Pre-term delivery in mothers receiving HIV treatment: more research needed before changing guidelines

Carole Leach-Lemens
Published: 16 August 2011

HIV-infected pregnant women on a protease-inhibitor (PI) triple antiretroviral regimen were twice as likely to have a preterm delivery compared to those on a triple nucleoside reverse transcriptase inhibitor (NRTI) regimen, Kathleen M. Powis and colleagues reported in a secondary analysis from a randomised clinical trial (the MmaBana study) in Botswana published in the August 15 edition of the Journal of Infectious Diseases.

But in an accompanying editorial Athena P. Kourtis and Mary Glenn Fowler caution that further research is needed before making recommendations about ART during pregnancy as a consequence of the risk of pre-term delivery attached to any regimen.

Antiretroviral therapy in HIV-infected women during pregnancy and breastfeeding for their own health and for the prevention of HIV transmission to their infants is recognised as a highly effective public health strategy.

The parent study from which this analysis is taken demonstrated that both a PI-based regimen and NRTI-based regimen started late in pregnancy (26-34 weeks into the pregnancy) were highly efficacious in reducing maternal viral load at delivery and transmission rates to the infant.

However, the link between a PI-based regimen and the risks of preterm delivery (before 37 weeks of pregnancy) continues to be controversial. Some observational studies have suggested a link, while others have not.

These studies have been primarily in resource-rich settings. In resource-poor settings any increase in preterm delivery can have an adverse effect on infant survival. Identifying risk factors for preterm delivery and determining the contributing role of PI-based regimens is essential for the development of guidelines.

The authors looked at risk factors for preterm delivery in a cohort of HIV-infected pregnant women with CD4 cell counts under 200 cells/mm3 randomised to get either PI-based regimen (lopinavir/ritonavir/zidovudine/lamivudine) or NRTI-based regimen (abacavir/zidovudine/lamivudine) at 26-34 weeks of pregnancy in a clinical trial to prevent mother-to-child transmission (MmaBana).

They also looked at maternal weight gain in late pregnancy and infant disease and death rates up to six months of age. Poor weight gain is known to be a risk factor for preterm delivery. PI-based regimens are known to cause gastrointestinal problems and adverse metabolic effects. Maternal change in body mass index (BMI) one month after the start of ART was compared according to treatment arm and delivery (preterm compared to term).

Among women (267) in the PI arm preterm delivery rates were higher than among the 263 women in the NRTI arm, 21.4% compared to 11.8%, p=.003. The PI-based regimen was the most significant risk factor, (OR: 2.03, 95% CI: 1.26-3.27, p=0.004).

While median change in body weight in those in the PI arm was lower than in the NRTI arm one month after starting ART, there was no significant association with preterm delivery.

Serious infant illness, defined as hospitalisation and death up to six months of age, did not differ by maternal regimen. Kourtis and Fowler suggest that this is probably because most preterm deliveries would have been near term, adding that this would have been useful data to include.

While acknowledging the importance of this first randomised controlled trial to show a difference between ART and the risk of preterm delivery, Kourtis and Fowler caution that the issue is not simple. The study, they add, has the potential to raise more questions than answers.

They note, as do the authors, that the MmaBana study was not powered to answer the question of preterm delivery. As such the sample size may not be adequate. In Botswana the preterm delivery rate in the general population is 20%. As a consequence, Kourtis and Fowler note that to validate an increase in the risk of preterm delivery to the degree seen in the PI arm would require a much larger sample size.

Kourtis and Fowler also highlight the unexplained finding of the considerable lower rate of preterm delivery in the NRTI arm.

Noting the high number of twin deliveries in the NRTI arm compared to the PI arm, Kourtis and Fowler state that while the differences cannot be linked to the type of ART regimen “this serves as a reminder that chance differences can be present even in randomly allocated groups.”

Kourtis and Fowler question whether late start of ART could be linked to preterm delivery: approximately one third of those who delivered preterm had had ART for less than thirty days. The highest rate of preterm delivery was seen in those who started ART later regardless of treatment arm.

They suggest that data from the same setting comprising women with similar baseline CD4 cell counts but who had been on a simplified prophylactic regimen, for example, zidovudine and/or single-dose nevirapine, would be helpful for comparison.

They further question the timing of enrolment that limits the ability to look at very preterm deliveries (less than 32 weeks of pregnancy), which are associated with more severe infant outcomes.

Kourtis and Fowler note that several randomised studies have suggested there may be a link between PI use and preterm delivery. However, they cite the Kesho Bora study as the only other study that “sheds light on the issue.” Over 800 women in Burkina Faso, Kenya and South Africa with CD4 cell counts between 200 and 500 cells/mm3 were randomised to a PI triple ART regimen orzidovudine and single dose nevirapine. ARVs were started between 28 and 36 weeks of pregnancy, except for single-dose nevirapine. No difference was seen in the preterm delivery rates.

Kourtis and Fowler stress the relevance and rigour of the findings in a field where more information is needed.

PI-based ART regimens are a critical component of PMTCT and treatment programmes with proven benefits for maternal and child health.

Kourtis and Fowler conclude further studies and careful consideration of the questions raised are needed before recommendations can be made regarding ARVs for PMTCT and the risk of preterm delivery.


Powis KM et al. Increased risk of preterm delivery among HIV-infected women randomised to protease versus nucleoside reverse transcriptase inhibitor-based HAART during pregnancy. JID 204: 506-14, doi: 10.1093/infdis/jir307, 2011.

Kourtis AP and Fowler MG, Editorial commentary, Antiretroviral use during pregnancy and risk of preterm delivery: more questions than answers. JID 204: 493-4, doi: 10.1093/infdis/jir318, 2011.

Community Consensus Statement on Access to HIV Treatment and its Use for Prevention

Together, we can make it happen

We can end HIV soon if people have equal access to HIV drugs as treatment and as PrEP, and have free choice over whether to take them.

Launched today, the Community Consensus Statement is a basic set of principles aimed at making sure that happens.

The Community Consensus Statement is a joint initiative of AVAC, EATG, MSMGF, GNP+, HIV i-Base, the International HIV/AIDS Alliance, ITPC and NAM/aidsmap

This content was checked for accuracy at the time it was written. It may have been superseded by more recent developments. NAM recommends checking whether this is the most current information when making decisions that may affect your health.

NAM’s information is intended to support, rather than replace, consultation with a healthcare professional. Talk to your doctor or another member of your healthcare team for advice tailored to your situation.