DOI: 10.1002/pds.70448 ISSN: 1053-8569

Prenatal Exposure to Modafinil and the Risk of Major Congenital Anomalies and Other Adverse Neonatal and Pediatric Outcomes

Melinda Kanoun, Naïm Bouazza, Jean‐Marc Treluyer, Mathis Collier

ABSTRACT

Objective

To assess the risk of major congenital anomalies (MCAs) and other adverse outcomes following prenatal exposure to modafinil.

Methods

This retrospective cohort study used the French national healthcare database (SNDS) to include singleton children born between January 2009 and September 2024 of women aged 15–55 years. Prenatal exposure to psychostimulants was defined as maternal dispensation during pregnancy. Children prenatally exposed to modafinil were compared with two control groups: children prenatally exposed to methylphenidate and children prenatally unexposed to any psychostimulant matched to the exposed group using propensity‐score matching (PSM). Outcomes included MCAs, neurodevelopmental disorders (NDs), and specialized consultations. Analyses included descriptive statistics, survival analysis, regression models, and dose–response analyses.

Results

Of 10 955 766 included children, 865 were prenatally exposed to modafinil and 950 to methylphenidate. Exposure to modafinil during the first trimester of pregnancy was statistically associated with increased risk of MCA compared to methylphenidate, although the confidence interval was wide and close to the null (aRR = 1.77 [1.01–3.07]). Comparison with PS‐matched unexposed population indicated a possible modest increase in risk, albeit with a wide confidence interval crossing the null (aRR = 1.23 [0.76–1.99]), showing no clear association. Occurrence of NDs (aHR = 0.96 [0.56–1.65]), and specialized consultations (aHR = 1.08 [0.91–1.28]) were similar in the modafinil and PSM unexposed groups but were higher in the methylphenidate group (NDs: aHR = 0.48 [0.29–0.80]; specialized consultations: aHR = 0.71 [0.59–0.85]).

Conclusion

This study shows no increase in neurodevelopmental risk, while a moderate MCA risk cannot be excluded.

More from our Archive