OUP user menu

Prevalence of Hepatitis C Virus Infection in Alcoholic Patients: Cohort Study and Systematic Review

Ignacio Novo-Veleiro, Cristina de la Calle, Susana Domínguez-Quibén, Isabel Pastor, Miguel Marcos, Francisco-Javier Laso
DOI: http://dx.doi.org/10.1093/alcalc/agt044 564-569 First published online: 20 May 2013

Abstract

Aims: Prevalence of chronic hepatitis C virus (HCV) infection among alcoholics is thought to be higher than in the general population, although prevalence rates reported are quite variable. Our study is aimed to analyze HCV prevalence in a cohort of alcoholics and to perform a systematic review on this topic. Patients and methods: A total of 396 alcoholic patients consecutively referred to our Alcoholism Unit were included. HCV infection status and other clinical variables were recorded for each patient. Variables associated with HCV infection were analyzed by means of logistic regression. Additionally, we performed a systematic review focused on previous studies on this topic. Results: Among our alcoholic patients, 14 of them (3.53%) had chronic HCV infection. Variables independently associated with HCV infection were female gender, injection drug use (IDU) and the presence of alcoholic liver disease (ALD). Twenty-four studies analyzing HCV prevalence in alcoholic patients were included in our systematic review, showing prevalence rates of HCV infection ranging from 2.1 to 51% and an average weighted prevalence of 16.32%. Conclusion: In our series, the prevalence rate of chronic HCV infection among alcoholic patients is lower than previously reported, which is probably explained by the relatively low number of patients with ALD or IDU in our sample. Prevalence rates previously published are quite different and the presence of ALD and/or IDU can act as confounding factors for HCV prevalence among alcoholics.

INTRODUCTION

Excessive alcohol consumption is a worldwide problem, which is associated with significant morbidity and mortality (WHO, 2012). Among the potential complications of heavy ethanol intake, alcoholic liver disease (ALD) is one of the most prevalent and severe consequences. This is highlighted by the fact that 18 new cases of ALD per 100,000 inhabitants are diagnosed in the USA every year (Yang et al., 2008). Regarding ALD pathogenesis, it is well known that ethanol-induced liver damage is augmented by other injurious factors such as hepatitis C virus (HCV) infection. The joint effect of alcohol and HCV increases the risk of development of liver cirrhosis (Hutchinson et al., 2005) and liver cancer (Donato et al., 1997), reduces the effectiveness of interferon therapy for HCV (Okazaki et al., 1994) treatment and is associated with higher mortality (Bedogni et al., 2008).

Further, it is traditionally assumed that the prevalence of HCV infection in alcoholic patients is much higher than in the general population, which is estimated between 0.5 and 2% in most European countries (WHO, 2002). The exact burden of HCV infection among alcoholic patients, however, is not known, since a wide range of prevalence rates have been reported [from 2.1% (De Silva et al., 1994) to 51% (Rosman et al., 1993)], which could be related to a different distribution of risk factors for HCV infection among studies (Pares et al., 1990; Nagata et al., 1993; Coelho-Little et al., 1995; Saigal et al., 2002).

Given the above-mentioned differences in reported prevalence rates among alcoholic patients, we decided to carry out an observational study in our setting as well as a systematic review on this topic.

PATIENTS AND METHODS

A total of 396 alcoholic patients consecutively attended at the Alcoholism Unit of the University Hospital of Salamanca were included in this study. All patients had a diagnosis of alcohol abuse or alcohol dependence according to DSM-IV-TR criteria (First, 2004). Patients had been mainly referred from Psychiatry and Internal Medicine Departments to evaluate the presence of alcoholic organ damage and further follow-up if appropriate (Avila et al., 2008). Demographic, epidemiological, clinical and analytical data, including HCV infection status, were collected at the first visit according to an established protocol. Quantification of alcohol consumption was done by means of a semi-structured questionnaire and was measured using the standard drink unit (SDU), which in Spain equals to 10 g of pure alcohol. All patients were screened for HCV infection by detection of HCV IgG antibodies using ELISA and, if positive, further confirmation by detection of circulating HCV-RNA using polymerase chain reaction (PCR) methods, according to current recommendations (Ghany et al., 2009). Patients with no detectable HCV IgG antibodies were considered as non-infected and patients with both IgG antibodies and serum HCV-RNA were considered as infected. ALD, if present, was classified as alcoholic liver cirrhosis, alcoholic hepatitis or alcoholic fatty liver in those cases with histological diagnosis or with unequivocal radiological or clinical signs. For the purpose of this study, patients with persistent abnormal liver function tests (>3 months) in whom liver biopsy was not performed for any reason (e.g. not clinically indicated) were classified as having undetermined hepatic disorder. This study was approved by the Ethics Committee of the University Hospital of Salamanca.

