Abstract
Background Spinal cord injury without radiographic evidence of trauma (SCIWORET) represents a significant portion of cervical spinal cord injuries (SCIs), particularly in elderly patients. Despite its clinical significance, the risk factors and their interactions remain poorly understood. This study aimed to identify key predictors of SCIWORET by analyzing anatomical and biomechanical factors in cervical SCI patients at a single tertiary emergency medical center. Methods We retrospectively analyzed consecutive patients with C3-C7 cervical SCI between April 2011 and November 2023. All patients underwent standardized neurological examination and comprehensive imaging studies, including whole-spine computed tomography (CT) and cervical magnetic resonance imaging (MRI). SCIWORET was defined as a neurological deficit without fracture, dislocation, or discoligamentous complex (DLC) injuries on imaging. We evaluated the presence of cervical canal stenosis (CCS), ossification of the posterior longitudinal ligament (OPLL), and diffuse idiopathic skeletal hyperostosis (DISH) using standardized criteria. Two experienced orthopedic surgeons independently assessed all imaging studies. Univariate logistic regression analysis was first examined to identify the associated factor of SCIWORET. We evaluated multicollinearity using variance inflation factors (VIFs) and correlation coefficients between CCS, OPLL, and DISH. Furthermore, multivariate logistic regression analysis was conducted to identify independent predictors of SCIWORET. Results Among the study population, 203 of 348 patients (58.3%) were diagnosed with SCIWORET. CCS was present in 174 of 348 patients (50.0%), with a significantly higher prevalence in the SCIWORET group (174 of 203 patients, or 85.7%) compared to the SCI with fracture group (78 of 145 patients, or 53.8%). OPLL was identified in 68 of 203 patients (33.5%), showing a higher prevalence in the SCIWORET group versus the SCI with fracture group (24 of 145 patients, or 16.6%). Regarding VIF, CCS, cervical OPLL, and DISH were 1.07, 1.12, and 1.19, respectively. Correlation analysis showed weak associations: CCS and DISH (r = 0.11), CCS and cervical OPLL (r = 0.25), and DISH to cervical OPLL (r = 0.33). Multivariate logistic regression analysis revealed that CCS (odds ratio (OR): 4.91, 95% CI: 2.78-8.70, p < 0.0001) and cervical OPLL (OR: 1.83, 95% CI: 1.01-3.29, p < 0.05) were independent predictors of SCIWORET, respectively. Conclusions CCS, with or without cervical OPLL, are independent predictors of SCIWORET. These findings emphasize the importance of comprehensive spinal evaluation in trauma patients, particularly in those presenting with neurological symptoms following low-energy trauma. Our results suggest that patients with pre-existing CCS or OPLL may be at increased risk for SCI, even in the absence of obvious radiographic abnormalities.