CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

Results: Of 10,558 mother-infant pairs, 8,994 (85.2%) met the inclusion criteria. The overall stillbirth rate was 25 per 1,000 deliveries (95% confidence interval 22–28). We found no significant association between timing of maternal ART initiation and stillbirth. Older maternal age, male sex of the infant, breech vaginal delivery, delivery at <34 weeks of gestation, and having any maternal obstetric complication were associated with increased odds of stillbirth. Delivery at a mission hospital or health center were associated with lower odds of stillbirth than deliveries at a central hospital. Conclusion: Pregnant women’s exposure to ART, regardless of time of initiation, was not associated with an increased risk of stillbirth. This finding suggests that any negative effects of antiretroviral drug exposure to the infant may be compensated by the benefit of ART on the health of the mother. Kartik K. Venkatesh 1 , Mona Farhad 2 , Terence Fenton 2 , Dhayendre Moodley 3 , Shilpa Naik 4 , Clemensia Nakabiito 5 , Lee Fairlie 6 , Bonus Makanani 7 , Friday Saidi 8 , Tsungai Chipato 9 , James A. McIntyre 10 , Mary Glenn Fowler 11 , Benjamin H. Chi 1 , Jeffrey S. Stringer 1 , for the PROMISE 1077BF team 1 University of North Carolina at Chapel Hill, Chapel Hill, NC, USA, 2 Harvard University, Boston, MA, USA, 3 University of KwaZulu-Natal, Durban, South Africa, 4 Byramjee Jeejeebhoy Government Medical College, Pune, India, 5 Makerere University, Kampala, Uganda, 6 University of the Witwatersrand, Johannesburg, South Africa, 7 University of Malawi, Blantyre, Malawi, 8 University of North Carolina Project–Malawi, Lilongwe, Malawi, 9 University of Zimbabwe, Harare, Zimbabwe, 10 Anova Health Institute, Johannesburg, South Africa, 11 Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA Background: The PROMISE trial found antiretroviral therapy (ART) to be associated with preterm birth (PTB) compared to zidovudine (ZDV) alone. PROMISE used newborn clinical exam to define PTB, since gold-standard ultrasound (US) dating was not universally available. We analyzed the association between ART and PTB in a subset of participants in whom fetal US was available. Methods: This analysis is restricted to singleton liveborn pregnancies with pre-randomization fetal US biometry. Our outcomes were PTB<37 weeks and <34 weeks. Exposures of interest were antiretroviral regimens in the trial’s 3 randomization groups: ZDV-alone, ZDV-based ART, and tenofovir (TDF)-based ART. We fit multivariable logistic regression models, adjusting for maternal characteristics, obstetric history, and HIV disease severity. Since earlier ultrasound dating is more accurate, we conducted a sensitivity analysis of women with US<24 weeks. For comparison, we also present results of an earlier analysis of all trial participants (Chi et al., CROI 2016). Results: Among 3,423 trial participants, 724 (21%) singleton pregnancies had gestational age dating by both newborn exam and fetal US. The median gestational age at US was 24.0 weeks (IQR: 19.0, 28.8); 99%were from Uganda, South Africa, or India. Overall, 46% of women were randomized to ZDV-alone, 44% to ZDV-based ART, and 10% to TDF-based ART (a lower proportion because of a mid-trial protocol change). PTB<37 weeks was 20% and PTB<34 weeks was 6%. In multivariable analysis, women receiving either ART regimen had significantly higher odds of PTB at both <37 and <34 weeks compared to ZDV- alone. Findings were similar when restricted to 353 women with US<24 weeks. The odds of PTB<37 weeks by randomization armwas generally consistent with prior analyses where gestational age was defined by newborn exam. However, our results differed when examining the PTB<34 outcome in this smaller subset: the association between ZDV-based ART and PTB < 34weeks became stronger, while a previously detectable difference between the two ART arms disappeared (Table). Conclusion: A subset analysis of PROMISE 1077BF among women with US dating reconfirmed a significant association between ART started in pregnancy and PTB. A significantly increased risk of PTB<34 weeks with ART was observed with US dating but not with newborn exam. This may be attributable to reduced misclassification with more accurate US gestational dating and warrants further research.

neonatal or infant deaths). Women receiving the text message intervention [adjusted odds ratio (aOR) 0.60, 95%CI (0.45-0.80)] and those who received both text messages and the CMM intervention [aOR 0.68 (0.55-0.86)] had lower odds of having an APO when compared to the control group. (Table1) Women on non- nucleoside reverse transcriptase inhibitors (NNRTI) based ART were less likely to experience an APO when compared to those on protease inhibitors (aOR 0.43, 95%CI 0.21-0.88). Women receiving Tenofovir were twice as likely to experience an APO when compared to women on Zidovudine (aOR 2.00, 95%CI 1.28-3.10). Other factors associated with increased odds of APO included age (aOR 1.14 per 5 years, 95%CI 1.01-1.29) and time on ART. Conclusion: This cohort of pregnant women on ART experienced high rates of adverse pregnancy outcomes, which were associated with age, type of ART, and duration on ART. Further understanding of the impact of ART and possible mitigating interventions to reduce adverse pregnancy outcomes in this population are needed.

755 PRETERM BIRTH AMONG WOMEN WITH ULTRASOUND-BASED GESTATIONAL DATING IN PROMISE 1077BF

Poster Abstracts

754 TIMING OF MATERNAL ANTIRETROVIRAL THERAPY INITIATION AND STILLBIRTH IN MALAWI Malango T. Msukwa 1 , Olivia Keiser 1 , Andreas Jahn 2 , Joep J. van Oosterhout 3 , Andrew Edmonds 4 , Nozgechi S. Phiri 1 , Ronald Manjomo 5 , Mary-Ann Davies 6 , Janne Estill 7 1 University of Geneva, Geneva, Switzerland, 2 Ministry of Health, Nsanje, Malawi, 3 Dignitas International, Zomba, Malawi, 4 University of North Carolina at Chapel Hill, Chapel Hill, NC, USA, 5 Baobab Health Trust, Lilongwe, Malawi, 6 University of Cape Town, Cape Town, South Africa, 7 University of Bern, Bern, Switzerland Background: Studies on the use of antiretroviral therapy (ART) during pregnancy in HIV-infected women suggest that in-utero ART exposure may be associated with adverse birth outcomes, including stillbirth, preterm delivery, and being small for gestational age. Despite efforts to understand the effect of ART exposure on birth outcomes in prevention of mother-to-child transmission programs, the association remains unclear in resource-limited settings. We assessed the association between timing of maternal ART initiation and stillbirth among HIV-infected pregnant women in Malawi’s Option B+ program. Methods: We conducted a cross-sectional study using routine program data frommaternity registers at 20 large health facilities. We included all women with data on maternal age and timing of ART initiation who delivered singleton live births or stillbirths at ≥28 weeks of gestation between January 1, 2012 and June 30, 2015. We defined stillbirth as an infant born with no sign of life at ≥28 weeks of gestation. We reported proportions of stillbirth by timing of maternal ART initiation (never initiated ART; before pregnancy; during 1st or 2nd trimester; during 3rd trimester or labor). We used logistic regression with cluster-based robust standard errors to account for clustering of women within health facilities to investigate the association between timing of ART initiation and stillbirth.

CROI 2019 291