Late Postnatal Transmission of HIV-1 in Breast-Fed Children: An Individual Patient Data Meta-Analysis

Document Type



Obstetrics and Gynaecology (East Africa)


Background. We analyzed individual patient data to determine the contribution of late postnatal transmission to the overall risk of mother-to-child transmission of human immunodeficiency virus type 1 (HIV-1) and the timing and determinants of late postnatal transmission.

Methods. Eligible trials were conducted where breast-feeding was common; included ⩾2 HIV-1 tests by 3 months, and, if follow-up continued, ⩾2 tests at 3-12 months; and regularly assessed infant-feeding modality Data on children born before January 2000 were analyzed.

Results. Of 4085 children from 9 trials (breast-fed singletons for whom HIV-1 testing was performed), 993 (24%) were definitively infected (placebo arms, 25.9%; treatment arms, 23.4%; =.08). Of 539 children with known timing of infection, 225 (42%) had late postnatal transmission. Late postnatal transmission occurred throughout breast-feeding. The estimated hazard function for time to late postnatal transmission was roughly constant. The cumulative probability of late postnatal transmission at 18 months was 9.3%. The overall risk of late postnatal transmission was 8.9 transmissions/100 child-years of breast-feeding and was significantly higher with lower maternal CD4+ cell counts and male sex.

Conclusions. Late postnatal transmission contributes substantially to overall mother-to-child transmission of HIV-1. The risk of late postnatal transmission is generally constant throughout breast-feeding, and late postnatal transmission is associated with a lower maternal CD4+ cell count and male sex. Biological and cultural mechanisms underlying the association between sex and late postnatal transmission should be further investigated. Interventions to decrease transmission of HIV-1 through breast-feeding are urgently needed.

Mother-to-child transmission of HIV-1 can occur in utero, during delivery, and postnatally through breastfeeding [1]. In settings where breast-feeding is the norm, a significant proportion of mother-to-child transmission of HIV-1 occurs through breast-feeding [2]. However, more information is needed with regard to the risk and timing of transmission through breast-feeding and potential risk factors for such transmission.

To inform the development of appropriate interventions to prevent transmission through breast-feeding in areas of the world where complete avoidance of breastfeeding is not, generally, feasible, we conducted an individual patient data meta-analysis of transmission of HIV-1 through breast-feeding. The objectives were the following: to estimate the contribution of late postnatal transmission of HIV-1 to the overall risk of mother-to-child transmission of HIV-1 and to characterize the timing and determinants of late postnatal transmission.


This work was published before the author joined Aga Khan University.

Publication (Name of Journal)

The Journal of Infectious Diseases