Article Text
Abstract
Objectives We sought to determine the prevalence of hepatitis C virus (HCV) infection among men who have sex with men (MSM) in Vancouver, Canada, and associations of risk behaviours with HCV serostatus.
Methods We used data from the ManCount Study, a cross-sectional survey of MSM selected through a venue-based, time-location sampling method. Bivariate analyses and multivariate logistic regression modelling were used to determine correlates of HCV seropositivity. Bivariate analyses of participants who reported no history of injection drug use (IDU) were used to explore sexual behaviours associated with HCV seropositivity.
Results HCV seroprevalence was 4.9% (56/1132). Among HCV-seropositive participants who responded to the question, 22.4% (11/49) were unaware of their HCV-seropositive status, 84.9% (45/53) reported a history of IDU and 60.7% (34/56) were HIV positive by dried blood spot. Multivariate modelling found previous IDU (adjusted OR (AOR): 26.30, 95% CI 11.15 to 62.03), receiving goods, drugs or money for sex (AOR 4.98, 95% CI 2.43 to 10.20) and current smoking (AOR 3.46, 95% CI 1.47 to 8.16) were associated with HCV seropositivity. Among MSM who reported no history of IDU, HCV seropositivity was associated with bleeding after receptive anal sex (p=0.001) and a previous diagnosis of gonorrhoea (p=0.007).
Conclusions HCV seroprevalence among a sample of MSM is higher than the general population and associated with a history of IDU. Among those who did not report IDU, we found evidence that suggests sexual exposure could be the route of transmission.
- HEPATITIS C
- GAY MEN
- SURVEILLANCE
Statistics from Altmetric.com
Introduction
Hepatitis C virus (HCV) is a bloodborne pathogen that is prevalent among people who have used injection drugs.1 Men who have sex with men (MSM) have also been reported to have a higher prevalence of HCV compared with the general population; in a review, the prevalence of HCV among MSM was as high as 22.7%.2 However, more recent studies have found a much lower burden (0.07–4.5%).3 ,4
The seroprevalence of HCV among MSM in Vancouver, Canada, is unknown. Studies from other Canadian cities found the seroprevalence of HCV among MSM was 1.9–5.3%.5–7
Recently, there have been outbreaks of HCV linked to sexual networks of HIV-positive MSM who did not report injection drug use (IDU), raising concerns about the potential for sexual transmission of HCV.8 Sexual behaviours associated with an increased risk of blood exposure are thought to be the mechanism for HCV transmission. Indeed, mucosal trauma (eg, group sex, fisting and use of sex toys), genital ulcer disease (eg, syphilis, herpes) and unprotected anal intercourse have been found to be correlated with HCV; however, the evidence is inconsistent.8
The objectives of this analysis were to describe the seroprevalence of HCV among MSM in Vancouver and association of HCV serostatus with risk behaviours. We also examined the association of sexual behaviours with HCV serostatus among participants who report no history of IDU to evaluate the hypothesis of sexual transmission of HCV.
Methods
We analysed data from the ManCount Study, a cross-sectional survey to monitor HIV and other sexually transmitted and bloodborne infections and associated risk factors among MSM. The details of the ManCount Study are described elsewhere.9
Men ≥18 years old who reported ever having had sex with other men were recruited between August 2008 and February 2009 at venues that catered to MSM. Participants completed an anonymous self-administered questionnaire and provided a dried blood spot (DBS) sample for HCV and HIV testing, among other tests.
HCV antibody testing was performed using the Ortho HCV V.3.0 EIA (Ortho-Clinical Diagnostics Inc, Raritan, New Jersey, USA).
Participants were included in this analysis if they provided a DBS sufficient for HCV testing. We conducted bivariate analyses of DBS HCV serostatus with sociodemographic characteristics and self-reported risk behaviour using the Fisher's exact or Wilcoxon rank-sum test. We also conducted a subanalysis among participants who reported no history of IDU (ie, never used injection drugs). For this study, IDU excludes steroid injections.
Logistic regression analysis was used to determine factors associated with HCV seropositivity. The multivariate model was selected by minimising the Akaike information criterion (AIC) while limiting the number of included variables to approximately 1 per 20 events.
All analyses were performed using SAS V.9.3 (SAS Corporation, Cary, North Carolina, USA).
Results
A total of 1169 participants enrolled in the study and 1139 (97.4%) provided a DBS for HCV testing. One thousand one hundred and thirty-two (96.8% of surveyed men) had HCV antibody results available and were included in this analysis.
Study participants had a median age of 33 years (IQR: 26–43). Most participants had an annual personal income of ≥$30 000 (662/1132, 58.5%) and some college or higher level of education (893/1126, 79.3%). The majority identified as North American or European (856/1132, 75.6%), while 77 (6.8%) identified as Asian, 45 (4.0%) as Aboriginal and 154 (13.6%) as other ethnicities (eg, African, Arab). Most participants described themselves as homosexual (900/1120, 80.4%) though 124 (11.1%) identified as bisexual, 20 (1.8%) as heterosexual, 34 (3.0%) as two-spirited, 26 (2.3%) as queer and 16 (1.4%) as other. (Two-spirit is a term used by some Aboriginal people that describes their sexual, gender and/or spiritual identity as having both feminine and masculine spirits.) Over a third of participants (419/1112, 37.7%) reported being current smokers. The majority of participants were HIV negative by DBS testing (930/1132, 82.2%) and reported having previous testing for HCV (743/1132, 65.6%).