Systematic review

We carried out a systematic review in order to identify observational studies published before December 2012 reporting the prevalence of HCV infection among alcoholic patients. To do so, we searched the Pubmed/Medline database with the following terms: alcohol OR alcoholism OR alcoholic liver disease AND hepatitis C virus and also retrieved additional studies by surveying references of other reports and by using the MEDLINE option ‘Related Articles’.

Statistical methods

Quantitative variables are presented as mean and standard deviation (SD) and qualitative variables are expressed in absolute and relative frequencies. Student's t-test was used for the comparison of quantitative variables, and χ2 test was used for the comparison of qualitative variables, together with Fisher's exact test if appropriate. Variables associated with the presence of HCV infection with P < 0.20 in univariate analysis were included on a multivariable logistic regression model. Odds ratio (OR) and 95% confidence intervals (CI) were reported. Statistical analysis was performed with SPSS program v. 19.0.

RESULTS

The baseline characteristics of the 396 patients included in our study are detailed in Table 1. Of them, 347 were men (87.62%) with a mean age of 44.81 years (SD = 11.54). Average alcohol consumption was 163.71 (106.06) SDUs/week (WHO, 2000). A total of 225 patients (56.8%) met criteria for alcohol dependence and 171 (43.2%) met criteria for alcohol abuse (First, 2004). The association between chronic alcohol consumption and tobacco use was very common since 333 patients (84.1%) were active smokers. Other substances used were cocaine (6.3%), heroin (2.3%) and hashish (1.3%). ALD was present in 120 patients (30.33%).

View this table:
Table 1.

Baseline characteristics and association of variables with the presence of HCV infection

VariablesTotal (n = 396)HCV infected (n = 14)Non-HCV infected (n = 382)P-value
Age (years)44.81 (11.54)47.50 (16.09)44.75 (11.35)0.536
Male gender347 (87.62)10 (71.42)336 (87.95)
Female gender49 (12.37)4 (28.57)46 (12.04)0.062
Rural residence74 (18.78)1 (7.14)73 (19.11)0.213
Current smoker333 (84.09)10 (71.42)323 (84.55)0.865
Former smoker49 (12.37)4 (28.57)45 (11.78)
Never smoker14 (3.53)0 (0)14 (3.66)
Family history of alcoholism203 (51.26)4 (28.57)199 (52.09)0.021
Alcohol dependence225 (56.81)10 (71.42)212 (55.49)0.283
Alcohol abuse171 (43.14)4 (28.58)170 (44.51)0.283
Addiction to other drugs43 (10.85)6 (42.85)37 (9.65)0.001
IDUs9 (2.27)3 (21.42)6 (1.57)<0.001
Major psychiatric illness46 (11.61)1 (7.14)45 (11.78)0.588
HBV infection46 (11.61)4 (28.57)42 (10.99)0.044
Liver disease120 (30.33)9 (64.28)111 (29.05)0.005
 Steatosis50 (12.62)0 (0)50 (13.08)
 Hepatitis6 (1.51)0 (0)6 (1.57)<0.001
 Undetermined42 (10.61)4 (28.57)38 (9.94)
 Cirrhosis22 (5.55)5 (35.71)17 (4.45)
SDU/week163.71 (106.06)198.36 (111.57)162.15 (105.88)0.253
Age of alcoholism onset (years)20.56 (9.73)23.45 (13.16)20.44 (9.59)0.469
AST (IU/l)48.87 (58.36)93.15 (81.99)47.23 (56.84)0.005
ALT (IU/l)46.21 (43.44)93.15 (60.74)44.39 (41.79)<0.001
ALP (IU/l)145.91 (99.92)254.31 (336.49)141.87 (76.68)<0.001
GGT (IU/l)140.25 (252.17)236.96 (214.29)136.68 (253.41)0.124
Hemoglobin (g/dl)15.11 (1.62)15.41 (1.12)15.11 (1.64)0.524
MCV (fl)96.21 (7.03)97.07 (6.29)96.19 (7.07)0.674
  • Quantitative variables are presented as mean (SD), while qualitative variables are presented as absolute frequency (percentage).

