Article Text
PDF
Abstract
Objective To determine the prevalence of newly diagnosed hepatitis C virus (HCV) and associated risk behaviours among men who have sex with men (MSM) in Manchester.
Method A survey among MSM attending four genitourinary medicine clinics in Manchester was carried out over 9 months in 2013. Participants were asked about recent sexual behaviour, recreational drug use and HIV status. All men were offered an HCV test.
Results Overall, 2030 MSM completed a questionnaire and accepted an HCV test. Of whom, 0.9% (18) were newly diagnosed with HCV, including 1.8% (13/735) of HIV-positive MSM, 0.7% (3/440) of MSM of unknown HIV status and 0.2% (2/855) of HIV-negative MSM. HCV positivity was significantly associated with HIV status (p<0.001). When compared with HIV-negative MSM, HIV-positive MSM had higher rates of sharing snorting drug equipment, injecting drugs/‘slamming’ and using recreational drugs (all p<0.05) but lower rates of five or more sexual partners and insertive unprotected anal intercourse (p<0.05). MSM newly diagnosed with HCV had significantly higher prevalence of unprotected sex, sex with someone HCV positive, fisting, group sex, ever injecting drugs/‘slamming’ and recreational drug use (p<0.002).
Conclusions In this survey, HIV-positive MSM had significantly different drug use behaviour which may explain the higher HCV burden. However, HCV was also associated with HIV-negative MSM engaging in high-risk sexual practices. All MSM attending sexual health clinics must have a risk assessment and HCV screening should be offered based on the risk. Further studies are warranted to explore the interplay between HCV and HIV risk associated with drug use versus sexual practices.
- HEPATITIS C
- HIV
- TESTING
- MEN
Statistics from Altmetric.com
Introduction
Hepatitis C virus (HCV) is a blood-borne virus with an estimated 214 000 (0.4%) persons living with chronic HCV infection in the UK; these individuals are at increased risk of severe liver disease and hepatocellular cancer.1 Persons who inject drugs are at greatest risk of HCV infection. However, since 2000 there has been an increase in the number of newly acquired HCV infections in HIV-positive men who have sex with men (MSM).2 ,3 This shift has also been observed across Europe and in the USA,4–6 and is thought to be driven by sexual behaviours, including unprotected anal intercourse (UAI), fisting (brachioproctic insertion), high numbers of sexual partners and associated with recreational drug use before and/or during sex.7–9 In light of these developments, experts recommended annual HCV testing of HIV-positive MSM. In addition, MSM reporting anorectal trauma or injury following sex, a history of recreational drug use or ‘chemsex’ or rectal lymphogranuloma venereum (LGV) should be tested for HCV.10
To date, much of the published research has focused on HCV infection of HIV-positive MSM, and while the prevalence of HCV in HIV-negative MSM in the UK and Europe is often similar to that of the general population,3 ,11 ,12 studies have continued to find sexual transmission of HCV in HIV-negative MSM13–15 and there remains a risk of bridging between HIV-positive and negative MSM. We evaluated the expansion of routine HCV testing for HIV-positive MSM to MSM who do not report an HIV diagnosis in four genitourinary medicine (GUM) clinics in Greater Manchester, to determine the additional yield of new diagnosis of current HCV infection, alongside a survey which asked questions about the sexual and drug-taking behavioural risk factors.
Method
Between 28 February and 15 December 2014, MSM (≥16 years) attending one of four GUM clinics within Greater Manchester were invited to complete a questionnaire. Participants were offered an HCV antibody test, with reactive samples tested for HCV RNA. Patients' unique identifiers and date of births were recorded by the clinics.
Self-reported measures
The survey was based on that used by the enhanced surveillance of newly acquired HCV in MSM.16 It collected information on self-reported sexual orientation, and whether they had engaged in any of the following in the previous 3 months: sex with someone known to have HCV, being fisted (unprotected), group sex, sharing sex toys, insertive UAI (iUAI) and receptive UAI (rUAI). Participants were asked to self-report their HIV status, and were able to answer either ‘positive’, ‘negative’ or ‘do not know’. Additionally, they were asked whether they had ever injected drugs (also known as ‘slamming’), ever shared the equipment used to snort drugs and taken recreational drugs, and had sex under their influence, in the previous 12 months. Persons who reported recreational drug use in the previous 12 months were able to indicated whether they had taken ecstasy, cocaine, ketamine, gammahydroxybutrate (GHB) speed, crack, heroin, lysergic acid diethylamide (LSD) methamphetamine, Viagra or ‘other (please specify)’.
