Hepatic steatosis in HIV-HCV coinfected patients in France: comparison with HCV monoinfected patients matched for body mass index and HCV genotype.

Abstract

Significance of steatosis in HIV-HCV coinfection remains controversial.
To compare the prevalence and predictors of hepatic steatosis between HIV-HCV and HCV patients matched for steatosis known determinants.
A total of 564 HCV-naive patients undergoing liver biopsy were studied: 137 with HIV-HCV coinfection and 427 with HCV monoinfection, among whom 137 were matched for age, gender, body mass index and HCV genotype.
Steatosis of any grade (67.1% vs. 41.6%, P < 0.0001), mixed steatosis (55.4% vs. 21.1%, P < 0.0001), severe histological activity (A2-A3: 78.1% vs. 55.5%, P < 0.0001) and severe fibrosis (F3-F4: 33.1% vs. 15.3%, P < 0.0001) were significantly more common in coinfected than in matched monoinfected patients. In multivariate analysis, steatosis was associated only with severe histological activity [odds ratio (OR): 3.1 (95% CI: 1.3-7.1)] in coinfected patients and with elevated body mass index [OR; 1.3 (1.1-1.5)], HCV genotype 3 [OR: 5.6 (2.3-13.9)], severe histological activity [OR: 3.1 (1.3-7.3)] and severe fibrosis [OR: 4.7 (1.3-17.3)] in monoinfected patients.
Steatosis is significantly more common and severe in HIV-HCV coinfected than in HCV monoinfected French patients, even after matching for body mass index and HCV genotype. Steatosis is associated only with severe histological activity in coinfected patients and with previously reported factors in monoinfected patients, thus suggesting different underlying mechanisms.

Full text

Page 1

Castera et al.

1
Hepatic steatosis in HIV-HCV coinfected patients in France: comparison with HCV
monoinfected patients matched for body mass index and HCV genotype
L. CASTÉRA (1, 2), M. A. LOKO (3), B. LE BAIL (4), P. COFFIE (3), V. DE LEDINGHEN
(1), P. TRIMOULET (5), M. WINNOCK (3), F. DABIS (3), D. NEAU (6) & THE GROUPE
D’EPIDÉMIOLOGIE CLINIQUE DU SIDA EN AQUITAINE (GECSA)*

(1) Department of Hepatology, Hôpital Haut Lévêque, Centre Hospitalier Universitaire CHU
Bordeaux, Pessac, France
(2) Department of Hepatology, Hopital St André, CHU Bordeaux, Bordeaux, France
(3) INSERM U593 ISPED, Université Victor Segalen Bordeaux-2, Bordeaux, France
(4) Department of Pathology, Hôpital Pellegrin, CHU Bordeaux, Bordeaux, France
(5) Department of Virology, Hôpital Pellegrin, CHU Bordeaux, Bordeaux, France
(6) Department of Infectious Diseases, Hôpital Pellegrin, CHU Bordeaux; Bordeaux, France

Corresponding author: Laurent Castera, MD, PhD
Service d’Hépato-Gastroenterologie,
C.H.U. de Bordeaux,
Hôpital Haut Lévêque,
Avenue Magellan,
33604 Pessac, France
Tel: 00 33 5 57 65 64 39
Tel: 00 33 5 57 65 64 45
e-mail: laurent.castera@chu-bordeaux.fr
Short title: Hepatic steatosis in HIV-HCV and matched HCV patients

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Alimentary Pharmacology and Therapeutics 2007;:epub ahead of print
This article is published in the Alimentary Pharmacology and Therapeutics'
OnlineAccepted collection on Blackwell Synergy. The article has been allocated
a unique Digital Optical Identifier (DOI), which will remain unchanged
throughout publication.
It must be emphasized that this article has not been edited, and changes are
inevitable during this process.
See http://www.blackwell-synergy.com/page/onlineaccepted for details.
Please cite this article as a "Postprint"; doi: 10.1111/j.1365-2036.2007.03533.x
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
HAL author manuscript inserm-00177055, version 1
HAL author manuscript

Page 2

Castera et al.

