Objective To compare the risks of pregnancy complications in women with repaired and unrepaired isolated ventricular septal defect (VSD).Design A retrospective multicentre study.Setting Tertiary centres in the Netherlands and Belgium.Methods Women were identified using two congenital heart disease registries. Eighty-eight women were identified who had experienced 202 pregnancies, including 46 miscarriages and nine terminations of pregnancy. Information on each completed pregnancy (n = 147; unrepaired VSD, n = 104; repaired VSD, n = 43) was obtained using medical records and telephone interviews. Data from the Generation R database (prospective cohort study; n = 9667) were used to determine the background risk (controls). Odds ratios and 95% CI were estimated using general estimation equation analysis adjusted for multiple pregnancies per woman, maternal age and parity status.Main outcome measures Adjusted odds ratios (AORs) for developing pregnancy complications in relation to corrective status.Results Pregnancies in women with an unrepaired VSD were associated with a higher risk of pre-eclampsia (AOR 4.59, 95% CI 2.01-10.5, P < 0.001) compared with controls. No differences were observed when comparing women with repaired VSD and controls. Pregnancies in women with repaired VSD were associated with a higher risk of premature labour (AOR 4.02, 95% CI 1.12-14.4, P = 0.03) and small-for-gestational-age (SGA) births (AOR 4.09, 95% CI 1.27-13.2, P = 0.02) compared with women with unrepaired VSD.Conclusions Women with unrepaired VSD are at increased risk of pre-eclampsia, which suggests that it is not a benign condition. In addition, women with repaired VSD are at increased risk of premature labour and SGA births compared with women with unrepaired VSD.