Is sexual contact a major mode of hepatitis C virus transmission?

Authors

  • Rania A. Tohme,

    Corresponding author
    1. Division of Viral Hepatitis, National Center for HIV/AIDS, Viral Hepatitis, STD, and TB Prevention, Atlanta, GA
    2. Epidemic Intelligence Service, Office of Workforce and Career Development, Centers for Disease Control and Prevention, Atlanta, GA
    • 1600 Clifton Road NE, Centers for Disease Control and Prevention Mailstop G-37, Atlanta, GA 30333
    Search for more papers by this author
    • fax: 404-718-8585

  • Scott D. Holmberg

    1. Division of Viral Hepatitis, National Center for HIV/AIDS, Viral Hepatitis, STD, and TB Prevention, Atlanta, GA
    Search for more papers by this author

  • Potential conflict of interest: Nothing to report.

Abstract

Medical opinion varies considerably regarding the transmission of hepatitis C virus (HCV) through sexual contact. Based on the study design, representativeness of the study population, and the methods used for case ascertainment, we analyzed 80 qualifying reports regarding the evidence for or against sexual transmission. Regarding heterosexual transmission, the weight of evidence is that there is no increased risk of sexual transmission of HCV among heterosexual couples in regular relationships. This risk increases among persons with multiple sexual partners (adjusted odds ratio [aOR] 2.2-2.9), but this association may be confounded by increased likelihood of injection drug use with increased number of partners. There appears to be a real increased risk for women coinfected with human immunodeficiency virus (HIV) or other sexually transmitted infections (aOR 3.3-3.9) and especially for HIV-infected gay men who are having sex with one another compared with HIV-uninfected men (aOR 4.1-5.7). HIV-infected gay men increase their risk of such transmission in association with practices that lead to mucosal trauma (multiple sexual partners, fisting, use of sex toys) and the presence of genital ulcerative disease. Conclusion: This review should inform, and not distract from, recommendations to reduce the risk of HCV transmission. Health care providers need to pay special attention to sexual transmission of HCV among HIV-infected individuals. HEPATOLOGY 2010

Hepatitis C virus (HCV) infection is a blood-borne infection transmitted mainly through injection drug use (IDU), blood transfusions, organ transplantations, accidental needle sticks, 1, 2 and other parenteral exposures, including inappropriate use or reuse of needles and syringes in health care settings. 3, 4 Sexual transmission is a controversial mode of HCV transmission that has received considerable attention among health care providers and the lay public. For example, in 2009, the Centers for Disease Control and Prevention's Division of Viral Hepatitis received over 2,600 telephone and email inquiries about hepatitis C. When looking at available data that capture responses to inquiries, transmission modes of hepatitis C and the sexual transmission of hepatitis C infection were among the top used responses (Centers for Disease Control and Prevention, unpublished reports).

The possibility of sexual transmission of HCV infection is supported by the isolation of HCV RNA from semen and cervical smears in some studies 5-7 but not others. 8, 9 Furthermore, although the sexual transmission of the same hepatitis C virus strain, as determined by molecular analysis, has been documented in some case reports and case series, 10-20 the magnitude of the risk varies depending on the quality of the study design, the likelihood of unmeasured parenteral routes of transmission, and the level of risk behavior of the study participants.

Given the conflicting evidence and the ongoing inquiries, we conducted a review of the literature to summarize the best available data on the risk of HCV transmission through sexual activity.

Abbreviations:

aOR, adjusted odds ratio; CI, confidence interval; HCV, hepatitis C virus; HIV, human immunodeficiency virus; IDU, injection drug use; MSM, men who have sex with men; STI, sexually transmitted infection.

Subjects and Methods

Study Identification.

Studies addressing the sexual transmission of hepatitis C were identified through a comprehensive literature search on PubMed of English-language articles published between 1995 and 2009. We excluded studies published prior to 1995 because hepatitis C case ascertainment depended on laboratory tests that were not as accurate as currently available ones. Search terms such as hepatitis C, HCV, sexual transmission, and men who have sex with men were used to identify potentially relevant papers. Cited references in relevant articles were also carefully assessed for inclusion.

Study Rating.

