A New Piece in the Puzzle of Antiretroviral Therapy in Pregnancy and Preterm Delivery Risk

Mother-to-child transmission (MTCT) is the most significant source of human immunodeficiency virus (HIV) infection in children worldwide, and recent years have seen some tremendous successes with respect to its prevention. The Joint United Nations Programme on HIV/ AIDS and partners have called for the ‘‘virtual elimination’’ of MTCT—that is, the reduction of the number of new pediatric HIV infections by 90% between 2009 and 2015 and reduction of MTCT rates to ,5% worldwide [1]. In resource-rich settings, high antenatal coverage, good access to and uptake of interventions for prevention of MTCT (PMTCT), and widespread use of combination antiretroviral therapy (ART) have contributed to the very low MTCT rates now reported (,1%–2%) [2–5]. Concerns about preterm delivery in HIV-positive women were first raised .20 years ago, with the initial concerns relating to the influence of HIV and its associated immunosuppression on pregnancy outcomes. A positive association between use of combination ART and preterm delivery was first reported in Europe in 2000 [6]. Subsequently, a number of other studies reporting significant associations of 1.5to 3.5-fold increased risk between antenatal combination ART and preterm delivery have added to the published literature on ART and pregnancy outcomes [7–14]. These studies varied in their approach (for example, with different ART reference groups used in the analyses) and in their ability to adjust for confounders; of note, some found the association to be limited to protease inhibitor (PI)–based combination ART. Last year, the first evidence from Africa of an association between preterm delivery and type of combination ART was published: a secondary analysis from the Mma Bana trial in Botswana demonstrated a higher preterm delivery rate in the group of pregnant women randomized to PI-containing triple antiretroviral regimens compared with those in the triple nucleoside reverse transcriptase inhibitor arm, with an odds ratio (OR) of 2.03 (95% confidence interval [CI], 1.26–3.27) [15]. However, other studies, predominantly carried out in the United States and South America, have not found an association between preterm delivery and combination ART use in pregnancy [16–19]. Kourtis and colleagues [20] performed a meta-analysis of studies carried out up to 2006, reporting no association between ART and increased risk of preterm delivery overall (OR, 1.01; 95% CI, .8–1.3); however, subgroup analyses showed an OR of 1.35 (95% CI, 1.1–1.7) for PIcontaining versus non-PI–containing ART and an OR of 1.7 (95% CI, 1.1–2.7) for ART started prepregnancy or in the first trimester versus later in pregnancy. Possible explanations for the conflicting evidence base include differences in case ascertainment, choice of ART reference group, unmeasured confounding, bias in indication for treatment, statistical power, and underlying population differences. Preterm delivery rates vary by world region, Europe having the lowest rate at an estimated 6% and Africa the highest at around 12% [21]. Events leading to preterm birth in the general population remain incompletely understood, and multiple pathways have been implicated, including inflammation and/or infection, maternal and/or fetal stress, abnormal uterine distension, and bleeding. There are a number of different factors associated with preterm delivery risk, including socioeconomic factors, ethnicity, maternal smoking, illicit drug use, maternal age, multiple gestation, maternal body mass index (BMI), previous preterm delivery, intrauterine infection, and bacterial vaginosis. It is estimated that in the general population, 40%–45% of preterm births are due to spontaneous preterm labor, Received 11 January 2012; accepted 17 January 2012; electronically published 28 March 2012. Correspondence: Claire Thorne, PhD, MRC Centre of Epidemiology for Child Health, UCL Institute of Child Health, University College London, 30 Guilford St, London WC1N 1EH, UK (claire.thorne@ucl.ac.uk). Clinical Infectious Diseases 2012;54(9):1361–3 The Author 2012. Published by Oxford University Press on behalf of the Infectious Diseases Society of America. All rights reserved. For Permissions, please e-mail: journals.permissions@ oup.com. DOI: 10.1093/cid/cis202