2

*Composition of the GECSA in charge of the ANRS CO3 Aquitaine Cohort:
Scientific committee: G. Chêne, F. Dabis, M. Dupon, M. Longy-Boursier, P. Morlat, JL.
Pellegrin, JM. Ragnaud and R. Salamon Chair.
Methodological coordination: F. Dabis, G. Chêne, R. Thiébaut, C. Lewden and S. Lawson-
Ayayi.
Medical coordination: M. Dupon, P. Mercié, JF. Moreau, P. Morlat, JL. Pellegrin, JM.
Ragnaud, N. Bernard, D. Lacoste, D. Malvy and D. Neau.
Data Management and Analysis: MJ. Blaizeau, M. Decoin, S. Delveaux, C. Hannapier, S.
Labarrère, V. Lavignolle-Aurillac, B. Uwamaliya-Nziyumvira, G. Palmer, D. Touchard, M.
Winnock, E. Balestre, A. Alioum, H. Jacqmin-Gadda and R. Thiébaut.
Participating physicians: Bordeaux University Hospital: P. Morlat, N. Bernard, M. Bonarek,
F. Bonnet, P. Gellie, D. Lacoste, K. Lacombe; M. Dupon, F. Dauchy, H. Dutronc, S. Lafarie;
M. Longy-Boursier, P. Mercié, D. Malvy, T. Pistonne, P. Thibaut; JM. Ragnaud, D.
Chambon, C. De La Taille, C. Cazorla, T. Galperine, D. Neau, A. Ochoa; JL.Pellegrin, JF.
Viallard, O. Caubet, E. Lazaro C. Nouts; P.Couzigou, L. Castera; H. Fleury, ME. Lafon, B.
Masquelier, I. Pellegrin; D. Breilh; JF. Moreau, P. Blanco. Dax Hospital: P. Loste, L.
Caunègre. Bayonne Hospital: F. Bonnal, S. Farbos, MC Gemain. Libourne Hospital:
J.Ceccaldi, S. Tchamgoué. Mont de Marsan Hospital: S. De Witte.

HAL author manuscript inserm-00177055, version 1

Page 3

Castera et al.

3
ABSTRACT
Background: Significance of steatosis in HIV-HCV coinfection remains controversial.
Aim: To compare the prevalence and predictors of hepatic steatosis between HIV-HCV and
HCV patients matched for steatosis known determinants.
Methods: 564 naive HCV patients undergoing liver biopsy were studied: 137 with HIV-HCV
coinfection and 427 with HCV monoinfection, among whom 137 were matched for age,
gender, BMI and HCV genotype.
Results: Steatosis of any grade (67.1% vs. 41.6%, p<0.0001), mixed steatosis (55.4% vs.
21.1%, p<0.0001), severe histological activity (A2-A3: 78.1% vs. 55.5%, p<0.0001), and
severe fibrosis (F3-F4: 33.1% vs. 15.3%, p<0.0001) were significantly more common in
coinfected than in matched monoinfected patients. In multivariate analysis, steatosis was
associated only with severe histological activity (odds ratio (OR): 3.1 (95% CI: 1.3-7.1)) in
coinfected patients and with elevated BMI (OR; 1.3 (1.1-1.5)), HCV genotype 3 (OR: 5.6
(2.3-13.9)), severe histological activity (OR: 3.1 (1.3-7.3)) and severe fibrosis (OR: 4.7 (1.3-
17.3)) in monoinfected patients.
Conclusions: Steatosis is significantly more common and severe in HIV-HCV coinfected
than in HCV monoinfected French patients, even after matching for BMI and HCV genotype.
Steatosis is associated only with severe histological activity in coinfected patients and with
previously reported factors in monoinfected patients, thus suggesting different underlying
mechanisms.

Key words: hepatic steatosis, HIV infection, HCV infection, BMI, HCV genotype
HAL author manuscript inserm-00177055, version 1

Page 4

Castera et al.