  • AST, aspartate aminotransferase; ALT, alanine aminotransferase; GGT, gamma-glutamyl-transpeptidase; ALP, alkaline phosphatase; MCV, mean corpuscular volume; HBV, hepatitis B virus; HCV, hepatitis C virus; SDU, standard drink unit; IDU, intravenous drug user.

As shown in Table 1, 14 patients (3.53%) had chronic HCV infection. Several variables were significantly associated with the presence of this disease in univariate analysis: alcoholic patients with HCV infection are more likely to have addiction to other drugs (P < 0.005) and to be current or past injection drug users (IDUs) (P < 0.001). This is emphasized by the fact that only two alcoholics had HCV infection among those who did not have addiction to other drugs or were not IDUs (n = 353). Moreover, patients with HCV infection were more likely to have liver damage and advanced forms of liver disease; indeed, 22.72% of the cirrhotic patients in our series had chronic HCV infection. Other demographic, analytical or clinical variables were not associated with HCV infection.

After performing a multivariable logistic regression analysis, we found that variables significantly and independently associated with HCV infection (P < 0.05) were female gender (OR = 4.86, 95% CI = 1.28–18.4), current or past IDU (OR = 27.09, 95% CI = 3.88–189.27) and ALD (OR = 3.84, 95% CI = 1.14–12.84). Collinearity was found between transaminase elevation and presence of liver disease and only this variable was included in the multivariable model.

Systematic review

Our search strategy initially found 335 articles, of which 290 were excluded after a review of the title and abstract. Thirty-five articles were fully read and 11 articles were excluded because of overlapping data or different study designs. Therefore, 24 studies analyzing the prevalence of chronic HCV infection in alcoholic patients were included in our systematic review.

These studies are summarized in Table 2, which shows that the prevalence of HCV is highly variable, ranging from 2.1 to 51%. Average weighted prevalence, including all studies and our own data, was 16.32%. It is of note, however, that if we consider only alcoholic patients without IDU or without liver disease, prevalence rates are much lower, even of 0% in several studies. In line with this, if we consider only those studies that provided HCV prevalence among patients with and without IDU (13 studies with a total of 2646 alcoholics), average weighted prevalence of HCV infection among alcoholics with no IDU was 6.6% and that among alcoholics with IDU was 72.8%. Further, if we focus on studies that provided data on ALD status (17 studies with 1737 patients), prevalence of HCV infection among patients without ALD or with mild ALD (steatosis) was 5.9%, whereas prevalence among alcoholics with severe ALD was 32.9%. It is important to remind that we could not extract data from patients without both ALD and IDU. Therefore, data from patients without ALD included patients with IDU and data from patients without IDU included patients with ALD.

View this table:
Table 2.

Studies found in our systematic review regarding prevalence of HCV infection in alcoholic patients

First authorYearnPrevalence of chronic HCV infection [n (%)]Prevalence of liver disease [n (%)]Prevalence of chronic HCV infection according to liver disease [n (%)]Parenteral drug use prevalence [n (%)]HCV in IDUs [n (%)]HCV in non-IDUs [n (%)]
Parés et al.199014435 (24.3)No liver disease: 45 (31.2)

Steatosis: 20 (13.9)

Hepatitis: 14 (9.7)

Cirrhosis: 65 (45.1)
No liver disease: 1 (2.2)

Steatosis: 4 (20.0)

Hepatitis: 6 (42.9)

Cirrhosis: 28 (43.1)
N/AN/AN/A
Brillanti et al.19914113 (31.7)No liver disease: 0 (0)

Steatosis: 11 (26.8)

Cirrhosis: 30 (73.2)
Steatosis: 5 (45.5)

