Interobserver reliability of the AO/OTA and Luo three-column classification systems for tibial plateau fractures and their impact on surgical approach selection.

Abstract

This study aims to compare the interobserver reliability of the traditional two-dimensional AO/OTA classification with the more recent three-dimensional Luo three-column classification for tibial plateau fractures. Furthermore, it evaluates the impact of both systems on the surgical approach selection, particularly examining Luo et al.'s hypothesis that the three-column classification encourages increased consideration of the posterior column during preoperative planning. However, this hypothesis has not been evaluated yet, leaving a research gap regarding its influence in practice on surgical approach selection. Fifteen cases of tibial plateau fractures were retrospectively analyzed by nine trauma surgeons using radiographs and CT scans. Fractures were classified according to the AO/OTA and Luo classifications, and preferred surgical approaches for definitive fixation were determined. Interobserver reliability was assessed using Fleiss' kappa and interpreted according to the categorical rating by Landis and Koch. Additionally, a chi-square test was performed to evaluate statistical significance in the surgical approach selection. Both classification systems showed overall substantial reliability (k_AO = 0.63; k_Luo = 0.67). The difference in agreement for surgical approach groups between the two classifications was 0.11 (k_{AO_approach} = 0.37; k_{Luo_approach} = 0.48). The posterior approach group was not selected significantly more often using the Luo three-column classification compared to the AO/OTA classification (p = 0.543). No significant difference in interobserver reliability or in the choice of surgical approach was observed between the AO/OTA and Luo classifications.