Abstracts

Rationale: Magnetic Resonance Imaging (MRI) has become an integral part of pre-surgical evaluation in patients with intractable epilepsy. However, a subset of patients with intractable temporal lobe epilepsy (TLE) has normal MRI. Previous studies have suggested that invasive ictal recording may be omitted in these patients. We investigated if invasive ictal recording provided additional information to their pre-surgical evaluations.Methods: We retrospectively identified 30 out of 302 patients with intractable TLE and normal MRI that underwent pre-surgical evaluation with subdural grid or depth electrodes between January 1991 and November 2006 at the University of Florida Comprehensive Epilepsy Center. All patients had been followed up for at least six months. The diagnosis of TLE and the localization of seizure onset were obtained from medical records. Localization by scalp ictal recording was divided into left-temporal, right-temporal, extra-temporal and bi-temporal. Localization by invasive ictal recording was classified as left/right mesial/lateral temporal, extra-temporal, bi-temporal and multifocal. Surgery was separated into left/right anterior temporal lobectomy (ATL), extra-temporal resection and no surgery. Primary outcome was measured by comparing the concordance of localization by scalp and invasive ictal recordings and surgery to determine if invasive ictal recording provided additional information.Results: Seven patients were excluded due to incomplete medical records. Of the remaining 23 patients, localizations by scalp ictal recording were: 10 left-temporal, 8 right-temporal, 3 extra-temporal and 2 bi-temporal. Invasive ictal recording was concordant in 55.6% (10/18) of the patients with unilateral temporal onset found on scalp ictal recording [40% (4/10) left-temporal and 75% (6/8) right-temporal]. Seven of the 23 total patients (30.4%) received different surgeries than suggested by scalp ictal recording and one patient (4.3%) was declined surgery because of the findings on invasive ictal recording. Thus, 34.8% (8/23) patients had their surgical plan altered by the results of invasive ictal recording. Median follow-up period was 20.5 months (range 6—48). 61.5% (8/13) of patients who received ATL and 75% (3/4) of patients who received extra-temporal resections remained seizure-free. Of the seven patients who received different surgeries based on invasive ictal recording, 71.4% (5/7) remained seizure-free.Conclusions: In contrast to previous reports, our study found that invasive ictal recording remains an integral component in the pre-surgical evaluation of patients with intractable TLE and normal MRI. Findings from invasive ictal recording changed the type of surgery in 34.8% of the patients (4 had different surgeries, 1 was declined surgery, and 3 had lateral temporal lobe resections in addition to ATL). Thus, omitting invasive ictal recording in the pre-surgical evaluation of patients with intractable TLE and normal MRI should be carefully considered.