Triple ART reduces maternal death, premature and still birth, Mozambique / Malawi study shows

Women in Malawi and Mozambique on triple ART for three months or longer before giving birth were thirteen times less likely to die than women with no treatment, Maria Cristina Marazzi and colleagues reported in a retrospective cohort study of over 3000 women in the 31 Drug Resource Enhancement Against AIDS and Malnutrition (DREAM) Centres published in the advance online edition of AIDS.

The protective effect of triple ART was progressive and also reflected in rates of abortion, stillbirth and prematurity. The effect was present at all CD4 levels and corresponded to the length of treatment.

Length of treatment, however, had no effect on low birth weight.

The benefits of ART for the prevention of mother-to-child transmission (PMTCT) of HIV are well documented.

In resource-poor settings women are more likely to get ART at an advanced disease stage leading some studies to link treatment with adverse pregnancy outcomes. However, in early large-scale studies AIDS was shown to be the key variable for adverse pregnancy outcomes. Use of triple ART can often represent a surrogate marker for advanced disease, note the authors.

Since 2002 the DREAM programme has offered triple ART for PMTCT in 31 centres in ten African countries. The decline in HIV MTCT and a corresponding decline in maternal and infant mortality has been significant resulting in an infant HIV-free survival rate of over 92% at 12 months of age.

To determine the effect of ART on maternal mortality and pregnancy outcomes the authors undertook a detailed review of the medical records of 3273 HIV-positive pregnant women at the DREAM centres in Malawi and Mozambique from July 2005 until December 2009.

The overall maternal mortality rate was 1.2% (42 out of 3273) of which 7.4% (5 out of 68) were in women on no treatment before birth; and 0.7% (10 out of 1370) in women on treatment before birth for more than 90 days.

The longer the mother was on treatment before birth, the greater the decrease in mortality.

While women with a lower CD4 cell count had a higher rate of mortality, length of treatment before birth had a 70% protective effect for all.

The authors highlight the misconceptions that pregnant women with high CD4 cell counts are not at risk of dying and that treatment can be delayed.

Studies have shown that in sub-Saharan Africa HIV-infected pregnant women with high viral loads and not necessarily low CD4 cell counts are at high risk, especially immediately after having given birth. Additional factors prevalent in resource-poor settings including anaemia, poor nutrition, tuberculosis and malaria all add to the risk of death.  In this study anaemia was an independent factor for death.

The authors note their study clearly shows the benefits of ART at all CD4 levels to reduce maternal mortality. And, in their cohort 50% of perinatal transmissions took place in women with CD4 cell counts over 350 cells/mm³ countering the argument of deferred treatment.

The overall rate of abortion and stillbirth was 5.2%. The relative risk in women with no treatment was 4.24. Women with less than 30 days of treatment before birth were two to three times more likely to abort or have a stillbirth. While the risks were greater in women with advanced disease, the longer a women was on treatment the lower the rate of abortion/stillbirth.

The authors found the decreased risk of prematurity to be the most closely linked to length of treatment and statistically significant (p<0.001). The overall rate was 19.1% of which 70% was in 10 women with no treatment and 8.5% in 1330 women with at least three months of antenatal treatment.

Prematurity rates in women with less than one month of antenatal treatment, regardless of CD4 cell count, were especially high, 58% and 56%, respectively.  The authors note the absence of ultrasound data may be a limitation in confirming the age of the foetus. However, in resource-poor settings physical examination is the norm, they add.

Maternal toxicities because of ART were uncommon and did not differ significantly from similar studies.

The authors conclude that extended antenatal ART was highly protective and  “should be provided, whenever possible, to all HIV-infected pregnant women in resource-limited settings for reduction of maternal mortality and improved pregnancy outcomes.”