Abstract
The objective of this study was to determine mother to child HIV transmission rates at different time points in a breastfeeding cohort enrolled in a single dose nevirapine program in Harare, Zimbabwe. Between 2002–2004, 434 HIV-positive mothers and their infants were recruited and followed up from delivery to 15 months. Infant blood specimens were collected for HIV testing at these time points. The majority of the patients (78%) received single dose nevirapine. The overall HIV transmission rate was 21.8% (17.8–25.8). Receiving single dose nevirapine was protective against HIV vertical transmission although statistically insignificant (relative risk: 0.76; 95% CI: 0.49–1.19). Breastfeeding was not found to be associated with HIV vertical transmission (P = 0.612). In this resource-limited setting, HIV transmission rates are high. Efforts to use more efficacious regimens to arrest HIV vertical transmission are required.
Introduction
Each year, HIV infects about 640,000 children worldwide, mainly due to mother-to-child transmission during pregnancy, at the time of delivery, and after birth through breastfeeding. 1 About 40% of transmissions occur during the breastfeeding period. 2 The World Health Organization (WHO) recommends the avoidance of all breastfeeding by HIV-positive mothers if replacement feeding is affordable, feasible, acceptable, sustainable and safe (AFASS). Otherwise exclusive breastfeeding should be followed by early cessation and replacement feeding if it is AFASS. 3
In resource-limited settings, breastfeeding could undermine the gains of single dose nevirapine in reducing mother-to-child transmission of HIV. The rate of postnatal HIV transmission among infants who were exposed to HIV, but uninfected at four weeks or later, was estimated in a meta-analysis from nine African countries at 8.9 infections/100 child-years of any breastfeeding remaining constant over time. 4,5 The rate of postnatal transmission among infants who were exclusively breastfed for up to three months in a Zimbabwean study was 5.1/100 child-years. 2 Few studies have examined the question of timing and the magnitude of HIV transmission during the breastfeeding periods as this requires relatively large sample sizes and repeated assessment of the HIV status of infants during the first two years of life.
At this time approximately 30% of the Zimbabwean women of childbearing age were HIV-infected. It was, therefore, important to determine HIV transmission rates at different time points in this breastfeeding population after the introduction of single dose nevirapine in a national programme. 6
Methods
Setting
The study was conducted at three primary maternal child health clinics in peri-urban areas around the capital city of Harare, Zimbabwe (namely, Epworth, Seke North and St Mary's).
Design
This was a prospective cohort study of HIV-positive mothers and children which was followed up for 15 months.
Between 2002 and 2003 pregnant women were enrolled into the Better Health for the African Mother and Child study from 36 weeks of gestation, after informed consent had been obtained. Pre- and post-test HIV counselling was offered as part of the national Prevention of Mother-to-Child Transmission of HIV (PMTCT) programme. Baseline characteristics included sociodemographic information, physical and gynaecological examination and full blood counts. All HIV-positive mothers were offered single dose nevirapine during labour (200 mg) and their infants were given single dose nevirapine (2 mg/kg) within 72 hours of delivery according to the national guidelines. 7
Follow-up
From the initial cohort of 479 HIV-positive mothers, 434 mother-baby pairs were eligible for this analysis. Forty-five mothers did not have delivery information and were excluded. Of the 434 initial mother-baby pairs, 281 infants were tested from delivery to 15 months. Blood samples of infants were collected at delivery, 6 weeks, 4 months, 9 months and 15 months for HIV testing. Feeding history was obtained from the caregivers. All HIV-infected babies were referred to the appropriate opportunistic infections clinic for treatment according to the national guidelines.
Laboratory methods
Infant blood samples were processed and tested for HIV with deoxy-ribonucleic acid polymerase chain reaction (DNA PCR) 1.5 (Roche Diagnostics, Indianapolis, USA), if aged less than 15 months of age, and rapid HIV antibody tests, Determine (Abbott Diagnostics, Illinois, USA), and Oraquick (Abbott Diagnostics) if 15 months or older. 8
Analysis of data
The data was entered and analysed using Statistical Package for Social Scientists (SPSS version 12.0, Chicago, USA) and Stata (version 9, College Station, Texas, USA). Simple proportions for categorical variables and means (standard deviations) for continuous variables were used for the description of baseline characteristics. Proportions were used for time point and cumulative HIV transmission rates. For the denominator for cumulative HIV transmission we considered infants who had undertaken at least one HIV test (408). Relative risk and their 95% confidence interval (CI) were calculated for the effects of nevirapine and breastfeeding on HIV transmission rates and P values of less than 0.05 were considered statistically significant.
The study was approved by the Ethical Review Committee in Norway and the Medical Research Council of Zimbabwe.
