OUP user menu

Management of Hepatitis C Virus Infection in Heavy Drinkers

Charlotte E. Costentin, Jean-Baptiste Trabut, Vincent Mallet, Stéphane Darbeda, Véronique Thépot, Bertrand Nalpas, Béatrice Badin de Montjoye, Béatrice Lavielle, Anaïs Vallet-Pichard, Philippe Sogni, Stanislas Pol
DOI: http://dx.doi.org/10.1093/alcalc/agt020 337-342 First published online: 21 March 2013


Aim: Optimal management of hepatitis C virus (HCV) infection is controversial in heavy drinkers. We compared the management of HCV infection of heavy drinkers with that of patients without a history of alcohol abuse. Methods: In a retrospective case–control study, 69 HCV-infected heavy drinkers [daily alcohol consumption at referral above 60 g/day, hereafter ‘alcohol group’] were compared with matched HCV-infected patients with low alcohol consumption (<40 g/day, ‘control group’). Results: Patients of the ‘alcohol group’ were younger (42 vs. 45 years, P = 0.05), more often male (69.6 vs. 56.5%, P = 0.11) and had been infected by intravenous drug use (85.5 vs. 45.0%, P < 0.0001). The percentage of patients with a recommendation for treatment according to the French 2002 consensus (bridging fibrosis or genotype 2 or 3) was 52 of 69 (75.4%) in both groups, while the proportion of patients treated was higher in the control group (71.0 vs. 44.9%, P = 0.002). In the ‘alcohol group’, patients had better access to treatment if they were employed or consumed 170 g/day or less at first referral. Sustained virological response (SVR) was obtained in 10 of 31 patients (32.3%) of the ‘alcohol group’ vs. 8 of 31 patients (25.8%) of the control group matched for genotype and type of treatment (P = 0.58). Conclusion: Heavy drinkers are less often considered for antiviral therapy compared with patients without a history of alcohol abuse. However, once treatment is actually initiated, SVR rates are comparable with those achieved in non-drinkers despite the continuation of alcohol consumption during therapy in some patients.


Chronic infection with the hepatitis C virus (HCV) is a leading cause of liver disease worldwide. Among chronic HCV carriers, however, only a fraction will eventually develop life-threatening complications such as cirrhosis, liver failure or hepatocellular carcinoma. A variety of co-factors—related to host characteristics, co-morbidities or lifestyle—affect the long-term outcome of HCV infection. Heavy alcohol consumption is one of the most important of those co-factors with a significant influence on the natural history of HCV infection (Safdar and Schiff, 2004).

Heavy alcohol consumption is common in patients with chronic hepatitis C and has a synergistic effect on fibrosis progression. Prevalence of HCV infection is higher in heavy drinkers (8 to 45%) (Parés et al., 1990; Caldwell et al., 1993; Mendenhall et al., 1993; Fong et al., 1994; Befrits et al., 1995; Coelho-Little et al., 1995; Srugo et al., 1998) than in the general population (<1% in France as in most western countries; Meffre et al., 2010). The higher prevalence of HCV infection in heavy drinkers is likely to be mainly related to the common intravenous drug use history, which has been switched to chronic alcohol intake (McHutchison et al., 1992; Mendenhall et al., 1993; Verbaan et al., 1993; Fong et al., 1994; Befrits et al., 1995; Galperim et al., 2006). An association has consistently been shown between high alcohol consumption and advanced fibrosis stage (Parés et al., 1990; Laskus et al., 1992; Caldwell et al., 1993; Fong et al., 1994; Befrits et al., 1995; Coelho-Little et al., 1995; Ostapowicz et al., 1998; Nevins et al., 1999; Khan and Yatsuhashi, 2000; Harris et al., 2001; Fabris et al., 2004; Hutchinson et al., 2005), especially when gender, age and suspected length of the disease are taken into account (Pessione et al., 1998; Pol et al., 1998; Wiley et al., 1998).

