Meta-analysis of sex difference in testosterone levels in umbilical cord blood

This meta-analysis reviewed published literature comparing human male and female umbilical cord total testosterone (T) levels. A total of 18 studies using 1,229 samples from 602 male and 627 female newborns were analysed using the RevMan 5 statistical package. Analysis using the inverse variance method based on a random-effects model revealed significantly higher cord T in boys than girls at a moderate effect size (Hedges' g = 0.57). There was significant heterogeneity between the 18 studies, although the five studies using direct assays showed no heterogeneity. For studies using extraction and chromatography, those that combined T from arterial and venous cord blood found a larger sex difference than those using only cord venous samples (Hedges' g = 0.94 vs 0.32); this suggests umbilical cord venous T is of maternal/placental origin and arterial T is of fetal origin. The wide range of T values between studies suggests high cross-reactivity in the assay methods reviewed.