Each article was evaluated based on the strength of the study design, representativeness of the study population, adjustment or control of potentially confounding HCV risk factors (such as IDU), and mode of ascertaining hepatitis C infection. The literature search yielded 535 published reports, 444 of which were excluded because they did not address the risk of sexual transmission of HCV. Of the remaining 91 reports, 11 were excluded because they were review papers (n = 8) or editorials and author responses (n = 3). Hence, a total of 80 articles were eligible for inclusion. The magnitude of the risk of sexual transmission of HCV was assessed by presenting the adjusted odds ratios (aORs) obtained from the studies that controlled for the most common routes of HCV transmission.

Results

Heterosexual Transmission.

Studies addressing heterosexual transmission of HCV distinguished among three types of sexual contacts: sexual contacts within regular partnerships; sexual contacts with multiple partners; and sexual contacts among persons with preexisting sexually transmitted infections (STIs) and/or human immunodeficiency virus (HIV). Table 1 summarizes major studies that assessed the risk of heterosexual transmission of HCV infection among these different groups.

Table 1. Selected Studies Assessing Heterosexual Transmission of Hepatitis C
StudyCountry (Years of Study)Study SampleSample Size (% HIV-Positive)HCV PrevalenceHCV Incidence (Cases per 100 Person-Years)Phylogenetic Testing*Sexual Transmission Reported as a Risk Factor (aOR [95% CI])
  • *

    To characterize specimens by molecular means.

  • No confirmatory HCV testing was performed.

Cohort studies
 Vandelli et al. (21)Italy (1991-2001)Monogamous couples967 (0%)3.4%0.037YesNo
 Marincovich et al. (22)Spain (1991-2001)Couples171 (0.6%)0.0NoNo
 Kao et al. (33)Taiwan (1990-1997)Spouses1120.233YesYes
Case-control studies
 Salleras et al. (42)Spain (1992)Pregnant women43 cases,172 controlsNoYes if unprotected sex with ≥2 sexual partners (2.9 [1.1-7.4])
 Mele et al. (43)Italy (1985-1996)Acute HCV surveillance data708 cases; 3,746 controls0.001NoYes if unprotected sex with multiple sexual partners (2.2 [1.6-3.0])
Cross-sectional studies
 Stroffolini et al. (25)Italy (1994-1996)Couples31110.3%NoNo
 Tahan et al. (26)Turkey (1999-2002)Spouses6002.0%NoNo
 Atome et al. (27)GabonHousehold1956.7%YesNo
 La Torre et al. (28)Italy (1995-2000)Household2598.9%NoNo
 Brusaferro et al. (29)Italy (1994-1995)Household51410.3%NoNo
 Caporaso et al. (30)Italy (1995)Household1,3797.3%NoNo
 Feldman et al. (44)United States (1990-1991)Women engaging in high-risk behavior502 (2.0%)1.6%NoYes (14.2 [1.8-642.5])
 Hershow et al. (48)United States (1994-1995)Women with HIV or at high risk for HIV infection296 (82.1%)42.0%NoYes (2.7 [1.1-7.0])
 Frederick et al. (49)United States (1994-2002)Non-IDU women3,636 (74.3%)31.5%NoYes among HIV-positive women (1.9 [1.2-2.9])

Several large prospective cohort studies did not show an increased risk for HCV transmission among heterosexual discordant couples (married or steady partners), even after 10 or more years of observation. 21-24 In these studies combined, there was no increased risk of sexual transmission of HCV, even after an estimated 750,000 vaginal and anal contacts between couples; accordingly, the probability of such transmission was less than 1 in 10 million sex contacts. Cross-sectional studies reported HCV prevalence rates among regular partners of infected persons varying between 2% and 10%. 21, 25, 26 However, no association was found between HCV infection and sexual transmission between partners in regular relationships after controlling for other risk factors. 25-32 Three studies documented the presence of the same virus in very few couples by molecular analysis and attributed this to sexual transmission of HCV, 33-35 but could not definitely exclude other common exposures.

