Mom’s Epilepsy Meds Did Not Impact Kids’ Language Development

Language development scores at age 2 were similar for children of women with epilepsy taking antiseizure medications and children with healthy mothers, a prospective observational study found.

There was no difference in the primary outcome of language domain (-0.5, 95% CI -4.1 to 3.2) between the two groups, based on an assessment using the Bayley Scales of Infant and Toddler Development, Third Edition (BSID-III), reported Kimford Meador, MD, of Stanford University in Palo Alto, California, and co-authors.

The interim report from the multicenter Maternal Outcomes and Neurodevelopmental Effects of Antiepileptic Drugs (MONEAD) study also found no differences in scores for the other four BSID-III domains — motor, cognitive, social-emotional, and general adaptive behavior, they noted in JAMA Neurology.

“This encouraging finding may be due to the use of newer [antiseizure medications] with lower risk of affecting the immature brain,” Meador and team wrote. “However, these findings must be interpreted within the context that neuropsychological assessments conducted at 2 years of age are not as strongly associated with adolescent/adult functioning as assessments performed in older children.”

“Exposure-dependent effects are expected for drug-induced teratogenic effects. Assessments at older ages and in other cohorts are critical,” they cautioned.

The MONEAD study is a continuation of the Neurodevelopmental Effects of Antiepileptic Drugs (NEAD) study. Meador and colleagues enrolled a new cohort who were taking antiseizure medications currently used at tertiary epilepsy centers.

The analysis included 292 children of women with epilepsy (median age of about 2, 53% girls) and 90 children of healthy women (median age of about 2, 52% boys).

Most of the mothers with epilepsy were taking lamotrigine (Lamictal; 46%) or levetiracetam (Keppra; 33.2%). About 45% were taking both.

Of note, secondary analyses revealed that higher maximum levels of antiseizure medications in the third trimester were associated with lower BSID-III scores for the motor domain (-5.6, 95% CI -10.7 to -0.5), and higher maximum doses of these drugs in the third trimester were associated with lower scores in the general adaptive domain (-1.4, 95% CI -2.8 to -0.05), Meador and team said.

Except for studies linking valproate with dose-dependent risks for both malformations and neurodevelopmental deficits, associations with other antiseizure medications have been less clear, though data support lower cognitive/behavioral risks for carbamazepine (Tegretol), lamotrigine, and levetiracetam, Meador and team noted.

Outcomes associated with levetiracetam “are the most important contribution, given the sparse information that has been available,” noted Torbjörn Tomson, MD, PhD, of the Karolinska Institutet in Stockholm, and Rebecca Bromley, ClinPsyD, PhD, of the University of Manchester in England, in an accompanying editorial.

Meador and colleagues also assessed risk factors for potential confounding in secondary analyses, including breastfeeding status, periconceptional use of folate, anxiety, depression, and perceived stress during pregnancy/postpartum through the child’s visit at age 2.

Of these risk factors, they found no negative association between breastfeeding and cognition at age 2; indeed, their data suggest that antiseizure medication blood levels are usually low in nursing infants of mothers using these drugs. Similarly, findings from the NEAD cohort showed no adverse cognitive effects of breastfeeding in children of mothers taking antiseizure medications at age 3, and positive effects on cognition in the same cohort assessed at age 6.

Periconceptional folate use was not significantly associated with reduced risk of language delay and autistic behaviors reported in prior studies of children of women with epilepsy, but the authors noted that these effects have been found to be stronger at age 6.

Increased maternal anxiety was associated with poorer child language scores, as was depression, although the effect did not reach significance in the full model. In addition, maternal anxiety, depression, and perceived stress were all associated with a child’s cognitive, social-emotional, and general adaptive domain outcomes.

Study limitations included a lack of randomization and that cognitive/behavioral assessments at age 2 did not provide associations as strong as those performed at older ages, Meador and co-authors acknowledged.

In addition, assessment of antiseizure medications as a single group is “somewhat problematic,” given their different modes of action, pharmacokinetic profiles, and level of documentation of therapeutic ranges, editorialists Tomson and Bromley noted.

While the use of participant antiseizure medication blood levels increased the sensitivity of the dose information, the editorialists said that unknowns included when samples were drawn, whether total or unbound levels were determined, and whether use of the highest blood levels or mean levels during the third trimester would be more appropriate.

“Further studies in our cohort in children of older ages and in other cohorts are needed to fully delineate the effects of [antiseizure medication] exposure on the immature brain,” Meador and colleagues concluded.

Last Updated June 07, 2021