A worldwide cross-sectional survey was recently conducted. The purpose of this study was to evaluate the Upper Cervical Injury Classification System developed by the AO Spine based on a global sample of AO Spine members with regard to its accuracy in classification, interobserver reliability, and intraobserver repeatability. 

Most previous attempts to categorize injuries to the upper cervical spine have relied on descriptors rather than a hierarchical damage progression. Furthermore, several anatomical parts within the upper cervical spine have been prioritized for classifying upper cervical spine injuries. Therefore, all injuries to the upper cervical spine are accounted for in the AO Spine Upper Cervical Injury Classification System, focusing on a hierarchical development from isolated bone injuries (type A) through fracture dislocations (type C). 

Using computed CT scans, 275 members of the AO Spine organization classified 25 cases of upper cervical spine injury according to the AO Spine Upper Cervical Classification System. The verification took place twice, at 3-week intervals. The Pearson χ2 test was used to determine statistical significance between validation sets, and descriptive statistics were computed to show what percentage of observations were in agreement with the gold standard. 

The consistency of results between observers and the consistency of results within an observer were calculated using kappa coefficients (κ). Classification accuracy for upper cervical spine injuries by AO Spine members was 79.7% on assessment 1 (AS1) and 78.7% on assessment 2 (AS2). Overall, AS1 and AS2 had strong interobserver reliability (κ=0.63 and κ=0.61, respectively), and intraobserver reproducibility was very high (κ=0.70). In addition, there was a greater agreement between observers about the injury site (AS1: κ= 0.85 and AS2: κ=0.83) than the injury category (AS1: κ= 0.59 and AS2: κ=0.57). 

Significant interobserver agreement and intraobserver repeatability were found in the worldwide validation of the AO Spine Upper Cervical Injury Classification System. These findings lend credence to the AO Spine Upper Cervical Injury Classification System’s claim of generalizability.

Source: journals.lww.com/spinejournal/Fulltext/2022/11150/Global_Validation_of_the_AO_Spine_Upper_Cervical.1.aspx