A potentially confounding factor in the sexual transmission of HCV in heterosexual couples is the duration of the relationship, an index of the number of sexual exposures to HCV from an infected partner. Whereas a few studies found an increased risk of acquiring HCV infection with a longer relationship, 28, 35-37 other larger studies that controlled for age did not find a significant association between the duration of the relationship and HCV infection. 26, 27, 38, 39 The higher prevalence of HCV infection in older couples may represent a cohort effect (in which couples of the same age might be exposed to common sources of infection or common practices, such as the reuse of nondisposable but contaminated medical equipment), as was reported in Spain 40 and Taiwan. 41

Unlike couples in regular relationships, persons having multiple sexual partners have more than twice the likelihood of acquiring HCV infection (aOR 2.2-2.9). 42-45 Notably, a study conducted among pregnant women in Spain showed that women who had unprotected sexual relations with two to four partners were almost three times (aOR 2.8, 95% confidence interval [CI] 1.1-6.9) more likely to acquire HCV than women with only one steady partner. 42

Data regarding heterosexual transmission of hepatitis C should be interpreted with caution, however. Three large Italian cross-sectional studies showed that the risk of spousal transmission could also be explained by the common practice of sharing syringes. 25, 30, 36 Furthermore, a recent analysis of acute HCV infections in the United States has indicated that increased numbers of sexual partners correlates with increased likelihood of injection drug use (Monina Klevens, Centers for Disease Control and Prevention, unpublished data).

The presence of preexisting STIs has also been found to increase the risk of acquiring HCV by heterosexual contact. 46, 47 A cross-sectional study in India showed that men infected with herpes simplex virus 2 were almost four times more likely to have HCV than men without herpes simplex virus 2 infection (aOR 3.85, 95% CI 1.18- 12.6). 47 Similarly, individuals with Trichomonas infection were much more likely to acquire HCV than individuals without an STI (aOR 3.3, 95% CI 1.7-6.3). 46

More unequivocal is the risk of heterosexual transmission to those who are infected with HIV. Two cross-sectional studies confirm a substantial increase in risk of acquiring HCV infection among heterosexual persons with preexisting HIV, particularly among those engaging in high-risk sexual behaviors and having unprotected sex with multiple sexual partners (Table 1). 48, 49 Notably, the large Women's Interagency HIV Study found that, controlling for IDU, HIV-infected women were still almost twice as likely as HIV-negative women to acquire HCV (aOR 1.9, 95% CI 1.2-2.9). 49 Likewise, a cross-sectional study among STD clinic attendees in Baltimore showed a four-fold increase in the risk of HCV infection among HIV-infected patients compared with those who were HIV-seronegative (aOR 4.4, 95% CI 1.9-10.3). 46 In a study of hemophilic men and their partners 23 in which unacknowledged IDU was unlikely to be a confounding variable, 6% of hemophiliac men who were coinfected with HIV compared with only 2% of the men infected with HCV alone transmitted HCV to their spouses. In contrast, a smaller cohort study did not show evidence of sexual transmission of HCV from partners who were both HCV/HIV-coinfected. 22

Homosexual Transmission.

Incidence rates of HCV infection among HIV-uninfected men who have sex with men (MSM) have varied between zero cases per 100 person-years in Amsterdam 50 to 1.5 cases per 1,000 person-years in the United Kindgdom. 51 The Omega Cohort Study, the only study that has included a large sample of MSM and controlled for all other HCV risk factors, did not find an increased risk for HCV infection among HIV-uninfected MSM, even among those engaging in risky sexual behaviors such as having multiple partners or unprotected anal sex 52; however, the observation time in this study was only 1 year, and few of the subjects engaged in such high-risk behavior. A prospective study from Australia showed lack of sexual transmission of HCV among HIV-negative MSM, 53 whereas another cohort study reported an HCV incidence of 0.11 per 100 person-years (95% CI 0.03-0.26) among HIV-negative MSM. 54 However, IDU was a common practice among these HCV-infected patients. Studies from Canada and Argentina also did not find an association between HCV infection and homosexual contact in HIV-uninfected men. 55, 56

The situation is entirely different for HIV-infected gay men, especially those who engage in high-risk and traumatic sex practices involving anal mucosal damage. Studies addressing the emerging public health problem of HCV in HIV-infected men are limited and are mainly from western Europe (Table 2), but they suggest that the incidence of HCV infection among HIV-positive MSM has been increasing. A cohort study in Amsterdam showed a significant increase in HCV incidence among HIV-infected MSM, from 0.08 cases per 100 person-years between 1984 and 1999 to 0.87 cases per 100 person-years between 2000 and 2003. 50 Similarly, it has been estimated that the incidence of acute HCV infections among HIV-infected MSM in the United Kingdom has increased by 20% every year since 2002. 57, 58 The French PRIMO cohort study also showed an increase in the incidence of HCV infection among HIV-infected individuals from 1.2 per 1,000 person-years before 2003 to 8.3 per 1,000 person-years after 2003. 59

