Mother-to-child transmission of HIV-1

Abstract

After reviewing the evidence on the relation of vertical transmission of HIV to stage of infection in the mother, I developed a stochastic model of transmission in which the probability of transmission per week is proportional to the virus load in the mother. The virus load in different stages of the infection is measured by viral RNA levels or tissue culture infectious virus levels in plasma. The constant of proportionality is assumed to be different for transmission during pregnancy, during parturition, and during breast-feeding. Using data on transmission from mothers who are in the primary stage of infection, I estimated the constant of proportionality and calculated the probability of transmission during pregnancy as a function of the time pregnancy starts in relation to the stage of the infection. For breast-feeding, I calculated the conditional probability of transmission by breast-feeding for 20 weeks, dependent on the infant escaping infection during pregnancy and parturition. As might be expected, the probabilities of transmission are highest if the mother is in the primary stage of infection or in late stages of the disease and is quite low when the mother is in the asymptomatic stage of the infection.

PIP: Based on a review of the literature on maternal-child HIV transmission, the author presents a stochastic model in which the probability of HIV transmission per week during pregnancy, parturition, and breast feeding is conceptualized as proportional to the mother's viral load at that stage. The model combines data on culturable virus levels and RNA copies in plasma as measures of maternal viral load. Through use of transmission data on women in the primary stage of infection, the probability of HIV transmission during pregnancy was calculated as a function of the time pregnancy starts in relation to the stage of infection. Also calculated was the conditional probability of transmission by breast feeding for 20 weeks, dependent on the infant escaping infection during pregnancy and delivery. Predicted transmission probabilities were quite low during the asymptomatic stage of maternal infection and highest when the mother was in the primary stage of infection or the late stages of the disease. RNA data emerged as a better basis for the calculations than culturable virus. The considerable variation by country in reported rates of maternal-child HIV transmission could be explained by changes in the mix of women in the primary, asymptomatic, and later stages of infection.