Potential conflict of interest: Dr. Bini received grants from Schering-Plough, Roche, Intermune, Pharmasset, Bristol-Myers Squibb, GlaxoSmithKline, and Gilead.
Little is known about hepatitis B virus (HBV) infection among patients with chronic hepatitis C virus (HCV) infection in the United States. We prospectively enrolled 1,257 patients with chronic HCV infection from two medical centers in New York City. A total of 61.5% (95% confidence interval, 58.8%–64.2%) had evidence of prior exposure to HBV (hepatitis B core antibody–positive), whereas 5.8% (95% confidence interval, 4.5%–7.1%) had dual infection with HBV (hepatitis B surface antigen–positive). Multivariable logistic regression analysis identified age <40 years, Asian race, injection drug use, and a greater number of lifetime sexual partners as independent risk factors for HBV-HCV dual infection. Liver biopsy results in 26 HBV-HCV–infected and 658 HCV-monoinfected patients showed that stage 3 or 4 fibrosis was significantly more common in those with HBV-HCV dual infection (84.6% versus 29.9%; P < 0.001). Patients infected with HBV and HCV had significantly lower median HCV RNA levels (1.3 versus 4.5 × 106 copies/mL; P < 0.001) and were less likely to have HCV RNA levels ≥5 × 106 copies/mL (12.3% versus 45.4%; P < 0.001) than those who had HCV monoinfection. All five patients with HBV-HCV dual infection who had undetectable HBV DNA levels had HCV RNA levels ≥5 × 106 copies/mL. Conclusion: American patients with chronic HCV infection should be tested for HBV, especially younger patients, Asians, injection drug users, and those with an increased number of lifetime sexual partners. The presence of severe liver disease and HBV-HCV viral interactions in patients with dual infection necessitates careful but aggressive clinical management, although the optimal strategy remains to be determined. (HEPATOLOGY 2010.)
There is growing interest in hepatitis B virus (HBV) and hepatitis C virus (HCV) dual infection for a number of reasons. First, several studies have shown that HBV-HCV dual infection is associated with faster hepatic fibrosis progression rates, more severe liver disease, a higher risk of progressing to cirrhosis, and a greater risk of hepatic decompensation as compared with HBV or HCV monoinfection.1–7 Second, HBV-HCV dual infection has been shown to significantly increase the risk of developing hepatocellular carcinoma as compared with HBV or HCV monoinfection.8, 9
Third, HBV and HCV are the most common causes of liver disease worldwide, with an estimated 350 million individuals who have chronic HBV infection and 170 million individuals who have chronic HCV infection.10–13 Although HBV-HCV dual infection is common due to shared routes of transmission, the worldwide prevalence of dual infection with HBV-HCV is unknown due to a lack of large population-based studies. Data from several published studies from Spain,1 Italy,14–16 Japan,17, 18 Taiwan,19 and Iran20 have shown that approximately 10%–15% of patients with chronic HBV infection are also infected with HCV.
The prevalence of HCV infection among patients with chronic HBV infection has been the subject of several published reports,1, 14–20 but far less is known about the proportion of patients with chronic HCV infection who are also infected with HBV, especially in the United States. In addition, although it is well known that the prevalences of HBV monoinfection and HCV monoinfection differ according to race/ethnicity,21–24 it is unclear whether there are racial/ethnic differences in the prevalence of HBV-HCV dual infection.
Finally, HBV-HCV dual infection is an important area to investigate because complex viral interactions between these two viruses have been reported. Interference between HBV and HCV in patients with HBV-HCV dual infection with inhibition of the replication of one of the viruses has been described in several studies,2, 17, 25–29 but this phenomenon is incompletely understood, and the mechanism by which this occurs is yet to be established.30, 31
Given the paucity of data on the epidemiology of HBV-HCV dual infection in the United States, we conducted this prospective study to determine the prevalence of hepatitis B surface antigen (HBsAg) seropositivity among a well-characterized cohort of Americans with chronic HCV infection, to evaluate whether the prevalence of dual infection with HBV differs according to race/ethnicity or other demographic characteristics, and to describe HBV-HCV viral interactions in this population.
ALT, alanine aminotransferase; CI, confidence interval; HBcAb, hepatitis B core antibody; HBeAg, hepatitis B e antigen; HBsAg, hepatitis B surface antigen; HBV, hepatitis B virus; HCV, hepatitis C virus; IQR, interquartile range; NHANES, National Health and Nutrition Examination Survey.
