Graduate student Georgetown University Washington, District of Columbia
Rationale: Patients with chronic epilepsy are known to exhibit a higher incidence of atypical language representation (25-30%) compared to control populations (5%) MRI abnormalities, early age of seizure onset, being under age seven, and left-handedness are all associated with increased rates of atypical language dominance. However, it is unclear to what degree seizure activity drives language to reorganize or if higher rates of atypical language are evident at the onset of seizures. Our study evaluates a cohort of pediatric epilepsy patients who were selected to control for factors other than seizure activity that might drive language reorganization (e.g., no MRI abnormality within language regions, right-handed, and recently diagnosed with epilepsy (≤ 2 years)). We hypothesized that rates of atypical language in this cohort would be similar to typically developing controls (TD). Methods: 34 participants (17 TD, mean age = 8.2 ± 1.8 years; 17 epilepsy patients, mean age = 8.0 ± 1.8 years) completed an fMRI language task (Auditory Description Decision Task) in 3T MRI. Epilepsy patients were scanned within two years of their epilepsy onset (mean lifetime seizure duration = 1.2 ± 0.53 years; range = 0.2-2.0 years). Images were processed in SPM12 using bilateral IFG and WA ROIs. Laterality Index (LI) was calculated to determine side and strength of language dominance. Descriptive statistics and t-tests were conducted to compare groups. We also examined whether age (groups defined by younger (age 4-6 yrs) and older (7-12 yrs)) or age of onset influenced results. Linear regression was used to test if the relationship between LI, and duration of epilepsy differed when age of onset was under or over age six. Results: Overall, mean LI values did not differ between epilepsy patients and TDs in either ROI (all p’s > 0.05; Figure 1), with 88.2% of subjects displaying left language dominance for IFG, and 91.2% displaying left language dominance for WA. Younger children (80% left lateralized IFG; 80% WA) did not differ from older children (91.7% left lateralized IFG; 96% WA) in either IFG or WA (all p’s > 0.05). Of the six children with atypical language dominance (4 TD; 2 patients), dominance was largely bilateral and no cases of right WA dominance was observed in this sample. Among epilepsy patients, there was a significant interaction such that the relationship between duration of epilepsy and LI differed by age of onset. For both IFG (p = 0.007) and WA (p = 0.002), in children with onset after age 6, duration of epilepsy predicted less left-lateralized language dominance compared to children with onset age six and under, where duration was not strongly correlated with LI (Figure 2). Conclusions: Language dominance is determined by several factors. Our findings suggest that in a cohort with only seizure activity as a main risk factor, language representation remains largely in the left hemisphere of the brain, at rates similar to control populations. When atypical language was evident, it was largely bilateral and in the frontal lobes which is a typical developmental pattern. A nuanced finding was that even in a recent onset group, duration of epilepsy may reduce left dominance as seizures persist only in those with age of onset after age six. These findings suggest that in a cohort without obvious factors that indicate reorganization of language (no MRI abnormality, no left handedness), language will largely remain in the left hemisphere, particularly in temporal regions and for early age of onset. However, language may reorganize partially by engaging the nondominant hemisphere as seizures persist in patients that have a later age of onset. This partial reorganization may reflect compensation rather than true language dominance. Funding: Please list any funding that was received in support of this abstract.: NINDS R01 NS44280 NICHD P30 HD40677 Click here to view image/table
