RESUMEN
Ankle injuries are responsible for more than 5 million emergency department visits each year. The AO and Lauge-Hansen classification systems are widely used in the clinical diagnosis of ankle injuries. This study aimed to analyze the intraobserver reliability and interobserver reproducibility of the AO and Lauge-Hansen classification systems. In addition, the authors explored the differences among physicians' classification responses and evaluated the clinical value for diagnosis. Fifty-six patients with an ankle injury with complete clinical and radiologic data were enrolled. The definition of injury type, the index score typing methods, and the specific study criteria were explained in detail. Five observers, who were orthopedic surgeons, determined the classifications according to both the AO and Lauge-Hansen systems. The classification was repeated 1 month later. Cronbach's alpha and Cohen's kappa test were used to determine interobserver reliability and intraobserver reproducibility. The physicians conducted 560 classifications (56 cases × 5 physicians × 2 times per patient). Average inter- and intraobserver kappa values for the AO system were 0.708 and 0.608, respectively. Average inter- and intraobserver kappa values for the Lauge-Hansen system were 0.402 and 0.398, respectively. Cronbach's alpha coefficient was 96.7% for the AO system and 76.0% for the Lauge-Hansen system. The Lauge-Hansen classification system is a comprehensive yet cumbersome system. Comparatively, the AO classification system is easier to understand. This study shows that the AO classification system has more reliability and reproducibility, and thus has more value in clinical practice, than the Lauge-Hansen classification system.