To evaluate the association of either propylthiouracil or methimazole treatment for hyperthyroidism during pregnancy with congenital malformations, relevant studies were identified by searching Medline, PubMed, the Cochrane Library and EMBASE. We intended to include randomized controlled trials, but no such trials were identified. Thus, we included cohort studies and case-control studies in this meta-analysis. A total of 7 studies were included in the meta-analyses. The results revealed an increased risk of birth defects among the group of pregnant women with hyperthyroidism treated with methimazole compared with the control group (odds ratio 1.76, 95% confidence interval 1.47-2.10) or the non-exposed group (odds ratio 1.71, 95% confidence interval 1.39-2.10). A maternal shift between methimazole and propylthiouracil was associated with an increased odds ratio of birth defects (odds ratio 1.88, 95% confidence interval 1.27-2.77). An equal risk of birth defects was observed between the group of pregnant women with hyperthyroidism treated with propylthiouracil and the non-exposed group (odds ratio 1.18, 95% confidence interval 0.97-1.42). There was only a slight trend towards an increased risk of congenital malformations in infants whose mothers were treated with propylthiouracil compared with in infants whose mothers were healthy controls (odds ratio 1.29, 95% confidence interval 1.07-1.55). The children of women receiving methimazole treatment showed an increased risk of adverse fetal outcomes relative to those of mothers receiving propylthiouracil treatment. We found that propylthiouracil was a safer choice for treating pregnant women with hyperthyroidism according to the risk of birth defects but that a shift between methimazole and propylthiouracil failed to provide protection against birth defects.