Multivariate neuropsychological prediction of seizure lateralization in temporal epilepsy surgical cases

Abstract

Purpose: Neuropsychological assessment can be of assistance in determining seizure lateralization in cases where EEG and MRI findings do not provide clear lateralizing data. While several studies have examined the lateralizing value of individual neuropsychological measures, clinicians are still in need of a statistically sound method that permits the incorporation of multiple neuropsychological variables to predict seizure lateralization in the individual patient.

Method: The present study investigated the lateralizing value of several commonly used neuropsychological measures in a large sample of patients (n = 217) who eventually underwent surgical resection to treat their epilepsy. Side of surgery was used to operationally define seizure lateralization. A comparison of the relative utility of a multivariate versus univariate approach to predict seizure lateralization was conducted in temporal epilepsy cases.

Results: The results provide evidence for the incremental validity of neuropsychological measures, other than memory and IQ tests, in the prediction of seizure lateralization in patients with medically intractable epilepsy. These data indicate that a multivariate approach increases the accuracy of prediction of seizure lateralization for temporal lobe epilepsy cases.

Conclusion: This study supports the use of a multivariate approach using neuropsychological measures to predict seizure lateralization in temporal epilepsy surgical candidates. Regression formulas are provided to enhance the clinical utility of these findings.