Patients and Methods
Subjects were recruited from the outpatient Gastroenterology, Primary Care, and Infectious Diseases clinics at the VA New York Harbor Healthcare System and Bellevue Hospital Center in New York City between September 1998 and July 2004. Patients were eligible for the study if they were ≥18 years of age, were seropositive for HCV antibody (Ortho HCV ELISA version 3.0; Ortho-Clinical Diagnostics, Inc., Raritan, NJ), and had detectable HCV RNA (COBAS Amplicor HCV Monitor Test version 2.0, Roche Molecular Systems, Branchburg, NJ).
Patients were excluded if they were coinfected with human immunodeficiency virus or if they were previously treated for HBV or HCV. All patients provided written informed consent, and the study was approved by the Institutional Review Board at our medical center.
A trained research assistant interviewed all patients and obtained detailed demographic and clinical information. Data collected on each patient included age, sex, self-reported race/ethnicity, current alcohol use, and risk factors for HBV and HCV infection, including injection drug use, blood transfusions prior to 1992, number of lifetime sexual partners, and sex with a same-sex partner.
Laboratory testing performed on each subject included complete blood count, serum electrolytes, liver profile, prothrombin time, HCV RNA testing, and HCV genotyping (Inno-LiPA HCV II assay; Innogenetics, Gent, Belgium), HBsAg, hepatitis B surface antibody, and hepatitis B core antibody (HBcAb). Patients with a positive HBsAg test had further testing for hepatitis B e antigen (HBeAg), hepatitis B e antibody, HBV DNA, hepatitis D antigen, and hepatitis D antibody. All HBV antigen, HBV antibody, and hepatitis D antibody testing was performed using commercially available assays (Abbott Laboratories, Abbott Park, IL), and HBV DNA testing was performed using the HBV NGI SuperQuant assay (National Genetics Institute, Los Angeles, CA; quantitative range, 100–5,000,000,000 copies/mL). Testing for hepatitis D antigen was performed using a commercially available assay (Focus Diagnostics, Inc., Cypress, CA).
Patients who agreed to a liver biopsy had the procedure performed percutaneously using a Jamshidi needle. Biopsy specimens were fixed in formalin, embedded in paraffin, and stained with hematoxylin-eosin and Masson trichrome stains. Biopsy specimens were evaluated by a board-certified staff pathologist who was blinded to the aims of this study. Necroinflammatory activity was scored from 0 to 18 using the Ishak grading system,32 whereas fibrosis was scored on a scale from 0 to 4 using the Scheuer staging system.33 Using the classification system for steatohepatitis proposed by Brunt et al.,34 steatosis was graded as 0 (none), 1 (<33% of hepatocytes), 2 (33%–66% of hepatocytes), or 3 (>66% of hepatocytes).
The primary outcome measure was the proportion of patients with chronic HCV who were also infected with HBV. For the primary outcome measure, patients were considered to have HBV dual infection if they had a positive HBsAg test.
Secondary outcome measures included the proportion of HCV-infected patients who have been exposed to HBV (HBcAb-positive), the proportion of HBV-HCV–infected subjects who were HBeAg-positive, the prevalence of HBV dual infection according to race/ethnicity and other demographic characteristics, the severity of liver disease in HBV-HCV infected subjects, and viral interactions between HBV and HCV.
To evaluate the severity of liver disease, we compared alanine aminotransferase (ALT) levels, the grade of inflammation (necroinflammatory score), the stage of fibrosis, the severity of steatosis, and fibrosis progression rates between HBV-HCV–infected patients and those with HCV monoinfection. Persistently normal ALT levels were defined as at least three normal ALT levels (<40 U/L) on separate occasions at least 1 month apart over a 6-month period.35, 36 Subjects were not considered as having persistently normal ALT levels if they ever had any documented ALT level above the upper limit of normal.
In patients with an identifiable risk factor for HCV infection, fibrosis progression rates were calculated by dividing the stage of fibrosis by the duration of HCV infection as described.37, 38 The duration of HCV infection was estimated from the date of first exposure to injection drug use or blood transfusion. Fibrosis progression rates are expressed as units of fibrosis per year. To evaluate for viral interactions, we compared the median HCV RNA levels between patients infected with both HBV and HCV with those who had HCV monoinfection.
Continuous data are expressed as the mean ± standard deviation for those variables that were normally distributed, and the median and interquartile range (IQR) (25th–75th percentile) for those with a nonnormal distribution. Continuous variables were compared using an unpaired t test or the Mann-Whitney U test, as appropriate. Categorical variables are expressed as proportions and were compared using the chi-square test or Fisher's exact test.