Regarding antiviral treatment, several studies have reported that a history of excessive alcohol consumption was associated with a lower rate of HCV eradication in patients receiving an interferon-based regimen (Okazaki et al., 1994; Mochida et al., 1996; Ohnishi et al., 1996; Loguercio et al., 2000; Tabone et al., 2002). It has been hypothesized, in accordance with in vivo data, that ethanol could have a direct impact on HCV replication and on interferon signal transduction (Pessione et al., 1998; Loguercio et al., 2000; Osna et al., 2005; McCartney et al., 2008). Guidelines, therefore, recommend a 6-month period of abstinence before initiation of antiviral therapy in HCV-infected patients (Dhumeaux et al., 2003; Sherman et al., 2007). However, data evaluating the real impact of ongoing alcohol consumption during interferon-based therapy are scarce and, to date, no study has demonstrated the beneficial effect of protracted abstinence before the introduction of antiviral therapy. On the one hand, in most studies reporting a negative impact of alcohol consumption on the outcome of interferon therapy, patients had stopped using alcohol before the beginning of antiviral therapy and were supposed to remain abstinent throughout treatment (Okazaki et al., 1994; Mochida et al., 1996; Ohnishi et al., 1996; Tabone et al., 2002). On the other hand, Anand et al. (2006) have shown that the negative impact of alcohol consumption on antiviral therapy might be entirely attributable to low adherence and premature termination of the treatment. Taken together, those data suggest that abstinence per se is not sufficient to improve the efficacy of antiviral treatment in patients with a history of heavy alcohol use and that efforts should also focus on the improvement of adherence to treatment.

It is, therefore, crucial to get more data on the management of HCV among alcoholics. The aim of this retrospective case–control study was to compare the management of HCV infection in a cohort of patients with heavy alcohol consumption at the time of referral with that in a cohort of HCV-infected patients without any history of alcohol abuse.



This was a retrospective case–control study including patients with HCV infection with a current history of alcohol abuse (case, hereafter ‘alcohol group’) and without a history of alcohol abuse (control). All patients have been referred to our Liver Unit between 1991 and 2009.

The inclusion criteria in the ‘alcohol group’ were: chronic HCV infection defined by detectable HCV RNA for at least 6 months; alcohol consumption ≥60 g/day for at least 1 year at the time of first referral; no previous antiviral treatment and available fibrosis evaluation performed at first referral, either with a liver biopsy or with a non-invasive test such as liver stiffness measurement or validated biomarkers (e.g. FibroTest®) according to the recommendations of the French health authorities (Fontaine et al., 2007).

Two control groups were constituted, for comparison with the ‘alcohol group’, in order to evaluate access to antiviral therapy (‘control/access group’) and response to treatment (‘control/response group’), respectively. The inclusion criteria in both control groups were identical to those in the ‘alcohol group’ (chronic HCV infection, fibrosis evaluation and no previous antiviral treatment) except for alcohol consumption that had to be below 40 g/day without previous periods of alcohol abuse. In each control group, patients were matched with those of the ‘alcohol group’ on a 1/1 ratio. In the ‘control/access group’, patients were matched according to stage of fibrosis (F0–1 vs. F2–3 vs. F4, in the METAVIR scoring system), period of the first evaluation (before 1998 vs. 1998 to 2001 vs. after 2001), genotype (1 or 4 vs. 2 or 3) and, when possible, gender and age. In the ‘control/response group’, patients were matched to the treated patients of the ‘alcohol group’ according to type of antiviral therapy (standard interferon monotherapy vs. standard interferon associated to ribavirin vs. pegylated interferon associated to ribavirin), genotype (1 vs. 2 or 3 vs. 4) and, when possible, to stage of fibrosis, gender and age.

Altogether, 69 patients were included in the ‘alcohol group’ [69.9% male, median age: 42 years] and in the ‘control/access group’ (56.5% male, median age: 45 years). The 31 treated patients of the ‘alcohol group’ (77.4% male, median age: 42 years) were matched to the 31 patients of the ‘control/response group’ (67.4% male, median age: 42 years).

The following data were recorded at the time of first referral: age, gender, route of HCV transmission, hepatitis B virus status [positivity or negativity HBs antigen (HBsAg) and of anti-HBc antibodies (HBcAb)], co-infection with HIV, aspartate transaminase (AST), alanine transaminase and gamma glutamyl transpeptidase (GGT) levels and, indication for antiviral treatment. Indication for antiviral treatment was defined according to the French 2002 consensus as follows (Dhumeaux et al., 2003): bridging fibrosis (i.e. METAVIR fibrosis stage ≥2) or genotype 2 or 3 whatever the fibrosis stage. In the ‘alcohol group’, data related to employment, housing, opiate substitution therapy and the existence of comorbidities (psychiatric diseases, diabetes, obesity and cardiovascular or neurological diseases) were also collected. During the follow-up, the number of antiviral treatment courses and, for each of them, the drugs used and their duration were recorded in both groups. Sustained virological response (SVR) was defined by undetectable HCV RNA 6 months after the end of therapy.

