CROI 2017 Abstract e-Book

Abstract eBook

Poster and Themed Discussion Abstracts

766 UPTAKE AND RETENTION IN CARE OF PREGNANT WOMEN STARTING OPTION B+ IN MAPUTO

Esmeralda Karajeanes 1 , Dulce Bila 1 , Orvalho Augusto 2 , Maria Lain 1 , Basilio Muianga 1 , Chadreque Muluana 3 , Nello Macuacua 1 , Victorino Chavane 1 , Paula Vaz 1 1 Fundação Ariel Glaser Contra o Sida Pediatrico, Maputo, Mozambique, 2 Univ Eduardo Mondlane, Maputo, Mozambique, 3 Direcção Provincial de Saude de Maputo, Maputo Provincia, Mozambique Background: Following the 2013 World Health Organization recommendation (WHO), Mozambique introduced the option B+ strategy to prevent mother to child transmission (PMTCT) of the human immunodeficiency virus (HIV). In Mozambique retention in care is still a challenge and undermines every effort made to control the epidemic. We compared attrition (defaulting from treatment or death) rates and retention in care of pregnant women (PW) initiating antiretroviral therapy (ART) on Option B+ with non-pregnant women of childbearing age initiating ART, followed at health facilities (HF) supported by Ariel Glaser Foundation in Maputo, Mozambique. Methods: A cohort analysis was carried out on data from the electronic ART patient tracking system. Anonymous data fromwomen who initiated ART between January 2014 and June 2015 was extracted. Descriptive statistics and Kaplan-Meier estimates were used to estimate retention. Cox proportional hazards regression was used to estimate the hazard- rate of attrition with robust cluster adjusted standard errors. Adjusting variables included HF’s patient volume, WHO staging, prophylactic cothrimoxazole (CTZ), CD4 count, age at ART initiation, education and marital status. Results: A total of 22079 women from 34 sites were included: 8316 were pregnant on B+ and 13763 non-pregnant. At ART initiation, PWwere younger than non-pregnant women (mean age 26.8 Vs. 32.4, P<0.001) and with less advanced disease (WHO I/II) (95.2% Vs 75.4%, P<0.0001). In both groups, women were more likely to be single, have primary education and be on TDF/3TC/EFV regimen. Lower cumulative retention was observed in B+ PW at 12 and 24 months (71.5 and 59.1 vs. 83.9 and 74.0, P < 0.001). Incidence of attrition was more likely to occur in the first 6 months with higher rate for B+ women (25.9/100 PY; 95% CI: 20.1 - 31.6 vs. 13.3/100 PY; 95% CI:10.1 - 16.5) (Figure 1). PW had higher risk of attrition (adjusted HR:1.65; 95% CI: 1.44–1.88); other risk factors included lower age (15-19) and illiteracy. Conclusion: The results from this study suggest that the benefits of ART to prevent MTCT through implementation of option B+ can be undermined by low retention rates in pregnant women on B+. Programmatic strategies should be implemented in order to improve retention in HIV positive pregnant women with particular emphasis to young girls.

Poster and Themed Discussion Abstracts

767 STRATEGIES FOR VIRAL-LOAD MONITORING DURING PREGNANCY IN RESOURCE-LIMITED SETTINGS Maia Lesosky 1 , Elaine J. Abrams 2 , Nei-Yuan M. Hsiao 3 , Landon Myer 1 , for the Maternal Child Health - AntiretroviralTherapy (MCH-ART) Study 1 Univ of Cape Town, Cape Town, South Africa, 2 Columbia Univ, New York, NY, USA, 3 NHLS, Cape Town, South Africa Background: Viral load (VL) monitoring is critical for effective antiretroviral therapy (ART). Pregnant women have unique monitoring needs due to MTCT risks and significant concerns about inadequate adherence, but there are few systematic insights into strategies for VL monitoring during pregnancy in resource-limited settings. Methods: We developed a multi-compartment Monte Carlo simulation of VL trajectories during gestation and postpartum, parameterised using South African data from the Maternal & Child Health-Antiretroviral Therapy (MCH-ART) study. The model allows variable distributions of: pre-ART VL, gestation at ART start, time to viral suppression (VS), gestation at delivery and loss of VS over time. We simulated a cohort of 10,000 women starting ART in pregnancy followed through delivery and evaluated the performance of VL monitoring schedules with testing at different times and frequencies (including point-of-care [POC] testing with same-day results) to predict VL>1000 cps/mL as a threshold for possible interventions at delivery/postnatally. Results: The model was parameterised with median (IQR) pre-ART log10 VL of 4.0 (3.3, 4.7), median gestation at ART start 20 weeks (16, 26), and 29% of individuals cumulatively experiencing VL>1000 cps/mL after VS by 6m postpartum. Four key findings emerge from applying different monitoring schemes to the simulated data: (i) if monitoring in pregnancy is based on current guidelines for non-pregnant adults with a first VL after 6m on ART, only 22% of women would be tested in pregnancy and 91% of all viraemia

CROI 2017 332