Data management
Completed paper questionnaires were sent to Public Health England, Colindale, London, and entered into OpenClinica. If participants completed the survey more than once, identified using patient number and date of birth, their most recent survey was used in analysis. A new diagnosis of HCV was reported separately by the clinics and subsequently added to patient's record. For patients to be classified as a new diagnosis of current HCV infection, they had to report no previous HCV diagnosis during their consultation, have no record of a prior positive anti-HCV result in the local laboratory prior to the study and test positive for HCV antibodies and RNA, indicative of an current infection, when tested between 28 February and 15 December 2014. A new diagnosis does not represent an incident infection, as we do not know when they acquired HCV.
Statistical analysis
Statistical analysis was carried out in Stata SE (V.13) and demographic and risk variables were explored by both HIV and newly diagnosed HCV status. When comparing variables by HCV status, those newly diagnosed with HCV were compared with all other MSM GUM attendees. We were unable to identify and exclude surveys completed by persons diagnosed with HCV prior to the study from the analysis, as this information was only collected by one clinic. As a result, the all other GUM attendee group will contain persons HCV negative, persons known to have current HCV infection and persons who have previously been HCV infected. The χ2 and Fisher's Exact tests were used to compare categorical variables and the Wilcoxon signed-rank test to explore differences in age. All proportions reported exclude unknowns and a p value of < 0.05 was considered significant. Multivariate analysis was not possible because of the small numbers of persons newly diagnosed with HCV but univariate logistic regressions were performed.
Ethics
This work was undertaken to evaluate a service evaluation within the four Greater Manchester GUM clinics. Routine HCV testing of all HIV-positive MSM is recommended by BASHH, and extending HCV testing to MSM who do not report an HIV diagnosis attending the four clinics was considered to be an expansion of existing services. While the HCV prevalence in HIV-negative MSM in the UK and in Europe is often similar to that of the general population, studies have continued to find sexual transmission of HCV in HIV-negative MSM13–15 and there remains a risk of bridging between HIV-positive and negative MSM. All MSM attending the clinics were offered an HCV test during this period, regardless of whether they completed a survey, but only those who completed the survey were included in the evaluation. The survey was freely available in participating clinics with no obligation for completion, and verbal information was given to the participant about the survey before completion, with a completed survey interpreted as implicit consent.
Results
Between 28 February and 15 December 2014, 2379 surveys were completed by MSM attending the four clinics, and following deduplication this equated to 2030 people.
Demographics
A total of 2030 MSM completed the questionnaire during the survey period (table 1). The majority of participants identified themselves as gay (88.3%) and the median age was 33 years (IQR: 27–42 years) (table 1). Where reported, 39.2% self-reported as HIV positive, with 77.7% reporting taking antiretroviral treatment. Self-reported HIV-positive MSM were, on average, older than HIV-negative MSM (31 years vs 38 years; p<0.001). Eighteen (0.9%) MSM were newly diagnosed with HCV. The highest proportion of new HCV diagnoses was in HIV-positive MSM (1.8%; 13/735), followed by MSM who did not report their HIV status (1.3%, 2/155), MSM who did not know their HIV status (0.4%; 1/285) and HIV-negative MSM (0.2%; 2/855). Self-reported HIV status was a significant predictor of a new HCV diagnosis (OR 7.8; 95% CI 1.7 to 34.1) (table 2).
Demographic characteristics, self-reported HIV status and newly diagnosed HCV status in MSM attending four Greater Manchester sexual health clinics
Sexual and drug-taking behaviour reported in MSM newly diagnosed with HCV
Sexual risk behaviours
One in five (20.0%, 397/1988) participants had five or more sexual partners in the previous 3 months, with higher proportion in HIV-negative MSM than HIV-positive MSM (25.2%: 213/844 vs 14.0%: 101/720; p<0.001).
