Pregnancy outcomes in infants exposed to maternal antiretrovirals in utero

Polly Clayden, HIV i-Base

Several posters at CROI 2010 showed findings from studies looking at outcomes in infants exposed to maternal antiretrovirals in utero.

Tenofovir exposure in DART

Enniie Chidziva and colleagues from the DART trial evaluated infants born to women mainly receiving tenofovir (TDF) based HAART in Uganda and Zimbabwe from 2004 to 2009. [1]

We have reported earlier results from the DART trial and pregnancy outcomes in previous issues of HTB. [2, 3]

During DART there were 223 live births with 6 infant deaths; 217 infants were alive two weeks after birth. Of these 129 (59%) were exposed to TDF in utero. Infants were evaluated in DART and in a separate follow up study.

The investigators reported that congenital abnormalities occurred in 7/217 (3%) of infants overall and 4/129 (3%) with TDF exposure. The abnormalities were: talipes 3 (2 with TDF exposure), cardiac 1, hydrocephalus 1(with TDF exposure), skin tag 1 (with TDF exposure) and undescended testes 1.

The majority 182/217 (84%) of infants were enrolled in the infant follow up study.  At their last visit they were a median age of 26 months (IQR 13–39); 69% were <12 months of age. The investigators noted that infants who were not enrolled in the follow up study were likely to have been born during the earlier part of the trial.

Prophylaxis was given to 152/182 (84%) of infants  (single-dose nevirapine 44%, AZT 18%, sd NVP+AZT 23%, other 15%). Of the 182 infants, 73 were ever breastfed for median 92 days (range 5-1186 days). Unadjusted HR for currently BF vs never BF 0.45 (95% CI 0.05–3.62) and for stopped vs never BF 0.7 (95% CI 0.19–2.57, p=0.59).

Of the 171 children tested, all were HIV-negative, 3 were lost to follow up and 8 died before testing. Fourteen children died at a median age of 9.4 months, giving 6% 12-month mortality. Of these, 8 had in utero TDF exposure, 6 were HIV-negative and 8 untested.

Only 4/386 creatinine and 7/310 phosphate measurements were abnormal, all were grade 1 and confined to 7 children of which 4 were exposed to TDF in utero (3 throughout pregnancy and one 61% of time in utero). There was no evidence of an effect of TDF in utero on growth after 48 weeks (p=0.31) and there were no bone fractures.

Additionally, an Italian cohort study reported by Alessandra Vigano and colleagues showed that exposure to TDF during the second and third trimesters of gestation, when bone formation occurs, does not impair bone mass and bone metabolism in HIV-negative children born to HIV-positive women. [2]

Mashi and Mma Bana

Two studies with data combined from the Mashi and Mma Bana PMTCT trials in Botswana looked at infant anaemia and birthweight respectively. [5, 6]

We have reported on both trials in previous issues of HTB. [7, 8, 9, 10, 11]

Scott Dryden-Peterson and colleagues compared the incidence of severe and life-threatening (grade 3 or 4, DAIDS 2004 toxicity tables) anaemia in breastfed (BF) infants exposed to HAART in utero with BF and formula fed (FF) infants exposed to AZT in utero in these trials.

Endpoints were incidence of first severe anaemia from birth to 7 months and the analyses used scheduled measurements of first born uninfected infants.

A total of 1788 infants met the inclusion criteria (1096 Mashi, 692 Mma Bana). Of this group, 743 were exposed to maternal HAART (AZT+3TC+LPV/r or AZT+3TC+NVP), one month of post natal AZT and BF (categorised as HAART+BF; 517 to in utero AZT, 6 months of post-natal AZT, and breastfeeding (AZT+BF); and 528 infants to in utero AZT, 1 month of post-natal AZT, and formula feeding (AZT+FF).

The investigators reported there were 126 infants with severe anaemia by 7 months with a cumulative incidence of 12.6 % (n=89) in HAART+BF, 5.4 % (n=26) in AZT+BF, and 2.3 % (n=110 in AZT+FF.

Severe anaemia occurred more frequently among HAART+BF infants than either AZT+BF infants (OR 2.51, 95% CI 1.59-3.95), or AZT+FF infants (OR 6.11, 95% CI 3.2-11.6), both p<0.0001.

In multivariate analysis, predictors of severe anaemia (AOR; 95%CI) were:  HAART+BF (2.4; 95%CI 1.5–3.8 and 5.7; 95%CI 3.0–10.7) compared to AZT+BF and AZT+FF, respectively; low birth weight <2.5 kg (2.4; 95%CI 1.5–3.9); and male sex (1.5; 95%CI 1.0–2.2). Maternal CD4, VL, haemaglobin, education, income, study site and gestation at delivery were not significantly associated with severe anaemia.

