BACKGROUND: Neonatal herpes is a condition with high morbidity and mortality. The greatest risk occurs when the mother acquires herpes simplex virus (HSV) towards the end of pregnancy. A study from Seattle has suggested that the risk of acquisition of HSV during pregnancy was 3.7%. In Australia, HSV-2 infection is less common in pregnant women than in the United States. Consequently we conducted a study to establish HSV seroprevalence and the rate of HSV seroconversion in this population. METHODS: The study was conducted at Westmead Hospital, Sydney, between June 1995 and April 1998. Women completed a questionnaire covering risk factors for the acquisition of genital herpes. A serum sample during pregnancy and a specimen of cord blood were obtained and tested for antibodies to HSV-2 using a type specific indirect enzyme linked immunosorbent assay (ELISA). Equivocal results were resolved by western blot. A subset of the paired sera was tested for antibodies to HSV-1. The data were analysed using SPSS. RESULTS: 326 of the 2616 (12.5%) women were HSV-2 seropositive. Three women (0.15%) acquired HSV-2 infection during pregnancy. None of the three babies of these mothers developed neonatal herpes. 416 maternal cord pairs were tested for HSV-1 antibodies and 330 (79.3%) were positive. No HSV-1 seroconversions occurred. CONCLUSIONS: In this population, HSV acquisition was uncommon (0.34% per year) and neonatal herpes was rare. A cost effective analysis suggested that type specific serology to screen pregnant women and their partners in low prevalence communities was not cost effective.