Abstract
No data exist on hepatitis B virus (HBV) incidence among African men who have sex with men (MSM). We tested plasma samples archived between 2005 and 2014 for HBV core antibody or surface antigen seroconversion in a cohort of 312 initially human immunodeficiency virus (HIV)-1-negative MSM with no evidence of prior HBV infection. Hepatitis B virus incidence was 6.0/100 person-years (95% confidence interval [CI], 3.9–9.1). Hepatitis B virus acquisition was associated with being uncircumcised (adjusted incidence rate ratio [aIRR], 5.0; 95% CI, 1.5–16.8), recent HIV-1 acquisition (aIRR, 2.9; 95% CI, 1.1–7.7), rape (aIRR, 5.0; 95% CI, 1.2–20.4), and any tertiary education (aIRR, 3.2; 95% CI, 1.1–9.7). African MSM have a substantial risk of HBV acquisition and require vaccination urgently.
Keywords: circumcision, HBV incidence, HIV-1 acquisition, Kenya, MSM
Chronic hepatitis B virus (HBV) infection, defined as persistence of HBV surface antigen (HBsAg) for greater than 6 months, is a leading cause of chronic liver disease and death worldwide [1]. In sub-Saharan Africa (SSA), the predominant mode of HBV transmission is horizontal, with a substantial number of infections acquired in early childhood [2, 3]. The estimated prevalence of chronic HBV infection in Kenya is 5.16% (95% confidence interval [CI], 4.86–5.48) [4]. Twenty-five percent of human immunodeficiency virus (HIV)-1-uninfected individuals with chronic HBV are estimated to develop cirrhosis, hepatocellular carcinoma (HCC), or both [1]. Approximately 40% of individuals who develop HBV-induced HCC in SSA do so before age 40 [5]. Although universal childhood HBV vaccination has been provided in Kenya since the early 2000s, adult male sex workers and other men who have sex with men (MSM) are not routinely offered HBV vaccination.
Human immunodeficiency virus-1 incidence among MSM in Kenya has been estimated at 8–10 per 100 person years [6, 7], with estimates as high as 35.2 (95% CI, 23.8–52.1) per 100 person years among MSM only [7]. Because HBV is more infectious than HIV-1 [8], it may be expected that HBV incidence is substantial among Kenyan MSM. However, no data exist on HBV incidence in this group, despite the need for such data to support HBV vaccination programming targeting MSM [9]. Therefore, we studied HBV incidence and associated risk factors for acquisition among initially HIV-1-negative, HBV-susceptible MSM enrolled in a cohort study in Coastal Kenya, all of whom were born before routine childhood vaccination began in Kenya.
METHODS
Study Design, Setting, and Population
We retrospectively tested stored plasma samples from HIV-1-negative MSM aged 18–49 years who were enrolled into a HIV-1 vaccine feasibility cohort study between 2005 and 2014 in Coastal Kenya [7]. All male cohort participants who reported anal sex with another man in the 3 months before enrollment and had stored blood samples available from enrollment and at least 1 follow-up visit through December 2014 were selected for the present study.
Cohort Procedures
Detailed cohort procedures are described elsewhere [7]. In brief, enrollment and follow-up included a face-to-face interview using a standardized behavioral questionnaire, HIV-1 testing and counseling, medical history, and physical examination at each visit. Follow-up was quarterly, or monthly when receptive anal intercourse (RAI) was reported. Men who acquired HIV-1 during follow-up were offered enrollment into an acute infection cohort with quarterly follow-up visits during the first 2 years and 6-monthly visits thereafter [10]. From May 25, 2009 onwards, HBV vaccination was provided to cohort participants as part of ongoing care and prevention services. Follow-up visits after the first documented HBV vaccination date were excluded from the study.
Laboratory Methods
We tested archived samples using 3 HBV serological assays: Murex HBsAg version 3 (Diasorin) for HBsAg, ETI-AB-AUK-3 (Diasorin) for hepatitis B surface antibody (anti-HBs), and Murex anti-HBc (Diasorin) for hepatitis B core antibody (anti-HBc). We first tested the last available follow-up sample. If this last sample was negative for all 3 markers, we did no further testing for that participant, assuming they would have negative serologies at baseline. If the last available sample was positive for any of the 3 HBV markers, we also tested the baseline sample. Any available in-between samples were tested for anti-HBc or HBsAg if the baseline sample was negative for all 3 markers and the last available sample was positive for anti-HBc or HBsAg.