All the patients were followed by a hepatologist during all the period covered by the study. In the ‘alcohol group’, patients had an additional specific follow-up by a multidisciplinary team comprising an addiction specialist, a psychologist and a social worker. The high incidence of psychiatric co-morbidities in this group and the management of interferon side effects also justified a joint follow-up with a psychiatrist for the majority of the patients. In the ‘control groups’, the need for psychosocial support was assessed on a case-by-case basis. Approval of the study was obtained from the local ethical board.

Statistical analysis

Quantitative parameters were described using median and interquartile range (IQR), while qualitative parameters were described as percentages. Quantitative values were compared using the Mann–Whitney test. Dichotomous parameters were compared using the χ2 test. The Fisher exact test was used when the sample was too small to use the χ2 test. Logistic binary regression was used for multivariate analyses after continuous variables were categorized (below or above the median). The statistical significance level was <5%. All tests were done using ‘BiostaTGV’ (http://marne.u707.jussieu.fr/biostatgv/) except for multivariate analyse that was done with the SPSS software.



The ‘alcohol group’ included 69 patients with excessive alcohol consumption and chronic HCV infection. To evaluate access to treatment, they were compared with 69 patients without excessive alcohol consumption (‘control/access group’). As shown in Table 1, patients in the ‘alcohol group’ were younger (42 vs. 45 years, P = 0.05) and more frequently infected through intravenous drug use (85.5 vs. 45.0%, P < 0.0001). No patient was HBsAg positive and the presence of anti-HBcAb was more common in the ‘alcohol group’ (54.7 vs. 35.5%, P = 0.03). Five patients were co-infected with HIV in the alcohol group and one in the ‘control/access group’ (P = 0.09). Patients were mainly infected with genotype 1 (50.7% in ‘alcohol group’ vs. 56.5% in ‘control/access group’), and genotype 3 (34.8 vs. 29.0%). Median alcohol consumption in the ‘alcohol group’ was 170 g/day (IQR: 115–250) at the time of referral.

View this table:
Table 1.

Baseline characteristics and access to antiviral treatment in HCV-infected heavy drinkers (‘alcohol group’) compared with HCV-infected patients without a history of alcohol abuse (‘control/access group’)

There was no difference in terms of fibrosis stage, as patients were matched according to this criterion: percentage of patient with significant fibrosis, i.e. METAVIR ≥ F2, was 63.8% in both groups. Percentage of cirrhosis was 43.4 and 37.7% in the ‘alcohol group’ and in the ‘control/access group’, respectively (P = 0.49). Median AST and GGT levels were higher in the ‘alcohol group’ (Table 1).

Access to treatment

As shown in Table 1, fewer treatments were initiated in the ‘alcohol group’ than in the ‘control/access group’ (44.9 vs. 71.0%, P = 0.002). Among the patients with an indication for antiviral therapy, the rate of treatment initiation was also lower in the ‘alcohol group’ than in the ‘control/access group’ (53.8 vs. 88.5%, P = 0.0001). A lower rate of treatment initiation in the ‘alcohol group’ was also observed among patients with advanced fibrosis (56.8 vs. 93.2%, P = 0.00008).

Patients in the ‘alcohol group’ were more likely to receive treatment if their alcohol consumption at first referral was below 170 g/day (OR = 3.40, P = 0.002) and/or if they were currently employed (OR = 3.20, P = 0.02), these items remaining significantly associated with treatment initiation in multivariate analysis (Table 2).

View this table:
Table 2.

Factors associated with access to antiviral therapy among HCV-infected heavy drinkers (‘alcohol group’)

In the ‘alcohol group’, the reasons for not treating despite indication (24 patients) were the following (several patients had more than one reason): mild fibrosis (METAVIR F1) in patients with genotype 3 (n = 4), absence of sustained abstinence (n = 11), psychiatric disorders (n = 5), patient's refusal (n = 5) and loss to follow-up (n = 5). Among the six patients of the ‘control/access group’ who did not receive treatment despite indication, the reasons not to treat were mild fibrosis in patients with genotype 3 (n = 3), patient's refusal (n = 2) and loss to follow-up (n = 1).

After one treatment failure, a second treatment was initiated in 50% of the ‘alcohol group’ and in 64.1% in the ‘control/access group’ (P = 0.30). At the end of follow-up, there was a non-significant trend toward a lower rate of viral clearance in the ‘alcohol group’ than in the ‘control/access group’ (17.4 vs. 24.6%, P = 0.29).