Cirrhosis: 10 (33.3)
N/AN/AN/A
Bode et al.19917318 (24.6)No liver disease: 0 (0)

Steatosis: 34 (46.5)

Cirrhosis: 39 (53.5)
Steatosis: 3 (8.8)

Cirrhosis: 15 (38.5)
N/AN/AN/A
Caldwell et al.19913711 (29.7)Not specified: 18 (48.6)

Cirrhosis: 19 (51.4)
Not specified: 5 (27.8)

Cirrhosis: 6 (31.6)
1 (2.7)N/AN/A
Nalpas et al.19926218 (29.0)No liver disease: 0 (0)

Not specified: 11 (17.5)

Steatosis: 8 (12.9)

Hepatitis: 9 (14.5)

Cirrhosis: 34 (54.8)
Not specified: 6 (54.5)

Steatosis: 2 (25.0)

Hepatitis: 3 (33.3)

Cirrhosis: 7 (20.6)
N/AN/AN/A
McHutchison et al.199213918 (12.9)No liver disease: 0 (0)

Not specified: 109 (78.4)

Steatosis: 1 (0.7)

Fibrosis: 6 (4.3)

Hepatitis: 10 (7.2)

Cirrhosis: 13 (9.3)
Not specified: 16 (14.7)

Steatosis: 0 (0)

Fibrosis: 0 (0)

Hepatitis: 1 (10.0)

Cirrhosis: 1 (7.7)
14 (10.1)6 (42.9)12 (9.6)
Nagata et al.199310031 (31.0)N/AN/AN/AN/AN/A
Verbaan et al.199331045 (14.5)N/AN/A52 (16.8)39 (75.0)6 (2.3)
Mendenhall et al.199328853 (18.4)No liver disease: 0 (0)

Not specified: 288 (100)
N/A25 (8.7)17 (68.0)36 (13.7)
Rosman et al.19934724 (51.1)No liver disease: 0 (0)

Steatosis: 9 (19.1)

Fibrosis: 20 (42.5)

Hepatitis: 5 (10.6)

Cirrhosis: 13 (27.6)
Steatosis: 0 (0)

Fibrosis: 18 (90.0)

Hepatitis: 1 (20.0)

Cirrhosis: 5 (38.5)
17 (36.2)16 (94.1)8 (26.6)
De Silva et al.1994471 (2.1)No liver disease: 0 (0)

Cirrhosis: 47 (100)
Cirrhosis: 1 (2.1)N/AN/AN/A
Jiang et al.19955017 (34.0)No liver disease: 12 (24.0)

Not specified: 26 (52.0)

Hepatitis: 5 (10.0)

Cirrhosis: 7 (14.0)
No liver disease: 0 (0)

Not specified: 10 (38.5)

Hepatitis: 4 (80.0)

Cirrhosis: 3 (42.9)
26 (52.0)16 (61.5)1 (4.2)
Befrits et al.199520129 (14.4)No liver disease: 161 (80.1)

Steatosis: 9 (4.5)

Hepatitis 6 (3.0)

Cirrhosis: 25 (12.4)
No liver disease: 13 (8.1)

Steatosis: 4 (44.4)

Hepatitis: 5 (83.3)

Cirrhosis: 7 (28.0)
21 (10.5)17 (80.9)12 (6.7)
Coelho-Little et al.199510020 (20.0)No liver disease: 59 (59.0)

Not specified: 31 (31.0)

Steatosis: 2 (2.0)

Hepatitis: 7 (7.0)

Cirrhosis: 1 (1.0)
No liver disease: 4 (6.8)

Not specified: 6 (19.4)

Steatosis: 2 (100)

Hepatitis: 7 (100)

Cirrhosis: 1 (100)
21 (21.0)16 (76.2)4 (5.1)
González Quintela et al.199518012 (6.7)No liver disease: 0 (0)

Not specified: 36 (20.0)

Steatosis: 23 (12.8)

Hepatitis: 28 (15.5)

Fibrosis: 27 (15.0)

Cirrhosis: 66 (36.6)
Not specified: 0 (0)

Steatosis: 0 (0)

Hepatitis: 0 (0)

Fibrosis: 1 (3.7)