Definitions used for this study were as follows: (1) HBV-susceptible, negative for all the 3 HBV markers at baseline or during follow-up; (2) resolved HBV-infection, positive for anti-HBc, with or without anti-HBs, but negative for HBsAg at baseline; (3) acute or chronic HBV-infection, positive for HBsAg at baseline; (4) incident HBV-infection, negative for all 3 HBV markers at baseline and positive for anti-HBc or HBsAg during follow-up, following the approach of Falade et al [11]); (5) possibly HBV-vaccinated, anti-HBs positive and negative for anti-HBc and HBsAg at baseline or during follow-up. The estimated date of infection for HBV was defined as the mid-point between a previously negative and subsequently positive anti HBc or HBsAg result [8, 11].
Detailed HIV-1 testing procedures are described elsewhere [7]. In brief, HIV-1 testing was performed at each study visit using 2 rapid antibody test kits (Determine, Abbott Laboratories; Unigold, Trinity Biotech) in parallel. Discordant rapid HIV-1 test results were resolved using an enzyme-linked immunohsorbent assay (ELISA) test (Genetic System HIV-1/2 plus O EIA; Bio-Rad). All HIV-1-negative samples were tested for p24 antigen (Vironostika HIV-1 p24 ELISA; Biomérieux), and pre- and post-seroconversion samples were tested for HIV-1 ribonucleic acid (Amplicor Monitor 1.5; Roche).
Data Analysis and Statistical Methods
We compared demographic and behavioral characteristics of men who returned for follow-up to those who did not return for follow-up after enrollment using Pearson’s χ2 test. Hepatitis B virus-susceptible MSM were compared with MSM who had resolved, acute, or chronic HBV infection at baseline using Pearson’s χ2 test, after excluding men whose serology results indicated possible HBV vaccination. Prevalence ratios were used to measure associations between participant characteristics and resolved, acute, or chronic HBV infection. We included factors associated with resolved, acute, or chronic HBV infection at P less than 0.2 in the bivariate analysis in a multivariable Poisson model to estimate adjusted prevalence ratios.
We included MSM who were HBV-susceptible at enrollment in the incidence analysis. The primary outcome was incident HBV infection, calculated as the number of seroconversions to a positive anti-HBc or HBsAg result divided by person-years of follow-up in each group of interest. Individual follow-up time was calculated from the date of study entry until the estimated date of incident HBV infection, the date of first documented HBV vaccine dose, or the date prior to an isolated positive anti-HBs result (i.e., possible HBV vaccination). We censored participants who were not vaccinated or became lost to follow-up at the last visit date in the study for which a stored sample was available for testing.
We evaluated potential risk factors that were fixed (demographics, baseline circumcision status) or time-dependent (genital washing, alcohol use, intravenous drug use, sexual behaviors, HIV status). Poisson models with robust standard errors were used to obtain population-averaged incidence rate ratios that accounted for correlation due to repeated measurements on the same subject over time. We included variables significant in bivariable analysis at P < .2 and a priori predictors (associated with resolved, acute, or chronic HBV infection at enrollment in the univariate analysis) including age, marital status, sex work, RAI, and any unprotected sex, in an initial multivariable model of potential predictors of HBV acquisition. We then dropped variables with P > .1 in this initial model that were not a priori predictors, to produce the final multivariable model. P values were 2-sided and significance was set at P ≤ .05. We used Stata 13.0 (StataCorp LP, College Station, TX) for analysis.
Ethical Considerations
The Kenya Medical Research Institute Ethics Review Committee approved the study. All participants provided written informed consent.
RESULTS
Study Population
Of 908 MSM screened for cohort enrollment between 2005 and 2014, 133 (14.6%) were HIV-1 positive and excluded. Of the 775 HIV-1-negative men enrolled, 198 (25.5%) did not return for any follow-up.
Compared to MSM who remained in follow-up, MSM who were lost to follow up after the enrollment visit were younger (60.1% vs 45.1% were 18–24 years of age, P = .001), less likely to have been married (89.4% vs 79.7% were never married, P = .002), and more likely to report insertive anal intercourse (IAI) (68.2% vs 58.8%, P = .019) (data not shown).
