Reliability of the New AOSpine Classification System for Upper Cervical Traumatic Injuries

Abstract BACKGROUND The new AOSpine Upper Cervical Classification System (UCCS) was recently proposed by the AOSpine Knowledge Forum Trauma team to standardize the treatment of upper cervical traumatic injuries (UCI). In this context, evaluating its reliability is paramount prior to clinical use. OBJECTIVE To evaluate the reliability of the new AOSpine UCCS. METHODS A total of 32 patients with UCI treated either nonoperatively or with surgery by one of the authors were included in the study. Injuries were classified based on the new AO UCCS according to site and injury type using computed tomography scan images in 3 planes by 8 researchers at 2 different times, with a minimum interval of 4 wk between assessments. Intra- and interobserver reliability was assessed using the kappa index (K). Treatment options suggested by the evaluators were also assessed. RESULTS Intraobserver agreement for sites ranged from 0.830 to 0.999, 0.691 to 0.983 for types, and 0.679 to 0.982 for the recommended treatment. Interobserver analysis at the first assessment was 0.862 for injury sites, 0.660 for types, and 0.585 for the treatment, and at the second assessment, it was 0.883 for injury sites, 0.603 for types, and 0.580 for the treatment. These results correspond to a high level of agreement of answers for the site and type analysis and a moderate agreement for the recommended treatment. CONCLUSION This study reported an acceptable reproducibility of the new AO UCCS and safety in recommending the treatment. Further clinical studies with a larger patient sample, multicenter and international, are necessary to sustain the universal and homogeneity quality of the new AO UCCS.