摘要
【目的】构建并验证高龄患者髋关节置换术后预后不良的预测模型。【方法】选取2019年2月至2022年7月在本院行髋关节置换术的85例高龄患者,随访1年,根据高龄患者髋关节置换术后预后情况分为预后不良组与预后良好组。比较两组临床资料;采用二元Logistic回归法分析高龄患者髋关节置换术后预后不良的影响因素;采用Logistic法构建高龄患者髋关节置换术后预后不良的预测模型;采用Hosmer-Lemeshow检验评估拟合度,绘制受试者工作特征(ROC)曲线评估预测模型区分度。【结果】随访1年,85例患者中失访2例,随访率为97.65%,其中预后不良22例,预后不良发生率为26.51%(22/83)。预后不良组手术严重度评分(P-POSSUM评分)≥15分、康复介入时间为术后7 d后、骨密度<-3 SD、控制营养状况(CONUT)评分≥5分、单核细胞与淋巴细胞比值(MLR)≥1.13比例及骨型碱性磷酸酶(BLAP)、甲状腺激素(PTH)水平均高于预后良好组(P<0.05)。二元Logistic回归分析结果显示:P-POSSUM评分、康复介入时间、骨密度、CONUT评分、MLR为高龄患者髋关节置换术后预后不良的影响因素(P<0.05)。以上述影响因素作为预测变量,建立列线图预测模型,各因素总分76~426分,对应风险率0.05~0.80。列线图模型的C-index指数为0.825,预测高龄患者髋关节置换术后预后不良风险的校正曲线趋近于理想曲线(P>0.05)。ROC曲线分析结果显示:列线图模型预测高龄患者髋关节置换术后预后不良的灵敏度为90.91%,特异度为88.52%,曲线下面积(AUC)为0.903。【结论】P-POSSUM评分、康复介入时间、骨密度、CONUT评分、MLR与高龄患者髋关节置换术后预后不良风险有关,构建列线图模型有助于早期甄别高龄患者髋关节置换术后预后不良的风险。
【Objective】To establish and validate a prediction model for poor prognosis after total hip replacement in older adult patients.【Methods】A total of 85 elderly patients who underwent hip arthroplasty from February 2019 to July 2022 were selected.After 1 year follow-up,patients were divided into the poor prognosis group and the good prognosis group according to the prognosis of older adult patients after hip arthroplasty.By comparing the clinical data of the two groups,the influencing factors of the poor prognosis after total hip replacement in elderly patients were analyzed by binary Logistic regression method,the prediction model of the poor prognosis after total hip replacement in older adult patients was constructed by Logistic method,and the fitting degree was evaluated by Hosmer-Lemeshow test.Receiver operating characteristic curve(ROC)was plotted to evaluate the prediction model differentiation.【Results】After 1 year of follow-up,2 cases of the 85 patients were lost to follow-up,the follow-up rate was 97.65%.Among them,22 patients had poor prognosis,the incidence of poor prognosis was 26.51%(22/83).The surgical severity score(P-POSSUM score)≥15 points,rehabilitation intervention time of 7 days after surgery,bone density<-3 SD,controlled nutritional status(CONUT)score≥5 points,monocyte-to-lymphocyte ratio(MLR)≥1.13,and bone alkaline phosphatase(BLAP)and thyroid hormone(PTH)levels in the poor prognosis group were significantly higher than those in the good prognosis group(P<0.05).Binary Logistic regression analysis showed that P-POSSUM score,rehabilitation intervention time,bone mineral density,CONUT score and MLR were the influencing factors for poor prognosis after total hip replacement in older adult patients(P<0.05).Taking the above influencing factors as predictive variables,a nomogram prediction model was established.The total score of each factor ranged from 76 to 426,and the corresponding risk rate ranged from 0.05 to 0.80.The C-index of the nomogram model was 0.825,and the correction curve for predicting the risk of poor prognosis after total hip replacement in older adult patients was close to the ideal curve(P>0.05).ROC curve results showed that the sensitivity and specificity of the nomogram model in predicting total poor prognosis after hip replacement in older adult patients were 90.91%and 88.52%,respectively.The area under the curve(AUC)was 0.903.【Conclusion】P-POSSUM score,rehabilitation intervention time,bone mineral density,CONUT score and MLR are associated with the risk of poor prognosis after total hip arthroplasty in older adult patients.The construction of a nomogram model is helpful for early screening of the risk of poor prognosis after total hip arthroplasty in older adult patients.
作者
易鑫明
万顺新
张杰
陈明波
YI Xinming;WAN Shunxin;ZHANG Jie(Department of Orthopedics,Affiliated Hospital of Xinyang Vocational and Technical College,Xinyang Henan 464000)
出处
《医学临床研究》
CAS
2024年第8期1148-1151,1155,共5页
Journal of Clinical Research
关键词
关节成形术
置换
髋
老年人
预后
Arthroplasty,Replacement,Hip
Aged
Prognosis