Out of 577 HIV-1-negative MSM with at least 1 follow-up visit, 525 (91%) had enrollment and follow-up samples available for hepatitis B serology testing. Of these 525 MSM, 136 (25.9%) were classified as having resolved HBV infection, 42 (8.0%) as having acute or chronic HBV infection, and 35 (6.7%) as having possibly been vaccinated, because their serology was positive for anti-HBs only. A total of 312 (59.4%) were HBV susceptible at baseline (details for all participants in Figure 1).
Figure 1.
Flow chart of men who have sex with men (MSM) cohort participants selected for hepatitis B virus (HBV) acquisition study, Kilifi, Kenya, 2005–2014. 1Missing either enrollment or follow-up sample. 2Defined as negative for HBV surface antigen (HBsAg), hepatitis B surface antibody (anti-HBs), and hepatitis B core antibody (anti-HBc) at baseline and positive for anti-HBc or HBsAg at any point during follow-up, regardless of whether exposure resulted in persistent HBV infection. 3Anti-HBs conversion without evidence of HBV exposure, possibly vaccinated outside our research cohort.
Hepatitis B Virus Prevalence and Risk Factors
Table 1 compares baseline demographic and behavioral characteristics of 312 HBV-susceptible and 178 MSM with resolved, acute, or chronic HBV-infections. Younger age, having never been married, sex work, RAI, and any unprotected sex were associated with HBV susceptibility. In multivariable analysis, only age remained statistically significantly associated with resolved, chronic, or acute HBV infection status, with younger men more likely to be HBV-susceptible (Supplemental Material, Table 1).
Table 1.
Baseline Demographic and Behavioral Characteristics of 490 HIV-1 Negative MSM in Kilifi, Kenya, 2005–2014
| Characteristics | Overall (n = 490) | HBV-Susceptible* (n = 312) |
Resolved, Acute, or Chronic HBV Infection* (n = 178) |
P Value† | |||
|---|---|---|---|---|---|---|---|
| n | (%) | n | (%) | n | (%) | ||
| Age group (years) | |||||||
| 18–24 | 254 | (51.8) | 178 | (57.1) | 76 | (42.7) | .001 |
| 25–34 | 188 | (38.4) | 113 | (36.2) | 75 | (42.1) | |
| 35+ | 48 | (9.8) | 21 | (6.7) | 27 | (15.2) | |
| Education | |||||||
| Primary/none | 249 | (50.8) | 149 | (47.8) | 100 | (56.2) | .193 |
| Secondary | 191 | (39.0) | 130 | (41.7) | 61 | (34.3) | |
| Higher/tertiary | 50 | (10.2) | 33 | (10.6) | 17 | (9.6) | |
| Marital status | |||||||
| Never married | 405 | (82.7) | 267 | (85.6) | 138 | (77.5) | .024 |
| Ever married | 85 | (17.3) | 45 | (14.4) | 40 | (22.5) | |
| Religion | |||||||
| Christian | 234 | (47.8) | 156 | (50.0) | 78 | (43.8) | .405 |
| Muslim | 146 | (29.8) | 90 | (28.8) | 56 | (31.5) | |
| Other/none | 110 | (22.4) | 66 | (21.2) | 44 | (24.7) | |
| Employment | |||||||
| None | 203 | (41.4) | 135 | (43.3) | 68 | (38.2) | .504 |
| Self | 218 | (44.5) | 133 | (42.6) | 85 | (47.8) | |
| Formal | 69 | (14.1) | 44 | (14.1) | 25 | (14.0) | |
| Circumcised | |||||||
| Yes | 456 | (93.1) | 294 | (94.2) | 162 | (91.0) | .177 |
| Sex partners past 3 months | |||||||
| Men and women | 306 | (62.4) | 190 | (60.9) | 116 | (65.2) | .348 |
| Men only | 184 | (37.6) | 122 | (39.1) | 62 | (34.8) | |
| Received payment for sex with cash, living expenses, or goods in past 3 months | |||||||
| Yes | 337 | (68.8) | 225 | (72.1) | 112 | (62.9) | .035 |
| Paid for sex in past 3 months | |||||||
| Yes | 157 | (32.0) | 98 | (31.4) | 59 | (33.1) | .692 |
| Any use of alcoholic beverages in past month | |||||||
| Yes | 320 | (65.3) | 206 | (66.0) | 114 | (64.0) | .658 |
