Sexual Transmission of Hepatitis C Virus Among HIV-Infected Men Who Have Sex with Men --- New York City, 2005--2010

In the United States, an estimated 3.2 million persons are living with hepatitis C virus (HCV) infection (1). HCV transmission occurs primarily through percutaneous exposure to blood, and persons who inject drugs are at greatest risk for infection. The role of sexual transmission of HCV has not been well defined. However, reports over the past decade, mainly from Europe, have implicated sexual transmission of HCV among human immunodeficiency virus (HIV)--infected men who have sex with men (MSM). In late 2005, two HIV-infected MSM, each with acute HCV infection that was suspected to have been acquired sexually, were evaluated at Mount Sinai Medical Center in New York City, prompting Mount Sinai to request referrals of similar patients (2). During 2005--2010, a total of 74 HIV-infected MSM with recently acquired HCV infection and no reported history of injection-drug use were evaluated. To examine the role of sexual transmission, a matched case-control study and viral analysis were conducted. Results from the case-control study showed that high-risk sexual behavior was the most likely mode of transmission among these men. Phylogenetic analyses revealed five clusters of closely related HCV variants, suggesting networks of transmission among these men. The findings underscore the importance of screening HIV-infected MSM for HCV, particularly those engaged in high-risk sexual behavior.

For this study, a case-patient was defined as an HIV-infected MSM examined at Mount Sinai during October 2005--December 2010 who had 1) a newly elevated alanine transferase (ALT) level, 2) a newly positive HCV-antibody test result, and 3) no other evident cause of the newly elevated ALT level. To the extent possible, positive HCV-antibody results were confirmed by HCV RNA testing. If no record was found of a previous negative HCV-antibody test, a finding of jaundice or an ALT elevation of more than 15-fold above the upper limit of normal (i.e., >450 U/L) also was required. To assess whether patients might have had a previous positive HCV test result unknown to the referring physicians, the date of the first positive HCV-antibody test of a subset of patients (24 men) was confirmed by the New York City Department of Health and Mental Hygiene through review of the hepatitis registry of HCV surveillance data. Providers of primary care to HIV-infected MSM in New York City (who, as part of care, routinely obtain ALT levels on their patients during HIV monitoring visits) were contacted by the lead investigator and asked to refer patients with newly elevated ALT levels to Mount Sinai as soon as possible. Reminders were provided periodically throughout the study period. A total of 35 HIV-care providers contributed information on their patients to this study.

Characteristics of Case-Patients

During October 2005--December 2010, Mount Sinai evaluated 74 HIV-infected MSM who reported no injection-drug use and had newly elevated ALT levels and a positive HCV antibody test result; 73 of 74 also had documented HCV viremia. Median age of the 74 patients was 39 years; 41 were non-Hispanic white, 14 non-Hispanic black, 18 Hispanic, and one Asian (Table 1). Median CD4+ cell count for the patients was 483 cells/µL (range: 66--1,258 cells/µL). Sixty patients (81%) were asymptomatic, and new HCV infection was detected solely because of new ALT elevation; 14 (19%) had jaundice at presentation. Median peak ALT level was 665 U/L (range: 72--5,291 U/L). No other cause for the patients' elevated ALT levels was found (e.g., no new infection with hepatitis A or B virus and no new drug therapy). Of the 74 patients, 65 (91%) had a previous negative HCV-antibody test result before detection of hepatitis (median: 12 months; range: 0--110 months).

Case-Control Study

To assess the role of sexual transmission of HCV, a matched case-control study was conducted beginning in July 2007. HIV-infected MSM examined at Mount Sinai during July 2007--December 2010 who were within 12 months of clinical onset of HCV infection and who reported no injection-drug use were recruited as case-patients. For each case-patient, 1--10 controls (i.e., HIV-infected MSM who did not have HCV infection, reported no injection-drug use, and matched by age [±5 years] and race/ethnicity) were recruited by Mount Sinai staff members from among the practices that referred case-patients during the enrollment period. In all, 22 case-patients and 53 control subjects were enrolled in the study.