4
INTRODUCTION
Hepatic steatosis, defined as lipid accumulation in the hepatocyte cytoplasm, is a frequent
histological finding in patients with chronic hepatitis C, occurring in more than 50 % of cases
[1] and that has been suggested to be associated with fibrosis progression [2-8]. Two distinct
forms of hepatocellular steatosis, “metabolic” and “viral”, can be observed in patients with
chronic HCV infection [9]. Classical metabolic risk factors for steatosis such as obesity,
diabetes type 2, hyperlipidaemia and excessive alcohol intake account for the vast majority of
cases of steatosis in patients infected with HCV non 3 genotypes [10-12]. In contrast, in
patients infected by HCV genotype 3, steatosis is thought to be induced by the virus itself
through a direct cytopathic effect [13].
Liver disease is an important and increasing cause of morbidity and mortality in HIV-
HCV coinfected patients [14, 15]. The prevalence and significance of steatosis in HIV-HCV
coinfected patients have been recently investigated, and it has been suggested that it may
contribute to the accelerated progression of liver disease observed in HIV-HCV coinfected
patients [16-21]. However, despite its potential clinical importance, conflicting results have
been published regarding factors associated with steatosis in the context of HIV-HCV
coinfection. Some studies [19, 21] suggest that predictors of steatosis in HIV-HCV coinfected
patients are the same than those observed in HCV monoinfected patients namely HCV
genotype 3 and high body mass index (BMI) whereas other do not [16-18, 20]. Such
discordant results may be explained by differences in the clinical features of studied
populations, patients’ recruitment biases or ethnic factors. In addition, it must be stressed that
two studies [18, 20] only from the US included a control group of HCV monoinfected patients
and that patients of this control group were not matched for steatosis known determinants,
BMI and HCV genotype.
HAL author manuscript inserm-00177055, version 1

Page 5

Castera et al.

5
The aim of this study was to compare the prevalence and predictors of steatosis between
HIV-HCV coinfected and HCV monoinfected French patients matched for steatosis known
determinants, BMI and HCV genotype as well as age and gender.
HAL author manuscript inserm-00177055, version 1

Page 6

Castera et al.

6
PATIENTS AND METHODS

Study population
This retrospective study included patients with HIV-HCV coinfection and HCV mono-
infection who were referred for liver biopsy prior to HCV antiviral therapy at the University
Hospital of Bordeaux between January 1999 and January 2005.
Inclusion criteria were: age over 18, positive serum antibodies to HCV by means of a
second- or third-generation HCV enzyme-linked immunosorbent assay (Ortho Diagnostic,
Raritan NJ, USA) and detectable serum HCV RNA (AmplicorTM HCV, Roche Molecular
Systems, Pleasanton, CA, USA).
Exclusion criteria were: coinfection with hepatitis B, other known causes of liver disease,
alcohol intake of more than 50g/day, previous treatment for HCV, decompensated cirrhosis,
and liver biopsy specimen <10 mm.
A total of 564 patients met those criteria. They comprised 427 HCV monoinfected patients
and 137 HIV-HCV coinfected patients. This latter group is a subgroup of 148 HIV-HCV
patients from the Aquitaine Cohort in whom we recently investigated the prevalence and
predictors of steatosis [22]. The ANRS CO3 Aquitaine Cohort is a prospective hospital-based
cohort of HIV-1-infected patients under routine clinical management which was initiated in
1987 in the Bordeaux University Hospital and four other public hospitals in the Aquitaine
region, Southwestern France, by the Groupe d’Epidémiologie Clinique du SIDA en Aquitaine
GECSA [23]. All adults who are in- or out- patients of the participating hospital wards with
HIV-1 infection, confirmed by Western-Blot testing, and who have given an informed consent
are enrolled in the cohort, whatever their clinical stage, gender or HIV transmission category.
Also, information from at least one follow-up visit after the baseline assessment, or a known
date of death must be available. At each hospital contact, a standardized questionnaire
including epidemiological, clinical, biological and therapeutical data is filled in by clinicians
HAL author manuscript inserm-00177055, version 1

Page 7

Castera et al.

7
for each patient included in the Cohort and entered into the database. The schedule of follow-
up visits is based on clinical practice, and an active search of patients lost to follow-up is
performed annually.
Among the 427 HCV mono-infected patients, 137 were matched to the 137 HIV-HCV
coinfected patients according to age, gender, BMI and HCV genotype.

Clinical, biological and virological parameters
The following data were recorded for all patients at the time of liver biopsy: age, sex, risk
factors for HCV infection, BMI, alaninz amonitransferas (ALT) and HCV genotype (Inno-
Lipa II HCV, Innogenetics, Ghent, Belgium). In addition, the following data were specifically
recorded for HIV-HCV coinfected patients: route of transmission of HIV infection,
antiretroviral therapy, median CD4 lymphocyte count, and plasma HIV RNA load.

