Prediction of survival and prognosis of maintenance hemodialysis patients based on nomogram
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
万方数据
目的:开发和验证一种易于使用的评估工具,用于预测维持性血液透析患者短期生存率.方法:回顾性分析2018年1月—2022年12月在长沙市第四医院透析中心行规律透析的维持性血液透析(maintenance hemodialysis,MHD)患者的临床资料.共纳入543名MHD患者,按照7:3的比例随机分为训练队列(380名)和验证队列(163名).使用最小绝对收缩和选择算子(least absolute shrinkage and selection operator,LASSO)回归和Cox多因素回归分析确定预测因素,并构建列线图.利用C指数、受试者工作曲线(receiver operating characteristic,ROC)、ROC曲线下面积(area under the curve,AUC)、校准曲线评估列线图性能.结果:Cox多因素回归分析结果显示年龄(HR=1.03,95％CI:1.01～1.05)、透析龄(HR=0.95,95％CI:0.93～0.96)、血清铁蛋白＜21.8 ng/mL(HR=0.68,95％CI:0.09～4.98)、274.66 ng/mL＜血清铁蛋白≤ 500 ng/mL(HR=1.72,95％CI:0.94～3.12)、血清铁蛋白＞500 ng/mL(HR=2.36,95％CI:1.50～3.72)有统计学意义.对列线图进行交叉验证,结果显示训练队列的C指数为0.802,验证队列的C指数为0.838,这说明该模型具有较好的区分度.训练队列和验证队列中的辨别能力较好(在训练队列和验证队列1年AUC分别为0.909和0.948;3年AUC分别为0.899和0.928).训练队列和验证队列的校准曲线显示校准良好,表明该模型具有较好的预测性能.结论:基于年龄、透析龄、血清铁蛋白这3个常见的临床指标,我们构建了 一个可以预测MHD患者1年和3年生存率的列线图,具有潜在的临床价值,可用于临床实践.
Objective To develop and validate an easy-to-use assessment tool for predicting short-term survival in main-tenance hemodialysis(MHD)patients.Method The clinical data of MHD patients undergoing regular dialysis at the Dialysis Center of the Fourth Hospital of Changsha City from January 2018 to December 2022 were retrospectively analyzed.A total of 543 MHD patients were enrolled and randomly allocated into the training cohort(380 patients)and the validation cohort(163 patients)at a 7:3 ratio.Least absolute shrinkage and selection operator(LASSO)regression and Cox multifactor regression analysis were employed to identify predictors and construct a nomogram.The nomogram's performance was assessed through the C-index,receiver operating curve(ROC),area under the curve(AUC),and calibration curve.Results Cox multivariate regression analysis revealed that age(HR=1.03,95％CI 1.01-1.05),duration of dialysis(HR=0.95,95％CI 0.93-0.96),se-rum ferritin＜21.8 ng/mL(HR=0.68,95％CI 0.09-4.98),274.66 ng/mL＜serum ferritin≤500 ng/mL(HR=1.72,95％CI 0.94-3.12),and serum ferritin＞500 ng/mL(HR=2.36,95％CI 1.50-3.72)were statistically significant.The cross-validation on the nomogram demonstrated a C-index of 0.802 for the training cohort and 0.838 for the validation cohort,indicating a strong discriminatory capability.Discrimination ability in both cohorts was favorable(the 1-year AUC in the training cohort and the validation cohort were 0.909 and 0.948,respectively;The 3-year AUC was 0.899 and 0.928,respectively).Calibration curves for both training and validation cohorts exhibited excellent alignment,underscoring the model's robust predictive performance.Conclusion Utilizing three widely employed clinical indicators—age,duration of dialysis,and serum ferritin.We developed a nomogram capable of predicting the 1-year and 3-year survival rates of MHD patients.This nomogram holds significant clinical potential and can be effectively applied in clinical practice.
NETL
NSTL
万方数据
