CROI 2020 Abstract eBook

Abstract eBook

Poster Abstracts

766 VIRAL LOAD MONITORING IN PREGNANCY TO PREDICT PERIPARTUM VIRAEMIA IN SOUTH AFRICA Jasantha Odayar 1 , Siti Kabanda 1 , Thokozile R. Malaba 1 , Maia Lesosky 1 , Landon Myer 1 1 University of Cape Town, Cape Town, South Africa Background: WHO guidance recommends VL monitoring in pregnant women on ART to help identify high-risk infants for enhanced prophylaxis but there are few data evaluating this approach in routine care. Methods: Data come from the screening procedures of a RCT of postpartum HIV care strategies at a large primary care clinic in Cape Town. In this setting VL monitoring takes place at the earlier of 12 weeks on ART, or 34 weeks’ gestation. In this context we identified consecutive HIV+ women initiating ART (TDF+FTC+EFV) who underwent VL testing during pregnancy, and for women with VL<400 copies/mL documented during pregnancy, repeated a VL within 4 weeks postpartum. All VL testing was done by the National Health Laboratory Services using the Abbott Realtime HIV-1 assay (Abbott Laboratories, Waltham, MA). We calculated sensitivity (SE), specificity (SP) and positive and negative likelihood ratios (LR+ and LR-) for antenatal VL<100 copies/ml in predicting peripartum VL<100 and <400 copies/mL, with sensitivity analyses examining subgroups of gestation at antenatal VL and prior ART exposure. Results: In 323 women (median age 28y, 40%with a history of previous ART), antenatal VL was taken at a median gestation of 33w (IQR 30-36), and at that time, 89.2% of women had a VL<100 copies/mL and 10.9% had a VL 100-400 copies/mL. Peripartum VL was taken at a median of 9 days (IQR 6-17) postpartum, at which point women were on ART for a median duration of 23w (IQR 18-28), and at that time 91.6%, 6.8% and 1.6% had VL<100, 100-400 and >400 copies/mL, respectively. The SE of antenatal VL<100 copies/mL in predicting peripartum VL<100 copies/mL was 0.95 (95% CI 0.92-0.97) and the corresponding SP was 0.74 (95% CI 0.54-0.89; LR+3.7, LR- 0.07). The ability of antenatal VL to predict peripartum VL was similar across strata of gestation at VL testing and history of ART use. Predictive performance was slightly weaker at thresholds of <400 copies/mL (LR+ 2.25, LR- 0.17). Conclusion: These novel data suggest that antenatal VL<100 copies/mL is a useful predictor of peripartum viraemia and may be used to target enhanced PMTCT interventions in this setting. The high SE and low LR- suggest few women who are virologically suppressed during antenatal care subsequently become viraemic peripartum.

767 HIGH VIRAL SUPPRESSION AMONG HIV-POSITIVE POSTPARTUM WOMEN: CLUSTER RANDOMIZED TRIAL Appolinaire Tiam 1 , Lauren Greenberg 1 , Vincent Tukei 1 , Thabelo Ramatlapeng 2 , Tsietso Mots'oane 2 , Matseliso Masitha 1 , Matsepeli Nchephe 2 , Mammatli Chabela 1 , Laura Guay 1 , Heather Hoffman 3 1 Elizabeth Glaser Pediatric AIDS Foundation, Washington, DC, USA, 2 Ministry of Health, Maseru, Lesotho, 3 George Washington University, Washington, DC, USA Background: HIV-positive women are particularly vulnerable to poor retention and ART adherence in the postpartum period with low viral suppression that poses risks to maternal health and to transmission of HIV to their infants. We assessed the effect of a multidisciplinary integrated management team intervention on viral suppression in a cohort of HIV-positive pregnant women in Lesotho. Methods: The IMPROVE cluster randomized study evaluated an intervention that included a multidisciplinary management teamwith maternal child health staff, village health workers, and peer mentor mothers to work together to support HIV-positive and negative women in uptake and retention in HIV and MCH services. Training together, using job aids, and adding early home based follow-up of new ANC attendees were included in the intervention. Twelve facilities were randomized to intervention or control arms. HIV-positive pregnant women were enrolled at their first ANC visit with prospective follow- up for at least 12 months postpartum. Study nurses conducted interviews with participants, extracted medical record information and collected dried (whole) blood spots from HIV-positive women for viral load testing. We compared viral load (VL) results at 12 months postpartum using Chi-square tests to test for differences between study arms. Results: 613 HIV-positive women were enrolled in the study, 308 in the interventional arm and 305 in the control arm. 570 women had delivery information, all of whomwere on ART at the time of delivery. VL results from 11-15 months postpartumwere available for 351 (57%) women. There was no difference in follow-up (pregnancy losses/stillbirths, transfer to facilities outside the district, and loss to follow-up) by study arm. Overall 325 (93%) women were suppressed with a VL <1000 copies/ml. A greater proportion of women in the intervention group had a suppressed VL (166/175, 95%) compared to women in the control arm (159/176, 90%) but the difference was not statistically significant (p=0.106). Significantly more women in the intervention group had an undetectable viral load (83% intervention vs. 72% control, p=0.016). Conclusion: The multi-component IMPROVE intervention led to a small but not significant increase in viral suppression in HIV positive women one year after delivery, with high rates of suppression in both arms. 768 CHANGES IN BONE MINERAL DURING AND AFTER LACTATION IN UGANDAN WOMEN ON OPTION B+ ART Florence Nabwire 1 , Ann Prentice 1 , MatthewM.Hamill 1 , Mary Glenn Fowler 2 , Josaphat Byamugisha 3 , Adeodata Kekitiinwa 4 , Gail R. Goldberg 1 1 MRC Elsie Widdowson Laboratory, Cambridge, UK, 2 Makerere University–Johns Hopkins University Research Collaboration, Kampala, Uganda, 3 Makerere

Poster Abstracts

CROI 2020 283