Histological studies and steatosis evaluation
Paraffin-embedded biopsy specimens of more than 10 mm were analyzed prospectively
during clinical management by an experienced pathologist unaware of clinical and biological
data except for the HCV or HIV status. Liver biopsy specimens were stained with
hematoxylin-eosin-saffran, or picroSirius red for collagen. Data were collected in a
standardized report form developed for hepatitis C before this specific study. In this report,
histological activity and fibrosis are scored according to the METAVIR classification [24].
Steatosis is also graded according to METAVIR as previously validated [24]: none; mild:
involving 1-10% of hepatocytes; moderate: involving 11-30% of hepatocytes; and severe:
>30% of hepatocytes. In addition, the pattern of steatosis was characterized as follows:
predominantly macrovesicular; predominantly microvesicular; mixed.

HAL author manuscript inserm-00177055, version 1

Page 8

Castera et al.

8
Statistical analysis
For univariate and multivariate statistical analysis, the outcome of interest was the
presence of steatosis. Variables considered for analyses of steatosis determining factors were
age at liver biopsy, gender, BMI, source of HCV infection, HCV genotype, ALT activity, use
of antiretroviral treatment before liver biopsy, histological activity and fibrosis.
Continuous variables were expressed by their mean ± standard deviation (SD) and
categorical variables by absolute figures and percentages. Mann-Whitney and Chi-2 tests
were used for statistical comparisons of unmatched quantitative and qualitative variables,
whereas Wilcoxon test and a conditional logistic regression were used for comparing
quantitative and qualitative matched variables, respectively. A p value<0.05 was considered
significant.
To assess the independent value of each parameter related to steatosis in univariate
analysis with p<0.25, a multivariate analysis was performed by means of a stepwise logistic
regression analysis. Statistical analyses were performed using SAS software (SAS institute
Inc, Cary, NC, USA, version 9.1).


HAL author manuscript inserm-00177055, version 1

Page 9

Castera et al.

9
RESULTS
Study population
The baseline characteristics of the 137 HIV-HCV coinfected patients as compared with
those of the 427 HCV monoinfected patients are shown in table 1. Among HIV-HCV
coinfected patients, the main source of HIV infection was intravenous drug use (68.0%). The
median CD4 count was 508 cells/mm3 (78-1644 cells/mm3). A total of 125 (91%) patients
were taking antiretroviral therapy: 122 (89%) were taking nucleoside reverse transcriptase
inhibitors (NRTI), 55 (40%) protease inhibitors (PI), and 50 (36.5%) non nucleoside reverse
transcriptase inhibitors (NNRTI). Plasma HIV RNA was undetectable (<50 copies/ml) in 60
patients (44%).
Coinfected patients differed from monoinfected patients in that there were significantly
younger, more often male, had lower BMI, were more likely to have a prior history of
intravenous drug use and to be HCV genotype 3-infected. They also had higher histological
activity grade and more severe fibrosis.

Prevalence, severity and patterns of steatosis in the 137 HIV-HCV coinfected patients
and the 427 HCV monoinfected patients
Steatosis of any grade was significantly more common (67.1 % vs. 49.6%, respectively) in
coinfected than in monoinfected patients (Figure 1A). Among the 92 HIV-HCV coinfected
and the 212 HCV monoinfected patients who had steatosis of any grade on liver biopsy, the
pattern of steatosis differed significantly between the two groups (Figure 1B). Although the
majority of HCV monoinfected patients had macrovesicular steatosis, coinfected patients
were more likely to have a mixed pattern of steatosis (55.4% vs. 23.6%, respectively) or
microvesicular steatosis (6.6% vs. 0.9%, respectively). Among the 92 HIV-HCV coinfected
patients with steatosis, those with macrovesicular steatosis did not differ from those with
HAL author manuscript inserm-00177055, version 1

Page 10

Castera et al.

10
mixed of microvesicular steatosis for most characteristics, except for HCV genotype
distribution (genotype 1: 54.3% vs. 31.6 %, respectively; p=0.01) (Table 2).


Factors associated with steatosis in the 427 monoinfected patients
In univariate analysis, steatosis was associated with age, BMI, ALT level, HCV genotype
3, severe histological activity and severe fibrosis (Table 3). In multivariate analysis, steatosis
was associated with BMI (OR: 1.2 95% CI 1.1-1.2, p<0.0001), HCV genotype 3 (OR: 3.0 1.5-
6.1, p=0.002), severe histological activity (A2-A3 OR: 1.7 1.1-2.6, p=0.02) and severe
fibrosis (F3-F4 OR: 3.8 2.2-6.8, p< 0.0001) (Table 3).

