HIV Articles  
2 Articles on Pregnancy in HIV: HAART & Congential Abnormalities; Mitochondrial Toxicity in Uninfected Newborns  
  Does Highly Active Antiretroviral Therapy Increase the Risk of Congenital Abnormalities in HIV-Infected Women?
JAIDS Journal of Acquired Immune Deficiency Syndromes
September 2005
Patel, Deven MSc; Thorne, Claire PhD; Fiore, Simona MSc; Newell, Marie-Louise PhD; European Collaborative Study, prepared by
To the Editor:
With increasing numbers of HIV-infected pregnant women receiving highly active antiretroviral therapy (HAART),1 concerns have been raised over the possible teratogenic effects related to exposure in early pregnancy. We have previously reported a 1.4% prevalence of congenital abnormalities in uninfected infants exposed to antiretroviral therapy (ART; mainly monotherapy and/or dual therapy), which is similar to that seen in those not exposed.2 It is unclear whether risk of congenital abnormalities is increased by first-trimester exposure or by use of HAART, however. In this analysis, we further assess the additional risk of antenatal use of HAART during pregnancy by investigating the prevalence of congenital abnormalities by earliest exposure to and type of treatment regimen.
The European Collaborative Study (ECS) is an ongoing cohort study in which infants of HIV-infected women identified during pregnancy are followed according to standard clinical and laboratory protocols.2,3 Information collected at enrollment and during pregnancy includes maternal ART use and timing of initiation, laboratory tests, illicit injecting drug use, and other sociodemographic characteristics. Gestational age was assessed by ultrasound and reported to the nearest completed week. Information on the child's health and development was collected according to standard laboratory and clinical protocols.2,3 Infants were assessed at birth by a senior obstetrician and/or senior pediatrician; additional perinatal assessments were made by experienced pediatricians, who were also responsible for reports on further pediatric follow-up visits.
To assess the outcome of pregnancies by type of treatment regimen, ART exposures were grouped into 4 categories: nucleoside analogue reverse transcriptase inhibitor(s) (NRTI) only; combinations of NRTI and protease inhibitor (PI); NRTI and nonnucleoside reverse transcriptase inhibitor (NNRTI); or a combination of NRTI, NNRTI, and PI. HAART was defined here as regimens including 3 or more drugs. Because organogenesis occurs in the first trimester, rates of congenital abnormalities among infants with second- and/or third-trimester exposures to ART were grouped together and compared with those among infants with first-trimester exposures. The first trimester of pregnancy was defined as up to and including 12 gestational weeks. Data entry was carried out using MS Access 2000, and analyses were performed using R 1.9.0 statistical software (R Foundation for Statistical Computing, Vienna, Austria, 2003) and SAS statistical software (version 8.02; SAS Institute, Cary, NC).
This analysis included data on 3740 mother-child pairs enrolled between 1986 and December 2003, of whom 1973 infants were exposed to ART in utero, including 602 exposed to HAART. Of the 1973 women who received antenatal ART, 789 (40%) received ART in the first trimester of pregnancy and had all initiated treatment before conception. Median maternal age at delivery was 28 years (range: 10-45 years). Most women were white (72%, 2700 of 3740 women), 21% (781 of 3740 women) were black (mainly from sub-Saharan Africa), and 7% (256 of 3740 women) were from other ethnic groups. Thirteen percent (493 of 3740 women) used illicit drugs during pregnancy. CD4 cell counts were routinely assessed since 1992 and were available for 2042 (55%) women, with a median CD4 count at delivery of 420 cells/mL (range: 0-2350 cells/mL). Median gestational age at delivery was 38 weeks (range: 22-43 weeks), and median birth weight was 2940 g (range: 500-5190 g).
Congenital abnormalities were recorded in 1.5% (55 of 3740 children), 31 of whom had been exposed to ART in utero and 14 in the first trimester of pregnancy. The abnormalities included ear malformations (2 children), cleft palate (2 children), ventricular septal defect (6 children), atrial septal defect (2 children), blood-clotting disorder (1 child), hydrocele in both testes (1 child), polydactyly (3 children), Down syndrome (3 children), polycystic kidney (3 children), hydronephrosis (2 children), esophageal atresia (2 children), ileostoma and enteritis (1 child), gastric perforation (1 child), hydrocephalus (1 child, who died 4 days after birth), cataract (1 child), and situs inversus transposition of arteries (1 child). Congenital abnormalities in the 24 unexposed children included cleft palate (1 child), ventricular septal defect (4 children), cardiac rhabdomyoma (1 child), tetralogy of Fallot (1 child), polydactyly (2 children), syndactyly (1 child), polycystic kidney (2 children), hydronephrosis (2 children), esophageal atresia (3 children), microcephalus (1 child), hydrocephalus (1 child), hip dysplasia (2 children), cataract (1 child), situs inversus transposition of arteries (1 child), and diaphragmatic hernia (1 child).