40.8% (849/2030) reported UAI in the previous 3 months, 29.2% (592/2030) reported iUAI and 31.0% (630/2030) reported rUAI over the same period. More HIV-negative MSM had engaged in iUAI than HIV-positive MSM (32.7%: 280/855 vs 27.9%: 205/735; p=0.036), but there was no difference in rUAI by HIV status (HIV−: 28.4%: 243/855 vs HIV+:32.4%: 238/735; p=0.087).
In the previous 3 months, newly diagnosed HCV-positive MSM were more likely than all other GUM attendees to report having had five or more sexual partners (44.4%: 8/18 vs 19.7%: 389/1970; p=0.016), iUAI and rUAI (iUAI: 61.1%: 11/18 vs 28.9%: 581/2012; p=0.007; rUAI: 66.7%: 12/18 vs 30.7%: 618/2012; p=0.003), sex with someone HCV positive (11.1%: 2/18 vs 1.4%: 28/2012; p=0.03), receptive fisting (27.8%: 5/18 vs 2.2%: 45/2012; p<0.001) and group sex (38.9%: 7/18 vs 9.8%: 197/2012; p=0.001).
Significant predictors of a new diagnosis of current HCV infection were sex with more than five partners (OR 3.3; 95% CI 1.3 to 8.3), iUAI (OR 3.9; 95% CI 1.5 to 10.0), rUAI (OR 4.5; 95% CI 1.7 to 12.1), sex with someone HCV positive (OR 8.9; 95% CI 1.9 to 40.4), being fisted (OR 16.8; 95% CI 5.7 to 49.2) and group sex in the previous 3 months (OR 5.9; 95% CI 2.2 to 15.3)(table 2).
Drug-taking risk behaviours
Over one in three (36.6%: 709/1938) MSM had used recreational drugs in the previous 12 months, of whom 60.9% (n=432) had sex under their influence. MSM reporting recreational drug use were younger (31 years vs 34 years; p<0.001). Recreational drug use and sex under their influence was more common in HIV-positive MSM compared with HIV-negative MSM (recreational drug use: 40.7%: 287/706 vs 31.3%: 254/811; p<0.001; sex under influence: 68.6%: 194/283 vs 57.4%: 143/249; p=0.008) and in MSM newly diagnosed with HCV when compared with all other GUM attendees (recreational drug use: 72.2%: 13/18 vs 36.3%: 696/1920; p=0.002; sex under influence: 100.0%: 13/13 vs 61.8%: 419/678; p=0.003). HIV-positive MSM were more likely to report cocaine, Viagra, ketamine, mephedrone, GHB, methamphetamine and speed use in the previous 12 months than HIV-negative MSM (all p<0.05).
All MSM newly diagnosed with HCV and who reported recreational drugs in the previous 12 months (n=13) had sex under their influence and MSM newly diagnosed with HCV were more likely to report ketamine, Viagra, mephedrone, GHB and methamphetamine use in the previous 12 months (all p<0.05)(figure 1).
Recreational drug use in the last 12 months in all men who have sex with men (MSM), by self-reported HIV-positive status and newly diagnosed with hepatitis C virus (HCV).
Injecting drug use was reported in 5.5% (108/1946) of MSM, of whom 17.6% (n=19) had ever shared needles. MSM had most commonly injected stimulants (86/108), followed by steroids (17/108) and opiates (10/108). The prevalence of injecting drug use was significantly higher in HIV-positive MSM than HIV-negative MSM (9.2%: 65/709 vs 3.3%: 27/819; p<0.001) and in MSM newly diagnosed with HCV when compared with all other GUM attendees (38.9%: 7/18 vs 5.2%: 101/1928; p<0.001). There was no difference in the type of drugs (stimulant, steroid or opiate) injected by HIV or newly diagnosed HCV status (all >0.05). There was no difference in needle sharing by HIV status (HIV−: 27.3%: 6/22 vs HIV+: 19.6%: 11/56; p=0.5) or by newly HCV diagnosed status (other GUM attendees: 19.5%: 17/87 vs newly diagnosed with HCV: 28.6%: 2/7; p=0.6).
Overall, 21.4% (416/1943) of MSM ever shared the equipment used to snort drugs and these men were younger than those who did not report sharing equipment (31 years vs 34 years: p<0.001). Having ever shared the equipment used to snort drugs was more prevalent in HIV-positive MSM than that in HIV-negative MSM (24.6%: 177/720 vs 18.9%: 153/811; p=0.007) and in MSM who were newly diagnosed with HCV when compared with all other GUM attendees (50.0%: 9/18 vs 21.1%: 407/1925; p=0.007).