All participants were asked to complete self-administered questionnaires regarding their sexual practices and drug-use behaviors during the 12 months preceding diagnosis (for case-patients) or preceding the questionnaire (for matched controls). To conduct a matched analysis, a conditional logistic regression of each variable (i.e., sexual practice or drug use behavior) was performed. Those variables that had a p value of ≤0.20 in the univariable analysis, as well as those previously associated with sexual transmission (3), were entered into a model and analyzed using multivariable conditional logistic regression (i.e., forward, backward, and stepwise) to determine which variables were independently associated with HCV infection.

Univariable results. Univariable analyses indicated that the HIV-infected MSM newly infected with HCV (case-patients) were significantly more likely than the HIV-infected MSM without HCV infection (matched controls) to have had receptive (matched odds ratio [mOR] = 24.87) or insertive (mOR = 2.62) anal intercourse with no condom and with ejaculation, practiced receptive (mOR = 10.08) or insertive (mOR = 7.90) fisting, used sex toys (mOR = 4.38), engaged in group sex (mOR = 19.28), engaged in sex while high on drugs (mOR = 11.37), previously had syphilis (mOR = 8.80) or gonorrhea (mOR = 5.02), and had sex while high on methamphetamine (mOR = 26.80) (Table 2). Because three variables (receptive anal intercourse, no condom, no ejaculation; sex while high on gamma hydroxybutyrate [GHB]; and sex while high on ketamine) yielded undefined ORs, the data were analyzed further using exact conditional logistic regression. Results showed that case-patients were significantly more likely than controls to report receptive anal intercourse with no condom and no ejaculation (mOR = 24.26) and sex while high on GHB (mOR = 16.34).

Multivariable results. Results from the multivariable analyses showed that receptive anal intercourse with no condom and with ejaculation of the partner (adjusted odds ratio [AOR] = 23.00) and sex while high on methamphetamine (AOR = 28.56) were both significantly related to acquiring HCV infection. Of all the practices and behaviors, having sex while using methamphetamine was most strongly associated with HCV infection (Table 2).

Results of Phylogenetic Analyses

Polymerase chain reaction and sequencing of a 470 base-pair region of NS5B from HCV strains recovered from 50 of the 74 men were conducted using methods described previously (4). Forty-seven of the 50 were genotype 1a, and three were genotype 1b. A maximum-likelihood phylogenetic tree was then created (5).* These analyses identified five clusters of closely related HCV variants from 26 (55%) of the 47 men with genotype 1a infections.

Reported by

Daniel S. Fierer, MD, Stephanie H. Factor, MD, Alison J. Uriel, MBBS, Damaris C. Carriero, MS, Douglas T. Dieterich, MD, Michael P. Mullen, MD, Arielle Klepper, Wouter van Seggelen, MSc, Kathryn Childs, MBBS, Andrea D. Branch, PhD, Dept of Medicine, Mount Sinai School of Medicine, New York, New York. Deborah Holtzman, PhD, John W. Ward, MD, Yury Khudyakov, PhD, Scott D. Holmberg, MD, Div of Viral Hepatitis, National Center for HIV/AIDS, Viral Hepatitis, STD, and TB Prevention, CDC. Corresponding contributor: Deborah Holtzman, dholtzman@cdc.gov, 404-718-8555.

Editorial Note

This report suggests high-risk sexual behavior as a cause of HCV transmission among HIV-infected MSM in New York City. Unprotected receptive anal intercourse with ejaculation and sex while high on methamphetamine were the most important predictors of HCV infection. Results from phylogenetic analyses suggest networks of HCV transmission among these men. The findings of high-risk sex, concurrent noninjection-drug use, and phylogenetic clustering are similar to those observed among cohorts of HIV-infected MSM with HCV infection in Northern Europe and Australia (4). A notable finding from this study and those in other countries is the association of noninjection, recreational drug use (e.g., methamphetamine use) with the acquisition of HCV infection.

