Racial and ethnic differences in epilepsy classification among probands in the Epilepsy Phenome/Genome Project (EPGP)

Little is known about the ethnic and racial differences in the prevalence of generalized and focal epilepsy among patients with non-acquired epilepsies. In this study, we examined epilepsy classification and race/ethnicity in 813 probands from sibling or parent-child pairs with epilepsy enrolled in the Epilepsy Genome/Phenome Project (EPGP). Subjects were classified as generalized epilepsy (GE), non-acquired focal epilepsy (NAFE), mixed epilepsy syndrome (both generalized and focal), and unclassifiable, based on consensus review of semiology and available clinical, electrophysiology, and neuroimaging data. In this cohort, 628 (77.2%) subjects identified exclusively as Caucasian/white and 65 (8.0%) subjects reported African ancestry, including subjects of mixed-race. Of the Caucasian/white subjects, 357 (56.8%) had GE, 207 (33.0%) had NAFE, 32 (5.1%) had a mixed syndrome, and 32 (5.1%) were unclassifiable. Among subjects of African ancestry, 28 (43.1%) had GE, 27 (41.5%) had NAFE, 2 (3.1%) had a mixed syndrome, and 8 (12.3%) were unclassifiable. There was a higher proportion of subjects with GE compared to other syndromes among Caucasians/whites compared to subjects with African ancestry (OR 1.74, 95% CI: 1.04-2.92, two-tailed Fisher's exact test, p=0.036). There was no difference in the rate of GE among subjects reporting Hispanic ethnicity (7.6% of total) when adjusted for race (Caucasian/white vs non-Caucasian/white; OR 0.65, 95% CI: 0.40-1.06, p>0.05). The proportion of participants with unclassifiable epilepsy was significantly greater in those of African-American descent. In a group of patients with epilepsy of unknown etiology and an affected first degree relative, GE is more common among Caucasian/white subjects than among those with African ancestry. These findings suggest there may be geographical differences in the distribution of epilepsy susceptibility genes and an effect of genetic background on epilepsy phenotype. However, the results should be interpreted with caution because of the low numbers of African-Americans in this cohort and more limited diagnostic data available for epilepsy classification in these subjects compared to Caucasians/whites.