Additional research will be needed to determine the prevalence and severity of steatosis among patients who are naive to ART, although such patients may also differ from those included in our study with respect to duration and severity of HIV infection. In addition, future research on ART and steatohepatitis must account for race, HCV genotype, body mass index, alcohol use, and other potential cofactors [36]. It will also be important in future studies to examine steatohepatitis by repeated liver biopsies, since our findings were based on longitudinal evaluation of ART but only a single liver biopsy result.
In conclusion, we observed hepatic steatosis in 40% of randomly selected, HIV-HCV-coinfected subjects with extensive ART exposure. Hepatic steatosis was associated with bridging fibrosis and/or cirrhosis, underscoring the potential contribution of steatosis to the pathogenesis of HCV-related liver disease in coinfected individuals. If confirmed, our findings suggest that weight reduction and discontinuation of stavudine should be investigated as measures that could reduce hepatic steatosis in affected or at-risk individuals.
Introduction
Liver disease is an important cause of morbidity and mortality in those infected with HIV-1 [1-3]. Indeed, since the advent of effective antiretroviral therapy (ART), hospitalizations for complications of end-stage liver disease have increased significantly, and, in some settings, liver disease has emerged as the leading cause of death among HIV-infected individuals [2,3]. Effective ART has also changed the types of liver disease that occur in HIV-infected individuals. While hepatic manifestations of opportunistic infections are now rare, chronic hepatitis C virus (HCV) infection is found in approximately one-third of HIV-infected individuals [4-6]. Furthermore the severity of HCV-related liver disease is greater in those infected with HIV than in those uninfected [7,8]. Among HIV-HCV-coinfected individuals, advanced immunosuppression has been associated more severe liver disease, supporting the hypothesis that reversal or prevention of immunosuppression with ART will slow the progression of HCV disease [9-11].
However, effective ART use may also contribute to liver disease. Grade 3-4 elevations in liver enzymes occur in 5-15% of individuals taking a new ART regimen, and more often in those coinfected with HCV [12,13]. Although ART may cause hepatitis through multiple mechanisms, the impact of nucleoside analogue reverse transcriptase inhibitors (NRTI) on hepatic mitochondria has been particularly severe in some cases leading to fulminant hepatomegaly and hepatic steatosis [14-16]. Furthermore, most HIV-1 protease inhibitors (PI) have been associated with metabolic abnormalities, hyperlipidemia, and decreased insulin sensitivity, which themselves have been linked to hepatic steatosis [17,18]. Among HCV-infected individuals without HIV, hepatic steatosis has been associated with an increased risk of HCV-related liver disease progression and decrease responsiveness to HCV therapy [17,19,20].
Despite its potential clinical importance, the actual prevalence of hepatic steatosis is not known in HIV-HCV-coinfected patients, nor is its relationship with ART, ART-associated metabolic abnormalities, and HCV-related liver disease. Accordingly, the objectives of this study were to ascertain the prevalence and severity of hepatic steatosis among ART-experienced, HIV-HCV-coinfected patients, to investigate if steatosis is associated with greater liver disease, and to identify factors that might contribute to the process.
Results
Study population
The median age was 38 years; 64% were male, 94% were African-American and median duration of HCV infection was ` 24 years. The median weight was 75 kg (165 lb) and 26% weighed more than 86 kg (190 lb). Hyperglycemia (glucose > 2.0 g/l) and hypercholesterolemia (cholesterol > 2.0 g/l) had been observed in 28% and 22% of patients, respectively. The majority of patients were infected with HCV genotype 1 (99%) and one patient had HCV genotype 3. Although few patients reported current alcohol use, 46% had been previously diagnosed with alcoholism or alcohol abuse by their medical provider. At the time of biopsy, 83 (74%) were taking ART; of whom, 74 (89%) received either a non-nucleoside analogue reverse transcriptase inhibitor (NNRTI)- or PI-based regimen. The median cumulative exposure to NRTI and PI was 5.8 and 3.7 years, respectively. Stavudine use was documented in 27 (24%) at the time of liver biopsy and 96 (86%) had previously used stavudine-containing ART. Among the 29 patients (26%) not receiving ART at the time of liver biopsy, the median time off ART was 1.23 years [interquartile range (IQR), 0.53 - 2.59].