Birthweight <2.5kg occurred in 103 (13.97%), 43 (8.4%) and 31 (5.9%) of infants in the HAART+BF, AZT+BF and AZT+FF groups respectively.

The investigators reported no differences in infant anaemia according to maternal HAART regimen. Microcytosis or hypochromia occurred in 39/89 (43.8%) infants in the HAART+BF group, with severe anaemia.

Patients with severe anaemia were treated with iron/multivitamin supplementation, and 10 infants (7.9%) received transfusions. Of those who improved to grade <3 with iron/multivitamin supplementation alone this occured in <30 days in 43 (34%), 31–90 days in 50 (39.7%) and >90 days in 18 (14.3%) infants. Three (2.4) infants died while grade 3–4 and 2 (1.6) were lost to follow up.

The investigators concluded: “The clinical implication of this finding requires further investigation to ensure that the established benefits of using HAART for MTCT prevention are maximised for all infants.”

The same research group looked at the impact of HAART and short course AZT on longitudinal growth in a subset of Mashi and Mma Bana infants. They noted that HAART for PMTCT may lead to lower birth weight but longitudinal effects of in utero exposure on infant growth have not been previously reported.

In this analysis, Kathleen Powis and colleagues evaluated breastfed, HIV-uninfected infants born >37 weeks and exposed in utero to at least two weeks of either HAART or AZT. Infants in the HAART-exposed group received postnatal AZT for 1 month. Infants in the AZT-exposed group received 6 months of AZT-prophylaxis during breastfeeding.

The investigators calculated gender-based weight-for-age, length-for-age, and weight-for-length z-scores were using WHO Child Growth Standards. They compared mean z-scores using the Student’s t- test and analysis of response profiles.

This analysis included 437 AZT-exposed infants from Mashi, and 592 HAART-exposed infants from Mma Bana.

Median maternal baseline CD4 counts were 393 and 337 cells/mm3 (p=<0.001) and median viral load 4.34 and 4.19 log copies/mL (p=0.04) for Mashi and Mma
Bana women, respectively  Demographics were similar between cohorts.

The median time of in utero AZT exposure was 5.7 weeks (range 2.0–10.9 weeks), and median in utero HAART exposure was 12.1 weeks (range 2.6–22.3 weeks).

Median birth weights were 3.1kg in AZT-exposed and 3.0 kg in HAART-exposed (p<0.001). HAART exposed infants had significantly lower mean weight-for-age, length-for-age, and weight-for-length z-scores (p<0.001, p = 0.02, and p= 0.007, respectively).

However, the investigators reported that by 3 months of age the infants’ median weight was no longer different by exposure group, and their weight remained
similar to 6 months. Mean weight-for-age differed over time by exposure group (p <0.001). Length-for-age remained lower in the HAART-exposed
group to 6 months of age, but weight-for-length improved significantly over time compared with AZT-exposed infants (p<0.001).

They noted that the proportions of infants with z-scores >2 standard deviations below the mean were not different between exposure groups.

These early developmental comparisons are useful and longer-term comparisons are planned. The investigators wrote: “The early correction of birth weight differences among HAART exposed infants is reassuring for programmes utilising maternal HAART for treatment and PMTCT.”


Both DART and the Botswana group continue to provide urgently needed and excellent data on maternal /infant health and outcomes.


Unless stated otherwise, all references are to the Programme and Abstracts of the 17th Conference on Retroviruses and Opportunistic Infections, 16-19 February 2010, San Francisco.

  1. Chidziva E et al. Outcomes in infants born to HIV-infected mothers receiving long-term ART in the DART trial, 2004 to 2009. Poster abstract 924.
  2. Pregnancy outcomes in the DART trial. HTB, April 2007.
  3. Pregnancy rates and outcomes among women in the DART trial. HTB, October 2009.
  4. Vigano A et al. Exposure during gestation to HAART, including tenofovir, does not impair bone status and metabolism in HIV-positive children born to HIV-positive mothers. Poster abstract 926.
  5. Dryden-Peterson S et al. Increased risk of severe but reversible anemia following perinatal exposure to maternal HAART in infants in Botswana. Poster abstract 927.
  6. Powis K et al. Impact of HAART and short-course zidovudine on longitudinal growth of HIV-exposed uninfected breastfed infants, Botswana. Poster abstract 928.
  7. Mashi study – late breakers on breastfeeding. HTB, April 2005.
  8. Risk factors for breastfeeding transmission, HTB, April 2008.
  9. Breast milk, HIV suppression and decreased mother to child transmission. HTB, October 2005.
  10. Response to nevirapine containing HAART following single dose nevirapine for PMTCT. HTB, February 2007.
  11. Reducing HIV transmission during breastfeeding. HTB, August 09.

Links to other websites are current at date of posting but not maintained.