| Receptive anal intercourse (RAI) in past 3 months | |||||||
| Yes | 303 | (61.8) | 203 | (65.1) | 100 | (56.2) | .052 |
| Insertive anal intercourse (IAI) past in 3 months | |||||||
| Yes | 288 | (58.8) | 183 | (58.7) | 105 | (59.0) | .942 |
| Sexual exposure and protection with condom in past week | |||||||
| No activity | 99 | (20.2) | 61 | (19.6) | 38 | (21.3) | .012 |
| All protected | 90 | (18.4) | 46 | (14.7) | 44 | (24.7) | |
| Any unprotected | 301 | (61.4) | 205 | (65.7) | 96 | (53.9) | |
| Total sex partners in past month | |||||||
| Less than 3 | 168 | (34.3) | 101 | (32.4) | 67 | (37.6) | .237 |
| Three or more | 322 | (65.7) | 211 | (67.6) | 111 | (62.4) | |
| Group sex in past 3 month | |||||||
| Yes | 83 | (16.9) | 52 | (16.7) | 31 | (17.4) | .832 |
| Been raped in past 3 month | |||||||
| Yes | 11 | (2.2) | 6 | (1.9) | 5 | (2.8) | .524 |
| Intravenous drug use in past 3 months | |||||||
| Yes | 8 | (1.6) | 4 | (1.3) | 4 | (2.2) | .417 |
| Genital washing with soap in past week | |||||||
| Yes | 399 | (81.4) | 252 | (80.8) | 147 | (82.6) | .619 |
| Sex with a menstruating female partner in past 3 months | |||||||
| Yes | 51 | (10.4) | 35 | (11.2) | 16 | (9.0) | .490 |
| Don’t know | 2 | (0.4) | 2 | (0.6) | 0 | (0.0) | |
Abbreviations: anti-HBc, hepatitis B core antibody; HBsAg, hepatitis B virus surface antigen; anti-HBs, hepatitis B surface antibody; HBV, hepatitis B virus; HIV, human immunodeficiency virus; MSM, men who have sex with men.
*HBV-susceptible was defined as negative for all 3 HBV markers studied (HBsAg, anti-HBs, anti-HBc); resolved HBV infection was defined as positive for anti-HBc (with or without anti-HBs); and acute or chronic HBV-infection was defined as positive for HBsAg. We excluded participants who were positive for HBV surface antibody only, who were considered to have possibly received HBV vaccine.
†Two-sided value based on χ2 test for proportions.
Hepatitis B Virus Incidence and Risk Factors
Overall, we followed 312 MSM for a median of 8.4 months (interquartile range, 2.8–19.5) contributing 369.0 person-years. Twenty-two MSM converted for anti-HBc or HBsAg during follow-up, giving a crude HBV incidence rate of 6.0 (95% CI, 3.9–9.1) per 100 person-years. Of these, 5 (22.7%) acquired HIV-1 at a median of 30 months (range, 13–36 months) before acquiring HBV infection, whereas the remaining 17 (77.3%) acquired HBV while remaining HIV-1 seronegative. Of 27 men who acquired HIV-1 during follow-up, 4 started lamivudine or tenofovir-containing antiretroviral therapy (ART) during follow-up, at 24 to 55 months after HIV-1 acquisition. One of these 4 MSM converted for anti-HBc, at ~12 months after ART initiation.
In the multivariable analysis adjusted for age, marital status, sex work, RAI, and any unprotected sex, uncircumcised status (adjusted incidence rate ratio [aIRR], 5.0; 95% CI 1.5–16.8), recent HIV-1 acquisition (aIRR, 2.9; 95% CI 1.1–7.7), rape (5.0; 95% CI 1.2–20.4) and any tertiary education (aIRR, 3.2; 95% CI 1.1–9.7) were associated with HBV acquisition (Table 2). Results did not change when using a time-updated circumcision variable to capture the status of 1 man who was circumcised during follow-up (and remained HBV susceptible at his last visit). Of note, the unadjusted association between uncircumcised status and HBV infection was stronger among MSM reporting IAI (incidence rate ratio [IRR], 4.6; 95% CI 1.0–20.4) than among men not reporting IAI (IRR, 3.5; 95% CI 0.9–14.3). The 8 incident HBV cases among men reporting RAI all occurred in circumcised MSM.
Table 2.