Prevalence and severity of hepatic steatosis
No hepatic steatosis was detected in 67 (60%) patients. Twenty-five patients (22%) had fat in < 5% of hepatocytes; 14 (13%) had fat in 5-29% of hepatocytes, and six subjects had fat in > 30% of hepatocytes (none had clinical signs or symptoms of lactic acidosis syndrome or hepatomegaly prior to or at the time of liver biopsy). Among the subset of patient not receiving ART at the time of liver biopsy, the prevalence of steatosis was 44% (13/29), which was similar to the prevalence observed among those receiving ART [38.5% (32/83)]. Bridging fibrosis or cirrhosis (MHAI F3) was present in 16 of 45 (36%) patients with hepatic steatosis compared with 11 of 67 (16%) of those without steatosis (P = 0.02) (Fig. 2a). In addition, subjects with steatosis had evidence of greater hepatic necroinflammatory activity. Twelve individuals with no hepatic steatosis (18%) had MHAI scores > 5 compared with 19 of those with steatosis (42%) (P = 0.005).
Determinants of hepatic steatosis
In univariate analysis, hepatic steatosis was associated with weight > 86 kg [odds ratio (OR), 2.8; 95% CI, 1.2-6.6] and hyperglycemia (OR for any glucose level > 2.0 g/l, 2.8; 95% CI, 1.2-6.5) (Table 2). Hepatic steatosis was also marginally associated with Caucasian race (OR, 4.06; 95% CI, 0.75-21.9), ever being exposed to stavudine (OR, 3.4; 95% CI, 0.9-12.6), cumulative exposure to PI (OR per year of exposure, 1.2; 95% CI, 1.0-1.44). No association was detected with hepatic steatosis and age, gender, past alcohol abuse, current alcohol use, HCV viral load, HCV genotype, CD4 cell count, serum total cholesterol level (P > 0.05). In multivariate analysis, independent determinants of hepatic steatosis were Caucasian race [adjusted OR (AOR), 11.2; 95% CI, 1.6-79.2], weight > 86 kg (AOR, 3.2; 95% CI, 1.2-8.2), hyperglycemia (OR for any glucose level > 2.0 g/l, 3.4; 95% CI, 1.4-8.5), and ever being exposed to stavudine (AOR, 5.1; 95% CI, 1.1-22.9).
Since in the multivariate model, stavudine use (and not PI use) remained associated with hepatic steatosis and because individuals were often exposed to both stavudine and a PI, combinations of these drugs were examined (Fig. 3). Hepatic steatosis was not detected in any subject who never received stavudine or a PI. Among those who were never exposed to stavudine, only three (19%) had evidence of hepatic steatosis compared with 42 (42%) of those exposed to stavudine. By comparison, 39 (46%) of 85 individuals exposed to a PI and stavudine had evidence hepatic steatosis. In fact, 39 of 45 (87%) cases of hepatic steatosis occurred in this group.
Methods
Study population and sample selection
The population for this study derived from HIV-HCV-coinfected members of the Johns Hopkins University (JHU) HIV clinic cohort in Baltimore, Maryland, USA. To obtain an unbiased estimate of the prevalence and severity of hepatic steatosis among HIV-HCV-coinfected individuals treated with ART, 137 subjects were randomly selected from a group of 630 HIV-HCV-coinfected individuals who had received ART for 2 years and had not yet received treatment for HCV infection. Of the 137 sampled, 25 individuals were excluded because of undetectable HCV RNA (n = 5), medical contraindications (n = 5), died before biopsy (n = 1), end-stage liver disease (n = 3), lost to follow-up (n = 5), and other (n = 6). After these exclusions, 112 individuals remained. These patients were not substantially different from other eligible individuals who were not selected, with respect to age, gender, liver enzymes, CD4 cell count, and HIV RNA level (P > 0.05, data not shown). However, patients in the random sample were slightly more likely than the other eligible individuals to be African-American and to have a history of alcohol abuse (data not shown).
