Evaluation and clinical application of the subaxial cervical spine injury classification system

Abstract

Objective To evaluate the interobserver reliability and intraobserver reproducibility of the subaxial cervical spine injury classification system (SLIC) and identify its guidance on clinical treatment of spinal cord injuries. Methods Thirty consecutive patients with subaxial cervical injury ad-mitted into our hospital between December 2007 and July 2008 were enrolled in this study and underwent neurological examinations followed by X-ray, CT, and MRI examinations. SLIC was evaluated in aspects of fracture morphology, disco-ligamentous complex (DLC) status and ncurologic status. According to the morphology in image, the fractures were classified into normal, compression, burst, distraction and trans-lation/rotation. According to injury severity, DLC was sorted into intact, indeterminate and disrupted types. Neurologic status was classified into intact, nerve root injury, complete/incomplete spinal cord in-jury and continuous spinal cord compression. The clinical and radiographic data of 30 patients with subax-ial cervical injury were evaluated prospectively by four groups of surgeons to count SLIC scores and decide treatment based on SLIC scores. The evaluation was done again three months later. Cohen' s unweighted kappa coefficients were calculated for the key parameters of each classification system ( morphology, neurologic status, DLCstatus, total score, and proposed management) to analyze the reliability and repreducibility of SLIC. Based on SLIC, the treatment method was selected and the neurological function recovery and the complications observed. Results The interrater Kappa statistics of all subgroups ( morphology, DLCstatus, total score, proposed management) were within the range of moderate to substantial reliability (0.47-0.69). Kappa coefficient was 0.83 for neurologic status, with high reliability. There was no statistical difference upon Kappa values in two evaluations. The interrater Kappa statistics of all sub-groups ( morphology, DLCstatus, total score, proposed management) were within the range of moderate to substantial reproducibility (0.53-0.78). Kappa coefficient was 0.89 for neurologic stares, with high repreducibility. The recovery rate of neuroiogie function was 79.2%. There was no aggravation of neurological status, with low incidence of complications. Conclusions SLIC has advantages of high reliability and reproducibility, simple use, exact evaluation and can facilitate decision-making in treatment of subaxial cervical spine injures. Key words: Spinal injuries; Cervical vertebrae; Diagnostic imaging; Classification system