Risk Factors Associated With HBV Incidence Among 312 MSM in Kilifi, Kenya, 2005–2014
| Characteristics | Incident HBV/100 PY* | Incidence/100 PY (95% CI) | Unadjusted IRR (95% CI) | P Value | Adjusted IRR (95% CI) | P Value |
|---|---|---|---|---|---|---|
| Age categories (years)† | ||||||
| 18–24 | 12/204.4 | 5.9 (3.3-10.3) | Reference | Reference | ||
| 25–34 | 9/133.4 | 6.7 (3.5-13.0) | 1.15 (0.48-2.75) | .755 | 1.09 (0.46-2.57) | .840 |
| 35+ | 1/31.3 | 3.2 (0.4–22.7) | 0.54 (0.07-4.33) | .564 | 0.74 (0.11-4.89) | .752 |
| Education | ||||||
| Primary/none | 9/185.0 | 4.9 (2.5-9.4) | Reference | Reference | ||
| Secondary | 9/160.9 | 5.6 (2.9-10.8) | 1.15 (0.46-2.92) | .762 | 1.00 (0.41-2.45) | .998 |
| Higher/tertiary | 4/23.2 | 17.2 (6.5-45.9) | 3.56 (1.15-11.00) | .027 | 3.18 (1.05-9.65) | .041 |
| Marital status† | ||||||
| Never married | 20/303.3 | 6.6 (4.3-10.2) | Reference | Reference | ||
| Ever married | 2/65.7 | 3.0 (0.8–12.2) | 0.46 (0.10-2.02) | .303 | 0.51 (0.12-2.08) | .346 |
| Religion | ||||||
| None/other | 3/51.0 | 5.9 (1.9-18.2) | Reference | – | – | |
| Christian | 14/184.9 | 7.6 (4.5-12.8) | 1.29 (0.36-4.66) | .700 | ||
| Muslim | 5/133.1 | 3.8 (1.6-9.0) | 0.64 (0.14-2.80) | .549 | ||
| Employment | ||||||
| None | 11/165.1 | 6.7 (3.7-12.0) | Reference | – | – | |
| Self | 8/138.6 | 5.8 (2.9-11.5) | 0.87 (0.35-2.16) | .764 | ||
| Formal | 3/65.3 | 4.6 (1.5-14.2) | 0.69 (0.18-2.57) | .578 | ||
| Circumcised | ||||||
| Yes | 18/349.4 | 5.2 (3.2-8.2) | Reference | Reference | ||
| No | 4/19.6 | 20.4 (7.7-54.3) | 4.00 (1.44-11.08) | .008 | 4.96 (1.47-16.75) | .010 |
| HIV-1 status | ||||||
| Seronegative | 17/325.2 | 5.2 (3.2-8.4) | Reference | Reference | ||
| Recent HIV acquisition (range, 13–36 months) | 5/43.8 | 11.4 (4.8-27.4) | 2.19 (0.83-5.74) | .111 | 2.88 (1.08-7.67) | .035 |
| Sex partners past 3 months | ||||||
| Men and women | 15/252.2 | 5.9 (3.6-9.9) | Reference | – | – | |
| Men only | 7/116.8 | 6.0 (2.9-12.6) | 1.01 (0.41-2.48) | .989 | ||
| Received payment for sex with cash, living expenses, or goods in past 3 months† | ||||||
| No | 11/159.6 | 6.9 (3.8-12.4) | Reference | Reference | ||
| Yes | 11/209.4 | 5.3 (2.9-9.5) | 0.76 (0.32-2.81) | .538 | 0.75 (0.31-1.85) | .537 |
| Paid for sex in past 3 months | ||||||
| No | 16/261.3 | 6.1 (3.8-10.0) | Reference | – | – | |
| Yes | 6/107.7 | 5.6 (2.5-12.4) | 0.91 (0.35-2.35) | .846 | ||
| Any use of alcoholic beverage in past month | ||||||
| No | 10/139.4 | 7.2 (3.9-13.3) | Reference | – | – | |
| Yes | 12/229.6 | 5.2 (3.0-9.2) | 0.73 (0.31-1.73) | .475 | ||
| Receptive anal intercourse (RAI) in past 3 months† | ||||||
| No | 14/203.8 | 6.9 (4.1-11.6) | Reference | Reference | ||
| Yes | 8/165.2 | 4.8 (2.4-9.7) | 0.70 (0.30-1.67) | .426 | 0.46 (0.19-1.13) | .091 |
| Insertive anal intercourse (IAI) in past 3 months‡ | ||||||
| No | 12/180.5 | 6.6 (3.8-11.7) | Reference | – | – | |
| Yes | 10/188.5 | 5.3 (2.9-9.9) | 0.80 (0.34-1.86) | .601 | ||
| Sexual exposure and protection with condoms in past week† | ||||||
| No activity | 6/120.3 | 5.0 (2.2-11.1) | Reference | Reference | ||
