Predictors of postoperative complications after oral cancer surgery: a comparative analysis of comorbidity indices and clinical risk factors.

OBJECTIVES: Postoperative complications remain a major challenge in the surgical management of oral squamous cell carcinoma (OSCC), particularly in the context of extensive resections and microvascular reconstruction. While comorbidity indices are commonly used for risk stratification, their predictive value relative to functional and procedural factors remains uncertain. MATERIALS AND METHODS: This retrospective cohort study included 401 consecutive patients undergoing curative-intent surgery for OSCC between 2017 and 2024. Postoperative complications within 30 days were graded according to the Clavien-Dindo classification. Preoperative comorbidity was assessed using the Charlson Comorbidity Index (CCI), Adult Comorbidity Evaluation-27 (ACE-27), Elixhauser Comorbidity Index (ECI), and the van Walraven-weighted ECI. Uni- and multivariate logistic regression analyses were performed to identify predictors of overall and major complications (Clavien-Dindo ≥ IIIb). RESULTS: Overall complications occurred in 73.1% of patients, with major complications in 28.4%. Operative duration ≥ 5.5 h was the strongest independent predictor of overall (OR 10.82; p < 0.001) and major complications (OR 2.20; p = 0.030). Impaired preoperative performance status (ECOG ≥ 1) independently predicted major complications (OR 1.50; p = 0.027). Among comorbidity indices, only the ECI was associated with overall complications (OR 1.48; p = 0.026), while all indices showed poor discrimination (AUC 0.52-0.59). In exploratory survival analyses, postoperative complications were associated with reduced overall survival in unadjusted Kaplan-Meier estimates but not in multivariable Cox regression. CONCLUSIONS: Procedural complexity and functional status outweigh traditional comorbidity indices in predicting postoperative morbidity after OSCC surgery. These findings support a shift toward perioperative risk assessment models that prioritize operative burden and physiological reserve over static comorbidity burden.