What is the risk of major congenital abnormalities among women on antiretroviral therapy?

More than 1.5 million HIV-infected women become pregnant yearly. An increasing percentage start antiretroviral therapy (ART) prior to pregnancy, resulting in a rapid rise in the number of first-trimester fetal ART exposures [1]. Longer duration of ART in pregnancy decreases risk of Mother to child transmission of HIV [2], but there are also potential teratogenic effects which include congenital malformations, stillbirth, growth restriction, and preterm birth. Therefore, studies evaluating the safety of antiretroviral medications (ARVs) in pregnancy have been designated as a research priority. In this respect, we applaud the by Berard et al. [3] study ‘Antiretroviral combination use during pregnancy and the risk of major congenital abnormalities’ published by Berard et al. (AIDS, vol 31:2267–2277) which observed no increased risk for congenital malformations in infants born to mothers prescribed first-trimester ART. The strengths of the study included the fact that it is population-based and collected many potential confounders. However, there are methodological limitations that compromise its ability to evaluate potential associations between ART and congenital malformations, notably an inappropriate comparison group (all women, regardless of HIV infection status), inappropriate measure of exposure (all ART regimens), and insufficient sample size of ART-exposed (n = 198). Studies of drug safety and efficacy are normally conducted in the study population for which the drug is indicated. However, Berard et al.[3] compare malformations among primarily HIV-infected women exposed to ART (90% of exposed group) to HIV-uninfected women (>99.9% of control group). The authors acknowledge significant imbalances in most background characteristics between the two groups, including several which are important risk factors for congenital malformations (such as comorbidities, coinfections, and substance use). A better comparison would have been to HIV-infected women without first-trimester ART exposure, but the study population included too few of these (n = 169) to make such a comparison stand alone. With low power for detecting associations with either overall or individual ARV exposures when restricting to HIV-infected women, the study provides little additional contribution to prior studies which have included thousands of ARV-exposed women [4,5]. With respect to the exposure measure utilized by Berard et al.[3], one must recognize that ART is not one drug, but typically three drugs prescribed together. Mechanisms of action vary across ARV drug classes and even within a single class, suggesting different biologic activity by individual drugs. Distinguishing the effect of individual ARVs requires careful study design and statistical analysis [6]. Despite the complexity of combination ARVs, the Berard study grouped all ART regimens together when evaluating congenital malformations, which lacks biologic plausibility and could mask a true association with any individual ARV. In addition, there is insufficient power to evaluate malformations individually or by organ system while adjusting for multiple potential confounding covariates – factors that could reasonably lead to both ART use and adverse birth outcomes. The adjustment for 10 covariates (despite <8 exposed cases for most outcomes) resulted in extreme changes between the unadjusted and adjusted relative risks, changing the direction of effect from adverse to protective for ART exposure. As a result of the low number of overall abnormalities among the exposed, risks of individual abnormalities presented in Table 3 of the study by Berard et al. [3] are suspect. In particular, the highlighted finding of increased risk of abnormalities of the small intestine with ART exposure was based on only one ART-exposed case. Given the sample size and rarity of outcome (0.04% of births), this result is more likely due to chance than a true increased risk. One final concern with the methodology used in this analysis is that the overall rate of malformations in the population (8.6%) is among the highest ever reported, more than two to four times higher than large studies from the United States and Canada [7–9]. This is particularly striking because this study excluded minor anomalies, anomalies in stillbirths, births to women taking known teratogens in the first trimester, and chromosomal causes of anomalies. Berard et al.[3] cite the founder effect in the Quebec province, with a small genetic pool, as the likely reason for this high rate of malformations. However, an alternative is that their methodology overestimates malformation rates. Inflated rates often arise from using International Classification of Diseases codes from medical records without expert review of physical exam findings [10], including reported anomalies that are not true abnormalities, or failing to exclude clinically insignificant variants of malformations such as cardiac septal defects which resolve spontaneously without intervention. Due to the concerns noted above, drawing conclusions about the teratogenicity of ART from this study is not possible. Further studies are needed to determine whether newer ARV drugs, or specific ARV combinations, are teratogens. Acknowledgements Conflicts of interest There are no conflicts of interest.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.