Five or more acute postoperative seizures predict hospital course and long-term seizure control after hemispherectomy

Purpose: Acute postoperative seizures (APOSs) are those that occur in the first 7 to 10 days after surgery, and previous studies in temporal lobe epilepsy patients support the notion that APOSs may foretell failure of long-term seizure control. It is unknown whether APOSs also predict seizure outcome or hospital course after hemispherectomy. Methods: Hemispherectomy patients (n = 114) were studied retrospectively and subdivided into the following groups: No APOSs, 1 to 5 APOSs, or >5 APOSs. Intensive care unit (ICU) nursing staff or family members reported and described the APOS events. APOS categories were compared with pre-and postsurgery clinical variables abstracted from the medical record. Results: APOSs occurred in 22.6% of hemispherectomy patients. Compared with the 0 and I to 5 APOS groups, patients with >5 APOS showed (a) longer seizure durations before surgery, (b) longer hospitalizations, (c) later oral food intake, (d) more frequent lumbar punctures, (e) worse seizure control at 0.5 and 1 year after surgery, (f) more antiepileptic drug (AED) use at 2 and 5 years after surgery, and (g) higher reoperation rate. No similar differences were found between the 0 and 1 to 5 APOS groups. The day of the APOS, whether the APOS was typical of preoperative seizures, and postsurgery scalp EEG did not predict long-term seizure control. APOS patients in the 1 to 5 and >5 groups had lower pre- and postsurgery Vineland developmental quotients compared with those without an APOS. Conclusions: Hemispherectomy patients with >5 APOSs had a more prolonged and complicated hospital course and worse postsurgery seizure control, more AED use, and higher reoperation rate than did patients with 0 or 1 to 5 APOSs. Thus the number of APOSs was a predictor of postsurgery seizure control and can be used to counsel patients and families about prognosis after hemispherectomy.