Table 2. Selected Studies Assessing Transmission of Hepatitis C Among MSM
StudyCountry (Year of Study)Sample SizeHCV PrevalenceHCV Incidence (Cases per 100 Person-Years)HIV PrevalencePhylogenetic Testing*Sexual Transmission Reported as a Risk Factor
  • *

    To characterize specimens by molecular means.

  • No confirmatory HCV testing was performed.

Cohort studies
 Van de Laar et al. (50)Netherlands (1984-2003)1,8361.3%0.07 (non-HIV, 0; HIV, 0.18)28%YesYes
 Alary et al. (52)Canada (1996-2001)1,0852.9%0.0380%NoNo
 Jin et al. (54)Australia (2001-2007)1,398 non-HIV; 245 HIVNon-HIV, 1.1% (range, 0.6-1.8); HIV, 9.4% (range,6.0-13.8)Non-HIV, 0.11 (range, 0.03-0.3); HIV, 0 (range, 0-1.5) NoYes among HIV-coinfected HCV- prevalent cases
 Ghosn et al. (59)France (1996-2005)3275.8%0.35100%NoYes
 Hammer et al. (60)United States (1997-2000)7542.1%06.6%NoNo
 Rauch et al. (75)Switzerland (1988-2004)7,89933%0.64100%NoYes
Case-control studies
 Danta et al. (69)United Kingdom (1999-2005)60 cases; 130 controls-100%YesYes
 Fierer et al. (72)United States (2005-2008)21 cases; 21 controls21 cases100%YesYes
Cross-sectional studies
 Matthews et al. (53)Australia (2004-2006)120120 cases21.7%NoYes among HIV-positive MSM
 Myers et al. (55)Canada3,3041.9%9%NoNo
 Urbanus et al. (61)Netherlands (2007-2008)6894.4%22.8%YesYes
 Turner et al. (70)United Kingdom (1999-2000)3083.6%100%NoYes
 Morin et al. (71)France (1999-2007)126 cases126 cases12%NoYes
 Ndimbie et al. (76)United States (1984-1985)1,0582.9%19.6%NoYes

Several longitudinal studies of HIV-infected MSM totaling more than 12,000 person-years of follow up have shown that these men are at much higher risk for sexually acquired HCV than HIV-uninfected MSM (aOR, 4.1 to 5.7). 50, 51, 60 Likewise, a large cross-sectional study in Amsterdam reported that HIV-infected MSM were almost 43 times (95% CI 8.49-215.1) more likely to acquire HCV infection than HIV-uninfected MSM. 61 HIV-positive men were much more likely to be coinfected with HCV in a few Australian studies, 53, 62 but IDU was also known to be widely prevalent among MSM in Australia. 53, 54, 62 A smaller cross-sectional study from the United States showed that HCV-infected MSM with were more likely to be coinfected with HIV than those who were HCV-negative (70% versus 29%). 63 Similarly, sexual transmission was the sole identified route of HCV infection among HIV-infected MSM in France. 59 Only a few cross-sectional studies have not shown an increased risk of HCV infection among HIV-infected MSM or found an association between HCV and HIV coinfection. 64-67

The practice of “serosorting” among HIV-infected MSM – unprotected sex between two HIV-infected men who are aware of their own and their partners' HCV infection (but not necessarily HCV infection) –has been commonly reported in recent studies. 15, 16, 68, 69 Many common themes or risks have been observed in the majority of studies that have addressed HCV infection among HIV-infected MSM: engaging in unprotected sex with multiple partners, 14-16, 20, 53, 69-74 anal fisting, use of sex toys, and the presence of genital ulcerative disease. 15-20, 50, 54, 61, 63, 69, 70, 74-78 In the only well-designed case control study thus far, participation in group sex was more prevalent among HIV-coinfected HCV cases compared with controls (88% versus 52%). 69 Participation in group sex significantly increased the odds for HCV infection (aOR 9.16, 95% CI 3.51-23.9) if it involved at least two of the following four sexual practices: receptive and insertive anal intercourse and receptive and insertive fisting. 69 Fisting significantly increased the risk of HCV infection among HIV-infected MSM by more than five-fold (aOR 6.27-12.6), 61, 70 as did the use of sex toys (78% versus 43%). 69 Fisting was highly correlated with use of sex toys, group sex, and bleeding in a cross-sectional study from Amsterdam. 61 In addition, the use of psychoactive substances was common among HCV/HIV-coinfected MSM and might have acted as a disinhibitor for risky sexual practices, leading to traumatic sex and mucosal damage. 55, 74 Thus, some researchers have concluded that the real risk of sexual transmission in HIV-infected persons results from blood-to-blood contact during sex. 78