Response to treatment

Each of the 31 patients of the ‘alcohol group’ was matched with a control patient according to the type of treatment received and to their genotype (‘control/response group’). ‘Alcohol group’ and ‘control/response group’ were also comparable according to other clinical characteristics (gender, age and stage of fibrosis, Table 3). Three patients of the alcohol group were lost to follow-up before the assessment of SVR. Those cases were considered treatment failures. The rate of SVR was similar in both groups (32.3% in the ‘alcohol group’ vs. 25.8% in the ‘control/response group’, P = 0.58). In the ‘alcohol group’, rate of premature treatment discontinuation was 12.9 vs. 10.0% in the ‘control/response group’ (P = 1.00).

View this table:
Table 3.

Response to HCV therapy in HCV-infected heavy drinkers (‘alcohol group’) compared with HCV-infected patients without a history of alcohol abuse (‘control/response group’)

Alcohol consumption during antiviral treatment

In the subgroup of 31 patients who received antiviral therapy, abstinence had been obtained before initiation of antiviral therapy in 23 patients but the length of abstinence was less than 6 months in 18 (Table 4). Twelve of those patients remained abstinent throughout the treatment. Eight patients did not stop alcohol consumption before initiation of antiviral therapy but most lowered their consumption during treatment. Among the 19 patients who consumed alcohol while on therapy, 9 consumed less than 40 g/day and 10 consumed 40 g/day or more.

View this table:
Table 4.

Sustained virological response according to alcohol consumption during antiviral treatment in the 31 treated patients of the ‘alcohol group’

There was a non-significant trend for a higher rate of SVR in patients who stopped alcohol consumption before treatment and among those who remained abstinent (Table 4). ‘High’ alcohol consumption (≥40 g/day) during therapy was not associated with a decrease of the SVR rate (3/10, 30%). In this cohort, 11 patients (35.5%) were abstinent 1 year after treatment.


This observational study compared the management of HCV infection in a group of patients with heavy alcohol consumption at the time of first referral (‘alcohol group’) with that in a group of patients with hepatitis C but no history of alcohol abuse. We observed that patients in the ‘alcohol group’ were less likely to receive antiviral therapy than patients without a history of alcohol abuse, even in the case of advanced liver disease with definite indication to treatment. However, patients of the ‘alcohol group’ who received antiviral treatment had a similar rate of SVR compared with controls.

In our study, failure to achieve sustained abstinence was the main reason to postpone treatment despite indication. This is in accordance with guidelines that recommend protracted abstinence before the introduction of antiviral treatment (Dhumeaux et al., 2003; Sherman et al., 2007). It should be emphasized, however, that most treatments were introduced after short-term abstinence (<6 months) and that some patients were treated despite ongoing alcohol consumption. Our results also show that it remains difficult to consider antiviral treatment in very heavy drinkers (≥170 g/day at first referral).

Besides alcohol consumption, reasons not to treat were absence of patient's adherence to the therapeutic project (treatment's refusal or loss to follow-up) and psychiatric co-morbidities. We also found that unemployed patients were less likely to be treated. It illustrates the fact that management of hepatitis C in heavy drinkers requires a global multidisciplinary approach, including hepatologists, addiction specialists, psychiatrists as well as psychosocial support. This kind of management was systematically proposed to all patients of the ‘alcohol group’ in this study, as our liver unit has a long experience of both hepatitis C treatment and alcohol addiction care. It might partly explain why, in our study, the rate of treatment initiation in heavy drinkers was higher than that reported by Anand et al. in a similar population setting (45 vs. 23%) but in a different country (Anand et al., 2006).

Our multidisciplinary approach probably also explains why we achieved similar SVR rates in our alcohol group compared with the control group. Several studies have reported lower rates of SVR in this population setting, and it has been hypothesized that ethanol could directly impair the antiviral activity of interferon (Okazaki et al., 1994; Mochida et al., 1996; Ohnishi et al., 1996; Loguercio et al., 2000; Tabone et al., 2002; McCartney et al., 2008). However, three studies recently reported results comparable with ours with satisfactory SVR rates in patients with a history of alcohol abuse (Bruggmann et al., 2010; Le Lan et al., 2012; Russell et al., 2012) probably due to a better rate of treatment completion. For example, Anand et al reported a 40% rate of treatment discontinuation among recent heavy drinkers, although only patients who managed to stop their alcohol consumption were considered for antiviral treatment in this study (Anand et al., 2006). In our study, we observed a lower rate of treatment discontinuation (13%) in the alcohol group, including abstinent and non-abstinent patients receiving antiviral treatment. Le Lan et al. also reported a low rate of treatment discontinuation (16%) in a similar population including abstinent and non-abstinent patients (Le Lan et al., 2012).