| All protected | 6/108.7 | 5.5 (2.5-12.3) | 1.11 (0.36-3.43) | .859 | 1.42 (0.45-4.53) | .549 |
| Any unprotected | 10/140.0 | 7.1 (3.8-13.3) | 1.44 (0.51-4.00) | .490 | 2.06 (0.67-6.34) | .208 |
| Total sex partners in past month | ||||||
| Less than 3 | 11/192.4 | 5.7 (3.2-10.3) | Reference | – | – | |
| Three or more | 11/176.6 | 6.2 (3.4-11.2) | 1.09 (0.48-2.48) | .838 | ||
| Group sex in past 3 months‡ | ||||||
| No | 18/333.2 | 5.4 (3.4-8.6) | Reference | |||
| Yes | 4/35.9 | 11.2 (4.2-29.7) | 2.06 (0.69-6.15) | .196 | ||
| Raped in past 3 months | ||||||
| No | 21/364.5 | 5.8 (3.8-8.8) | Reference | Reference | ||
| Yes | 1/4.5 | 22.0 (3.1–156.2) | 3.82 (0.92-15.80) | .064 | 4.95 (1.20-20.42) | .027 |
| Intravenous drug use in past 3 months | ||||||
| No | 22/368.0 | 6.0 (3.9-9.1) | – | – | – | – |
| Yes | 0/1.0 | 0 | ||||
| Genital washing with soap in past week‡ | ||||||
| No | 7/71.2 | 9.8 (4.7-20.6) | Reference | – | – | |
| Yes | 15/297.8 | 5.0 (3.0-8.4) | 0.51 (0.21-1.25) | .140 | ||
| Sex with a menstruating female partner in past 3 months | ||||||
| No | 20/341.0 | 5.9 (3.8-9.1) | Reference | – | – | |
| Yes | 1/24.9 | 4.0 (0.6–28.5) | 0.68 (0.09-5.01) | .701 | ||
| Don’t know | 1/3.1 | 32.2 (4.5-228.7) | 5.49 (0.67-44.96) | .113 | ||
Abbreviations: aIRR, adjusted incidence rate ratios; CI, confidence interval; anti-HBc, hepatitis B core antibody; HBsAg, hepatitis B virus surface antigen; anti-HBs, hepatitis B surface antibody; HBV, hepatitis B virus; HIV-1, human immunodeficiency virus; IRR, incidence rate ratios; MSM, men who have sex with men; PY, person-years.
*Incident HBV infection was defined as being negative for HBsAg, anti-HBs, and anti-HBc at baseline and positive for anti-HBc or HBsAg at any point during follow-up, regardless of whether exposure resulted in persistent HBV infection.
†Considered a priori in the initial multivariable model.
‡Not included in the final model.
DISCUSSION
In this population of MSM in coastal Kenya, approximately 60% were susceptible to HBV infection at enrollment. Susceptible men were young and reported high-risk sexual behavior that put them at risk for both HIV-1 and HBV acquisition. We found an overall incidence of HBV core or HBsAg seroconversion of 6.0 per 100 person-years. Recent HIV-1 acquisition, having been raped and any tertiary education were strongly associated with HBV acquisition risk, whereas circumcised MSM had lower HBV acquisition rates.
A high HBV incidence was observed in our cohort among MSM compared with that reported in the Multicenter AIDS Cohort Study (MACS) cohort in the United States (~0.96 per 100 person-years) [11] and in Amsterdam (~1.0–2.0 per 100 person-years) [8]. Of note, 31% of men in the MACS study had received at least 1 dose of HBV vaccine, increasing to 60% by the end of the observation period. In our analysis, we excluded men positive for anti-HBs only at baseline and censored initially HBV-susceptible men at their first HBV vaccine dose, so our higher estimate reflects HBV incidence in an unvaccinated cohort.
