影响踝关节骨折术后疗效的多因素回归分析

发布时间：2019-05-14 08:52

【摘要】：目的:分析影响踝关节术后疗效的各种相关影响因素,总结相关经验,为临床医师治疗踝关节骨折提供参考。方法:分析于2012年6月~2016年3月就诊于兰州大学第二医院骨科并行手术治疗的踝关节骨折患者123例,根据患者的随访资料,选取可能影响踝关节骨折手术疗效的因素,(1)一般资料和既往情况:性别、年龄、是否患有糖尿病、吸烟史;(2)受伤情况:受伤类型、骨折分型(AO-Danish-Weber)、开放性损伤、骨折部位皮肤张力性水疱、血管损伤、神经损伤;(3)术前准备:踝关节CT检查、是否行跟骨牵引、受伤时间至行手术时间;(4)手术情况:手术方式;(5)术后康复及并发症:术后服用钙剂、术后下地负重时间、伤口愈合情况。先行各可能相关因素的单因素Logistic回归分析,对于分类资料单因素Logistic回归等同于卡方检验,以确定单因素对结果的影响而不考虑各因素之间的相互影响;根据单因素Logistic回归分析结果筛选出有统计学差异的影响因素,再将这些可能的影响因素纳入多因素Logistic逐步回归分析,以筛选出主要影响踝关节骨折术后疗效的因素。结果:123例患者术后均获得6~24月随访,平均时间(15.19±6.12)月,根据AO-Danish-Weber骨折分型将骨折分类,其中A型骨折34例,B型骨折45例,C型骨折44例;末次随访时根据Mazur踝关节功能评分标准评价,优30例,良47例,可29例,差17例,根据单因素Logistic回归分析发现年龄、受伤类型、骨折分型、是否行跟骨牵引、是否行踝关节CT检查、是否为开放性损伤、手术方式差异有统计学意义,P值、OR值及95%可信区间分别为:0.001,2.686(1.575~4.579);0.001,3.862(2.245~6.646);0.001,4.871(2.606~9.103);0.010,3.000(1.306~6.893);0.004,3.052(1.427~6.528);0.002,0.262(0.112~0.614);0.001,3.974(2.065~7.649)。再行这些因素的多因素logistic逐步回归分析发现年龄、受伤类型、骨折分型、是否行跟骨牵引、是否有踝关节CT、是否为开放性损伤、手术方式差异有统计学意义,均为影响踝关节骨折术后踝关节功能的影响因素。P值、OR值及95%可信区间分别为0.014,2.688(1.225~5.898);0.004,3.334(1.453~7.651);0.025,3.048(1.148~8.092);0.027,4.099(1.174~14.313);0.039,3.499(1.064~11.509);0.023,0.176(0.039~0.784);0.049,2.798(1.003~7.809)。结论:年龄、骨折分型、受伤类型、是否为开放性损伤、是否行跟骨骨牵引、踝关节CT检查、手术方式为影响踝关节骨折术后关节功能的主要影响因素。正确判断骨折的类型、完善术前检查明确骨折位置、骨折块的大小、有无伴随韧带损伤,积极行术前辅助治疗,改善软组织条件,可以增强术后踝关节功能,最大可能的防止创伤性关节炎的发生。
[Abstract]:Objective: to analyze the related factors affecting the curative effect of ankle joint operation, and to summarize the relevant experience, so as to provide reference for clinicians in the treatment of ankle fracture. Methods: 123 patients with ankle fracture treated by orthopaedic surgery in the second Hospital of Lanzhou University from June 2012 to March 2016 were analyzed. According to the follow-up data of the patients, the factors that might affect the curative effect of ankle fracture were selected. (I) General information and past information: sex, age, diabetes, history of smoking; (2) injury: injury type, fracture classification (AO-Danish-Weber), open injury, skin tension blister at fracture site, vascular injury, nerve injury; (3) preoperative preparation: ankle joint CT examination, calcaneal traction, injury time to operation time, (4) operation condition: operation method; (5) postoperative rehabilitation and complications: taking calcium after operation, loading time after operation, wound healing. According to the single factor Logistic regression analysis of each possible related factor, the single factor Logistic regression of classified data is equivalent to chi-square test in order to determine the influence of single factor on the result without considering the interaction between each factor. According to the results of univariate Logistic regression analysis, the influencing factors with statistical difference were selected, and then these possible influencing factors were included in multivariate Logistic stepwise regression analysis in order to screen out the main factors that affected the postoperative curative effect of ankle fracture. Results: all the 123 patients were followed up for 6 ~ 24 months with an average time of (15.19 卤6.12) months. According to the classification of AO-Danish-Weber fracture, 34 cases of type A fracture, 45 cases of type B fracture and 44 cases of type C fracture were classified. According to Mazur ankle function score, 30 cases were excellent, 47 cases were good, 29 cases were fair and 17 cases were poor. According to univariate Logistic regression analysis, age, type of injury, fracture type and calcaneal traction were found. Whether the ankle joint was examined by CT and whether it was open or not, the difference of operation mode was statistically significant. The P value, OR value and 95% confidence interval were 0.0012.686 (1.575 鈮，

本文编号：2476576