CURRENTLY AVAILABLE CLASSIFICATION SYSTEMS FOR LOWER CERVICAL SPINE INJURIES. PART 2. SYSTEMATIC REVIEW OF STUDIES ON THE RELIABILITY AND REPRODUCIBILITY OF EXISTING SCALES

Abstract

The study objective is to review the Russian and foreign studies and to identify an optimal classification system for lower cervical spine injuries. Materials and methods. This systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA). We conducted a search for articles published in English (PubMed database) and Russian (eLIBRARY.ru). The inclusion criteria were as follows: available full text, patient age ≥18 years, and information on one of the validation phases for classifications according to L. Audige et al. Results. A total of 30 articles were eligible. Of them, 3 studies were published in Russian (by one group of authors); however, they didn’t contain required statistical parameters and had duplicated data; therefore, they were excluded from the analysis. Out of 27 articles published in English, 8 articles met all the criteria and were included into the systematic review. The AOSpine and Subaxial Injury Classification Systems demonstrated the highest reliability and reproducibility of the results. The Allen–Fergusson classification has lower intraobserver and interobserver agreement coefficients, but it can give a clearer visual representation of injuries. We also assessed J. Harris classification system. The reliability of the scale developed by С. Argenson et al. was not evaluated. The analyzed publications contained no data for full evaluation of the Cervical Spine Injury Severity Score. Our analysis clearly demonstrated the need for a more thorough evaluation of all available scales and classifications. This study should be multicenter and involve experts with different levels of experience (from residents to experienced spinal surgeons). Moreover, it should analyze not only the reproducibility of individual classifications, but also the aspects of learning and the relationship between individual scales and systems. The main study limitations included insufficient number of publications, small sample sizes, heterogeneity of groups, and differences in the experience of experts. Conclusion. The AOSpine and Subaxial Injury Classification Systems are the most reliable classification systems. However, the data available in literature is not sufficient for a full comparison of all existing scales and systems. Further multicenter studies on the reliability of classifications are needed to select an optimal one.