Development and validation of an MRI-Based nomogram for predicting neurological recovery after acute cervical spinal cord injury.

STUDY

DESIGN

Retrospective cohort study.

OBJECTIVES

Acute cervical spinal cord injury (CSCI) leads to severe and variable neurological outcomes. Early, accurate prognosis is crucial for clinical decisions and patient counseling. This study aimed to identify key prognostic factors and develop a reliable, imaging-based nomogram for predicting neurological recovery after surgical treatment.

SETTING

Northern Jiangsu People's Hospital Affiliated to Yangzhou University, Yangzhou, Jiangsu, China.

METHODS

In a retrospective cohort of 103 surgically treated acute traumatic CSCI patients, neurological recovery was defined as ≥1 grade improvement on the ASIA Impairment Scale at 6 months. We analyzed clinical variables and MRI parameters: intramedullary edema length (IEL), intramedullary hemorrhage length (IHL), and maximum spinal cord compression (MSCC). Independent predictors were identified via multivariate logistic regression. A nomogram was built and internally validated using the C-index, calibration curves, ROC analysis, and DCA.

RESULTS

Neurological recovery occurred in 73 patients (70.9%). Multivariate analysis identified IEL (OR = 0.904), IHL (OR = 0.808), and MSCC (OR = 0.812) as independent predictors (all p < 0.05). IEL showed a threshold effect at 48 mm, beyond which recovery probability declined markedly. The nomogram demonstrated excellent predictive performance, with C-indices of 0.969 (modeling) and 0.936 (validation). ROC and DCA confirmed its superior accuracy and clinical utility over single parameters.

CONCLUSIONS

IEL, IHL, and MSCC are key imaging-based predictors of recovery after acute CSCI. The developed nomogram provides an accurate, practical tool for individualized prognosis, supporting precision clinical management. MRI-based parameters including intramedullary edema length (IEL), intramedullary hemorrhage length (IHL), and maximum spinal cord compression (MSCC) were integrated into a nomogram model to predict neurological recovery in patients with acute cervical spinal cord injury.