Objective: The present study aimed to investigate the adverse effects of the antithyroid drugs propylthiouracil (PTU) and methimazole (MMI)/carbimazole (CMZ) in treating hyperthyroidism. Methods: Qualitative analysis was performed for studies identified in a literature search up to April 20, 2019, and 30 studies were selected for meta-analysis. The study designs included case-control, randomized controlled, and retrospective cohort. Patients were in four age groups: childhood, gestating mothers, older adults, and other ages, and all were receiving PTU or MMI/CMZ. Adverse reactions to MMI/CMZ and PTU were evaluated and compared. Results: Odds of liver function injury were higher in the PTU group than in the MMI/CMZ group (odds ratio [OR], 2.40; 95% confidence interval [CI], 1.16 to 4.96; P = .02). Odds of elevated transaminase were much higher in the PTU group than in the MMI/CMZ group (OR, 3.96; 95% CI, 2.49 to 6.28; P<.00001). No significant between-group differences were found in odds of elevated bilirubin, agranulocytosis, rash, or urticaria; incidence of other adverse events; or in children. Odds of birth defects during the first trimester of pregnancy were higher in the MMI/CMZ group than in the PTU group (OR, 1.29; 95% CI, 1.09 to 1.53; P = .003). Conclusion: The impact of PTU on liver injury and transaminase levels is greater than that of MMI/CMZ, but no significant between-group differences are found in the drugs' effects on bilirubin, agranulocytosis and rash, urticaria, or in children. In treating pregnancy-related hyperthyroidism, PTU should be used in the first trimester and MMI reserved for use in late pregnancy. Abbreviations: ALT = alanine aminotransferase; ATD = antithyroid drug; CI = confidence interval; CMZ = carbimazole; GD = Graves disease; MMI = methimazole; MTU = methylthiouracil; NOS = Newcastle-Ottawa Scale; OR = odds ratio; PTU = propylthiouracil; RAI = radioactive iodine.