Sexual transmission of HCV is considered to be an inefficient and rare mode of transmission (6). However, concurrent HIV infection results in increased HCV RNA levels (viral load) (7), which are thought to increase infectiousness of HCV acquired through sexual contact. Of further concern among persons who are coinfected is that HIV accelerates HCV disease progression, even in its early stages (2). End-stage liver disease and hepatocellular carcinoma, both usually resulting from chronic HCV infection, are now leading causes of death not attributable to acquired immunodeficiency syndrome (AIDS) among HIV-infected persons in the United States (8).

The findings in this report are subject to at least three limitations. First, recall of events such as ejaculation by sex partner up to 12 months before HCV diagnosis can be imperfect. For example, the findings should not be interpreted to definitively exclude acquisition of HCV by some men through unprotected receptive anal intercourse without ejaculation, even though this variable did not exert a significant independent effect on HCV infection in the multivariable analysis. Second, refusal to acknowledge injection-drug use is not uncommon, and other types of stigmatizing risk behavior also might be underreported. Such social desirability bias was addressed by using a self-administered questionnaire and assuring each patient that his responses would not be shared with his primary-care provider. Finally, study investigators relied on patient referrals from HIV-care providers outside Mount Sinai, and referral bias might have occurred; however, the number of referring providers was fairly sizable (n = 35).

Sexual transmission of HCV among HIV-infected MSM is more widespread than this one study demonstrates. A recent U.S. report described HCV-antibody seroconversions among HIV-infected MSM without a history of injection-drug use (9). A recent European report that examined a group of studies, primarily from Europe, found substantial increases, particularly during 2002--2007, in the incidence of HCV infection among HIV-infected MSM, demonstrating just how serious the epidemic has become among these men (10). Hepatitis C should be added to the list of infections spread among HIV-infected MSM who have sex with HCV-infected partners. HIV-infected patients should be counseled and reminded that unprotected sex between HIV-infected partners can transmit other infections, including HCV. In addition to HCV screening for MSM newly diagnosed with HIV, routine HCV screening using both ALT and antibody testing should be considered for HIV-infected MSM, particularly those with high-risk sexual behaviors or concomitant ulcerative sexually transmitted diseases (e.g., syphilis and herpes simplex virus).† Finally, newly diagnosed HCV infections among HIV-infected MSM should be reported to state and local health authorities.

References

1. Armstrong GL, Wasley A, Simard EP, McQuillan GM, Kuhnert WL, Alter MJ. The prevalence of hepatitis C virus infection in the United States, 1999 through 2002. Ann Intern Med 2006;144:705--14.
2. Fierer DS, Uriel AJ, Carriero DC, et al. Liver fibrosis during an outbreak of acute hepatitis C virus infection in HIV-infected men: a prospective cohort study. J Infect Dis 2008;198:683--6.
3. Danta M, Brown D, Bhagani S, et al. Recent epidemic of acute hepatitis C virus in HIV-positive men who have sex with men linked to high-risk sexual behaviours. AIDS 2007;21:983--91.
4. van de Laar T, Pybus O, Bruisten S, et al. Evidence of a large, international network of HCV transmission in HIV-positive men who have sex with men. Gastroenterology 2009;136:1609--17.
5. Felsentein J. Evolutionary trees from DNA sequences: a maximum likelihood approach. J Mol Evol 1981;17:368--76.
6. Tohme RA, Holmberg SD. Is sexual contact a major mode of hepatitis C virus transmission? Hepatol 2010;52:1497--505.
7. Matthews-Greer JM, Cladito GC, Adley SD, et al. Comparison of hepatitis C viral loads in patients with or without human immunodeficiency virus. Clin Diagn Lab Immunol 2001;8:690--4.
8. Palella FJ Jr, Baker RK, Moorman AC, et al. Mortality in the highly active antiretroviral therapy era: changing causes of death and disease in the HIV outpatient study. J Acquir Immune Defic Syndr 2006;43:35--41.
9. Taylor LE, Holubar M, Wu K, et al. Incident hepatitis C virus infection among US HIV-infected men enrolled in clinical trials. Clin Infect Dis 2011;52:812--8.
10. van der Helm JJ, Prins M, del Amo, et al. The hepatitis C epidemic among HIV-positive MSM: incidence estimates from 1990 to 2007. AIDS 2011;25:1083--91.