Abstracts

Rationale: Interictal fast ripples (FR, 250-500 Hz) are proposed biomarkers of the epileptogenic zone and may predict outcomes following resective epilepsy surgery. Retrospective studies using intraoperative or extraoperative electrocorticography (ECoG) have linked complete resection of FR-generating cortex to postoperative seizure freedom. The value of FR has not been systematically contrasted with conventional ECoG abnormalities, including focal slowing, paroxysmal fast activity, intermittent spike discharges, continuous epileptiform discharges, focal attenuation, and intraoperative seizures. Methods: The study cohort included 60 consecutive pediatric patients who underwent epilepsy surgery with intraoperative ECoG (and without extended extraoperative ECoG). Using test metrics (sensitivity, specificity, positive and negative predictive value, and accuracy), we compared the value of competing ECoG abnormalities in the prediction of postoperative seizure freedom. The extent to which each ECoG abnormality independently predicted time to first postoperative seizure was ascertained by multivariate Cox proportional hazards regression. Results: FR were among the most prevalent (80% of studies contained FR) of ECoG abnormalities. In contrast to other measured ECoG abnormalities, complete FR resection demonstrated the most favorable combination of positive predictive value (100%), negative predictive value (81%), sensitivity (42.9%), and specificity (100%) in the prediction of post-operative seizure-freedom. See Table. Similarly, in multivariate survival analyses, time to first postoperative seizure was independently predicted by incomplete resection of cortex generating FR (HR = 31.9, 95%CI 8.6 - 118.9, p < 0.001) and focal slowing (HR 5.1, 95%CI 1.4 - 18.8, p = 0.016). See Figure. Conclusions: The incomplete resection of FR-generating cortex is a substantial and specific predictor of postoperative seizures. With a high detection rate during brief intraoperative ECoG and favorable test characteristics, FR represents a useful and feasible adjunct to standard methods for identification of the epileptogenic zone. The time has come for a carefully blinded trial to evaluate the incremental value of FR localization in the surgical treatment of intractable focal (or multifocal) epilepsy.