CROI 2018 Abstract eBook

Abstract eBook

Poster Abstracts

830 PRECONCEPTIONAL ART AND SPONTANEOUS PRETERM BIRTH IN AN URBAN ZAMBIAN COHORT Joan T. Price 1 , Bellington Vwalika 2 , Jennifer Winston 1 , Margaret P. Kasaro 1 , Humphrey Mwape 1 , Benjamin H. Chi 1 , Elizabeth M. Stringer 1 , Bethany Freeman 1 , Courtney Gravett 3 , Katelyn J. Rittenhouse 1 , Lungowe Njobvu 1 , Julie A. Nelson 1 , Angela Kashuba 1 , Jeffrey S. Stringer 1 1 University of North Carolina Chapel Hill, Chapel Hill, NC, USA, 2 University of Zambia, Lusaka, Zambia, 3 Global Alliance to Prevent Prematurity and Stillbirth, Seattle, WA, USA Background: HIV and its treatment are associated with preterm birth (PTB) and other adverse birth outcomes. The underlying mechanism(s) may be related to timing of antiretroviral therapy (ART) relative to conception, specific drug exposures, or both. Methods: The Zambia Preterm Birth Prevention Study is ongoing at the University Teaching Hospital in Lusaka. Participants receive early ultrasound dating, lab testing, midtrimester cervical length measurement, serial fetal growth monitoring, and careful phenotyping of each adverse birth outcome. For the present analysis, we defined PTB as delivery prior to 37 gestational weeks and differentiated between spontaneous and indicated PTB phenotypes. We confirmed timing of ART exposure among HIV+ women by testing drug concentrations at first antenatal visit with liquid chromatography/mass spectrometry. Results: Between Aug-2015 and Aug-2017, we enrolled 1,425 pregnant women, of whom 929 have delivered. Median maternal age was 27 years (IQR: 23,32); median BMI was 23.9 kg/m2 (IQR: 21.4,27.4); 34%were nulliparous; 30% had ≥1 prior PTB; 2.3% had short cervix (<2.5cm); and 23%were HIV+. Of 946 infants born to date (17 sets of twins), 153 (16%) were small for gestational age (<10%ile) and 27 (2.9%) were stillborn. Of 129 (14%) PTBs, 70 (54%) were spontaneous (i.e., spontaneous labor or membrane rupture prior to labor), 27 (21%) were indicated (i.e., provider-initiated), and 32 (25%) could not be clearly classified. Among 216 HIV+ parturients, 121 (56%) were on preconceptional ART (predominantly TDF/XTC/EFV) and 108 (50%) had viral load <40 copies/mL at first antenatal visit. Factors associated with PTB in multivariable Poisson regression were: short cervix (ARR 2.18, 95% CI 1.16–4.12), twin gestation (ARR 5.01, 95% CI 3.22–7.81), prior PTB (ARR 2.31, 95% CI 1.48 – 3.62), and preconceptional ART (ARR 1.62, 95% CI 1.03–2.55). In a sensitivity analysis limited to the spontaneous PTB phenotype, the risk associated with preconceptional ART was amplified (ARR 1.93, 95% CI 1.03–3.61). Preconceptional ART was not statistically associated with asymmetric fetal growth restriction at 32 weeks, small for gestational age at birth, or stillbirth. Conclusion: In this African cohort established specifically to study PTB and where the predominant ART exposure is TDF/XTC/EFV, women continuing preconceptional ART are at higher risk of PTB than HIV-negative women and those initiating ART in pregnancy. The risk of preconceptional ART did not extend to other adverse birth outcomes.

829 BIRTH DEFECTS AMONG OFFSPRING OF HIV-INFECTED & UNINFECTED WOMEN IN KAMPALA, UGANDA Philippa Musoke 1 , Daniel M. Mumpe 1 , Ayoub Kakande 2 , Jolly Nankunda 3 , Diana Valencia 4 , Joyce Namale 1 , Evelyn Nabunya 3 , Josaphat Byagumisha 5 , Doreen Birabwa-Male 3 , Monica Nolan 1 , Dhelia Williamson 4 , Linda Barlow-Mosha 1 1 Makerere University–Johns Hopkins University Research Collaboration, Kampala, Uganda, 2 Makerere University, Kampala, Uganda, 3 Mulago National Referral Hospital, Kampala, Uganda, 4 CDC, Atlanta, GA, USA, 5 Makerere University College of Health Sciences, Kampala, Uganda Background: Universal antiretroviral treatment (ART) is being scaled up for all HIV-infected women; however, the risk for birth defects (BD) related to HIV and ART exposure during pregnancy is not well documented. We compared the prevalence of birth defects in offspring of HIV-infected women (on and off ART) and HIV negative women enrolled in the hospital-based birth defect surveillance (BDS) project in Kampala, Uganda. Methods: Hospital-based BDS was implemented at Mulago Hospital in Kampala. All informative live births (LB), stillbirths (SB) and abortions were included. Maternal history, HIV status and ART regimens were documented based on medical records and interviews. Eligible births were examined by midwives for selected major structural external birth defects. The prevalence of birth defects and 95% confidence intervals (95% CI) were calculated for HIV-infected (on and off ART) and HIV negative women. Data collected from August 2015 through May 2017 are presented. Association of BDs with ART was determined by logistic regression. Results: A total of 43,293 births were included with 38,527 (89.0%) delivered by HIV negative, 4,634 (10.7%) HIV-infected women and 132(0.3%) of unknown HIV status. 4,407 (95.1%) HIV-infected women were on ART with 3,403 (77.2%) on TDF/3TC/EFV, 403 (9.1%) on AZT/3TC+NVP, 320 (7.3%) on TDF/3TC/NVP, and 281(6.4%) on other regimens. 475(1.1%) birth defects of interest were identified: 431(1.1%) to HIV negative women and 44(0.9%) to HIV-infected women, with an OR of 0.8(0.6-1.1), p value= 0.31. 41/44 (93%) women with birth defect-affected pregnancies were on ART. Birth defect prevalence in offspring of infected women on ART [41(0.9%)] vs those not on ART [3(1.3%)] was similar [OR 0.6(0.2-2.1) p=0.55]. Prevalence of BD/1,000 births in offspring of those women who initiated ART prior to conception and during the 1st trimester (10.4) was similar to those who initiated ART later (7.3) [OR 1.4 (0.7-2.8) p= 0.3] and did not differ across ART regimens [X2(3) = 4.1, p=0.25]. Overall, common birth defects (n) included talipes equinovarus (51), hypospadias (48), neural tube defects (36), and total limb reduction (24). Conclusion: The prevalence of birth defects in offspring of HIV-infected women was similar to those who were uninfected. There was no difference in BD by ART, timing of initiation or specific ART regimen. These findings are reassuring but these data only had adequate numbers to examine all birth defects combined and not individual defects or categories.

Poster Abstracts

CROI 2018 313