Fifty-six (4.9%) participants were HCV seropositive. Of those with available data, 22.4% (11/49) were unaware they were HCV seropositive. Participants who were HCV seropositive tended to be older than those who were HCV seronegative (median age 40 years old vs 33 years old, p<0.001), had less education (14/54, 25.9% had less than a high school education vs 68/1072, 6.3%, p<0.001) and have lower incomes (39/56, 69.6% had an annual income <$30 000 vs 431/1076, 40.1%, p<0.001). A greater proportion of HCV-seropositive individuals identified as bisexual (12/56, 21.4% vs 112/1064, 10.5%, p<0.001) and were current cigarette smokers (45/56, 80.4% vs 374/1056, 35.4%, p<0.001). Thirty-four of 56 (60.7%) participants were both HCV and HIV seropositive by DBS. HCV-positive individuals were also much more likely to report a history of IDU, use of non-injection party drugs, sex trade involvement and recent sex with a woman (p<0.001 for all) (table 1).
In our final multivariate logistic regression model, HCV seropositivity was associated with past or current IDU (adjusted OR (AOR) 26.30, 95% CI 11.15 to 62.03), receiving goods, drugs or money in exchange for sex (AOR 4.98, 95% CI 2.43 to 10.20) and being a current smoker (AOR 3.46, 95% CI 1.47 to 8.16). Notably, multivariate models which replaced current smoking status with either ‘use of party drugs within 2 h of or during sex’ or ‘had sex with a woman in the past 6 months’ variables produced similar but slightly higher AIC scores for our final model (data not shown).
In our subanalysis of the 926 participants who reported no history of IDU, eight (0.9%) were HCV seropositive. Compared with those who were HCV seronegative, HCV-seropositive individuals were older (median age, 56.5 years vs 33 years, p=0.002) and were more likely to report being HIV positive (75.0% vs 14.6%, p<0.001), report Aboriginal ethnicity (37.5% vs 2.7%, p<0.001) and self-identify with sexual orientations other than homosexual (50.0% vs 17.0%, p<0.001). HCV-seropositive individuals were also more likely to report ever bleeding with receptive anal sex (71.4% vs 19.5%, p=0.001) and a previous diagnosis of gonorrhoea (66.7% vs 21.1%, p=0.007) but were not more likely to report riskier sexual behaviours in the last 6 months, such as receptive fisting (14.3% vs 3.5%, p=0.130), use of sex toys (14.3% vs 31.0%, p=0.341) and group sex (28.6% vs 33.7%, p=0.775).
Discussion
We found a HCV seroprevalence of 4.9% among a sample of MSM in Vancouver, similar to the seroprevalence found in other Canadian cities of 5.3%.7 As found in other studies, previous IDU is most strongly associated with HCV seropositivity.1 ,3 We found that 84.9% (45/56) of HCV-seropositive MSM reported a history of IDU, compared with an estimated 58% who report the same in the general population.10 This difference in IDU behaviour likely accounts for much of the higher burden of HCV among MSM in our study compared with the general Canadian population (estimated at 0.78%).10
However, even among MSM who reported no history of IDU, we found a HCV seroprevalence of 0.9%, higher than the estimated 0.3–0.4% for the general Canadian population who have not used injection drugs.10 Within this group, HCV infection was more common among those with a previous diagnosis of gonorrhoea, suggesting a similar route of transmission for HCV in these individuals. Additionally, HCV-seropositive MSM who reported no history of IDU were more likely to report bleeding after receptive anal intercourse, consistent with the hypothesis that mucosal damage during sexual intercourse may be associated with acquisition of HCV.8
Notably, 16.8% (34/202) of HIV-positive individuals were HCV seropositive. Among those with no history of IDU, 75% (6/8) of HCV-seropositive participants were also HIV positive, suggesting that HIV infection may increase the efficiency of HCV transmission8 or that seroadaptive behaviours of HIV-positive MSM may increase the risk of HCV by increasing the likelihood of unprotected sexual intercourse. The semen of HIV-positive individuals have also been found to have higher viral loads of HCV compared with HIV-negative individuals.11
In the logistic regression model, being a current smoker was found to increase the odds of HCV seropositivity. Current smoking has been associated with use of other substances, including IDU,12 which likely accounts for much of this relationship. Current smoking status may be a surrogate for other substance use and should prompt health care providers to ask further about IDU. Non-injection party drug use was also associated with HCV seropositivity. This may reflect the influence of psychoactive substances on disinhibiting behaviour, leading to higher risk sexual practices (such as unprotected anal intercourse).8 ,13
One limitation of our study was that venue-based sampling was used to recruit participants, and thus may not be representative of the entire MSM population in Vancouver. Further statistical analysis which adjusted for the frequency of venue attendance estimated the seroprevalence of HCV to be 7%.14 Also, as a cross-sectional survey, risk factors examined may not reflect behaviours at the time HCV infection occurred. Our behavioural data were based on self-report and our analyses are subject to social desirability biases and therefore misclassification. Our subanalysis exploring the hypothesis of sexual transmission of HCV was limited by the small number who did not report a history of IDU.
Acknowledgments
We thank Warren Michelow, members of the community advisory board, volunteer interviewers, the venues who participated in the study and the ManCount Study participants.
References
Footnotes
Handling editor Jackie A Cassell
Twitter Follow Mark Gilbert at @mpjgilbert
Collaborators ManCount Study Team.
Contributors DM, RG, RH, SO-C, TW and MG contributed to the overall planning, conduct and reporting of the ManCount Study. JW and SK contributed to the analysis and reporting of the study. JB contributed to the reporting of the study.
Funding The ManCount Survey was funded by the Public Health Agency of Canada.
Competing interests None.
Ethics approval Health Canada/Public Health Agency of Canada and University of British Columbia.
Provenance and peer review Not commissioned; externally peer reviewed.