维持性血液透析患者透析中低血压风险预测模型的构建

目的 分析维持性血液透析（maintenance hemodialysis,MHD）患者发生透析中低血压（intradialytic hypotension,IDH）的影响因素,构建MHD患者发生IDH的风险预测模型。 方法 选取福建省5所三级甲等医院行MHD的患者475例。通过Lasso回归筛选预测因子,采用多因素Logistic回归分析建立MHD患者发生IDH的风险预测模型,并使用列线图展示模型。采用ROC曲线、校准曲线评价模型的区分度及校准度,使用Bootstrap法对模型进行内部验证。结果 身高、原发病类型、超滤量、透析前收缩压及近期（≤1个月）频发IDH是MHD患者发生IDH的影响因素。列线图模型的ROC曲线下面积为0.830,灵敏度为71.8%,特异度为79.9%,约登指数为0.517,内部验证C-统计量为0.814。结论 该研究构建的模型预测效果较好,可为临床护理人员识别MHD患者发生IDH及实施预防性护理提供借鉴。

Development of a risk prediction model for intradialytic hypotension in patients with maintenance hemodialysis

Objective To explore risk factors of intradialytic hypotension（IDH） in patients with maintenance hemodialysis（MHD） and to develop a risk prediction model of it. Methods Totally 475 patients with MHD in 5 tertiary hospitals were selected. Lasso regression analysis was used to select predictors,and multivariate logistic regression analysis was used to develop the risk prediction model,and this was presented with nomogram. ROC curve and calibration curve were used to assess the discrimination and calibration of the model,and Bootstrap method was used for internal validation. Results The multivariate regression analysis showed that height,primary disease types,ultrafiltration,predialysis systolic blood pressure,and frequent intradialytic hypotension within 1 month were risk factors of IDH in patients with MHD. The area under the ROC curve of the nomogram model was 0.830,with the sensitivity of 71.8% and the specificity of 79.9%. The youden index was 0.517. The model internal verification showed the C-index of 0.814. Conclusion The model has satisfactory effects,which can provide references for nurses to identify high-risk patients and implement preventive care.