Results
Figure 1 shows the follow-up of infants over 15 months. Eighty-nine infants became HIV-infected; 192 were not HIV-infected by 15 months; and 26 died without any testing. Out of 434 mother-infant pairs, 65% were followed. The baseline characteristics of the mothers and infants are summarized in Table 1 which shows the decline in the number of infants who were breastfeed from 6 weeks to 9 months.
Follow-up of infants born to HIV-positive mothers
Baseline characteristics of HIV positive women and their infants
Intrauterine HIV transmission
As shown in Figure 1 and Table 2, 373 had delivery HIV results and 16 were HIV-positive; hence estimated intrauterine HIV transmission was 4.3 (2.2–6.4%).
Time point and cumulative HIV transmission rates up to 15 months
Intrapartum and early breastfeeding HIV transmission
Of the 329 infants that were HIV-negative at delivery with an HIV result at 6 weeks, five became HIV-positive. This gives an estimated intrapartum transmission rate of 1.5 (0.2–2.8%) as shown in Figure 1. Seventy-three percent (241) of 329 mother-baby pairs received single dose nevirapine. These included four of the five infants who were HIV-infected at 6 weeks. The other one had no information on nevirapine use. The 49 that failed to get nevirapine were not HIV-infected at 6 weeks.
Overall transmission
Table 2 shows time point transmission rates with a decline at 9 months after delivery and also shows cumulative HIV transmission rates. The overall transmission is 21.8% in the patients that were followed. Although not statistically significant, mother-child pairs who received nevirapine were less likely to transmit HIV overall: 20.7% versus 27.1% (relative risk: 0.76, 95% CI: 0.49–1.19). The mother-baby pairs who did not receive nevirapine were more likely to deliver outside the study sites (P < 0.001). There was no baseline difference between those who received nevirapine and those who did not, except that mothers who received nevirapine were more likely to be married (P = 0.008). There was no statistical difference in HIV transmission rates between infants who breastfed and those who did not (P = 0.612).
Discussion
The observed intrauterine HIV transmission rate of 4.3% is rather low in this cohort from Harare, Zimbabwe. At this time, CD4 counts and viral loads were not offered to women on the national programme because of cost. Another local study quoted a figure of 9.4% (CI: 7.7–11.5%) before the use of antiretroviral drugs. 9 The intrapartum transmission rate of 1.5% is even lower and could be due to the effectiveness of single dose nevirapine; however, the nevirapine efficacy at reducing intrapartum and early breastfeeding HIV transmission was difficult to demonstrate as the children of mother-infant pairs who did not receive nevirapine remained uninfected at six weeks of age (Figure 1). Nevirapine is effective at reducing intrapartum HIV transmission rates, although a study demonstrated its effectiveness at 18 months as well which was not evident in our study. 10 Prior to the use of single dose nevirapine in resource-limited settings, Zijena et al. reported the HIV intrapartum and early breastfeeding transmission rates of 16% using similar intervals and that, in the first 6 months of life, intrauterine and intrapartum HIV vertical transmission contribute more than three quarters of overall transmission. 9 Our study demonstrates the change in the proportions of intrapartum and early breastfeeding HIV transmission rates.
At 6 weeks our transmission rate of 6.1% is comparable to those in South Africa between 2002 and 2004, where the operational effectiveness of single dose nevirapine was evaluated. Rates of early HIV transmission, at between 3–4 weeks, ranged from 8.6 to 13.7%. 11 However, Quaghebeur et al. reported a HIV transmission rate of 18.1% in Mombasa at 14 weeks, in a breastfeeding population using nevirapine. 12 At 17 weeks our rate was 6.9%. Tanzania reported a HIV infection risk at 4 months of 3.8% (1.6–6.1%) which increased to 17.9% (11.2–42.5%) by 24 months in breastfeeding infants. 13 These results are similar to our findings. 13
Patients were counselled to cease breastfeeding at 6 months in this programme but 62.5% of mothers interviewed at 9 months after delivery continued to breastfeed. However, breastfeeding was not significantly associated with HIV transmission. Cumulative HIV transmission rates could have been underestimated because not all infants were tested at every visit. Performing longitudinal studies in developing areas is very challenging. One major problem is that many mothers and their infants are lost to follow-up. Some of the factors that affect loss to follow-up rates may also affect mother-child transmission risk. For a child to be brought back for evaluation in a clinical study, the parents and the child must be alive, remain in the same geographical area, be able to take time from their occupation and to find transportation. 14 The gradual loss to follow-up is a limitation of our findings although we maintained a follow-up rate of 65%.
Conclusion
Our study shows that, with the single dose nevirapine regimen, postnatal transmission is the major contributor to vertical HIV transmission in our setting. With this regimen, overall transmission is still high. It is a matter of urgency that better methods of reducing vertical transmission are used in resource-limited settings where replacement feeding is not AFASS.