Timing and Benefit of Early Versus Delayed Reoperation in Recurrent Glioblastoma: A Systematic Review and...

Background: The prognostic relevance of surgical timing at glioblastoma recurrence remains uncertain, and definitions of early versus delayed reoperation vary widely. Whether earlier surgery provides meaningful survival or functional benefit has not been clearly established. Methods: Databases including PubMed, Embase, Scopus, and Web of Science were searched from inception to May 2025. Eighteen observational studies met the inclusion criteria, fourteen of which provided extractable hazard ratios for survival. The primary outcome was overall survival after reoperation; secondary outcomes included functional status (ΔKPS or discharge home) and major postoperative complications. Random-effects models with Hartung-Knapp adjustment were used, with subgroup analyses stratified by KPS, extent of resection, and eloquence. Results: Across 2267 reoperated patients from 14 survival studies, earlier reoperation was associated with significantly longer survival (pooled HR 0.86; 95% CI 0.78-0.95). Subgroup analyses showed stronger effects in patients with KPS ≥ 70 (HR 0.81; 95% CI 0.72-0.92), non-eloquent tumors (HR 0.84; 95% CI 0.75-0.94), and near-total/gross-total resection (HR 0.79; 95% CI 0.68-0.93). Functional outcomes were pooled from 9 studies (n = 1182), demonstrating higher odds of postoperative stability or improvement with early surgery (OR 1.28; 95% CI 1.12-1.46). Major complications were reported in 9 studies (n = 1344) and did not differ between groups (OR 0.98; 95% CI 0.81-1.19). Sensitivity analyses and influence diagnostics showed consistent effect estimates and no undue single-study influence. Conclusions: Earlier reoperation for recurrent glioblastoma is associated with improved survival and better functional outcomes without increased morbidity in appropriately selected patients. Surgical timing should be incorporated into multidisciplinary planning. Prospective studies with standardized timing definitions and time-dependent modeling are needed to validate these findings.