Z-drug use in the first trimester and risk of congenital malformations in the U.S.

Background: Sleep disorders are common in pregnancy and often treated by non-benzodiazepine sedative hypnotics (also known as “z-drugs”). However, there is limited evidence on the fetal safety of z-drugs.

Objectives: To evaluate the association between first-trimester exposure to z-drugs and congenital malformations.

Methods: This U.S.-based cohort study utilized data of pregnancies linked to live-born infants from nationwide Medicaid data (2000-2018) and the MarketScan Research Database (2003-2020). We defined major congenital malformations using validated algorithms in maternal and infant claims and compared the risks of any major congenital malformation and organ-specific malformations in pregnancies with ≥1 dispensing of z-drugs (zaleplon, eszopiclone, or zolpidem) in the first trimester to that in those unexposed. Relative risks (RR) and 95% confidence intervals (CI) were estimated with log-binomial regression models, and propensity score fine stratification weights were used in the adjusted analysis to control for potential confounders, including demographics, comorbidities, healthcare utilization, and concomitant medication use. We utilized indirect standardization to combine the RRs from the two data sources. To minimize the potential for exposure misclassification, a sensitivity analysis with z-drug exposure defined as ≥2 prescriptions in the first trimester was conducted.

Results: We identified 11,652 of 2,506,106 pregnancies in Medicaid and 10,925 of 1,784,143 pregnancies in MarketScan with first-trimester z-drug exposure. Compared to those unexposed, pregnancies exposed to z-drugs in the first trimester were older and had a higher prevalence of psychiatric conditions and greater use of concomitant medications. The absolute risk of any major congenital malformation was 4.0% (95% CI: 3.7%-4.4%) for exposed and 3.3% (95% CI: 3.2%-3.3%) for unexposed pregnancies in Medicaid; the risks for exposed and unexposed pregnancies in MarketScan were 4.2% (95% CI: 3.8%-4.5%) and 3.8% (95% CI: 3.7%-3.8%), respectively. The crude pooled RR for any major malformation was 1.18 (95% CI: 1.10-1.25); the adjusted pooled RR was 1.01 (95% CI: 0.95-1.08). Exposure to z-drugs did not appear to increase the risk of organ-specific malformations except for neural tube defects (adjusted pooled RR=1.69, 95% CI: 1.02-2.80). Results were consistent in the sensitivity analysis focused on filling ≥2 prescriptions.

Conclusions: Our findings suggest that z-drug exposure in the first trimester does not elevate the risk of congenital malformations overall. Additional evidence is needed to evaluate the signal observed for neural tube defects.