肾移植术后1年内非计划再入院Nomogram构建与评价:基于Lasso-Logistic回归模型。

IF 0.8 4区 医学 Q4 IMMUNOLOGY
Weiwei Cao , Bei Ding , Kejing Zhu , Li Ma , Minghuan Zhong , Yulin Niu
{"title":"肾移植术后1年内非计划再入院Nomogram构建与评价:基于Lasso-Logistic回归模型。","authors":"Weiwei Cao ,&nbsp;Bei Ding ,&nbsp;Kejing Zhu ,&nbsp;Li Ma ,&nbsp;Minghuan Zhong ,&nbsp;Yulin Niu","doi":"10.1016/j.transproceed.2025.05.012","DOIUrl":null,"url":null,"abstract":"<div><h3>Objective</h3><div>To develop and evaluate a risk prediction model for unplanned readmission within 1 year following kidney transplantation using Lasso-logistic regression.</div></div><div><h3>Methods</h3><div>Clinical data of kidney transplant recipients from the Department of Organ Transplantation at the Affiliated Hospital of Guizhou Medical University, spanning April 2017 to June 2023, were retrospectively analyzed. Initially, Lasso regression analysis was used to select predictive variables. Subsequently, logistic regression analysis was employed to construct a risk prediction model, which was presented as a nomogram. Bootstrap repeated sampling was conducted 1000 times for internal model validation. The comprehensive efficacy of the prediction model was assessed from four dimensions: discrimination, fit, calibration, and clinical benefit.</div></div><div><h3>Results</h3><div>The incidence of unplanned readmission within 1-year post-transplant was 36.48%. Serum creatinine, cystatin C, albumin, serum potassium, serum magnesium, drinking history, rejection, and length of stay were the predictors of unplanned readmission within 1 year after renal transplantation. The comprehensive ability of the risk prediction model for unplanned readmission within 1 year after renal transplantation was as follows: The area under the receiver operating characteristic curve of the nomogram model was 0.715 (95% CI: 0.673-0.757). The internal validation results showed that the corrected C-index was 0.700. The positive predictive value of the model was 0.533 and the negative predictive value was 0.785. The Hosmer–Lemeshow goodness of fit test result was <em>χ</em>² = 4.941, <em>P</em> = 0.764, indicating a satisfactory fit of the model. In the calibration curve, the actual fitting curve was well-fitted to the standard curve, and the model calibration ability was acceptable. The clinical decision curve confirmed the clinical value of the model and its positive impact on actual decision-making.</div></div><div><h3>Conclusions</h3><div>The constructed model demonstrates considerable predictive value for unplanned readmission within 1 year after kidney transplantation. It serves as a valuable tool for early clinical warning, enabling healthcare professionals to formulate personalized preventive strategies based on identified risk factors.</div></div>","PeriodicalId":23246,"journal":{"name":"Transplantation proceedings","volume":"57 6","pages":"Pages 989-1000"},"PeriodicalIF":0.8000,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Construction and Evaluation of a Nomogram for Unplanned Readmission Within 1 Year After Kidney Transplantation: Based on Lasso-Logistic Regression Model\",\"authors\":\"Weiwei Cao ,&nbsp;Bei Ding ,&nbsp;Kejing Zhu ,&nbsp;Li Ma ,&nbsp;Minghuan Zhong ,&nbsp;Yulin Niu\",\"doi\":\"10.1016/j.transproceed.2025.05.012\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><h3>Objective</h3><div>To develop and evaluate a risk prediction model for unplanned readmission within 1 year following kidney transplantation using Lasso-logistic regression.</div></div><div><h3>Methods</h3><div>Clinical data of kidney transplant recipients from the Department of Organ Transplantation at the Affiliated Hospital of Guizhou Medical University, spanning April 2017 to June 2023, were retrospectively analyzed. Initially, Lasso regression analysis was used to select predictive variables. Subsequently, logistic regression analysis was employed to construct a risk prediction model, which was presented as a nomogram. Bootstrap repeated sampling was conducted 1000 times for internal model validation. The comprehensive efficacy of the prediction model was assessed from four dimensions: discrimination, fit, calibration, and clinical benefit.</div></div><div><h3>Results</h3><div>The incidence of unplanned readmission within 1-year post-transplant was 36.48%. Serum creatinine, cystatin C, albumin, serum potassium, serum magnesium, drinking history, rejection, and length of stay were the predictors of unplanned readmission within 1 year after renal transplantation. The comprehensive ability of the risk prediction model for unplanned readmission within 1 year after renal transplantation was as follows: The area under the receiver operating characteristic curve of the nomogram model was 0.715 (95% CI: 0.673-0.757). The internal validation results showed that the corrected C-index was 0.700. The positive predictive value of the model was 0.533 and the negative predictive value was 0.785. The Hosmer–Lemeshow goodness of fit test result was <em>χ</em>² = 4.941, <em>P</em> = 0.764, indicating a satisfactory fit of the model. In the calibration curve, the actual fitting curve was well-fitted to the standard curve, and the model calibration ability was acceptable. The clinical decision curve confirmed the clinical value of the model and its positive impact on actual decision-making.</div></div><div><h3>Conclusions</h3><div>The constructed model demonstrates considerable predictive value for unplanned readmission within 1 year after kidney transplantation. It serves as a valuable tool for early clinical warning, enabling healthcare professionals to formulate personalized preventive strategies based on identified risk factors.</div></div>\",\"PeriodicalId\":23246,\"journal\":{\"name\":\"Transplantation proceedings\",\"volume\":\"57 6\",\"pages\":\"Pages 989-1000\"},\"PeriodicalIF\":0.8000,\"publicationDate\":\"2025-07-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Transplantation proceedings\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S0041134525002866\",\"RegionNum\":4,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q4\",\"JCRName\":\"IMMUNOLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Transplantation proceedings","FirstCategoryId":"3","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0041134525002866","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"IMMUNOLOGY","Score":null,"Total":0}
引用次数: 0

