Prediction of seizure outcome with presurgical intracarotid amobarbital procedure, mesial temporal sclerosis on MRI, and PET in surgical candidates with temporal lobe epilepsy.

Objective: To determine the relative predictive value of the intracarotid amobarbital procedure (IAP), mesial temporal sclerosis (MTS) on magnetic resonance imaging (MRI), and positron emission tomography (PET) for long-term (3-years) seizure outcome following neurosurgery for temporal lobe epilepsy (TLE).

Methods: Data from 88 patients with TLE were analyzed. We examined demographic, clinical, and presurgical workup variables in relation to absolute seizure outcome (freedom vs. recurrence) within a three-year monitoring period following surgery. Presurgical variables were also examined as longitudinal predictors of time to seizure recurrence within the three-year postsurgical interval using univariate Cox regression models. Predictor survival curves were compared using Mantel-Cox (log-rank) tests.

Results: Side of resection, resection type (selective vs. non-selective), hemisphere of language dominance, and epilepsy duration were not associated with seizure outcome. PET and MRI were also unrelated to seizure outcome. Incorrectly lateralizing IAP was associated with higher odds of seizure recurrence (OR = 3.93, p = .018, 95% CI = 1.20, 12.94). Additionally, survival analyses indicated that incorrectly lateralized IAP was a longitudinal predictor of earlier seizure recurrence (HR = 2.84, p = .023, 95% CI = 1.15-6.98). Log-rank analyses revealed that, as opposed to those with expected IAP memory lateralization, patients with incorrect IAP lateralization experienced postsurgical seizure relapse about seven months earlier.

Significance: IAP that does not lateralize to the presumed hemisphere of seizure focus is associated with increased risk of seizure recurrence, as well as shortened time-to-seizure relapse following surgery for TLE.