The HBV acquisition risk in this study was 5-fold higher in those who were uncircumcised compared with circumcised MSM. To the best of our knowledge, our study is the first to report an association between HBV acquisition risk and lack of circumcision among MSM in Africa. A cross-sectional study conducted among MSM from Argentina showed no significant association between male circumcision and prior HBV infection, defined as a positive HBsAg or anti-HBc test, either overall or among MSM who reported no RAI [12]. Some studies of the association between circumcision status and HBV infection in other populations have reported an increased risk of HBV due to circumcision as one of several unsafe medical procedures or traditional practices examined [13]. It is possible that safe surgical removal of the foreskin reduces sexual HBV acquisition through similar mechanisms as for HIV-1 infection [14]. That no association has been found between circumcision and prevalent HBV infection in studies in SSA could be explained by the large number of HBV infections acquired during childhood [3] and the failure to evaluate HBV acquisition prospectively. Our finding that circumcision may protect against HBV acquisition needs verification in other populations with sexual HBV acquisition risk, including uncircumcised men who report IAI.
Men who have sex with men who had acquired HIV-1 in the past 1–3 years were 3-fold more likely to acquire HBV compared with men who remained HIV-1 negative during follow-up. This is consistent with a cohort study conducted among MSM in the United States that documented higher rates of incident HBV (defined as positive HBsAg or anti-HBc results during follow-up) in men who were HIV-infected at baseline, compared with men without HIV infection at baseline [11]. In that study, individuals who acquired HIV-1 during follow-up were reclassified as HIV-infected and not categorized separately as having recent HIV-1 acquisition. Although the strong association of HBV acquisition with recent HIV-1 acquisition in our study of initially HIV-1-negative men remained after adjusting for sexual risk behavior, it may be possible that men who acquired HIV-1 during follow-up were at higher risk for sexual HBV acquisition due to unmeasured confounders. In our study, most MSM who acquired HIV-1 were not immediately treated with ART because Kenyan guidelines during the study period required a CD4 count of 350 or below to initiate therapy. In future prospective studies in which HIV-1-positive men will be immediately treated, it is likely that HBV risk will be reduced [15].
We observed a 5-fold higher risk of HBV acquisition among men who were raped. Because there was only 1 event among men who reported rape, CIs for the risk estimate were wide; however, it is plausible that damage to mucosal surfaces during rape may enhance HBV acquisition.
The rate of HBV infection among men with any tertiary education was 3 times that of MSM with primary or no formal education. Of note, a lower proportion of MSM with tertiary education reported any unprotected sex compared to MSM with lower education (15.2% vs. 38.0%, P = .008) (data not shown). The association we report between HBV risk and tertiary education may be due to under-reporting of sexual risk behavior among MSM with tertiary education, or to unmeasured confounders.
This study had 3 main limitations. First, self-reported sexual behavior may have been subject to recall bias, and social desirability bias may have led to underreporting of risky sexual behavior resulting to reduced effect sizes. Second, our sample and the number of HBV acquisition events were small. This likely reduced power and affected our ability to test multiple associations. Third, the MSM we studied are not representative of all Kenyan MSM; they all reported high-risk sexual behavior and most participated in transactional sex. Thus, the HBV incidence we observed may overestimate HBV incidence among MSM in Kenya.
Universal childhood HBV vaccination was introduced in Kenya in 2002 [16]. However, MSM born before this date did not receive HBV vaccination during childhood and hence are susceptible to HBV infection later in life. They and other at-risk populations such as female sex workers and injection drug users should be targeted for HBV vaccination.
CONCLUSIONS
In conclusion, we documented high HBV acquisition rates among MSM in Coastal Kenya, associated with recent HIV-1 acquisition, uncircumcised status rape, and tertiary education. Kenyan MSM who are at high risk for HIV-1 acquisition and for rape [17] should be prioritized for HBV vaccination. Given our findings on decreased HBV-infection risk in circumcised men, consideration should also be given to the promotion of circumcision among MSM who report IAI.
Supplementary Data
Supplementary materials are available at Open Forum Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.
Supplementary Material
Acknowledgments
Disclaimer. The contents are the responsibility of the study authors and do not necessarily reflect the views of United States Agency for International Development (USAID), the National Institutes of Health (NIH), the United States Government, or the Wellcome Trust. This report was published with permission from the Director of KEMRI.
Financial support. We acknowledge support from the University of California, San Francisco’s International Traineeships in AIDS Prevention Studies, US National Institute of Mental Health ([NIMH] Grant R25MH064712). This work was supported by the International AIDS Vaccine Initiative and the UW Center for AIDS Research, an NIH-funded program (P30 AI027757), which is supported by the following NIH Institutes and Centers (National Institute of Allergy and Infectious Diseases, National Cancer Institute, NIMH, National Institute on Drug Abuse, Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Heart, Lung, and Blood Institute, and National Center for Complementary and Alternative Medicine). S.M.G. was supported by NIMH grant R34 MH09946. The KEMRI-Wellcome Trust Research Programme at the Centre for Geographical Medicine Research-Kilifi is supported by core funding from the Wellcome Trust (Grant 077092). This study was also made possible by the generous support of the American people through the USAID.