Prevalence, severity and patterns of steatosis in the 137 HIV-HCV coinfected patients
and the 137 matched HCV monoinfected patients
The baseline characteristics according to HIV status of the 137 matched patients are
shown in Table 4. Coinfected patients were more likely to have a prior history of injection
drug use, higher ALT levels, more severe histological activity and more severe fibrosis.
Steatosis of any grade was significantly more common in coinfected patients than in
monoinfected patients (67.1% vs 41.6%, p<0.0001) (Figure 2A). Among the 92 HIV-HCV
coinfected and the 57 HCV monoinfected patients who had steatosis of any grade on liver
biopsy, the type of steatosis differed significantly between the two groups (Figure 2B).
Although the majority of HCV monoinfected patients had macrovesicular steatosis,
coinfected patients were more likely to have a mixed pattern of steatosis (55.4% vs. 21.1%,
respectively) or microvesicular steatosis (6.6% vs. 1.7%, respectively).

HAL author manuscript inserm-00177055, version 1

Page 11

Castera et al.

11
Factors associated with steatosis in the 137 HIV-HCV coinfected patients and the 137
matched HCV monoinfected patients
In univariate analysis, steatosis was associated with severe histological activity A2-3, in
coinfected patients whereas it was associated with elevated BMI, HCV genotype 3, severe
histological activity (A2-A3) and severe fibrosis (F3-F4) in monoinfected patients (Table 5).
In multivariate analysis, steatosis was associated with severe histological activity (A2-3 OR:
3.1 95% CI 1.3-7.1, p=0.008) in coinfected patients whereas it was associated with elevated
BMI (OR: 1.3 1.1-1.5; p=0.005), HCV genotype 3 (OR: 5.6 2.3-13.9; p=0.0002), severe
histological activity (OR: 3.1 1.3-7.3; p=0.009) and severe fibrosis (OR: 4.7 1.3-17.3; p=0.02)
in monoinfected patients (Table 5).

HAL author manuscript inserm-00177055, version 1

Page 12

Castera et al.

12
DISCUSSION
The results of the present study, conducted in a large cohort of French patients with
chronic hepatitis C, show that, even after adjusting by design for potential confounding
factors HCV genotype and BMI, hepatic steatosis is significantly more common and more
severe in HIV-HCV coinfected than in HCV monoinfected patients. In multivariate analysis,
steatosis was associated only with severe histological activity in coinfected patients whereas
it was associated with already reported factors (increased BMI, HCV genotype 3, severe
histological activity and severe fibrosis) in monoinfected patients.
Steatosis has been reported in HIV-HCV coinfected patients with a prevalence ranging
from 40% to 72% [16-20]. The 67% prevalence in our study is close to those reported in
another French study [19] and in two recent American studies [20, 21]. A strength of the
present study is that coinfected patients were compared to a control group of HCV
monoinfected patients. So far, only two studies used a similar methodology [18, 20]. Both
were conducted in American patients but obtained discrepant results: Monto et al. [18] found
that steatosis was less common in coinfected patients than in monoinfected patients, 47% vs.
59%, respectively whereas Gaslightwala et al. [20] found that steatosis was more common in
coinfected patients than in monoinfected patients (72.1% vs. 52%, respectively). Difference in
the studied populations might account for this discrepancy.
It must be stressed that the baseline characteristics age, gender, BMI and proportion of
patients infected by HCV genotype 3 of our population of HIV-HCV coinfected patients are
similar to those reported in the Ribavic study, a multicenter French trial having included 395
coinfected patients [19]. In addition, the baseline characteristics as well as the prevalence and
severity of steatosis in our control group of 427 HCV monoinfected patients are also similar
to those previously reported [25, 26]. Finally, the factors independently associated with
steatosis in multivariate analysis BMI, HCV genotype 3, severe activity and severe fibrosis
HAL author manuscript inserm-00177055, version 1

Page 13

Castera et al.