Univariate analyses were used to identify factors associated with HBV-HCV dual infection. Subsequently, a multivariable logistic regression model was created using all significant variables identified in the univariate analyses (P < 0.05). The strength of the association between covariates and presence of HBV-HCV dual infection are expressed as the odds ratio with 95% confidence interval (CI). Statistical analysis was performed using SPSS software version 15.0 for Windows (SPSS Inc., Chicago, IL); a two-tailed P value of <0.05 was considered statistically significant.
Baseline Characteristics of HCV-Infected Subjects.
A total of 1,257 patients with chronic HCV infection were enrolled in the study, including 982 patients from the VA New York Harbor Healthcare System and 275 from Bellevue Hospital Center. The baseline demographic and clinical characteristics of the subjects are shown in Table 1. The majority of subjects were male and the population was racially/ethnically diverse. Current or prior injection drug use was the most common risk factor for HCV infection identified, and nearly half of the subjects had more than 25 lifetime sexual partners. As expected, the majority of patients were infected with HCV genotype 1.
Table 1. Demographic and Clinical Characteristics of the 1,257 Patients with Chronic HCV Infection
Of the 1,257 subjects with chronic HCV infection, 773 (61.5%; 95% CI, 58.8%–64.2%) had evidence of prior exposure to HBV (HBcAb-positive), whereas 73 (5.8%; 95% CI, 4.5%–7.1%) had dual infection with HBV (HBsAg-positive). All of the HBcAb-positive subjects were positive for total antibody against the HBV core antigen, but none were HBc immunoglobulin M–positive. The prevalence of HBV dual infection was significantly higher in patients enrolled at the VA New York Harbor Healthcare System than those subjects seen at Bellevue Hospital Center (6.5% versus 3.3% [P = 0.04]). Among the 73 patients with HBV-HCV dual infection, 54 (74.0%; 95% CI, 63.7%–84.3%) were HBeAg-positive and 2 (2.7%; 95% CI, 1.1%–6.6%) were hepatitis D antigen–positive.
To further evaluate the epidemiology of HBV infection among patients with chronic HCV infection, we determined the prevalence of HBV-HCV dual infection according to race/ethnicity and other select demographic and clinical characteristics (Table 2). The prevalence of HBV dual infection differed significantly according to age, race/ethnicity, injection drug use, and the number of lifetime sexual partners. HBV dual infection was highest in those <40 years of age, whereas it was lowest in those who were 50–59 years of age. Asians had the highest prevalence of HBV dual infection followed by blacks. The prevalence of HBV infection was strongly associated with injection drug use and increased as the number of lifetime sexual partners increased.
Table 2. Prevalence of HBV-HCV Dual infection in the 1,257 Patients with Chronic HCV Infection According to Select Demographic and Clinical Characteristics
Number of Subjects Tested
Proportion of Subjects Infected with HBV-HCV
Hispanic or Latino
Born in the United States
Current alcohol use
Injection drug use
Transfusion prior to 1992
Number of lifetime sexual partners
Sex with same-sex partner
Using multivariable logistic regression analysis, we found that HBV-HCV dual infection was independently associated with age, race/ethnicity, injection drug use, and the number of lifetime sexual partners (Table 3). The strongest factor was an increased number of lifetime sexual partners, followed closely by injection drug use and Asian race.
Table 3. Factors Associated with HBV-HCV Dual Infection in Multivariable Logistic Regression Analysis
Odds Ratio (95% CI) of Being Infected with HBV-HCV
Hispanic or Latino
Injection drug use
Number of lifetime sexual partners
Comparison of Asian and Non-Asian HBV-HCV–Infected Patients.
In order to evaluate differences between Asian and non-Asian patients with HBV-HCV dual infection, we compared the virologic characteristics of these two groups of patients (Table 4). Compared with non-Asian patients, Asians had lower HCV RNA levels and HBV DNA levels. In contrast, there were no differences in HCV genotypes or the proportion of subjects who were HBeAg-positive between the two groups.
Table 4. Virologic Characteristics of the 73 Asian and Non-Asian Patients with HBV-HCV Dual Infection
The median ALT level was 75.0 U/L (IQR, 56.5–85.5 U/L) in patients with HBV-HCV dual infection and was 49.0 U/L (IQR, 31.0–88.0 U/L) in patients with HCV monoinfection (P = 0.006). In addition, HBV-HCV–infected patients were significantly less likely to have persistently normal ALT levels than were those with HCV monoinfection (15.1% versus 38.7%; P < 0.001).
