Study Examines if Prenatal Exposure to Sedatives Is Linked to Psych Disorders

Prenatal exposure to sedative drugs used for anxiety and insomnia was not associated with an increased risk of psychiatric or neurodevelopmental conditions in children when accounting for familial factors, a large population-based cohort study from South Korea indicated.

An analysis of 3.8 million children initially suggested that prenatal exposure to benzodiazepines or Z-hypnotics was tied to a higher risk of psychiatric disorders compared with no exposure over a median 6.85-year follow-up, but this association disappeared in a sibling-controlled analysis (HR 0.99, 95% CI 0.94-1.04).

This suggested that familial or environmental factors — rather than the medication itself — likely drove the initial findings, noted Ju-Young Shin, PhD, of Sungkyunkwan University in Seoul, and colleagues.

Furthermore, there were no increased risks for 12 individual disorders, the authors reported in The BMJ.

Interestingly, there were reduced risks for schizophrenia spectrum disorder (HR 0.49, 95% CI 0.27-0.89) and eating disorders (HR 0.55, 95% CI 0.31-0.97) in the sibling analysis. However, Shin and team urged caution in interpreting these findings, noting that the small number of cases limited the study’s statistical power for these outcomes.

The 12 disorders included were identified via ICD-10 codes and encompassed a broad range of general psychiatric (also including substance use, spectrum disorder, mood disorder, anxiety disorder, personality disorder) and neurodevelopmental (intellectual disability, autism, attention deficit-hyperactivity disorder, and tic or behavioral disorder) conditions.

The findings offer reassurance about the use of these drugs during pregnancy, the authors said. Benzodiazepines and Z-hypnotics are used in roughly 2% of pregnancies worldwide, with prescriptions often increasing during the third trimester.

The safety of these medications has long been a subject of conflicting research. Some studies have found no risk for congenital malformations, while others have pointed to potential risks for miscarriages and other adverse pregnancy outcomes. The “neurodevelopmental consequences of prenatal exposure to benzodiazepines or Z-hypnotics remain uncertain,” Shin and colleagues wrote.

“Clinicians face a challenging dilemma when treating these conditions in pregnant women: untreated maternal psychiatric symptoms can adversely affect both mother and child, but drug treatments are used cautiously because of potential fetal risks,” they noted. “Given the high rate of unplanned pregnancies, a substantial potential exists for inadvertent exposure during the early stages of gestation.”

These findings will help to inform individualized risk-benefit discussions when considering sedative use in pregnancy, they added.

While the results are reassuring, Hyesung Lee, PhD, of Kangwon National University College of Medicine in Chuncheon-si, South Korea, and colleagues argued for “careful rather than uncritical interpretation” in an accompanying editorial.

“This study covers psychiatric diagnoses among liveborn offspring and cannot be taken as establishing overall perinatal safety — it does not inform risks of pregnancy loss, congenital malformations, or neonatal outcomes that also influence treatment decisions,” they wrote. “Reassurance should therefore remain outcome specific rather than global.”

They maintained that non-drug interventions, such as cognitive behavioral therapy, remain the first-line treatment for anxiety and insomnia during pregnancy. If medication is necessary, clinicians should remain mindful of potential signals around prolonged use or late-pregnancy exposure.

In secondary analyses, Shin and team found trends towards elevated hazard ratios associated with exposure during the second half of pregnancy and exposure of ≥30 days. However, neither association reached statistical significance.

For this study, the authors utilized South Korea’s National Health Information Database and included children born in 2010-2022 and followed them through 2023. Of 3.8 million children, 2.5% were exposed to benzodiazepines or Z-hypnotics during pregnancy, and the rest were unexposed. Among the unexposed group, 147,307 children were born to past users of benzodiazepines or Z-hypnotics.

The sibling-controlled analysis — designed to account for shared genetic and environmental factors — included sibling pairs in which one was exposed (n=33,370) and the other was not (n=37,574).

A higher proportion of exposed children had mothers with psychiatric conditions and more frequent healthcare encounters for psychiatric conditions.

By age 13, the cumulative incidence of any disorder was 19.2% in exposed children, 13.8% in unexposed children, and 16.5% in the past-user group.

The researchers relied on prescription data, which does not necessarily mean the medication was taken. Also, the follow-up period may have been too short to fully capture disorders that typically manifest in late adolescence, such as schizophrenia or certain personality disorders.