13
are also consistent with those of previous reports and of a recent meta-analysis in 3068
patients [27], lending further validity to our findings.
Interestingly, when compared with the total group of 427 HCV monoinfected patients, our
coinfected patients were younger, more often male, had lower BMI, and were more likely to
be infected with HCV genotype 3. Also, steatosis was significantly more common and more
severe in coinfected patients, a finding in line with those reported by Gaslightwala et al. [20].
However, one should be cautious when interpreting such findings because the observed
difference might be related to differences in the distribution of well known risk factors for
steatosis such as HCV genotype 3 infection and increased BMI, between coinfected and
monoinfected patients. The only way to avoid such a bias is to match coinfected patients with
monoinfected patients according to these two risk factors, which had not been done in any
study so far.
An important finding of the present study, is that, even after matching for BMI and HCV
genotype, steatosis remained significantly more common and more severe in coinfected
patients than in monoinfected patients. Also, coinfected patients were more likely to have a
mixed pattern of macrovesicular and microvesicular steatosis than monoinfected patients
before and after matching, a finding consistent with those of two recent studies [20, 21].
Although the reason for this finding is unclear, it suggests that different underlying
mechanisms may be responsible for steatosis in coinfected patients. Indeed, microvesicular
steatosis may be induced by NRTI which inhibit mitochondrial DNA polymerase gamma
[28]. Consistent also with the hypothesis of different underlying mechanisms is the fact that
factors independently associated with steatosis in multivariate analysis differed between
coinfected and monoinfected patients. Indeed, steatosis was associated only with severe
histological activity in coinfected patients whereas it was associated with previously reported
HAL author manuscript inserm-00177055, version 1

Page 14

Castera et al.

14
risk factors such as BMI, HCV genotype 3, severe activity and fibrosis in monoinfected
patients.
In HCV monoinfected patients, two distinct forms of steatosis can be observed:
“metabolic” steatosis associated with risk factors such as obesity, diabetes type 2,
hyperlipidaemia and excessive alcohol intake and “viral” steatosis in patients infected by
HCV genotype 3, through a direct cytopathic effect. The direct responsibility of HCV in the
pathogenesis of steatosis has been suggested by several lines of evidence: 1) the independent
association with HCV genotype 3 infection in most studies [9, 26, 29]; 2) the correlation
between severity of steatosis and HCV genotype 3 replication levels while no such
relationship has been found with other HCV genotypes [2, 7, 13, 30, 31]; 3) the disappearance
of steatosis in HCV genotype 3 infected patients with sustained viral clearance after antiviral
therapy [7, 31-33]; 4) finally, the development of hepatic steatosis in transgenic mouse lines
[34] or in cell lines [35] expressing HCV core protein and the inhibition of microsomal
triglyceride transfer protein, a key enzyme for hepatic lipoprotein assembly and secretion, in
both transgenic mouse [36] and human liver [37].
Interestingly, despite the significant proportion of coinfected patients with HCV genotype
3 infection (30.7%) in our study, which is higher than that reported in studies from the US
[16-18, 20] but similar to that reported in a recent French study [19], no correlation was found
between genotype 3 and steatosis in coinfected patients. Histological activity was the only
factor associated with steatosis. An association between histological activity and steatosis has
been reported in HCV monoinfected patients [27, 38] as well as in HIV-HCV coinfected
patients [17]. Several mechanisms may account for the relationship between steatosis and
histological activity. HCV infection is associated with increased cytokines production
enhancing histological inflammation and leading to increased lipid peroxydation [39]. Also,
HCV core protein has been shown to induce oxidative stress in vitro and in vivo [40]. In HIV-
HAL author manuscript inserm-00177055, version 1

Page 15

Castera et al.

15
HCV coinfected patients, histological inflammation may be related not only to HCV infection
but also to the use of antiretroviral drugs. Antiretroviral therapy has clearly been implicated in
the development of steatosis in patients with HIV [41] and HIV-HCV coinfection [17, 21].
Although no relationship was found between steatosis and antiretroviral treatment in the
present study, we cannot rule out the implication of the latter. Indeed, given the fact that
nearly all patients were on antiretroviral therapy, we cannot exclude the fact that we did not
have enough statistical power to demonstrate a significant relationship between steatosis and
antiretroviral treatment. Finally, the retrospective design of our study did not allow to take
into consideration the cumulative duration of exposure to antiretroviral drug which may also
play an important role.
In conclusion, steatosis is significantly more common and more severe in HIV-HCV
coinfected than in HCV monoinfected French patients, even after matching for BMI and HCV
genotype. Steatosis is associated only with severe histological activity in coinfected patients
and with already reported factors BMI, HCV genotype 3, severe histological activity and
fibrosis in monoinfected patients. These results suggest different underlying mechanisms
which deserve further investigation.

HAL author manuscript inserm-00177055, version 1