A liver biopsy was performed in 684 (54.4%) of the 1,257 subjects enrolled in this study, including 26 patients with HBV-HCV dual infection and 658 of those with HCV monoinfection. The median necroinflammatory score was significantly higher in HBV-HCV–infected patients than in those with HCV monoinfection (8.0 [IQR, 6.0–10.0] versus 5.0 [IQR, 4.0–7.0]; P < 0.001). HBV-HCV–infected patients also had more advanced fibrosis than those with HCV monoinfection (Fig. 1). Stage 3 or 4 fibrosis was present in 84.6% of patients with HBV-HCV infection and 29.9% of those with HCV monoinfection (P < 0.001).
Among subjects with a presumed known date of HCV infection, the median fibrosis progression rate was 0.15 U/year (IQR, 0.13–0.17) in HBV-HCV–infected patients and this was significantly higher than the fibrosis progression rate of 0.07 U/year (IQR, 0.03–0.10) in those with HCV monoinfection (P < 0.001).
In addition to having more severe necroinflammation and fibrosis, HBV-HCV–infected patients had more severe steatosis than those with HCV monoinfection (Fig. 2). Steatosis of any grade (96.2% versus 54.0%; P < 0.001) and grade 2 or 3 steatosis (61.5% versus 22.2%; P < 0.001) were significantly more common in HBV-HCV–infected patients than in those infected with HCV alone.
To evaluate whether the greater severity of steatosis in HBV-HCV–infected patients than in those with HCV monoinfection was due to differences in metabolic factors, we compared body mass index, total cholesterol, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, serum triglycerides, and serum glucose levels between the two groups of subjects and found no significant differences (data not shown).
HBV-HCV Viral Interactions.
Among the entire cohort of 1,257 patients, HBV-HCV–infected patients had significantly lower median HCV RNA levels (1.3 [IQR, 0.2–2.5] versus 4.5 [IQR, 1.2–5.0] × 106 copies/mL; P < 0.001) and were less likely to have HCV RNA levels ≥5 × 106 copies/mL (12.3% versus 45.4%; P < 0.001) than those with HCV monoinfection. The proportion of patients infected with HCV genotype 1 did not differ between the HBV-HCV–infected patients and the HCV–monoinfected subjects (71.2% versus 79.1%; P = 0.11).
The median HBV viral load among the 73 patients with HBV-HCV–infection was 0.9 (IQR, 0.2–8.8) × 105 copies/mL. As expected, patients who were HBeAg-positive had significantly higher median HBV levels than those who were HBeAg-negative (2.1 [IQR, 0.2–13.0] versus 0.2 [IQR, 0.06–0.7] × 105 copies/mL; P = 0.001). Five of the 73 patients with HBV-HCV dual infection (6.8%) had undetectable HBV DNA levels. Of these five patients, four were non-Hispanic blacks and one was a non-Hispanic white patient. It is interesting to note that all five of the HBV-HCV–infected patients with undetectable HBV DNA levels had HCV RNA levels ≥5 × 106 copies/mL.
In the present study of 1,257 subjects with chronic HCV infection, we found that 5.8% were also infected with HBV. According to data from the 1999–2002 National Health and Nutrition Examination Survey (NHANES), there are an estimated 3.2 million individuals with chronic HCV infection in the United States.24 Therefore, there are potentially a large number of Americans with chronic HCV infection who have dual infection with HBV.
Although the burden of chronic HCV infection in the United States has been well studied,23, 24 it is unknown how many Americans with chronic HCV infection are also infected with HBV, because nearly all studies of HBV-HCV–infected patients have been performed outside of the United States. In a landmark paper from the NHANES III study, Alter et al. reported that 25.5% of HCV-infected Americans had been exposed to HBV (HBcAb positive). However, that paper, as well as the most recent paper from the 1999–2002 NHANES study, did not report the proportion of HCV-infected Americans who were also infected with HBV (HBsAg-positive).23, 24
Our study builds upon these prior studies by evaluating the proportion of HCV-infected American patients who have been exposed to HBV, as well as the proportion of subjects who have HBV dual infection. The prevalence of prior exposure to HBV among our subjects with chronic HCV infection (61.5%) is much higher than the prevalence of HBV exposure among HCV-infected patients reported in NHANES III (25.5%)23 but similar to other cohorts of HCV-infected subjects.39, 40 The reasons for this difference are likely due to differences in the populations evaluated.
