中国血液净化

中国血液净化 ›› 2022, Vol. 21 ›› Issue (06): 408-412.doi: 10.3969/j.issn.1671-4091.2022.06.006

• 临床研究 • 上一篇    下一篇

维持性血液透析患者透析中低血压风险预测模型的构建及验证

郭雪梅   朱朕男   罗佳懿    张和平   敬剑英   

  1. 川北医学院附属医院肾内科
  • 收稿日期:2021-11-11 修回日期:2022-03-16 出版日期:2022-06-12 发布日期:2022-06-12
  • 通讯作者: 郭雪梅 752685221@qq.com E-mail:752685221@qq.com
  • 基金资助:
    四川省卫生健康委重点研究项目(19ZD006)

Construction and validation of a prediction model for the risk of intradialytic hypotension in maintenance hemodialysis patients

  • Received:2021-11-11 Revised:2022-03-16 Online:2022-06-12 Published:2022-06-12
  • Contact: GUO Xue-mei 752685221@qq.com E-mail:752685221@qq.com

PDF

365

摘要： 目的 探讨维持性血液透析(maintenance hemodialysis，MHD)患者发生透析中低血压(intradialytic hypotension，IDH)的影响因素，建立风险预测模型，并对预测效果进行评价。 方法 选取2020年9月～2021年8月在川北医学院附属医院血液净化中心行MHD的222例患者为研究对象，按照6:4的比例随机分为建模组和验证组，参照美国肾脏病基金会肾脏病预后质量指南IDH定义，将建模组分为低血压组和无低血压组，采用单因素分析IDH发生的危险因素，将P＜0.05的因素进行多因素Logistic回归分析，构建风险预测模型，采用Hosmer-Lemeshow检验模型的拟合优度，ROC曲线检测模型的预测价值，验证组对该模型进行验证。 结果 维持性血液透析患者透析中低血压发生率为40.09%，其中，透析前收缩压(OR =0.959，95% CI：0.932～0.988，P =0.005)、血红蛋白(OR=0.962，95% CI：0.929～0.996，    P=0.030)、血镁(OR =79.558，95% CI：2.644～2393.594，P =0.012)、合并糖尿病(OR=16.066，95% CI：4.914～52.530，P＜0.001)、合并高血压(OR =5.221，95% CI：1.661～16.409，P=0.005)是透析中低血压的影响因素。Hosmer-Lemeshow检验P =0.718，ROC曲线下面积(AUC)为0.922，最佳临界值为0.436，灵敏度为0.83，特异度为0.887，约登指数为0.717，实际应用的正确率为83.15%。 结论 该模型能较好地预测维持性血液透析患者透析中低血压的发生风险，为临床医护人员评估维持性血液透析患者透析中低血压提供参考。

关键词: 血液透析, 透析中低血压, 影响因素, 风险预测模型

Abstract: Objective  To investigate the influencing factors for intradialytic hypotension (IDH) in maintenance hemodialysis (MHD) patients, then to construct a risk prediction model and to evaluate the prediction effect of this model.  Methods  A total of 222 MHD patients in the Blood Purification Center of The Affiliated Hospital of North Sichuan Medical College from September 2020 to August 2021 were enrolled in this study as the study subjects. They were randomly divided into modeling group and modeling validation group in the patient number ratio of 6:4. Patients in the modeling group were further divided into IDH group and non-IDH group based on the definition of IDH in the National Kidney Foundation Kidney Disease Prognostic Quality Guidelines. Univariate analysis was used to analyze the risk factors for IDH. Multivariate logistic regression analysis was performed to obtain the factors with P<0.05 for construction of the risk prediction model. Hosmer-Lemeshow test model was used to test the goodness of fit of the model. ROC curve was used to examine the predictive value of the model. The effectiveness of the constructed model was then evaluated in the modeling validation group.  Results  The incidence of IDH was 40.09% in the MDH patients. Predialysis systolic blood pressure (OR=0.959, 95% CI=0.932~0.988, P=0.005), hemoglobin (OR=0.962, 95% CI=0.929~0.996, P=0.030), magnesium (OR=79.558, 95% CI=2.644~2393.594, P=0.012), diabetes (OR=16.066, 95% CI=4.914~52.530, P<0.001), and hypertension (OR=5.221, 95% CI=1.661~16.409, P=0.005) were the influencing factors for IDH. Hosmer-Lemeshow test gave the result of P=0.718. When the optimal cut-off value was set at 0.436, the area under the ROC curve (AUC) was 0.922, the sensitivity was 0.83 and the specificity was 0.887. The Youden index was 0.717, and the correct rate of practical application was 83.15%.  Conclusion The model we constructed can better predict the risk of IDH in MDH patients and provides a reference for clinicians to evaluate IDH in MHD patients.

Key words: Hemodialysis, Intradialytic hypotension, Influencing factor, Risk prediction model

中图分类号: