BACKGROUND: Little is known about the antecedents of microcephaly in early childhood among children born at extremely low gestational age. AIM: To identify some of the antecedents of microcephaly at age two years among children born before the 28th week of gestation.

STUDY DESIGN: Observational cohort study.

SUBJECTS: 1004 infants born before the 28th week of gestation.

OUTCOME MEASURES: Head circumference Z-scores of <-2 and >or=-2, <-1.

RESULTS: Risk of microcephaly and a less severely restricted head circumference decreased monotonically with increasing gestational age. After adjusting for gestational age and other potential confounders, the risk of microcephaly at age 2 years was increased if microcephaly was present at birth [odds ratio: 8.8 ((95% confidence interval: 3.7, 21)], alpha hemolytic Streptococci were recovered from the placenta parenchyma [2.9 (1.2, 6.9)], the child was a boy [2.8 (1.6, 4.9)], and the child's mother was not married [2.5 (1.5, 4.3)]. Antecedents associated not with microcephaly, but with a less extreme reduction in head circumference were recovery of Propionibacterium sp from the placenta parenchyma [2.9 (1.5, 5.5)], tobacco exposure [2.0 (1.4, 3.0)], and increased syncytial knots in the placenta [2.0 (1.2, 3.2)].

CONCLUSIONS: Although microcephaly at birth predicts a small head circumference at 2 years among children born much before term, pregnancy and maternal characteristics provide supplemental information about the risk of a small head circumference years later. Two findings appear to be novel. Tobacco exposure during pregnancy, and organisms recovered from the placenta predict reduced head circumference at age two years.