Abstracts

We have previously demonstrated disparities in the use of surgical treatment for temporal lobe epilepsy (TLE) between African-Americans (AA) compared to non-hispanic Caucasians (NHC), as well as the existence of a possible relationship between race and surgical outcome following surgery for intractable TLE due to mesial temporal sclerosis (MTS). The objective of the present study was to evaluate possible predictors of seizure recurrence in all cases of TLE, as well as short- and long-term outcome, and their relationship with the patients[apos] racial profile. Data was obtained from the University of Alabama - Epilepsy Center database (1985-2001). The sample consisted of all patients with a diagnosis of medically intractable TLE who underwent surgery. Seizure recurrence was tabulated at 7 days, 2 and 6 months, 1, 2, 3, 4, 5, and 6 years following surgery. Other information included: age at evaluation, date of birth, date of surgery, age of seizure onset, gender, side of surgery (left or right), seizure frequency, and pathology of surgical specimens. Logistic regression analysis was used to model the presence of seizure recurrence after surgery for all patients. Initial univariate analysis generated odds ratios (ORs) for seizure recurrence after temporal lobectomy for AA relative to NHC at 2, 3, 4, 5, and 6 years post-surgery. Subsequent multivariate analysis generated ORs with the inclusion of all the variables. 370 patients (210 females) underwent surgical treatment for TLE. No significant differences were found between AA and NHC in age, age at onset, gender, pathology, side of surgery, seizure frequency, and immediate postoperative seizures (p values [gt] .10). No differences were found between males and females, groups of age at presentation ([lt] 25, 25-34, and [gt] 34 y), side of surgery (left vs. right), seizure frequency ([lt] 4, 4 to 9, and [gt] 9 seizures per month), underlying pathology (MTS, gliosis, normal tissue, and other) and age of onset ([lt] 5, 5 to 14, and [gt] 14 yo); with the exception of the occurrence of immediate postoperative seizures, which was a strong predictor of late seizure recurrence (ORs= 1.2-3.5). Logistic regression analysis revealed that AA were more likely than NHC to have seizure recurrence after surgery at 6m, 1, 2, 3, 4, 5, and 6 years following surgery (ORs=1.1-2.1). After potential confounders were controlled, this finding did not change (ORs=1.1-2.2). Seizures in the immediate postoperative period are a strong predictor of later seizure recurrence. Our data also suggest that race may be an important factor related to seizure outcome.