Cotrimoxazole prophylaxis significantly improves birth outcomes in women with HIV according to an analysis of a mother-to-child HIV transmission (MTCT) study in Lusaka, Zambia. The serendipitous finding was observed after cotrimoxazole prophylaxis began to be offered routinely in Zambia to HIV-infected women with advanced disease.
The study was presented by Dr Jan Walter of Columbia University, at the Thirteenth Conference on Retroviruses and Opportunistic Infections in Denver, and can be downloaded as an mp3 file here.
Background
In Africa, HIV-infected mothers, particularly women with advanced disease, have a high risk of adverse birth outcomes, including still births, miscarriages, preterm deliveries, low birth weight babies, and infant mortality. Although the underlying causes are unclear, they may include a higher risk of maternal infections that are harmful to the infants.
Cotrimoxazole prophylaxis, which prevents a variety of bacterial and protozoal infections, began to be routinely offered as part of HIV care in November 2003 for HIV-infected women with CD4 cell counts below 200 copies/mm3 — including pregnant women after 14 weeks of gestation. This happened to occur right in the middle of a large MTCT study that Dr Walter and colleagues were conducting in Lusaka, Zambia.
The MTCT study evaluated the effect of early cessation of breastfeeding on HIV transmission. The study enrolled over 1400 women with birth outcome data between May 2001 and April 2005 — more than enough patients to explore whether the introduction of cotrimoxazole had any impact on birth outcomes. Standard care offered to women in this study included malaria prophylaxis, multivitamins and single dose nevirapine during delivery.
Women treated with cotrimoxazole were compared to those who gave birth before the introduction of cotrimoxazole prophylaxis. Women who received antiretroviral therapy in the study were excluded from the analysis and logistic regression was performed to adjust for possible changes in other risk factors that may have occurred over this time period.
Of the women involved in the study, 330 had CD4 cell counts below 200 copies/mm3. Out of these, 97 delivered after cotrimoxazole was introduced. Those with obvious risk factors for adverse birth outcomes such as multiple births (twins or more), maternal deaths, etc., were excluded from the analysis.
Results
Cotrimoxazole prophylaxis is associated with several improvements in pregnancy outcomes:
- The percentage of preterm births (≤34 weeks estimated gestation) declined from 31% to 18% (p = 0.04).
- Clinical chorioamnionitis declined from 6% to 0 (p = 0.04).
- Birth weights increased — the mean birth weight was on average 100g greater although this finding did not quite reach statistical significance (p = 0.15).
- There was a reduction in neonatal mortality (infant death ≤28 days) from 9% to 0 (p = 0.01).
After adjustment for other risk factors that might have changed over the period (including poverty, maternal education, hemoglobin, low weight, and at what time during pregnancy women enrolled into the study), delivery after the introduction of the cotrimoxazole program was associated with about half the risk of preterm birth (OR=0.5 95% CI: 0.2, 0.9). One factor which did differ significantly before and after November 2003 was food insecurity — due to local drought and famine — but this was not found to significantly affect outcomes.
As a second control group for temporal differences other than cotrimoxazole prophylaxis that could have affected the outcomes, births in women who did not receive cotrimoxazole because of higher CD4 cells counts were compared before and after November 2003 — but no significant difference was observed. The percentage with preterm birth went from 21% to 22% and neonatal mortality was 3% and 2.4%. It is also interesting to note that these outcomes are not much different from those in women with lower CD4 cell counts on cotrimoxazole.
"Treatment was most beneficial for the ones most severely affected by HIV," said Dr. Walter. To illustrate, he noted that the increase in birth weight was most pronounced among the women with the lowest CD4 cell counts. Usually, birth weights in women with CD4 cell counts below 100 copies mm3 are significantly lower than in women with higher CD4 cell counts (above 200 copies mm>sup>3). However, after the introduction of cotrimoxazole prophylaxis, there was no significant difference in birth weight by CD4 cell count.
Dr Walter offered a couple of possible explanations of cotrimoxazole’s effect. "These benefits might arise from a reduction in preterm birth, which is a strong risk factor of neonatal mortality and could explain up to 80% of neonatal deaths in our cohort, or might arise from direct effect of by the reduction of maternal infections and the infant exposure to these during delivery or the early neonatal period," he said.
However, the study was not prospectively controlled and could not explore the potential for cotrimoxazole-related side effects. While the results show that, on the whole cotrimoxazole appears to be beneficial in this context, in pregnant women, cotrimoxazole can reduce levels of folic acid or folate (a B vitamin). Folic acid deficiency during the first trimester of pregnancy is associated with neural tube birth defects, so during the first trimester of pregnancy, cotrimoxazole should either be avoided entirely or co-administered with additional folate.
Walter J et al. Cotrimoxazole prophylaxis and adverse birth outcomes among HIV-infected women in Lusaka, Zambia. Thirteenth Conference on Retroviruses and Opportunistic Infections, Denver, abstract 126, 2006.