Reporting antiviral treatment outcomes among patients with ongoing alcohol consumption during therapy is also of interest. In the ‘alcohol group’, two thirds of the patients consumed alcohol during antiviral therapy, whether they had begun treatment without prior withdrawal or resumed alcohol consumption during therapy. Compared with the ‘control group’, alcohol consumption (even above 30 g/day) did not clearly impact on the rate of SVR. There was, however, a trend toward a better outcome in patients who remained abstinent throughout treatment. In the same line of evidence, Le Lan et al. have reported a significantly higher rate of SVR in former heavy drinkers who remained abstinent throughout antiviral therapy compared with patients with ongoing alcohol consumption (Le Lan et al., 2012).

It is noteworthy that most heavy drinkers at the beginning of the study had reduced their consumption before initiation of therapy or during treatment. Moreover, in more than one-third of the patients, antiviral therapy had led to sustained abstinence. Management of chronic hepatitis C could influence the patient's awareness of his addiction, as knowledge of HCV infection and antiviral treatment have both been shown to conduct patients to reduce their alcohol consumption (Nalpas et al., 2001; Russell et al., 2012). Thus, to start with the management of chronic hepatitis C instead of waiting for a substantial length of abstinence that might never be obtained could be a strategy to get the patient involved in a process of change and to reduce his alcohol consumption.

It is important to bear in mind that interferon can modulate the immune system and hence induce flare-ups of acute alcoholic hepatitis (Zylberberg et al., 1999). As a consequence, antiviral treatment in heavy drinkers with advanced liver disease must be cautiously evaluated. Nevertheless, we believe that the so-called ‘6-months rule’ of abstinence from alcohol before considering antiviral treatment recommended by the French 2002 consensus (Dhumeaux et al., 2003) is excessive. In the present study, most of the patients had started antiviral treatment before achieving this duration of abstinence with satisfactory results in terms of SVR.

Our work has limitations. First, it was a retrospective study without formalized data collection. This could have biased the evaluation of alcohol consumption and the analysis of its interactions with treatment efficacy. Second, the number of patients studied was insufficient to draw definite conclusion from subgroups analyses. For example, despite the fact that the rate of SVR was more than 2-fold higher in former heavy drinkers who had remained completely abstinent during antiviral therapy than in patients who had consumed alcohol during therapy, this difference did not reach statistical significance. Third, the study was not controlled for the type of intervention received by heavy drinkers and the conclusions we draw on the positive impact of our multidisciplinary approach is mainly based on a comparison with previously published data. We cannot exclude that confounding factors could have biased such a comparison. Fourth, we do not provide data on the evolution of liver disease after antiviral treatment. If clinical benefits associated with HCV eradication in heavy drinkers have never been specifically studied, epidemiological data clearly show the harmful effect of ongoing HCV infection in this population. For example, while the odds ratio associated with heavy drinking alone are 9 and 7 for decompensated cirrhosis and hepatocellular carcinoma, respectively, the corresponding figures are 147 and 109 in patients with both heavy drinking and HCV infection (Corrao and Aricò, 1998; Donato et al., 2002). As a consequence, one may speculate that successful antiviral treatment is likely to have a positive impact on the rate of liver-related clinical events in this population.

In conclusion, although alcohol abstinence must remain the primary target in the care of heavy drinkers infected with HCV, our work suggests that it should be, per se, neither a sufficient nor an indispensable prerequisite for the introduction of antiviral therapy. First, we have shown that a multidisciplinary approach including psychosocial support might improve access to treatment, completion of therapy as well as the rate of SVR in HCV-carriers with a history of alcohol abuse. Second, it seems possible, with such an approach, to achieve SVR in some patients with ongoing alcohol consumption during treatment. Moreover, antiviral therapy can be a starting point for change leading the patient to abstinence or at least to reduce his alcohol consumption. Additional work is obviously needed on these topics, but our results should already encourage all professionals involved in the management of patients struggling with alcohol to screen for HCV infection and to consider this population for antiviral treatment.

Conflict of interest statement. None declared.


  • C.E.C. and J.-B.T. have equally contributed and are the joint first authors.


View Abstract