Significant predictors of a new diagnosis of current HCV infection were recreational drug use (OR 4.6; 95% CI 1.6 to 12.9) and sex under their influence (OR 10.9; 95% CI 3.1 to 38.3) in the previous 12 months, a history of injecting drug use (OR 11.5; 95% CI 4.4 to 30.3) and sharing the equipment to snort drugs (OR 3.7; 95% CI 1.5 to 9.5). Recreational drugs, used in the previous 12 months, significantly associated with new diagnosis of current HCV infection include ketamine (OR 4.4; 95% CI 1.5 to 12.4), GHB (OR 5.7; 95% CI 1.8 to 17.7), methamphetamine (OR 13.7; 95% CI 4.3 to 43.5), mephedrone (OR 4.5; 95% CI 1.4 to 13.8) and Viagra (OR 7.7; 95% CI 3.0 to 20.2).
MSM who had ever injected drugs or reported recreational drug use in the previous 12 months were more likely to report sex with more than five partners, iUAI, rUAI, sex with someone HCV positive, being fisted and group sex in the previous 3 months, and having ever shared the equipment used to snort drugs (table 3)(all p<0.05).
Sexual and drug-taking behaviour by injecting drug use (IDU) status and recreational drug use status
Discussion
Over a 9-month period, February–December 2014, 18 MSM attending any one of four GUM clinics in Greater Manchester were newly diagnosed with HCV, the majority of whom self-reported as HIV positive (n=13). Overall, 0.9% of MSM were newly diagnosed with current HCV infection, 1.8% in self-reported HIV-positive MSM, 1.3% in MSM who did not report their HIV status, 0.4% in MSM who did not know their HIV status and 0.2% in HIV-negative MSM. The rate of newly diagnosed HCV in men who did not report their HIV status was similar to that found in HIV-positive MSM, whereas the rate in MSM who did not know their HIV status was more similar to HIV-negative MSM. A new diagnosis of current HCV infection was associated with a number of high-risk behaviours, including sex with five or more partners, iUAI, rUAI, fisting, group sex in the previous 3 months, a history of injecting drug use and recreational drug use and sex under their influence in the previous 12 months. Self-reported HIV-negative MSM were more likely to report having sex with five or more partners and iUAI in the previous 3 months, whereas self-reported HIV-positive MSM were more likely to report recreational drug use, sex under their influence and ever injecting drugs. Those HIV/HCV co-infected had the same risks as all MSM newly diagnosed with HCV.
The most important limitation in our study was the inability to identify clinic attendees who were already diagnosed with current HCV infection from the other GUM attendee group, as we only received this information from one of the clinics. In this clinic, 1.5% of persons were diagnosed with current HCV infection, with an overall HCV prevalence of 2.7%. When we retested the associations for newly diagnosed current HCV infection within this clinic, excluding all known current HCV infections from the analysis, all associations remained significant, apart from sex with five or more people in the previous 3 months. Additionally, this information would have been of particular interest as 0.4% of MSM of unknown, undisclosed and negative HIV status were newly diagnosed with HCV, which is the same as the prevalence of chronic HCV in the UK.17 As a result, we were unable to see whether the overall prevalence of HCV infection in MSM in this group was higher than the national estimate. Second, it is worth noting that HIV-positive MSM were over-represented in our study; however, our large sample size allows us to be confident of our comparisons by HIV status. Third, as this survey is limited to MSM attending GUM clinics and completing the survey, the finding may not be generalisable to the whole MSM population or all HIV-positive MSM. Despite these limitations, and the relatively small numbers of MSM newly diagnosed with HCV, we found several significant differences between those newly diagnosed with HCV and the other GUM attendees which can be used to direct HCV testing to MSM most at risk.
Direct comparisons with similar studies are difficult, as the time frames around behaviours being questioned are different. Despite this, our results are generally consistent with other studies within the UK,11 ,13 ,18–20 with elevated rates of higher risk sexual and drug-taking behaviour, such as recreational drug use and sex under their influence, and HCV infection being significantly more likely in HIV-positive MSM. Interestingly, we found HIV-positive MSM less likely to report sex with more than five partners and iUAI in the previous 3 months than HIV-negative MSM, but HIV-positive MSM reported higher overall rates of UAI.