It is worth noting that phylogenetic analyses in cohort studies 50, 75 and in an international network of MSM 68 have also indicated concordance of gene sequences in recovered HCV strains. The difficult-to-treat genotypes 1a and 4d were usually recovered. 17, 18, 50, 59, 61, 68, 69, 74, 79

Discussion

This review focused on the risk of sexual transmission of hepatitis C infection and distinguished between heterosexual and homosexual contact. The real risk for sexual transmission appeared to be predominantly related to HIV infection: of all the practices considered in this review, the clearest and least equivocal risk behavior was unprotected sex between HIV-infected partners, particularly HIV-positive MSM.

Since the last review of this topic in Hepatology, 80 the most notable trend has been a growing number of reports from European and American cities, indicating an increase in incidence and prevalence of HCV infection among HIV-infected MSM. 14, 18, 50, 54, 55, 57-59, 61, 69, 74, 75 This increase has mostly been reported after the introduction of highly active antiretroviral therapy treatment, which may contribute to risky sexual behavior due to the belief among these men that treatment will eliminate further risk of HIV infection. It has also been postulated that a greater proportion of HIV-infected MSM have been engaging in serosorting, 15, 16, 18, 74, 81-83 which is thought to be the source of a specific pan-national cluster of HCV infection in Europe. 68 In addition, the presence of several activities and conditions that disrupt anal mucosal integrity (traumatic sex, sex with visible blood, genital ulcerative diseases, use of sex toys) were frequently noted in these instances of putative sexual transmission.

The use of molecular sequencing in some case reports has led some researchers to conclude that HCV transmission between spouses was caused by sexual contact. However, this finding does not preclude that the virus might have been transmitted through unacknowledged needle (or other sharp object) sharing. 31, 84, 85 In fact, when the risk of spousal HCV transmission was analyzed in Italy, this resulted from the common practice of sharing syringes. 25, 30, 36 Although phylogenetic analysis is a useful laboratory technique to demonstrate genetic similarities or variations in recovered viruses, it does not obviate the role of careful epidemiological analysis.

The studies included in our review had several limitations. A major limitation common to all the studies was the unavoidable reliance on self reports for the ascertainment of IDU. Unacknowledged or unascertained IDU among men and women with multiple sex partners undoubtedly confounds all analyses of association of HCV infection with number of sex partners. Another limitation is that the risk from and exposure to other sharp objects as potential vehicles for transmission cannot be excluded. 63, 65, 66, 70 Furthermore, prospective cohort studies of heterosexual couples in a single-partner relationship may have preselected persons who would be unlikely to transmit the virus—that is, if transmission of HCV occurred in one of the first sexual encounters, choosing discordant couples for analysis (those who have not previously transmitted) may represent a selection bias.

Despite these limitations, studies could be categorized and evaluated as to their quality and credibility and conclusions drawn. The use of condoms and refraining from high-risk sexual behavior is definitely indicated among persons who have HIV infection or another STI or who are not in a single-partner relationship. Health care providers need to pay special attention to HIV-infected MSM. Initial testing for HCV is recommended for all individuals in the United States who are entering HIV care, 86 but annual or other regular testing should receive serious consideration. This review should form a basis for appropriate health messages to inform susceptible individuals of the real risks of HCV infection rather than distract them with highly unlikely sources of transmission.

Acknowledgements

We thank Cynthia Jorgensen, Division of Viral Hepatitis, Centers for Disease Control and Prevention, for information on the volume of e-mail, and telephone inquiries about HCV transmission.

Ancillary

Advertisement