In a recent meta-analysis, Kourtis et al. concluded that overall antiretroviral therapy in pregnancy was not associated with premature delivery, although they reported that longer duration of treatment and regimens that included a protease inhibitor (PI) might increase the risk of prematurity . By contrast, we subsequently reported a statistically significant 1.5-fold increased risk of prematurity among women on highly active antiretroviral therapy (HAART) with or without a PI, compared with women on monotherapy or dual therapy . In another recent study, Schulte et al. reported a similar significant association, but limited to women on HAART with a PI and compared with women on dual therapy (adjusted odds ratio = 1.21) .
As Kourtis et al. noted, there was a significant degree of heterogeneity between the studies included in their meta-analysis, suggesting that bias or confounding could have been a problem. The studies they included were conducted over different time periods and in quite different populations, and sometimes had different inclusion criteria. It is questionable, therefore, whether a meta-analysis was appropriate.
Furthermore, there is a fundamental problem with the comparison of pregnancies that are exposed and unexposed to antiretroviral therapy. Since the majority of diagnosed HIV-infected women in resource-rich settings are now offered antiretroviral therapy in pregnancy (and very few decline), premature delivery itself is often the reason for not receiving treatment. This could lead to a failure to identify a real association between antiretroviral therapy and prematurity.
Differences in treatment classification could also be a problem; in the meta-analysis, regimens containing two antiretroviral drugs were combined with those containing three or more. Dual therapy is no longer commonly used, as it has lower potency than HAART (three or more drugs, generally from at least two different drug classes). The inclusion of studies where combination therapy consists mainly of dual therapy (e.g. Mandelbrot et al. 2001 ) rather than HAART could therefore attenuate any association with exposure to more potent combinations of drugs.
Finally, in several of the studies included in the meta-analysis [4–7], only crude prematurity rates were available, and estimates were therefore not adjusted for potential confounders such as maternal injecting drug use, HIV-related symptoms or prior preterm delivery. Since the studies were carried out over different time periods, the baseline characteristics of the women in these different studies and in the different treatment groups are likely to have varied substantially, and adjusted estimates are essential.
Investigation of the association between antiretroviral therapy and prematurity in any single observational study is fraught with difficulties. The potential for bias and/or confounding in such studies means that any meta-analysis needs to be interpreted with caution.
1. Kourtis AP, Schmid CH, Jamieson DJ, Lau J. Use of antiretroviral therapy in pregnant HIV-infected women and the risk of premature delivery: a meta-analysis. AIDS 2007; 21:607–615.
2. Townsend CL, Cortina-Borja M, Peckham CS, Tookey PA. Antiretroviral therapy and premature delivery in diagnosed HIV-infected women in the United Kingdom and Ireland. AIDS 2007; 21:1019–1026.
3. Schulte J, Dominguez K, Sukalac T, Bohannon B, Fowler MG. Declines in low birth weight and preterm birth among infants who were born to HIV-infected women during an era of increased use of maternal antiretroviral drugs: Pediatric Spectrum of HIV Disease, 1989–2004. Pediatrics 2007; 119:E900–E906.
4. Mandelbrot L, Landreau-Mascaro A, Rekacewicz C, Berrebi A, Benifla JL, Burgard M, et al
. Lamivudine-zidovudine combination for prevention of maternal-infant transmission of HIV-1. JAMA 2001; 285:2083–2093.
5. Cooper ER, Charurat M, Mofenson L, Hanson IC, Pitt J, Diaz C, et al
. Combination antiretroviral strategies for the treatment of pregnant HIV-1-infected women and prevention of perinatal HIV-1 transmission. J Acquir Immune Defic Syndr 2002; 29:484–494.
6. Bellon Cano JM, Sanchez-Ramon S, Ciria L, Leon JA, Gurbindo D, Fortuny C, et al
. The effects on infants of potent antiretroviral therapy during pregnancy: a report from Spain. Med Sci Monit 2004; 10:CR179–CR184.
7. Lorenzi P, Spicher VM, Laubereau B, Hirschel B, Kind C, Rudin C, et al
. Antiretroviral therapies in pregnancy: maternal, fetal and neonatal effects. Swiss HIV Cohort Study, the Swiss Collaborative HIV and Pregnancy Study, and the Swiss Neonatal HIV Study. AIDS 1998; 12:F241–F247.