No increase in any particular abnormality with the use of ART or HAART during pregnancy was apparent. Among all 3740 infants, 1.4% of those not exposed to antenatal ART (24 of 1767 children) had a congenital abnormality versus 1.6% of those exposed (31 of 1973 children; x2 = 0.16; Yates corrected, P = 0.69). There was also no discernible pattern between the type and timing of ART during pregnancy (see Table 1). The prevalence of congenital abnormalities in children exposed to any antenatal ART in the first trimester (14 of 789 children [1.8%], 95% confidence interval [CI]: 0.97-3.0) was similar to that of children exposed in the second or third trimester (17 of 1184 children [1.4%], 95% CI: 0.84-2.3) (x2 = 0.17; P = 0.68). Among the 789 infants with first-trimester exposure, there was no difference in prevalence of congenital abnormalities among infants exposed to HAART (11 of 546 children [2.0%], 95% CI: 1.0-3.6) compared with those exposed to monotherapy and/or dual therapy (3 of 243 children [1.2%], 95% CI: 0.26-3.6) (Fisher exact test, P = 0.57).
In multi-variable logistic regression analysis involving 3471 mother-child pairs with information on all variables included, the presence of congenital abnormalities was not associated with the use of monotherapy or dual therapy or HAART during pregnancy after adjusting for maternal age at delivery and injecting drug use during pregnancy (adjusted odds ratios [AORs] = 1.05 [95% CI: 0.49-2.2; P = 0.91] and 1.22 [95% CI: 0.57-2.6; P = 0.61] for monotherapy or dual therapy and HAART, respectively, with no antenatal ART as the reference category). Rerunning the logistic regression, with time period instead of ART exposure (grouped as <1994 [pre-ART prophylaxis era], 1994-1996 [monotherapy or dual therapy era], and >1996 [HAART era]) similarly showed a lack of association with the prevalence of congenital abnormalities (data not shown). These findings confirm the lack of evidence for an increase in the prevalence of congenital abnormalities as a result of HAART in this population. In a subanalysis among 1810 mother-child pairs with antenatal ART use adjusting for maternal age at delivery, there was no increased risk of congenital abnormalities associated with first-trimester exposure to ART compared with later exposures (AOR = 1.33, [95% CI: 0.64-2.8; P = 0.44]). None of the mothers of children with congenital abnormalities used injecting drugs during pregnancy, precluding adjustment for this variable here.
Although efavirenz should be avoided in pregnant women and in women planning a pregnancy because of potential teratogenicity,4,5 19 women in this cohort received efavirenz-containing HAART regimens at the time they became pregnant and continued taking the drug for a median of 40 days into their pregnancy (range: 24-106 days). No congenital abnormalities (0%, 95% CI: 0-17.6) were reported in this group.
In our study we found a similar pattern and prevalence of congenital abnormalities among infants exposed to antenatal ART and those who were not, and this was also true for exposure to HAART. Furthermore, there was no evidence to suggest that exposure to first-trimester ART increases the risk of congenital abnormalities. As such, our findings are consistent with those of the Antiretroviral Pregnancy Registry.5,6 As a long-running birth-cohort study, however, we benefit from a large number of mother-child pairs (exposed and nonexposed) and are not subject to the potential for ascertainment and reporting bias that may limit interpretation of registry data. Although a small risk cannot be excluded, such data should reassure the increasing number of HIV-infected women becoming pregnant while taking HAART. Further monitoring and research are necessary, however, particularly to assess the teratogenic risk of use of combinations of antiretroviral and other drugs at the time of conception or in early pregnancy.
Long-Term Mitochondrial Toxicity in HIV-Uninfected Infants Born to HIV-Infected Mothers
JAIDS Journal of Acquired Immune Deficiency Syndromes June 2003; 33(2):175-183
  icon paper stack View Older Articles   Back to Top