维持性血液透析患者骨骼肌质量下降预测模型的构建及对护理工作的启示

目的 构建维持性血液透析患者骨骼肌质量下降的预测模型,并验证模型的预测效果。方法 采用便利抽样法,选取2020年5月—8月在连云港市某三级甲等医院的304例维持性血液透析患者作为模型构建组,将骨骼肌质量下降组（n=74）和骨骼肌质量未下降组（n=230）的各项指标进行单因素分析,采用Logistic回归分析构建预测模型,并对模型进行简化,通过受试者工作特征曲线评估模型的预测效果。选取徐州市某三级甲等医院的309例维持性血液透析患者进行模型预测效果的外部验证。结果 该研究最终纳入性别、BMI、小腿围、透析前肌酐4个影响因素,并构建了预测模型,受试者工作特征曲线下面积为0.911,简化模型的受试者工作特征曲线下面积为0.911,实际应用的正确率为75.40%。结论 该研究构建的模型具有良好的预测效果,指标简单易得,可为医护人员早期识别并及时对高风险患者采取预防性干预措施提供借鉴。

Establishment and enlightenment of a predictive model for decreased skeletal muscle mass in maintenance hemodialysis patients

Objective To construct a predictive model of decreased skeletal muscle mass in maintenance hemodialysis patients and to apply it in clinical practice to evaluate the effects of the model. Methods A total of 304 maintenance hemodialysis patients in a tertiary hospital in Lianyungang were selected as the training set by convenience sampling method,and various indicators of the decreased skeletal muscle mass group（n=74） and the non-decreased skeletal muscle mass group（n=230） were analyzed by univariate analysis. Logistic regression analysis was used to construct the prediction model and the model was simplified it. The area under the ROC curve was used to test the model effects,and 309 maintenance hemodialysis patients in a tertiary hospital in Xuzhou were selected to externally verity the effects of the model. Results The study finally included 4 factors,namely gender,BMI,calf circumference and predialysis creatinine,to construct the risk prediction model. The area under the ROC curve of this model was 0.911;the area under the ROC curve of the simplified model was 0.911;the accuracy rate of practical application was 75.40%. Conclusion The risk prediction model constructed in this study can predict the risk of decreased skeletal muscle mass in patients undergoing maintenance hemodialysis,and all the indicators are simple and easy to obtain,thus providing references for medical staff to carry out early indenfication and take preventive care interventions for high-risk patients in the early stage.