7th IAS Conference on HIV Pathogenesis, Treatment and Prevention (IAS 2013)


MOLBPE30 - Poster Exhibition

Undetectable viral load is associated with increased unprotected anal intercourse in gay serodiscordant couples

B.R. Bavinton1, F. Jin1, I. Zablotska1, G. Prestage1,2, A. Grulich1, The Opposites Attract Study Team

1The University of New South Wales, The Kirby Institute, Sydney, Australia, 2La Trobe University, The Australian Research Centre in Sex, Health and Society, Melbourne, Australia

Background: In serodiscordant heterosexual couples, undetectable viral load is associated with a substantially reduced HIV transmission risk. However, no studies in gay male serodiscordant couples (SDCs) have been reported. We present data on viral load (VL) and risk behaviour in a study of HIV transmission in gay male SDCs.
Methods: Opposites Attract is an ongoing longitudinal cohort study of gay male SDCs recruited via clinics and community-based advertising. HIV-Positive Partners (HPPs) have VL tested at baseline, while HIV-Negative Partners (HNPs) have HIV antibody tests and complete an online computer-assisted self-interview within 48 hours of the study visit, reporting sexual behaviour and partner's perceived VL. The association between unprotected anal intercourse (UAI) and VL was examined with logistic regression and Wilcoxon rank-sum test.
Results: As of April 2013, 64 couples were enrolled. At baseline, 84.4% (n=54) of HPPs were taking ART and 76.6% (n=49) had undetectable VL. Overall, 68.8% (n=44) of HNPs reported any UAI with their partners: 59.4% of HNPs reported any insertive UAI, 37.5% reported receptive UAI without ejaculation, and 21.9% reported receptive UAI with ejaculation. Over three-quarters (78.1%) of HNPs believed their HPP's last VL test result to be undetectable. While the most recent VL test may have been several weeks prior, the HNP's perception of his partner's VL was mostly in accord with the baseline test results. In the couples with perceived undetectable VL, 78.0% reported UAI in the last three months. In comparison, only 35.7% of couples in which the HPP's VL was perceived to be detectable (n=8) or where the VL result was not known (n=3) reported UAI (OR=0.16, 95%CI=0.04-0.56, p=0.005). Overall, the median number of UAI acts in the last three months was 4 (range=0-183, mean=18.5, SD=34.0). The median number of UAI acts was 6.5 in couples where the HNP believed his partner's VL was undetectable. This compared to 0 in couples where the perceived VL was detectable or was not known (p=0.007).
Conclusion: Among Australian gay male SDCs, perceived undetectable VL is strongly related to increased practice of UAI. Studies of HIV transmission risk in this population are an urgent research priority.

Download the e-Poster

Back to the Programme-at-a-Glance