Reproductive history and the risk of neonatal sepsis.

It was recently suggested that a previous abortion increases the risk of intrapartum infection in a following pregnancy. The authors hypothesised that abortion also could be associated with a higher risk of neonatal sepsis. A case-control study of neonatal sepsis was conducted using the Washington State Birth Registry. Cases of sepsis were selected among singleton livebirths during the period 1984-90, and compared with a control group for the occurrence of spontaneous or induced abortion in previous pregnancies. The risk estimates were calculated using a stratified analysis. After exclusion of primigravidae, the age-adjusted odds ratio (OR) was 1.68, with a 95% confidence interval (CI) 1.33, 2.11 for previous spontaneous abortion, and 2.20 (95% CI 1.73, 2.79) for induced abortion, compared with previous livebirth. After exclusion of nulliparous women, the OR decreased to 1.19 (95% CI 0.90, 1.58) for spontaneous abortion and 1.45 (95% CI 1.03, 2.04) for induced abortion. After controlling for the effect of parity, induced abortion is associated with an increased risk of neonatal sepsis in a subsequent pregnancy, but the association between spontaneous abortion and sepsis is small and non-significant. The authors suggest that the procedures involved in a therapeutic abortion might produce a latent, sub-clinical infection that persists until the next pregnancy, and is then transmitted to the newborn.