Potential conflicts of interest. All authors: No reported conflicts.
All authors have submitted the ICMJE Form for Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.
References
- 1. Hoffmann CJ, Thio CL. Clinical implications of HIV and hepatitis B co-infection in Asia and Africa. Lancet Infect Dis 2007; 7:402–9. [DOI] [PubMed] [Google Scholar]
- 2. Kiire CF. Hepatitis B infection in sub-Saharan Africa. The African Regional Study Group. Vaccine 1990; 8Suppl: S107–12. [DOI] [PubMed] [Google Scholar]
- 3. Matthews PC, Geretti AM, Goulder PJ, Klenerman P. Epidemiology and impact of HIV coinfection with hepatitis B and hepatitis C viruses in sub-Saharan Africa. J Clin Virol 2014; 61:20–33. [DOI] [PubMed] [Google Scholar]
- 4. Schweitzer A, Horn J, Mikolajczyk RT, et al. Estimations of worldwide prevalence of chronic hepatitis B virus infection: a systematic review of data published between 1965 and 2013. Lancet 2015; 386:1546–55. [DOI] [PubMed] [Google Scholar]
- 5. Yang JD, Gyedu A, Afihene MY, et al. Hepatocellular carcinoma occurs at an earlier age in Africans, particularly in association with chronic hepatitis B. Am J Gastroenterol 2015; 110:1629–31. [DOI] [PubMed] [Google Scholar]
- 6. McKinnon LR, Gakii G, Juno JA, et al. High HIV risk in a cohort of male sex workers from Nairobi, Kenya. Sex Transm Infect 2014; 90:237–42. [DOI] [PubMed] [Google Scholar]
- 7. Sanders EJ, Okuku HS, Smith AD, et al. High HIV-1 incidence, correlates of HIV-1 acquisition, and high viral loads following seroconversion among MSM. AIDS 2013; 27:437–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. van Houdt R, Bruisten SM, Geskus RB, et al. Ongoing transmission of a single hepatitis B virus strain among men having sex with men in Amsterdam. J Viral Hepat 2010; 17:108–14. [DOI] [PubMed] [Google Scholar]
- 9. Beyrer C, Sullivan PS, Sanchez J, et al. A call to action for comprehensive HIV services for men who have sex with men. Lancet 2012; 380:424–38. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Amornkul PN, Karita E, Kamali A, et al. Disease progression by infecting HIV-1 subtype in a seroconverter cohort in sub-Saharan Africa. AIDS 2013; 27:2775–86. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Falade-Nwulia O, Seaberg EC, Snider AE, et al. Incident hepatitis B virus infection in HIV-infected and HIV-uninfected men who have sex with men from pre-HAART to HAART periods: a cohort study. Ann Intern Med 2015; 163:673–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Heuft MM, Houba SM, van den Berk GE, et al. Protective effect of hepatitis B virus-active antiretroviral therapy against primary hepatitis B virus infection. AIDS 2014; 28:999–1005. [DOI] [PubMed] [Google Scholar]
- 13. Pando MA, Balan IC, Dolezal C, et al. Low frequency of male circumcision and unwillingness to be circumcised among MSM in Buenos Aires, Argentina: association with sexually transmitted infections. J Int AIDS Soc 2013; 16:18500. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Aziz S, Khanani R, Noorulain W, Rajper J. Frequency of hepatitis B and C in rural and periurban Sindh. J Pak Med Assoc 2010; 60:853–7. [PubMed] [Google Scholar]
- 15. Morris BJ, Wamai RG. Biological basis for the protective effect conferred by male circumcision against HIV infection. Int J STD AIDS 2012; 23:153–9. [DOI] [PubMed] [Google Scholar]
- 16. Guidelines for the Treatment of Chronic Hepatitis B and C Viral Infections in Kenya. Available at: http://apps.who.int/medicinedocs/documents/s21972en/s21972en.pdf Accessed 19 October 2016. [Google Scholar]
- 17. Micheni M, Rogers S, Wahome E, et al. Risk of sexual, physical and verbal assaults on men who have sex with men and female sex workers in coastal Kenya. AIDS 2015; 29Suppl 3:S231–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.