Data on demographics, drug and alcohol use, and HIV medications were obtained at the time of biopsy; audio-computer assisted self-interview (ACASI) was used to obtain additional information on alcohol use and other socially sensitive exposures. Designation of current and past alcohol use was based on physician diagnosis as well as the ACASI result and the validated Alcohol Use Disorders Identification Test (AUDIT), which includes a 10-item screening scale designed to identify a broad spectrum of hazardous or harmful alcohol users. AUDIT was administered via ACASI at the time of the liver biopsy. All participants gave a blood sample for assessment of laboratory parameters. Additional information on prescribed medications and laboratory parameters preceding the liver biopsy was obtained from the JHU clinical and laboratory database. As previously described, data on patient demographics, social practices, clinical and laboratory parameters, and prescribed ART and other medications were collected and abstracted from patient charts by trained personnel and transferred electronically from the JHU laboratory database at enrollment and at subsequent 6-month intervals [21]. Validity checks were performed periodically on a 5% sample of abstracted fields in the database; to date errors have only been found in 0.2% of variable fields. The study was approved by the JHU Joint Committee on Clinical Investigation and written informed consent was obtained for all participants. The US National Institutes of Health, which funded this research, played no role in the design or execution of the study.
Laboratory testingM
Patients had standard laboratory assessments at each visit performed by licensed clinical laboratories including a complete blood cell count, serum chemistry panels, alanine aminotransferase (ALT), aspartate aminotransferase (AST), CD4 cell count and plasma HIV RNA level (reverse transcriptase polymerase chain reaction). HCV testing was routinely performed on all cohort participants by a licensed clinical laboratory using a second- or third-generation enzyme immunoassay (EIA 2.0: Abbott Laboratories, Abbott Park, Illinois, USA; EIA 3.0: Ortho Diagnostics, Raritan, New Jersey, USA) and confirmatory HCV RNA testing (COBAS AMPLICOR MONITOR assay, Roche Diagnostic Systems, Indianapolis, Indiana, USA). Individuals also had quantitative testing for HBV DNA (polymerase chain reaction).
Liver histology
A transcutaneous liver biopsy was performed using an 18-gauge needle. Liver tissue was then fixed in 10% formalin and paraffin-embedded sections were stained with hematoxylin-eosin and trichrome stains. Slides were evaluated by a single pathologist (M. Torbenson), deemed adequate based on specimen size and number of portal tracts and scored according to the Ishak modified histological activity index (MHAI) scoring system [22]. The MHAI uses a 7-point scale for fibrosis stage and an 18-point scale for necroinflammatory activity. Fibrosis was classified as: F0, no fibrosis; F1-F2, fibrosis with or without septae; F3-F4, fibrosis/bridging, F5, marked bridging with occasional nodules; F6, cirrhosis. Activity was a composite score of periportal or periseptal interface hepatitis, confluent necrosis, focal lytic necrosis, apoptosis and focal inflammation, and portal inflammation. Hepatic steatosis was classified on a 5-point scale: 0, none; 1, steatosis involving < 5% of hepatocytes; 2, 5-29%; 3, 30-60%; 4 > 60%.
Statistical analysis TOP
Univariate and multivariate logistic regression analyses were used to identify determinants of hepatic steatosis. Correlates of hepatic steatosis were first examined separately comparing those with fat involving < 5% of hepatocytes and those with fat involving 5% of hepatocytes. Because no significant differences were observed between these two groups, these individuals were analyzed together and compared with those with no fat in the liver.
Factors considered in analysis included general demographics, alcohol abuse, duration of HCV infection, past and current ART, liver enzymes, HIV RNA level, CD4 cell count, previous ART-associated hepatotoxicity, necroinflammatory activity, and hepatic steatosis. Duration of HCV infection was estimated as time since first drug use. In order to capture the collective experience of an individual, ALT, AST, HIV RNA and CD4 cell count were analyzed both at the time of biopsy and also as cumulative variables. Accordingly, ALT and AST were examined as the cumulative proportion of ALT and AST > 5~ upper limit of normal (ULN). Individuals with > 33% of ALT and AST measurements > 2.5~ ULN were compared with those who had 33% of their values > 2.5~ ULN. HIV RNA and CD4 cell count were analyzed as the proportion of HIV RNA measurements that were undetectable and the proportion of CD4 cell counts that were < 200 ~ 106 cells/l. Two-sided P values < 0.05 were deemed statistically significant. Analysis was performed using SAS, version 8.12 (Cary, North Carolina, USA).
|
|
| |
|
|
|
