CROI 2015 Program and Abstracts

Abstract Listing

Poster Abstracts

delivery was associated with higher pre-ART VL (relative odds [RO] 0.39 for a 1-log increase in pre-ART VL, p<0.001); later GA at ART initiation (RO, 0.87 for a 1-week increase in GA at ART initiation, p<0.001); and inversely associated with higher pre-ART CD4 cell counts (RO, 1.08 for a 50-unit increase in pre-ART CD4 cell count, p=0.025).

Conclusions: These data provide novel evidence on VS after ART initiation in pregnancy in African populations using a standardised first-line regimen. The rapid early declines in VL to <3 log within a month on ART in most women are encouraging. However one-quarter of the cohort still had detectable VL at the time of delivery, demonstrating the importance of early initiation of ART in pregnancy. 865 Maternal Viral Load in the Context of PMTCT B+Within the Kabeho Study in Kigali Emily A. Bobrow 1 ; Placidie Mugwaneza 2 ; Gilles F. Ndayisaba 3 ; Dieudonne Ndatimana 3 ; Michelle Gill 1 ; Heather J. Hoffman 4 ; Cyprien Baribwira 5 ; Laura Guay 1 ; Anita Asiimwe 6 Kabeho StudyTeam 1 Elizabeth Glaser Pediatric AIDS Foundation, Washington, DC, US; 2 Ministry of Health, Kigali, Rwanda; 3 Elizabeth Glaser Pediatric AIDS Foundation, Kigali, Rwanda; 4 George Washington University Milken Institute School of Public Health, Washington, DC, US; 5 University of Maryland, School of Medicine, Kigali, Rwanda; 6 Rwanda University Teaching Hospitals, Kigali, Rwanda Background: In April 2012, Rwanda started to implement a policy to initiate HIV-positive pregnant women on lifelong antiretroviral treatment (ART) (‘Option B+’). In April 2013, EGPAF and the Ministry of Health began the Kigali Antiretroviral and Breastfeeding Assessment for the Elimination of HIV (Kabeho) Study. The study will determine 18 and 24 month HIV-free survival of a cohort of HIV-exposed children in the PMTCT program. Methods: From April 2013-January 2014, 608 HIV-positive women on triple drug ART in the third trimester of pregnancy or within two weeks post-delivery were enrolled in the observational prospective cohort from 14 health facilities in Kigali. After providing written informed consent, women underwent enrollment, including HIV and ART-related history and adherence, and a blood draw for viral load (VL) testing by RNA-PCR (Roche). Results: The median time women knew their HIV-positive status was 38.0 months (IQR 4.7–83.5). The most common ARV regimen (56.6%, 344/608) was TDF/3TC/EFV. Overall, 35.2% (n=214) of women reported taking another regimen previously; 21.5% (n=130) due to PMTCT during an earlier pregnancy. At enrollment, women were on ART for a median of 13.4 months (IQR 2.96–48.8); median time on current ART was 9.2 months (IQR 2.3-34.8). The adherence rate based on a 3-day ART recall was 90.9%. Side effects were reported in the past month by 17.5% (n=105) of women, with dizziness as most common (n=53). Half of women (52.2%, 316/606) had undetectable VL. Figure 1 shows the distribution of VL by ART duration. Logistic regression using GEE (N=579) indicates women were more likely to have a detectable VL if they had no education (AOR=2.21, 95% CI: 1.31, 3.73), reported side effects in the past month (AOR=1.96, 95% CI: 1.37, 2.81), and had been on ART less than four months, when compared to those with ART exposure from 4-12 months (AOR=3.98, 95% CI: 2.11, 7.50), 12-24 months (AOR=6.04, 95% CI: 2.47, 14.76), and 24-36 months (AOR=5.57, 95% CI: 2.38, 13.05). VL slightly decreased beyond 36 months on ART (AOR=3.56, 95% CI: 1.69, 7.50).

Poster Abstracts

Figure 1. Distribution of viral loads stratified by time on ART.

521

CROI 2015