Poster #33 - The association between maternal report of antiepileptic drug use during pregnancy and offspring neurodevelopmental problems

• In Event: 3-001 - Poster Session 2 during Breakfast

Abstract

Despite clinical guidelines warning against antiepileptic drug (AED) use during pregnancy (Harden et al., 2009), there are critical questions that remain unanswered surrounding the association between prenatal exposure to AEDs and offspring neurodevelopmental problems (NDPs), including attention deficit/hyperactivity disorder (ADHD) and autism spectrum disorders (ASD). Research has linked AEDs, particularly valproate, to NDPs and documented biological plausibility of a causal effect (Nicolai, 2008; Veiby et al., 2013). For instance, models of prenatal and postnatal exposure to AEDs in rats have resulted in neural and behavioral anomalies consistent with neurodevelopmental dysfunction (Veiby et al., 2013), and studies of human exposure have suggested that at least some AEDs cross the placental barrier (Froescher et al., 1984). However, researchers have been unable to 1) estimate associations between several specific drugs and offspring outcomes with adequate precision and 2) rule out alternative explanations for associations (e.g., confounding by indication; Coe et al., 2009; Jacoby et al., 2005). The aim of the current study was to examine associations using 1) a large cohort that enabled estimation of associations with specific drugs and 2) adjustment for many measured plausible confounding factors that previous studies have not accounted for (e.g., maternal bipolar diagnosis).

We analyzed data from a cohort of N=1,669,228 individuals born in Sweden from 1996-2009 and followed through 2012. Each individual in Sweden is assigned a unique registration number, which allowed us to track people over time in several linked national registers (Ludvigsson et al, 2009). Thus, we were able to combine information from medical birth records, specialist outpatient diagnoses, prescription drug dispensations, and parental sociodemographic data. Four indices of maternal self-reported AED use during the first trimester of pregnancy (e.g., any AED, lamotrigine, carbamazepine, and valproate use) were used as predictors of offspring diagnosis of ADHD and ASD. We used Cox proportional hazard regression models to obtain parameter estimates for these predictors, adjusting for several important maternal, paternal, and pregnancy characteristics.

Estimates of association (and 95% confidence intervals) presented in Table 1 are consistent with prior research (Nicolai, 2008). Notably, we observed strong associations between maternal report of AED use in first trimester and offspring NDPs, particularly ASD. Further, covariate adjustment reduced, but did not eliminate observed associations. While this is the largest study that we are aware of to examine these associations, future research must look at specific indications for use (e.g., epilepsy, bipolar diagnosis) and explore the role of additional confounding factors.

Authors

• Kelsey Wiggs, Indiana University
  Presenting Author

• Martin Rickert, Indiana University

• Ayesha Sujan, Indiana University, Bloomington

• Paul Lichtenstein, Karolinska Institute

• Henrik Larsson, Örebro University

• Sara Oberg, Karolinska Institute

• Brian D'Onofrio, Indiana University