Cirrhosis: 11 (16.7)
0 (0)12 (6.7)
Curciarello et al.199611523 (20.0)N/AN/AN/AN/AN/A
Sata et al.1996252103 (41.3)No liver disease: 0 (0)

Steatosis: 23 (9.1)

Hepatitis: 27 (10.7)

Fibrosis: 69 (27.4)

Not specified: 50 (19.8)

Cirrhosis: 55 (21.8)

Hepatocarcinoma: 28 (11.1)
Steatosis: 1 (4.3)

Hepatitis: 6 (22.2)

Fibrosis: 7 (10.1)

Not specified: 42 (84.0)

Cirrhosis: 20 (36.4)

Hepatocarcinoma: 28 (100)
N/AN/AN/A
Rosman et al.199619655 (28.1)N/AN/A50 (25.5)41 (82.0)14 (9.6)
Srugo et al.199849638 (7.6)N/AN/A5 (1.0)2 (40.0)36 (7.3)
Kwon et al.200016214 (8.6)No liver disease: 0 (0)

Cirrhosis: 162 (100)
Cirrhosis: 14 (8.6)1 (0.6)1 (100)13 (8.1)
Tanaka et al.200014361 (42.6)No liver disease: 0 (0)

Steatosis: 7 (4.9)

Hepatitis: 24 (16.7)

Fibrosis: 18 (12.6)

Chronic hepatitis: 39 (27.3)

Cirrhosis: 42 (29.4)

Hepatocarcinoma: 13 (9.1)
Steatosis: 0 (0)

Hepatitis: 2 (8.3)

Fibrosis: 1 (5.6)

Chronic hepatitis: 27 (69.2)

Cirrhosis: 24 (57.1)

Hepatocarcinoma: 7 (53.8)
N/AN/AN/A
Saigal et al.200221020 (9.5)No liver disease: 0 (0)

Cirrhosis: 210 (100)
Cirrhosis: 20 (9.5)N/AN/AN/A
Galperim et al.200611417 (14.9)N/AN/A15 (13.2)13 (86.7)4 (4.1)
Tsui et al.20066354984 (15.5)N/AN/AN/AN/AN/A
  • N/A, non-available. IDUs, intravenous drug users.

DISCUSSION

In the present study, we have found a low prevalence of chronic HCV infection (3.53%) in a cohort of 396 consecutive alcoholic patients that had been attended at a specific unit. This is in contrast with previous reports in this field, since many authors have found prevalence rates of HCV infection around 10–40% among alcoholics (Pares et al., 1990; Bode et al., 1991; Brillanti et al., 1991; Caldwell et al., 1991; Mchutchison et al., 1992; Nalpas et al., 1992; Mendenhall et al., 1993; Nagata et al., 1993; Rosman et al., 1993, 1996; Verbaan et al., 1993; Befrits et al., 1995; Coelho-Little et al., 1995; Jiang et al., 1995; Curciarello et al., 1996; Sata et al., 1996; Tanaka et al., 2000; Galperim et al., 2006; Tsui et al., 2006). In order to further analyze this discrepancy, we have systematically reviewed previous studies on this topic and we have found that prevalence rates are quite variable, particularly according to the presence or absence of either liver disease or injection drug use. Therefore, these two factors may act as confounders for the real prevalence of HCV infection among alcoholics.

Regarding the presence of liver disease, our systematic review revealed that many previous studies have included a large number of patients with liver cirrhosis or other types of ALD (Pares et al., 1990; Bode et al., 1991; Brillanti et al., 1991; Caldwell et al., 1991; Mchutchison et al., 1992; Nalpas et al., 1992; Mendenhall et al., 1993; Nagata et al., 1993; Rosman et al., 1993, 1996; Coelho-Little et al., 1995; Jiang et al., 1995; Sata et al., 1996; Tanaka et al., 2000). In some of these studies, the presence of ALD could have determined a selection bias, since patients were even included because of the presence of this disease (Bode et al., 1991; Brillanti et al., 1991; Mchutchison et al., 1992; Nalpas et al., 1992; Mendenhall et al., 1993; Nagata et al., 1993; Rosman et al., 1993; Sata et al., 1996; Kwon et al., 2000; Tanaka et al., 2000; Saigal et al., 2002). Since HCV infection is a co-factor of ethanol for the development of liver damage, it is clear that alcoholics with liver disease are more likely to have chronic HCV infection than alcoholics without liver damage. Further, patients with excessive alcohol consumption and ALD are more likely to seek medical care in the presence of manifestations of liver damage (e.g. elevated liver function tests), thus favoring the detection of HCV infection in alcoholics with liver disease. Conversely, alcoholics without liver disease may be less likely to be included in studies regarding HCV prevalence since they may not be receiving specialized medical care. In any case, a limitation of our study is that we do not have liver histology for all patients, and therefore we must be cautious about conclusions regarding patients with ALD.