摘要

目的:应用Lasso-logistic回归建立肾移植术后1年内意外再入院的风险预测模型并进行评价。方法:回顾性分析2017年4月至2023年6月贵州医科大学附属医院器官移植科肾移植受者的临床资料。初步采用Lasso回归分析选择预测变量。随后,采用逻辑回归分析构建风险预测模型,该模型以nomogram表示。Bootstrap重复采样1000次,进行内部模型验证。从辨别性、拟合性、校正性和临床获益四个维度评价预测模型的综合疗效。结果:移植后1年内意外再入院发生率为36.48%。血清肌酐、胱抑素C、白蛋白、血清钾、血清镁、饮酒史、排斥反应和住院时间是肾移植术后1年内意外再入院的预测因素。肾移植术后1年内非计划再入院风险预测模型综合能力为:nomogram模型受者工作特征曲线下面积为0.715 (95% CI: 0.673-0.757)。内部验证结果表明,修正后的C-index为0.700。模型的阳性预测值为0.533,阴性预测值为0.785。Hosmer-Lemeshow拟合优度检验结果χ²= 4.941,P = 0.764,模型拟合满意。在标定曲线中,实际拟合曲线与标准曲线拟合良好,模型标定能力可接受。临床决策曲线证实了该模型的临床价值及其对实际决策的积极影响。结论:所建立的模型对肾移植术后1年内非计划再入院具有相当的预测价值。它是早期临床预警的宝贵工具,使医护人员能够根据已确定的风险因素制定个性化的预防策略。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Construction and Evaluation of a Nomogram for Unplanned Readmission Within 1 Year After Kidney Transplantation: Based on Lasso-Logistic Regression Model

Objective

To develop and evaluate a risk prediction model for unplanned readmission within 1 year following kidney transplantation using Lasso-logistic regression.

Methods

Clinical data of kidney transplant recipients from the Department of Organ Transplantation at the Affiliated Hospital of Guizhou Medical University, spanning April 2017 to June 2023, were retrospectively analyzed. Initially, Lasso regression analysis was used to select predictive variables. Subsequently, logistic regression analysis was employed to construct a risk prediction model, which was presented as a nomogram. Bootstrap repeated sampling was conducted 1000 times for internal model validation. The comprehensive efficacy of the prediction model was assessed from four dimensions: discrimination, fit, calibration, and clinical benefit.

Results

The incidence of unplanned readmission within 1-year post-transplant was 36.48%. Serum creatinine, cystatin C, albumin, serum potassium, serum magnesium, drinking history, rejection, and length of stay were the predictors of unplanned readmission within 1 year after renal transplantation. The comprehensive ability of the risk prediction model for unplanned readmission within 1 year after renal transplantation was as follows: The area under the receiver operating characteristic curve of the nomogram model was 0.715 (95% CI: 0.673-0.757). The internal validation results showed that the corrected C-index was 0.700. The positive predictive value of the model was 0.533 and the negative predictive value was 0.785. The Hosmer–Lemeshow goodness of fit test result was χ² = 4.941, P = 0.764, indicating a satisfactory fit of the model. In the calibration curve, the actual fitting curve was well-fitted to the standard curve, and the model calibration ability was acceptable. The clinical decision curve confirmed the clinical value of the model and its positive impact on actual decision-making.

Conclusions

The constructed model demonstrates considerable predictive value for unplanned readmission within 1 year after kidney transplantation. It serves as a valuable tool for early clinical warning, enabling healthcare professionals to formulate personalized preventive strategies based on identified risk factors.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
Transplantation proceedings
Transplantation proceedings 医学-免疫学
CiteScore
1.70
自引率
0.00%
发文量
502
审稿时长
60 days
期刊介绍: Transplantation Proceedings publishes several different categories of manuscripts, all of which undergo extensive peer review by recognized authorities in the field prior to their acceptance for publication. The first type of manuscripts consists of sets of papers providing an in-depth expression of the current state of the art in various rapidly developing components of world transplantation biology and medicine. These manuscripts emanate from congresses of the affiliated transplantation societies, from Symposia sponsored by the Societies, as well as special Conferences and Workshops covering related topics. Transplantation Proceedings also publishes several special sections including publication of Clinical Transplantation Proceedings, being rapid original contributions of preclinical and clinical experiences. These manuscripts undergo review by members of the Editorial Board. Original basic or clinical science articles, clinical trials and case studies can be submitted to the journal?s open access companion title Transplantation Reports.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信