Flagship mother-to-child transmission prevention programme in Lusaka only 30% effective

A free city-wide programme to prevent mother-to-child transmission (MTCT) in Lusaka, Zambia is only effectively reaching 30% of mother-child pairs, according to the results of a paper published in the September 1^st issue of the journal Clinical Infectious Diseases. The most striking findings were that anonymously identified HIV-positive women were significantly less likely to accept an HIV test than their HIV-negative counterparts, and that almost one-in-three of the women who knew they were HIV-positive (32%) did not actually take the single dose of nevirapine (Viramune) during labour.

Each year, more than 10,000 babies are born to HIV-infected mothers in the Zambian capital of Lusaka. For the past five years, Dr Jeffrey Stringer, from the University of Alabama, Birmingham, and colleagues from the United States and Zambia have been working with the Zambian Government to provide a universal MTCT prevention programme. The programme, which was put into place in November 2001, is considered a flagship of MTCT prevention for high prevalence urban sub-Saharan African centres.

All women who receive antenatal care in the 24 clinics of the Lusaka District are offered HIV testing, and those who are antibody positive receive post-test counselling and a single dose nevirapine tablet which they are told to take during labour. When the women present at one of the ten clinics with delivery facilities, as well as the University Teaching Hospital, they are also asked if they need a replacement nevirapine tablet prior to delivery. After delivery, the new-born is given nevirapine by clinic or hospital staff.

Lack of resources for viral load testing of infants means that novel methods needed to be used to ascertain the effectiveness of this programme, the results of which serve as a surrogate for the number of neonatal HIV infections prevented.

The investigators collected anonymous umbilical cord blood specimens from the discarded placentas of all live public sector deliveries during the 12-week period, June 7^th to August 31^st 2003. Since both the mother's HIV antibodies are found in the umbilical cord, and nevirapine crosses the placenta within minutes of its rapid oral absorption, umbilical cord blood specimens were regarded by the investigators as objective and sensitive measures of both HIV antibody status and nevirapine adherence. Direct observation was used to determine infant nevirapine dosing.

Over the 12 week period 10,384 women gave birth to live infants in Lusaka's public sector facilities. Cord blood specimens were obtained from 10,194 (98%). No differences were seen between the mothers from whom specimens were or were not obtained in terms of age, number pregnancies and HIV testing. Of these 10,194 women, 8787 (86%) obtained their antenatal care from one of the 24 Lusaka District clinics, and data from these women make up the surveillance data analysed by the researchers.

The women were aged a median of 23 (range 12-50) years old, with a median number of two (range 1-13) pregnancies. HIV prevalence is high in this population, and cord blood testing found that 2257 (26%) of these women were HIV-positive.

Of the 8787 women in total, 7204 (82%) had been offered HIV testing in antenatal care. HIV prevalence was found to be similar among those who had been offered testing (1854 of 7204, 26%) compared with those who had not (403 of 1583, 25%; p=0.9).

However, only 5149 (59% of the total, and 71% of those who had been offered testing) actually accepted an HIV test. The researchers found a significant difference in HIV status between those who accepted and those who did not accept an HIV test. HIV prevalence was significantly lower amongst those women who accepted an HIV test (1250 of 5149, 24%) compared with those women who refused to test (604 of 2055, 29%; p

Of the 5149 women who accepted HIV testing, 5129 (99%) received a result. Cord blood tests revealed that 1246 (24%) of the women were HIV-positive. However, only 1112 (89%) of the anonymously-tested HIV-positive women received a positive test result in the clinic.

The remaining 134 comprised false-negative antenatal tests, false-positive cord blood tests, clerical errors at the clinic, and seroconversions between the first test and delivery. In their discussion the investigators estimated that approximately 10% of these were due to errors in the surveillance exercise, "and that the remaining majority fall into a category of de facto seroconversions," that could have happened due to lab or clerical error, or actual recent infection.

Of the 1112 cord blood HIV-positive women who were given an HIV-positive result in the clinic, only 751 (68%) had nevirapine detected in the cord blood. This suggests that 32% of women were non-adherent to nevirapine during labour. The investigators found that adherent women had a shorter average time between HIV testing than non-adherent women (96±46 vs. 104±46 days; p=0.006), but no differences were seen between age or number of pregnancies.

Of the 751 infants born to HIV-positive mothers who took nevirapine during labour, 675 (90%) received directly observed nevirapine by a health care worker.

With only 675 mother-infant pairs correctly identified as needing MTCT prevention, and receiving both doses of nevirapine out of 2257 infants born to all HIV-positive mothers in this surveillance population, population coverage for the Lusaka MTCT prevention programme was found to be just 30%.

Reasons why 70% of mother-child pairs did not receive effective care include:

• Mother not offered HIV testing during antenatal care (n=403, 18%)
• Mother refused HIV testing (n=604, 27%)
• Mother did not receive test result and/or nevirapine (n=4,
• Mother's test result recorded as negative (n=134, 6%)
• Mother did not take nevirapine during labour (n=361, 16%)
• Infant not dosed with nevirapine (n=76, 3%)

"Our study," write the investigators, "demonstrates that failures can and do occur at each step along this cascade [of events], resulting in reduced coverage and diminished programme effectiveness."

During their discussion, the investigators point out that while debates continue regarding the most effective antiretroviral regimen for the prevention of MTCT, their study found that "in absolute terms 403 HIV-infected mothers are lost due to failure to offer testing, compared with only 361 due to non-adherence...[and therefore] resources may be better spent in maximizing coverage of simple and moderately effective interventions such as nevirapine, graduating later to more efficacious drug regimens."

The investigators point out that they have already begun to attempt to address some of the shortcomings revealed by this study. These include "clinic-level incentives to ensure all women are offered services, community efforts to encourage HIV testing and nevirapine adherence, and quality assurance programmes for the small clinical labs that are performing the HIV rapid testing."