Another potential confounding factor is IDU, a well-known risk factor for HCV infection regardless of ethanol intake. Since a high proportion of IDU patients have been included in some studies, the presence of this risk factor makes it difficult to establish the role of alcoholism itself in HCV prevalence (Caldwell et al., 1991; Rosman et al., 1993; Verbaan et al., 1993; Befrits et al., 1995; Jiang et al., 1995; Galperim et al., 2006).

In the present study, only 9 patients had previous IDU and 120 patients had any form of liver disease (30.3%). Therefore, the relatively low frequency of these two variables when compared with other studies could contribute to explain the lower prevalence of HCV infection that we have found in our study (Pares et al., 1990; Bode et al., 1991; Brillanti et al., 1991; Caldwell et al., 1991; Mchutchison et al., 1992; Nalpas et al., 1992; Mendenhall et al., 1993; Nagata et al., 1993; Rosman et al., 1993, 1996; Verbaan et al., 1993; Befrits et al., 1995; Coelho-Little et al., 1995; Jiang et al., 1995; Curciarello et al., 1996; Sata et al., 1996; Tanaka et al., 2000; Galperim et al., 2006; Tsui et al., 2006). Since patients with ALD or IDU are not overrepresented in our sample, we believe that our work offers a realistic and global view of the prevalence of HCV infection among patients with excessive alcohol consumption. Actually, if we consider only those alcoholic patients who do not have ALD or IDU, the prevalence of chronic HCV infection in this subgroup is as low as 1.09%. This prevalence is similar to HCV infection prevalence in Spain, which ranges between 1 and 2.6% (Bruguera and Forns, 2006), and even lower than seroprevalence of blood donors in our hospital (2–3%) (unpublished data, 2012).

It is well established that HCV infection in alcoholic patients is a key factor that could contribute to ethanol-induced liver damage and to the development of severe forms of liver disease (Hutchinson et al., 2005). However, evidence for alcoholism as a risk factor per se for acquisition of HCV infection is lacking (Verbaan et al., 1993; De Silva et al., 1994; Gonzalez Quintela et al., 1995; Jiang et al., 1995; Srugo et al., 1998; Galperim et al., 2006). It is true that excessive ethanol intake may promote some activities such as IDU, but reports regarding this issue are controversial (Brillanti et al., 1991; Caldwell et al., 1991; Nalpas et al., 1992; Verbaan et al., 1993; Jiang et al., 1995). Therefore, further studies are needed in order to clarify the role of excessive alcohol consumption as a risk factor for HCV infection. Regardless of other recommendations for the screening of HCV infection (Smith et al., 2012), it is also unclear whether all alcoholic patients should be screened for HCV infection or only those with factors associated with HCV infection. In line with this, liver disease and IDU appear to be the main conditions associated with HCV infection in alcoholic patients, as our study shows.

In summary, in our series, we have found prevalence rates of chronic HCV infection in alcoholic patients without liver disease or prior intravenous drug use close to those of the general population. Prevalence rates of HCV infection reported in other series are quite variable and both the presence of liver disease and previous or current injection drug use may act as confounding factors for HCV prevalence among alcoholic patients.

Funding

This work has been partially supported by project PI10/01692 from the Spanish Ministry of Science and Innovation, Instituto de Salud Carlos III (ISCIII) and the European Union FEDER funds (“Una manera de hacer Europa”) and by project GRS 531/A/10 from Junta de Castilla y León. Conflict of interest statement. None declared.

Footnotes

  • M.M. and F.J.L. are both senior authors of this paper.

References

View Abstract