Absolute and Relative Predictor Values of Some Non-Invasive and Invasive Studies for the Outcome of Anterior Temporal Lobectomy

Abstract

Background There is controversy in the literature regarding the strategy used to obtain better outcomes after performing an anterior temporal lobectomy (ATL). Some investigators prefer to reduce the risks and costs of the predictor studies despite the fact that the number of patients cured after ATL (no seizures with no medication) is relatively small. Other investigators prefer to attempt a total cure in all patients by using all available predictor studies regardless of risks and costs. The latter strategy was the aim of the present work. Methods The absolute (%) and relative (ratio differences) predictor values of non-invasive and invasive studies for the outcome 24 months post-ATL were determined on 22 patients suffering from intractable non-lesional temporal lobe epilepsy. Results Under these conditions, 11 (50%) patients had excellent outcomes (seizure-free, no medication), 8 (36%), good outcomes (only auras with medication), and 3 (13.6%), poor outcomes (1 with >70%-seizure reduction and two with no changes in seizure frequency). Predictors of excellent (vs. poor outcome) include the following: complex partial seizures (CXP) alone or associated with secondary generalized tonic-clonic seizures; ictal motionless stare and postictal amnesia; abnormal bitemporal spikes (prominent ipsilateral to ATL) and secondary bilateral synchrony EEG activities; focal hippocampal ictal EEG activities, and the presence of focal anterior temporal delta EEG activity 3 months after ATL. Predictors of poor (vs. excellent) outcomes include the following: CXP associated with other generalized seizure types; CTS, MRI, and EEG extratemporal abnormalities, and generalized basotemporal ictal EEG activities and the presence of seizures and focal anterior temporal spikes 3 months after ATL. Conclusions The present study corroborates that no single predictive study (including non-invasive MRI and invasive ictal EEG activity) is predictive of the success or failure of ATL. Rather, a concordant combination of non-invasive and invasive studies is more likely to be predictive of a high probability of success. The high efficiency of ATL (86% of patients seizure-free) was accomplished by using all available predictor studies.