牛津膝关节评分与膝关节单髁置换术术后满意度的相关性研究

Correlation between Oxford knee score and satisfaction after unicompartmental knee arthroplasty

目的:比较手术前后牛津膝关节评分与膝关节单髁置换术后患者满意度的相关性以及确定不同满意度对应的牛津膝关节评分阈值。方法:选取2017年1月至2018年6月行膝关节单髁置换术患者88例,其中男24例,女64例;年龄52~77(65.39±7.33)岁。所有患者在术前和术后6个月完成牛津膝关节评分问卷以及满意度调查问卷。采用Spearman秩相关检验分析比较牛津膝关节评分和满意度之间的相关性,以满意度为金标准,绘制ROC曲线判断手术前后牛津膝关节评分对患者满意度的评判价值,并计算约登指数探索其最佳阈值。结果:术前牛津膝关节评分和满意度之间无明显相关性(r=-0.058,P>0.05);术后牛津膝关节评分和满意度之间呈中度正相关(r=0.51,P<0.05),即术后牛津膝关节评分越高,患者满意度越高。术前牛津膝关节评分的ROC曲线下面积为0.55[95%CI(0.40,0.70),P>0.05],无统计学意义;术后牛津膝关节评分的ROC曲线下面积为0.95[95%CI (0.90,0.99),P<0.05],有统计学意义。术后牛津膝关节评分阈值为35.5时,评判满意度的敏感度为0.86,特异度为1.00,约登指数最大,为0.86。以35.5分为阈值比较高分组与低分组患者满意率,差异有统计学意义(P<0.001)。结论:术前牛津膝关节评分与膝关节单髁置换术后患者满意度相关性差,其在预测患者满意度方面不具有准确性,术后牛津膝关节评分与患者满意度具有明显相关性,术后牛津膝关节评分在35.5分及以上的患者可认为对手术效果满意。同时,结果证实牛津膝关节评分在对全膝关节置换术和膝关节单髁置换术后患者主观满意度的评价具有相同的阈值。

Objective:To compare correlation between preoperative and postoperative Oxford knee scores and patient satisfaction after unicompartmental knee arthroplasty and to identify Oxford knee score threshold for different satisfactions.Methods:Totally 88 patients with unicompartmental knee arthroplasty were enrolled from January 2017 to June 2018,including 24 males and 64 females,aged from 52 to 77 years old with an average of(65.39±7.33) years old.All patients completed Oxford Knee Score Questionnaire and Satisfaction Questionnaire before operation and 6 months after operation.Correlation between Oxford knee score and satisfaction was analyzed by Spearman rank correlation test,the satisfaction degree was used as the gold standard,and ROC curves were drawn to judge the value of knee score on patient satisfaction before and after operation.Yoden index were used to explore its optimal threshold.Results:There was no significant correlation between preoperative Oxford knee score and satisfaction before operation(r=-0.058,P>0.05).There was a moderate positive correlation between postoperative Oxford knee score and satisfaction(r=0.51,P<0.05),it means that the higher Oxford knee score after surgery,the higher the patients satisfaction.The area under ROC.curve for preoperative Oxford knee score was 0.55 [95%CI(0.40,0.70),P>0.05],which had not significant difference;the area under ROC curve of postoperative Oxford knee joint score was 0.95[95%CI(0.90,0.99),P<0.05],which had significant difference.When postoperative Oxford knee score was 35.5,the sensitivity of satisfaction was 0.86,the specificity was 1.00,and Yoden index was the largest(0.86),which could be used as a threshold for judging patient satisfaction,and had significant differences.Conclusion:Preoperative Oxford knee score was poorly correlated with patient satisfaction after knee joint replacement.It was not correct in predicting patient satisfaction.Postoperative Oxford knee score was significantly correlated with patient satisfaction.Patients with a knee score of 35.5 or above may be considered satisfied with surgical outcome.At the same time,we confirm that Oxford knee score has the same threshold for evaluation of subjective satisfaction after total knee arthroplasty and unicompartmental knee arthroplasty.