DOI: https://doi.org/10.36347/sasjm.2026.v12i10.010

Pages: 1070-1077

Background: After elective intracranial surgery, immediate extubation enables early neurological assessment, but it is not always safe. Planned postoperative ventilation may be justified when neurological status, airway protection, gas exchange, or operative conditions are unfavorable. This review evaluates the incidence of delayed extubation, its predictors, and its associations with major postoperative outcomes. Methods: PubMed/MEDLINE and Europe PMC were searched from inception to 21 September 2026. Adult cohorts comparing extubation in the operating room or post-anesthesia care unit with later extubation after transfer to intensive care were eligible. Random-effects models produced pooled proportions and odds ratios. Risk of bias was assessed using an adapted Newcastle-Ottawa Scale, and certainty was judged according to GRADE principles. Results: Four cohorts including 1,345 patients were included, of whom 632 underwent delayed extubation. The pooled incidence was 42.3% (95% CI 27.7–58.4; I² = 96.0%). In two infratentorial cohorts, preoperative hydrocephalus (OR 3.01; 95% CI 1.25–7.21), lower cranial nerve dysfunction (OR 3.97; 2.15–7.34), and blood loss ≥1,000 mL (OR 5.88; 1.74–19.92) were associated with delayed extubation. Delayed extubation was also associated with unfavorable neurological outcome (OR 3.74; 1.30–10.80) and intensive care stay >24 hours (OR 7.62; 4.75–12.23), without a statistically demonstrated difference in reintubation (OR 1.52; 0.49–4.77). Conclusion: Delayed extubation is common and mainly reflects neurological and operative complexity. The observed associations do not prove that delayed extubation is itself harmful because all data are observational and highly susceptible to confounding by indication. A protocolized, individualized, and repeatedly reassessed decision appears preferable to a uniform strategy.