Objective To assess the association between maternal cytomegalovirus (CMV) antibodies in mid-pregnancy and preeclampsia.Design Nested case-control study.Setting Pregnancies registered in the Norwegian Mother and Child Cohort Study (MoBa): a large population-based pregnancy cohort (1999)(2000)(2001)(2002)(2003)(2004)(2005)(2006).Sample A cohort of 1500 women with pre-eclampsia and 1000 healthy pregnant women.Methods Plasma samples and pregnancy-related information were provided by the MoBa. Antibody status (CMV IgG and CMV IgM) and levels (CMV IgG) at 17-18 weeks of gestation were determined by enzyme-linked immunosorbent assay (ELISA).Main outcome measure A diagnosis of pre-eclampsia, as defined in the Medical Birth Registry of Norway.Results There was no evidence of an effect of CMV IgG seropositivity on the likelihood of developing pre-eclampsia, and CMV IgG antibody levels among women who were seropositive did not differ between groups. Adjusted for maternal age, parity and smoking, the odds ratio for pre-eclampsia in women seropositive for CMV IgG was 0.89 (95% CI 0.74-1.05; P = 0.17). The proportions of women who were seropositive for IgM did not differ between women with pre-eclampsia and women who were healthy (P = 0.98). Among nulliparous women, the proportion of women who were seropositive for CMV IgG was slightly lower among women with pre-eclampsia (53.5%) than among healthy women (59.8%) (P = 0.03). Subgroup analyses were performed for women with early or late onset pre-eclampsia, with preterm delivery and/or with neonates that were small for gestational age, but antibody status did not differ between pre-eclampsia subtypes and controls.Conclusions The presence of maternal antibodies to CMV was not associated with pre-eclampsia in our study. The results suggest that CMV infection is unlikely to be a major cause of preeclampsia.