Diagnostic and prognostic value of noninvasive long-term video-electroencephalographic monitoring in epilepsy surgery: A systematic review and meta-analysis from the E-PILEPSY consortium

Abstract

Objective: The European Union–funded E-PILEPSY network (now continuing within the European Reference Network for rare and complex epilepsies [EpiCARE]) aims to harmonize and optimize presurgical diagnostic procedures by creating and implementing evidence-based guidelines across Europe. The present study evaluates the current evidence on the diagnostic accuracy of long-term video-electroencephalographic monitoring (LTM) in identifying the epileptogenic zone in epilepsy surgery candidates. Methods: MEDLINE, Embase, CENTRAL, and ClinicalTrials.gov were searched for relevant articles. First, we used random-effects meta-analytical models to calculate pooled estimates of sensitivity and specificity with respect to postsurgical seizure freedom. In a second phase, we analyzed individual patient data in an exploratory fashion, assessing diagnostic accuracy within lesional and nonlesional temporal lobe epilepsy (TLE) and extratemporal lobe epilepsy (ETLE) patients. We also evaluated seizure freedom rate in the presence of “localizing” or “nonlocalizing” LTM within each group. The quality of evidence was assessed using the QUADAS-2 tool and the GRADE approach. Results: Ninety-four studies were eligible. Forty-four were included in sensitivity meta-analysis and 34 in specificity meta-analysis. Pooled sensitivity was 0.70 (95% confidence interval [CI] = 0.60-0.80) and specificity was 0.40 (95% CI = 0.27-0.54). Subgroup analysis was based on individual data of 534 patients (41% men). In lesional TLE patients, sensitivity was 0.85 (95% CI = 0.81-0.89) and specificity was −0.19 (95% CI = 0.13-0.28). In lesional ETLE patients, a sensitivity of 0.47 (95% CI = 0.36-0.58) and specificity of 0.35 (95% CI = 0.21-0.53) were observed. In lesional TLE, if LTM was localizing and concordant with resection site, the seizure freedom rate was 247 of 333 (74%), whereas in lesional ETLE it was 34 of 56 (61%). The quality of evidence was assigned as “very low.”. Significance: Long-term video-electroencephalographic monitoring is associated with moderate sensitivity and low specificity in identification of the epileptogenic zone. Sensitivity is remarkably higher in lesional TLE compared to lesional ETLE. Substantial heterogeneity across the studies indicates the need for improved design and quality of reporting.