HCV was associated with behaviours, both sexual and drug related, that put persons at increased risk of HCV infection. Seven out of 18 MSM newly diagnosed with HCV reported ever injecting drugs, two of whom reported sharing needles. The increasing risk of HCV infection through sexual transmission is consistent with studies within the UK.7 ,8 However, while most MSM newly diagnosed with HCV did not report injecting drug use, a significant number did report sex under the influence of recreational drugs (n=13). Recreational drug use, particularly drugs associated with chemsex, continues to be linked with UAI and a higher number of sexual partners, both important factors in HCV and HIV transmission.21 ,22 Although HIV-positive MSM reported lower rates of sex with five or more partners and iUAI, over a quarter (194/706) had sex under the influence of recreational drugs, which lower inhibitions, can leave users unaware of their surroundings and may contribute to the higher numbers of newly diagnosed HCV among HIV-positive MSM. This highlights the need for multifunctional clinics within sexual health, which are able to educate patients on how to reduce the harm from drug taking and provide safe injecting packs, such as the 56 Dean Street and the Burrell Street Clinics in London.
Using the average number of MSM attending the clinics each month, we estimate survey uptake to be approximately 50% of all MSM attendees. However, clinics reported that HCV testing was very acceptable to clinic attendees, with high levels of HCV test uptake (>70%). It is important to note that if all those newly diagnosed with HCV disclosed the behaviours they reported in the survey during their GUM appointment 16 out of 18 would have been tested according to BASHH guidelines, as a result of their recreational drug use and/or HIV infection.10 The two MSM not covered by the BASHH guidelines reported iUAI and between 5 and 10 sexual partners in the previous 3 months, but no other risk factors. One did not know his HIV status and the other did not report their HIV status. As a result, they may not be tested for HCV unless they reported sex associated with trauma or injury or tested positive for LGV. This suggests that expanding routine testing to MSM who do not report an HIV diagnosis may not be necessary to detect the majority of undiagnosed current HCV infections, but it might be worthwhile considering offering HCV testing to attendees who report high numbers or recent sexual partners and UAI.
Our results suggest, alongside anecdotal reports from clinicians, that HCV testing is acceptable to MSM attending the four Greater Manchester GUM clinics, regardless of HIV status, and we identified that the diagnosis of HCV infection in MSM continues to be associated with HIV infection, sex with five or more partners, UAI, sex with partners HCV positive, being fisted and group sex in the previous 3 months, a history of injecting drug use and recreational drug use in the past year, in particular sex under their influence. A comprehensive risk assessment, which includes questions on sexual and drug-taking behaviours, of MSM attending GUM services and targeted HCV testing based on this risk should be conducted to ensure early diagnosis and to change behaviour, reducing onward transmission. Further research is needed to understand whether HCV among MSM is mainly a function of sexual behaviour, injecting drug use or both, and tailor prevention and intervention strategies accordingly.
Key messages
The proportion of men who have sex with men (MSM) newly diagnosed with current hepatitis C virus (HCV) infection was 0.9% overall, 1.8% in HIV-positive MSM, 1.3% in MSM who did not report their HIV status, 0.4% in MSM who did not know their HIV status and 0.2% in those HIV negative.
HCV infection in MSM continues to be associated with HIV infection, unprotected anal intercourse, fisting, sex with HCV-positive partners, group sex, recreational drug use and a history of injecting drug use.
Comprehensive risk assessments, with targeted HCV testing based on the risk assessment, are likely to be more efficient than screening all MSM in genitourinary medicine services in Manchester.
References
Footnotes
Handling editor Jackie A Cassell
Contributors SH, BG, CW, SA, CS and VL collected the data, which was sent to GI and SL at Public Health England. GI undertook the analysis and had access to the complete dataset, RS, SL and VL supervised GI. GI drafted the paper and all authors provided critical input to the manuscript and approved all revisions.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement The data is held by Public Health England (PHE) in London. The data holds personal information which means that it cannot be freely available in a public repository, but readers may contact Georgina Ireland (georgina.ireland@phe.gov.uk) at PHE to make a data request.