SRS87 Patey Prize Entrant - The Suspected Cauda Equina Syndrome Score (SuCESS): development and validation of a clinical triage tool

Abstract Background Cauda equina syndrome (CES) is a serious spinal emergency. Yet, over 80% of urgent MRIs for suspected CES show no compression, creating strain on imaging services and delaying care. We aimed to develop and validate a bedside clinical triage tool—SuCESS (Suspected Cauda Equina Syndrome Score)—to safely rule out CES and improve accuracy. Methods We retrospectively derived SuCESS using 259 patients referred to a UK tertiary spinal unit (2016–2018). Six variables—saddle anaesthesia, reduced perianal sensation, motor weakness, bilateral sciatica, bladder post-void residual volume (PVR), and urethral catheterisation—were identified through logistic regression, random forest, and XGBoost, then combined into an eight-point score. Penalised regression reduced overfitting. External validation was performed in two independent cohorts from 2020 and 2023 (n = 444). Primary outcomes were sensitivity and negative predictive value (NPV) for surgically confirmed CES. Model performance was evaluated with area under the curve (AUC), Brier score, calibration, and decision curve analysis. Results At a threshold of ≥3.0, SuCESS achieved 100% sensitivity and 100% NPV across all cohorts, identifying all 64 confirmed CES cases. Specificity rose from 37% in the development cohort to 54% in validation. AUC improved from 0.70 to 0.85, with Brier scores improving from 0.26 to 0.13. Use of the score could have safely deferred 38.7% of urgent MRIs, based on bedside data alone. Conclusions SuCESS is the first validated CES triage tool to demonstrate perfect sensitivity. It provides a pragmatic framework for reducing unnecessary MRI requests while maintaining diagnostic safety in spinal referrals.