Development of the TORS severe outcome score: a preoperative risk stratification tool for postoperative morbidity in oropharyngeal cancer.

BACKGROUND

Transoral robotic surgery (TORS) is increasingly adopted for the treatment of oropharyngeal squamous cell carcinoma (OPSCC), with minimizing treatment-related morbidity remaining a key priority. We developed the TORS Severe Outcome Score (TSOS), a preliminary preoperative risk stratification framework to identify candidate predictors of severe postoperative morbidity following TORS.

METHODS

A retrospective cohort study was conducted using the TriNetX Research Network. Adults with OPSCC who underwent TORS within three months of diagnosis between 2015 and 2024 were identified. The composite outcome included postoperative hemorrhage, tracheostomy, or gastrostomy placement within 30 days of surgery. Cox proportional hazards regression identified independent predictors of severe postoperative events and was used to generate a point-based risk score. Candidate point thresholds were evaluated using sensitivity, specificity, and Youden index for the composite outcome and each individual endpoint.

RESULTS

A total of 1,592 patients were included. Severe postoperative events occurred in 250 patients (15.7%), including hemorrhage in 93 (5.8%), tracheostomy in 71 (4.5%), and gastrostomy in 132 (8.3%). Independent predictors included advanced T stage (T3, p = 0.029; T4, p = 0.001), body mass index (BMI) < 22 (p < 0.001), smoking (p < 0.001), preoperative dysphagia (p < 0.001), and hyperlipidemia (p = 0.024). A threshold of ≥ 24 points demonstrated the highest discriminatory performance for the composite outcome. Outcome-specific cutoffs were ≥ 24 for hemorrhage, ≥14 for tracheostomy, and ≥ 24 for gastrostomy placement. Conclusions In this derivation study, TSOS identifies candidate preoperative risk factors associated with severe postoperative morbidity following TORS for OPSCC. Pending external validation, these findings may inform preoperative counseling and multidisciplinary risk assessment.