CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

is needed to optimize the timing of monitoring in P&BF women to improve outcomes.

1 University of California San Francisco, San Francisco, CA, USA, 2 University of California Los Angeles, Los Angeles, CA, USA, 3 Harvard University, Boston, MA, USA, 4 University of Zimbabwe, Harare, Zimbabwe, 5 Malawi Coll of Med–Johns Hopkins Univ Rsr Proj, Blantyre, Malawi, 6 Makerere University–Johns Hopkins University Research Collaboration, Kampala, Uganda, 7 University of North Carolina Project–Malawi, Lilongwe, Malawi, 8 St. Jude Children’s Research Hospital, Memphis, TN, USA, 9 Johns Hopkins University, Baltimore, MD, USA Background: Adherence to antiretroviral therapy (ART) throughout pregnancy and breastfeeding is crucial for both maternal health and prevention of perinatal and sexual transmission. Tenofovir (TFV) concentrations in hair, reflecting long-term cumulative exposure, have been examined as an adherence metric for PrEP but have not yet been examined among persons living with HIV. We examined hair TFV levels in breastfeeding women on TFV disoproxil fumarate (TDF)/emtricitabine (FTC)-based ART over time, predictors of these levels, and the association of hair levels with viral suppression. Methods: Women in the IMPAACT PROMISE 1077BF Study who were on ART during both pregnancy and breastfeeding are included in this analysis. From 2013-2016, hair samples were collected at 1 week (6-14 days) postpartum, 6, 14, 26 weeks and every 3 months through breastfeeding up to 18 months. For women on TDF/FTC ≥30 days, hair TFV levels were measured by liquid chromatography/tandemmass spectrometry. Using generalized estimating equations, we estimated the impact of hair TFV levels on viral suppression (plasma HIV RNA <400copies/mL) over time via logistic regression and assessed predictors of hair TFV levels via linear regression. Results: Hair TFV levels were measured at 374 visits in 71 women who breastfed a median of 14 months (interquartile range [IQR] 12-15). Median weeks on ART at delivery was 12 (IQR 7-17); median age 26 years (IQR 22-30). After ≥90 days on ART, 18/69 (26%) ever experienced viremia (median 8907 copies/mL, range 444-244,984); 8 (12%) had >1 measure ≥400. Each doubling of TFV level was associated with 2.53 times the odds of viral suppression (95%CI: 1.51-4.25, p=0.0004), adjusted for age and time since delivery. The strongest predictor of hair TFV levels was time since delivery. Compared to 0-3 months postpartum, TFV levels were 1.38 fold higher (95%CI 1.09-1.76) in months 3-6, 1.65 fold higher (95%CI 1.31-2.07) in months 6-12 and 1.45 fold higher (95%CI 1.12-1.89) after 12 months (Figure). We did not identify other factors meaningfully associated with TFV levels. Conclusion: We present the first report examining hair TFV levels among people living with HIV on TDF/FTC-based ART, here in breastfeeding women up to 18 months postpartum. Hair TFV levels strongly predicted viral suppression. Average hair TFV levels were lowest in the first 3 months postpartum, suggesting the need for intensified adherence support in this major transition period to preserve maternal health and prevent perinatal transmission.

765 MOBILITY AND THE 1-YEAR POSTPARTUM MATERNAL MORTALITY IN HIV-POSITIVE PREGNANT WOMEN Hae-Young Kim 1 , Adrian Dobra 2 , Frank Tanser 1 1 Africa Health Research Institute, Mtubatuba, South Africa, 2 University of Washington, Seattle, WA, USA Background: There is increasing evidence that mobile population living with HIV might experience disengagement from health services and worse health outcomes. We sought to characterize pregnant women’s mobility patterns and its association with maternal mortality. Methods: All pregnant women aged ≥15 years were followed up to 1 year after delivery using one of Africa’s largest ongoing population-based cohorts between January 2003 and December 2016 in rural KwaZulu-Natal, South Africa. Changes in residency and household membership were recorded during biannual household surveys. External migration was defined as moving-in or -out of the surveillance area during pregnancy or in the first-year postpartum period. Maternal death was ascertained with the closest care giver via verbal autopsy based on the INDEPTH/WHO questionnaire. Of those with unknown HIV status, women whose death were attributable to AIDS or TB were considered as HIV-positive and the others as HIV-negative in a sensitivity analysis. Multiple cox regression models were used. Results: Of 30,291 pregnant women, 3,339 were HIV-positive while 10,958 were HIV-negative and 15,994 had unknown HIV status at delivery. There were 27 unique mobility patterns- 64% and 13% of women always resided inside or outside of the study area, respectively. Of 23%women who had externally migrated at least once, 39% delivered outside the study area. The overall maternal death rates were 0.6/1000 person-years (PY) among HIV-negative postpartumwomen. Of HIV-positive postpartumwomen, maternal mortality rates were 7.5/1000 PY among who always resided within the study area, compared to 22.6/1000 PY among those who externally migrated and delivered outside the area (p<0.001). HIV-positive pregnant women who externally migrated and delivered outside the study area had a 2.82 times higher hazard of maternal mortality (95% CI: 1.04-7.69) after adjusting for age, parity, time period (before or after 2010) and other sociodemographic factors. In the sensitivity analysis, HIV-positive women who had external migration had a six times higher hazard of mortality than HIV-negative women who always resided within the study area, adjusting for maternal HIV status. Conclusion: A substantial portion of peripartumwomen moved within- country around the time of delivery and experienced a significantly higher risk of mortality, likely due to disengagement from health services. Interventions to address linkage to and retention in care among migrating pregnant women are urgently needed. 766 TENOFOVIR HAIR LEVELS RISE OVER THE POSTPARTUM PERIOD AND HIGHLY PREDICT VIRAL LOADS Pamela M. Murnane 1 , Peter Bacchetti 1 , Judith S. Currier 2 , Sean Brummel 3 , Hideaki Okochi 1 , Nhi Phung 1 , Karen Kuncze 1 , Risa M. Hoffman 2 , Teacler Nematadzira 4 , Dean Soko 5 , Maxensia Owor 6 , Friday Saidi 7 , Patricia M. Flynn 8 , Mary Glenn Fowler 9 , Monica Gandhi 1

Poster Abstracts

CROI 2019 296