Abstracts

Positive Predictive Value of Presurgical Test Combinations for Good Outcome from Temporal Lobectomy.

Abstract number : 2.285
Submission category :
Year : 2001
Submission ID : 1711
Source : www.aesnet.org
Presentation date : 12/1/2001 12:00:00 AM
Published date : Dec 1, 2001, 06:00 AM

Authors :
N.B. Fountain, MD, Neurology Department, University of Virginia, Charlottesville, VA; A. Prasad, MD, Neurology Department, University of Virginia, Charlottesville, VA; E.H. Bertram, MD, Neurology Department, University of Virginia, Charlottesville, VA; M.

RATIONALE: Outcome from anterior temporal lobectomy (ATL) for temporal lobe epilepsy (TLE) is generally good, but identification of positive prognostic indicators may allow better patient selection and avoid intracranial monitoring. We hypothesized that the outcome improves with a greater number of concordant presurgical tests.
METHODS: A consecutive series of 136 patients were retrospectively reviewed. Selection criteria included; first-time ATL for TLE at our institution between 1991-2000, having undergone all examined tests, and had at least 1 year of follow-up. Patient records were reviewed for patient characteristics (age, sex, duration of epilepsy, side and type of surgery), presurgical test results (routine and prolonged interictal EEG, ictal EEG, MRI, and Wada), pathology results, and outcome at one year. Side of surgery was taken as the reference for concordance of findings. Concordance followed standardized definitions. Positive predictive value (PPV; the proportion of patients with a positive finding who had a class I outcome) and sensitivity (the proportion of patients with a class I outcome who had a positive finding) were calculated. Fisher[ssquote]s exact test was used to compare patients with concordant findings to all others.
RESULTS: Overall, 74% had class I outcome. Statistically significant PPV were found for concordance of mesial temporal sclerosis (MTS) on MRI (81%, p=.02), and appropriate memory asymmetry on Wada (80%, p=.02). PPV improved with a combination of concordant MTS on MRI and Wada (86%, p[lt].01), and improved further by concordance of MTS on MRI, Wada, and scalp EEG ictal onset (91%, p[lt].001). Interictal EEG findings did not statistically influence outcome. Independent bilateral ictal onset from intracranial recording was associated with a poor outcome (29%, p[lt].01). Other individual tests and combinations of tests did not yield statistically significant associations with good outcome. Sensitivity and specificity were uniformly low.
CONCLUSIONS: Concordance of MTS on MRI, memory asymmetry on Wada testing, and unilateral scalp EEG ictal onset have a very high statistically significant PPV. MTS on MRI and Wada testing alone have a high PPV, even when ictal scalp EEG is not concordant, which suggests that patients with this pattern of findings may not need intracranial monitoring.
Support: NINDS grant NS01992 (NBF).