目的：总结踝关节骨折CT表现及特点，分析CT图像与常用踝关节骨折分型之间的关系。方法： 回顾性调查369例成人踝关节骨折，分析踝穴上1 cm水平横断面CT图像及其特点，进行踝关节骨折X线分型，分析CT图像与骨折X线分型之间的关系。结果： 踝穴上1 cm水平横断面CT图像共有8种表现形式，以a、b、c分别指代腓骨骨折、后踝骨折和骨间韧带（interosseous tibiofibular ligament，IOL）断裂，369例踝关节骨折CT图像包括：0度（腓骨、后踝、IOL均完整）40例；Ⅰa度（腓骨骨折、后踝完整、IOL完整）60例，Ⅰb度（腓骨完整、后踝骨折、IOL完整）3例，Ⅰc度（腓骨完整、后踝完整、IOL断裂）26例；Ⅱab度（腓骨骨折、后踝骨折、IOL断裂）163例，Ⅱac度（腓骨骨折、后踝完整、IOL断裂）6例，Ⅱbc度（腓骨完整、后踝骨折、IOL断裂）61例；Ⅲ度（腓骨骨折、后踝骨折、IOL断裂）10例。根据DanisWeber骨折分型：A型骨折18例，B型骨折238例，C型骨折94例，腓骨无骨折19例。分析CT图像，A、B、C型骨折IOL断裂的发生率分别为0、5.9%、88.3%；CT图像与DanisWeber分型之间存在相关性，IOL断裂的发生率随DanisWeber分型严重程度而改变，经秩相关检验，差异有统计学意义（Spearman R=0.781，P<0.001）。CT图像能发现X线骨折分型无法判断的IOL断裂，其发生率是5.9%。结论： 踝穴上1 cm水平横断面CT图像在术前能够清楚判断下胫腓联合IOL损伤情况，与DanisWeber骨折分型存在良好的相关性，并能发现某些X线片中无法判断的IOL断裂情况。

Objective： To summarize the CT features of ankle fracture and to analyze the relationship between the CT images and the most commonly used ankle fracture classification. Methods: With 369 cases of adult ankle fractures analyzed retrospectively, the CT images 1 cm above the ankle joint and its characteristics, the Danis-Weber classification of ankle fracture were studied, and so was the relationship between CT images and the fracture classification. Results: There were 8 forms of CT images. With a, b, and c referred to the fibular fracture, posterior malleolar fracture and interosseous tibiofibular ligament (IOL) rupture respectively. 369 CT imges had 40 cases of 0 degree injury (fibula, posterior malleolus, IOL all intact); 60 cases of Ⅰa degree injury (fibular fracture, posterior malleolus and IOL intact), 3 cases of Ⅰb degree injury (fibula intact, posterior malleolus fracture, IOL intact), 26 cases of Ⅰc degree injury (fibula and posterior malleolus intact, IOL rupture); 163 cases of Ⅱab degree injury (fibula and posterior malleolus fractures, IOL intact), 6 cases of Ⅱac degree injury (fibular fracture, posterior malleolus intact, IOL rupture), 61 cases of Ⅱbc degree injury (fibula intact, posterior malleollar fracture, IOL rupture); 10 cases of Ⅲ degree injury (fibular fracture, posterior malleollar fracture and IOL rupture). According to the Danis-Weber classification, there were 18 cases of type A, 238 cases of type B, 94 cases of type C, and 19 cases without fibular fracture. The prevalence of IOL rupture were 0, 5.9%, and 88.3% in types A, B, and C respectively. There was a correlation between the CT image and Danis-Weber classification, the incidence of IOL rupture was changed with the severity of Danis-Weber classification, and the difference was statistically significant after the rank correlation test (Spearman R=0.781, P<0.001). IOL rupture not determined by the fracture classification was found with the CT images and the incidence was 5.9%. Conclusion: Cross-sectional CT images 1 cm above the ankle joint can clearly determine the IOL injury pre-operatively with a good correlation with the Danis-Weber fracture classification, IOL rupture unrecognized with the fracture classification can also be noticed with the CT image.