In addition to evaluating the proportion of patients with chronic HCV infection who had dual infection with HBV, we found several important factors that were associated with HBV-HCV infection. Multivariable logistic regression analysis identified age <40 years, Asian race, injection drug use, and a greater number of lifetime sexual partners as the only independent risk factors for dual infection with HBV. The marked racial/ethnic differences in the prevalence of HBV infection among subjects with chronic HCV infection are in agreement with a prior study from California showing that HBV-HCV infection was highest among Asians (11.0%) versus Caucasians (5.6%), African Americans (3.9%), or Latinos (2.4%).41 Taken together, these findings may be helpful in identifying HCV-infected patients who are likely to also be infected with HBV.
Another interesting and clinically important finding of our study was the marked differences in the severity of liver disease between the two groups. Compared with HCV-monoinfected patients, patients with HBV-HCV dual infection had higher ALT levels, a higher necroinflammatory score, more advanced fibrosis, a faster fibrosis progression rate, and more severe steatosis. Although the finding that HBV-HCV dual infection is associated with more advanced liver disease than HBV or HCV monoinfection has been described,1–7 we are unaware of any studies that have evaluated the grade of hepatic steatosis in this population. The reasons for the greater severity of steatosis in the patients with HBV-HCV dual infection are unknown. Metabolic causes are not a likely explanation because there were no differences in body mass index or serum levels of cholesterol, triglycerides, or glucose between the two groups. However, we did not evaluate insulin resistance and this should be explored in future studies. An alternate explanation for the greater severity of steatosis in patients dually infected with HBV and HCV is that infection with both viruses had an additive or synergistic effect in producing steatosis. This latter hypothesis is supported by studies demonstrating that HBV and HCV can each be independently associated with hepatic steatosis, although the association between HBV infection and hepatic steatosis is controversial and may be mediated by host metabolic factors rather than viral factors.42–45
The presence of severe liver disease in patients with HBV-HCV dual infection, along with data showing that HBV-HCV–infected patients are at very high risk of developing hepatocellular carcinoma,8, 9 suggest that HBV-HCV–infected patients should be offered a liver biopsy, aggressive treatment of the dominant virus, and screening for hepatocellular carcinoma. To date, however, there are no published guidelines for the management of HBV-HCV–infected patients.
Patients infected with both HBV and HCV had significantly lower median HCV RNA levels and were less likely to have HCV RNA levels ≥5 × 106 copies/mL than those with HCV monoinfection. In addition, all five HBV-HCV–infected patients with undetectable HBV DNA levels had HCV RNA levels ≥5 × 106 copies/mL. The lower median HCV RNA levels in HBV-HCV–infected patients provides evidence for a suppressive effect of HBV on HCV, whereas the high HCV RNA levels in the patients with undetectable HBV DNA provides evidence that HCV can also exert a suppressive effect on HBV. These findings confirm previous reports and provide additional evidence for the complex viral interactions between HBV and HCV.2, 18, 46–48 However, two recent studies demonstrated the lack of viral interactions between HBV and HCV in vitro, suggesting that the reciprocal viral suppression observed in vivo may be mediated by host innate and/or adaptive immune responses.30, 31
The strengths of our study include the prospective study design, collection of detailed demographic and clinical data, the large sample size, the racial/ethnic diversity of our patients, and the availability of liver biopsy results in a large number of subjects.
There are, however, several limitations of our study that should be considered when interpreting our findings. First, the majority of our patients were men and were enrolled at two medical centers in New York City. Therefore, our findings may not be generalizable to women or to other clinical settings. Second, the study was cross–sectional, and a longitudinal assessment of HBV-HCV viremia would provide a more comprehensive picture of viral interactions, because this is likely a dynamic process, as demonstrated by Raimondo et al.48 Third, this cross-sectional study did not evaluate treatment or long-term outcomes of HBV-HCV dual infection. Fourth, a liver biopsy was not performed in all subjects, and it is possible that the liver histology in those who had a biopsy may not be representative of all HBV-HCV–infected subjects. Finally, for cost reasons, patients who were HBsAg-negative did not have HBV DNA testing performed. Therefore, patients with occult HBV infection (HBsAg-negative with detectable HBV DNA) could have been missed, and we may have underestimated the true prevalence of HBV-HCV dual infection.
In conclusion, HBV infection is a major public health problem and is an important cause of liver disease in Americans with chronic HCV infection. Increased public and physician awareness is needed in order to identify, evaluate, and effectively treat patients who are at high risk for HBV-HCV dual infection. Population-based studies to accurately determine the prevalence of HBV-HCV dual infection in the United States as well as additional research to further understand the complex viral interactions that exist in HBV-HCV–infected patients are needed.