TOPLINE

Paternal use of valproate during sperm development (3 months before conception) was not associated with an increased risk for neurodevelopmental disorders (NDDs) or congenital malformations compared with no valproate use in the offspring, a large Taiwanese cohort study shows.

METHODOLOGY

• A nationwide birth cohort study analyzed data from Taiwan's National Health Insurance Research Database, including 2,583,503 individuals born between 2001 and 2016 and followed until 2021.
• 6987 individuals were exposed to paternal use of any antiseizure medication (ASM) during spermatogenesis and 1701 were exposed particularly to paternal use of valproate.
• The study comprised 3 analyses: population-based analysis adjusted for confounding factors, population-based analysis restricted to children (n = 31,984) with epilepsy-affected fathers, and analysis in which siblings with paternal ASM use during spermatogenesis were compared with unexposed siblings (548 pairs).
• Primary outcomes included NDDs (autism spectrum disorder [ASD], attention-deficit/hyperactivity disorder [ADHD], tic disorders, and intellectual disability) identified using medical records, with congenital malformations (defined as those diagnosed within 90 days after birth).
• Covariates included offspring sex, birth year, parental age, maternal lifestyle factors during pregnancy, delivery characteristics, socioeconomic status, and parental history of psychiatric disorders and epilepsy.

TAKEAWAY

• In adjusted population-based analysis and analysis restricted to offspring with epilepsy-affected fathers, paternal use of ASMs or valproate during spermatogenesis was not associated with a risk for NDDs including ASD, ADHD, tic disorders, or intellectual disability in the offspring.
• Paternal valproate use was not associated with congenital malformations in the offspring in any of the analyses. Paternal use of any ASM (adjusted odds ratio [aOR], 1.24; P = .024) or phenytoin (aOR, 1.39; P = .025) was associated with congenital malformations in the offspring in analysis restricted to epilepsy-affected fathers.
• In sibling-comparison analysis, paternal use of ASMs or valproate was not associated with NDDs or congenital malformations.
• When parental psychiatric history was excluded, a significant association was observed between paternal ASM use and ADHD (adjusted hazard ratio [aHR], 1.23; 95% CI, 1.06-1.43), suggesting potential confounding by familial psychiatric factors. In population analysis limited to paternal valproate monotherapy, a significant association was found with ADHD in offspring (aHR, 1.28; P = .004), but no association was found in the analysis restricted to children with epilepsy-affected fathers or in the sibling comparison analysis.

IN PRACTICE

"These findings may contribute to ongoing debates about the management of valproate in men of reproductive age," the investigators wrote.

"A pragmatic clinical position would inform male patients of reproductive age that a possible risk has been raised, but that the strongest available evidence does not currently support it; it would not mandate switching of patients who are well controlled on valproate; it would, where reasonable alternatives exist and the patient is in agreement, consider other agents first; and it would reserve broader changes in practice for the moment more definitive data emerge," the author of an accompanying editorial wrote.

SOURCE

The study was led by Yen-Chen Anne Feng, Institute of Health Data Analytics and Statistics, College of Public Health, National Taiwan University, Taipei. The editorial was authored by P. Emanuela Voinescu, Massachusetts General Brigham–Brigham and Women's Hospital, Harvard Medical School, Boston. Both articles were published online on August 19 in Neurology.

LIMITATIONS

The limited follow-up duration may not have captured all NDD diagnoses, particularly those presenting later in life. Formal neuropsychological testing was unavailable, potentially missing mild neurodevelopmental deficits. The study did not examine dose-dependent effects of ASMs or adjust for family history of intellectual disability. Sibling analyses did not consider time-varying and non-shared familial confounding factors and may have been subject to carryover effects from families with affected older children.

DISCLOSURES

The study was funded by the National Science and Technology Council, Chang Gung Memorial Hospital, and the Ministry of Education. The editorial received no specific funding. The study investigators and author of the editorial reported no relevant conflicts of interest.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.