Seizure outcomes after resection of temporal encephalocele in patients with drug-resistant epilepsy: A systematic review and meta-analysis.

OBJECTIVE

Temporal encephaloceles (TEs) are seen in patients with drug-resistant epilepsy (DRE); yet they are also common incidental findings. Variability in institutional pre-surgical epilepsy practices and interpretation of epileptogenic network localization contributes to bias in existing epilepsy cohorts with TE, and therefore the relevance of TE in DRE remains controversial. We sought to estimate effect sizes and sample sizes necessary to demonstrate clinically relevant improvements in seizure outcome with different surgical approaches.

METHODS

We searched Medline, Embase, and Cochrane to identify studies reporting the outcomes of epilepsy surgery after 12 months in patients with DRE and TE. The main outcome was seizure freedom or favorable seizure outcome. We also assessed the rates of seizure freedom among patients with DRE, TE, and the following covariables: use of intracranial electroencephalography (iEEG), side of the encephalocele, sex, and type of surgical resection (anterior temporal lobectomy [ATL] vs lesionectomy). Random-effects meta-analysis was used to calculate the proportion of patients attaining seizure outcomes.

RESULTS

We identified 332 studies, of which 15 (282 patients) met inclusion criteria for meta-analysis. Seizure-freedom rate was 65% (95% confidence interval [CI] 53-76), whereas the favorable outcome rate was 78% (95% CI 70-85). There was no significant interstudy heterogeneity. Patients with TE undergoing iEEG (risk ratio [RR] 0.80, 95% CI 0.74-0.87) had a lower chance of a favorable seizure outcome. A power analysis estimated a sample size of 28 932 patients with TE (13 764 with ATL) necessary to determine a significant difference in seizure freedom between limited resection and ATL.

SIGNIFICANCE

Retrospective cohort studies demonstrate good outcomes after TE resection regardless of the extent of resection. Prohibitively large sample sizes necessary to observe outcome differences between surgical approaches and presurgical predictors indicate that improved biomarkers and mechanistic understanding of TE epileptogenicity are needed.