Purpose: Antidepressant use during pregnancy has been increasing in recent years. We evaluated whether in utero exposure to antidepressants increased the risk of childhood cancer.
Methods: This population-based cohort study using national registers in Denmark comprised 915 128 liveborn singletons during 1998-2012. We categorised children into three mutually exclusive exposure groups according to maternal redemption of an antidepressant prescription from 2 years before pregnancy until delivery of the index child: Unexposed (N = 863 033), prior user (use before but not during pregnancy) (N = 30 607), and use during pregnancy (N = 21 488). The children were followed from birth until first diagnosis of cancer, death, emigration, or December 31, 2012, whichever came first. The children were followed maximum 14.9 years and contributed to 6.9 × 106 person-years at risk. We estimated hazard ratios (HRs) of cancer using Cox regression with 95% confidence intervals (CIs).
Results: In total, 1298 (0.1%) children were diagnosed with cancer. Antidepressant use during pregnancy was not associated with a significantly increased risk of childhood cancer in general; the HR was 1.03 (95% CI, 0.63-1.68), compared to children born by mothers who discontinued antidepressant use prior to pregnancy. The association between in utero exposure to antidepressants and childhood cancer did not depend on type or duration of antidepressant use. There was no strong evidence indicating a higher risk of leukaemia or nervous system tumours among children exposed to antidepressants in utero.
Conclusion: Antidepressant use during pregnancy was not significantly associated with childhood cancer in general nor with leukaemia or nervous system tumours in specific.
Keywords: antidepressant; childhood cancer; depression; epidemiology; in utero; pharmacoepidemiology; register-based.
Copyright © 2017 John Wiley & Sons, Ltd.