The Suspected Cauda Equina Syndrome Score (SuCESS) : development and validation of a clinical triage tool
Elie Najjar1, Shahbaz Khan1, Rawan Masarwa1
1Spinal Surgery Unit, Queen's Medical Centre, Nottingham University Hospitals NHS Trust, Nottingham, UK.
Aims:
Cauda equina syndrome (CES) is a rare but critical spinal emergency. However, over 80% of urgent MRIs for suspected CES show no cauda equina compression. This diagnostic uncertainty places strain on imaging services and delays care. We aimed to develop and validate a clinical triage tool, the Suspected Cauda Equina Syndrome Score (SuCESS), to safely exclude CES and improve the accuracy of triage.
Methods:
We retrospectively developed the SuCESS score using the records of 259 patients who had been referred to a UK tertiary spinal unit between 2016 and 2018. Six clinical variables - saddle anaesthesia, reduced perianal sensation, motor weakness, bilateral sciatica, bladder post-void residual volume (PVR), and urethral catheterization - were identified using multivariable logistic regression, random forest, and XGBoost, and combined into an eight-point scoring system. Penalized regression was applied to minimize overfitting. External validation was undertaken using two temporally independent cohorts from 2020 and 2023 (n = 444). Primary outcomes were sensitivity and negative predictive value (NPV) for surgically confirmed CES. Model performance was assessed using area under the curve (AUC), Brier scores, calibration plots, and decision curve analysis.
Results:
At a threshold of ≥ 3.0, SuCESS achieved 100% sensitivity and 100% negative predictive value across all cohorts, correctly identifying all 64 cases of confirmed CES. Specificity increased from 37% in the development cohort to 54% after validation. AUC improved from 0.70 to 0.85, and Brier scores improved from 0.26 to 0.13. Use of the score might have safely deferred 38.7% of urgent MRIs, based solely on bedside clinical data.
Conclusion:
SuCESS is the first validated triage tool for CES to show perfect sensitivity. It offers a pragmatic and safe framework for reducing unnecessary MRI requests and for improving diagnostic accuracy in urgent spinal referrals.
