Construction and validation of a Klotho-based machine learning model for predicting all-cause mortality in chronic kidney disease

Yating WANG; Jiachuan XIONG; Jinghong ZHAO

doi:10.16016/j.2097-0927.202312065

您当前的位置：

首页 >

文章列表页 >

Construction and validation of a Klotho-based machine learning model for predicting all-cause mortality in chronic kidney disease

Clinical Medicine | 更新时间：2026-01-09

- Construction and validation of a Klotho-based machine learning model for predicting all-cause mortality in chronic kidney disease
- Journal of Army Medical University Vol. 46, Issue 8, Pages: 859-867(2024)
- 作者机构：
  
  400037 重庆，陆军军医大学(第三军医大学) 第二附属医院肾内科，全军肾脏病中心，重庆市慢性肾脏病防治重点实验室
- 作者简介：
- 基金信息：
  
  the Key Project of Joint Regional Innovation and Development Fund of National Natural Science Foundation of China(U22A20279);the Key Project of Technological Innovation and Application Development of Chongqing(CSTB2023TIAD-KPX0060)
- DOI：10.16016/j.2097-0927.202312065
  CLC： R195.4;R319;R692.07
- Received：12 December 2023，
  
  Revised：2024-02-02，
  
  Published：30 April 2024
- 稿件说明：
移动端阅览
Yating WANG, Jiachuan XIONG, Jinghong ZHAO. Construction and validation of a Klotho-based machine learning model for predicting all-cause mortality in chronic kidney disease[J]. Journal of Army Medical University, 2024, 46(8): 859-867.
DOI：

Yating WANG, Jiachuan XIONG, Jinghong ZHAO. Construction and validation of a Klotho-based machine learning model for predicting all-cause mortality in chronic kidney disease[J]. Journal of Army Medical University, 2024, 46(8): 859-867. DOI： 10.16016/j.2097-0927.202312065.

摘要

目的

构建并验证包含血清Klotho蛋白在内的机器学习(machine learning，ML)模型来预测慢性肾脏病(chronic kidney disease，CKD)患者全因死亡的风险。

方法

研究采用回顾性队列研究设计方案。选取2012年2月7日至2019年10月18日我科收治的CKD1~5期非透析成年患者病例资料，按照7∶3分为训练集和内部验证集。将47项常规临床特征(包括血清Klotho蛋白)用于为模型提供变量信息，在训练集中利用单因素Cox回归筛选可能的风险因素、Lasso-Cox回归模型筛选风险因素、多因素Cox逐步回归构建列线图风险预测模型，内部验证评估模型性能。

结果

共400例病例资料纳入分析，训练集280例，验证集120例。训练集有52例死亡，228例存活；验证集有21例死亡，99例存活。全因死亡风险预测模型显示5年生存率受试者工作曲线下面积(area under the curve，AUC)在训练集和验证集分别为0.760(95%

：0.676，0.844)和0.788(95%

：0.679，0.897)，模型总体C指数在训练集和验证集分别为0.755(95%

：0.685，0.826)和0.720 (95%

：0.614

0.826)。单因素Cox回归分析结果显示，年龄、心血管疾病病史、胱抑素C、碱性磷酸酶、白蛋白、嗜酸性粒细胞、血红蛋白、补体C3、钙、C反应蛋白、肿瘤坏死因子-α(TNF-α)以及血清Klotho蛋白可能是全因死亡的预测因子(

0.05)。多因素Cox逐步回归最终筛选出年龄、白蛋白、补体C3及血清Klotho蛋白共4个独立预测因子(

0.05)，模型最终纳入了年龄、白蛋白、补体C3、血清Klotho蛋白共4个指标。

结论

成功构建并验证了基于血清Klotho蛋白的慢性肾脏病全因死亡预测机器学习模型；年龄大是CKD患者全因死亡的危险因素，白蛋白、补体C3及血清Klotho蛋白高是CKD患者全因死亡的保护因素。

Abstract

Objective

To develop and validate a machine learning (ML) model based on serum Klotho protein that can accurately predict all-cause mortality in chronic kidney disease (CKD) patients.

Methods

A retrospective cohort trial was conducted on all the non-dialysis adult patients diagnosed with CKD stages 1~5 in our department from February 7

2012 to October 18

2019. They were assigned into a training set and an internal validation set in a ratio of 7∶3. A total of 47 clinical features

including serum Klotho protein level

were used as variables to inform these models. Based on the training set

univariate Cox regression model was employed to screen out the possible risk factors for all-cause mortality

and Lasso-Cox regression model was further applied for the screening. Then multivariate Cox stepwise regression analysis was conducted to develop a nomogram risk prediction model for all-cause mortality

and the model performance was evaluated through internal validation.

Results

There were totally 400 patients enrolled in this trial

and 280 of them (including 52 dead and 228 survival) were assigned into the training set and other 120 (including 21 dead and 99 survival)into the validation set. For the constructed 5-year all-cause mortality risk prediction model

the area under the curve (AUC) value was 0.760 (95%

: 0.676~0.844) in the training set and 0.788 (95%

: 0.679~0.897) in the validation set

and the overall C-index was 0.755 (95%

: 0.685~0.826) and 0.720 (95%

: 0.614~0.826)

respectively in the 2 sets. Univariate Cox regression analysis showed that age

history of cardiovascular disease(CVD)

cystatin C(Cys-C)

alkaline phosphatase (ALP)

albumin

eosinophil (EOS) count

hemoglobin (Hb)

complement C3

calcium

C-reactive protein (CRP)

TNF-α and serum Klotho protein may be predictive factors for all-cause mortality (

0.05). Multivariate Cox stepwise regression analysis finally screened age

albumin

complement C3 and serum Klotho protein as independent predictors (

0.05). Based on these 4 predictors

a risk prediction model for all-cause mortality was constructed and validated.

Conclusion

A Klotho-based risk ML model for predicting all-cause mortality in CKD patients is successfully developed and validated. Advanced age is a risk factor

and higher albumin

complement C3 and serum Klotho protein levels are protective factors for all-cause mortality in CKD patients.

关键词

Keywords

references

HILL N R, FATOBA S T, OKE J L, et al . Global prevalence of chronic kidney disease—A systematic review and meta-analysis[J ] . PLoS One, 2016, 11(7): e0158765. DOI: 10.1371/journal.pone.0158765.

JAGER K J, KOVESDY C, LANGHAM R, et al . A single number for advocacy and communication-worldwide more than 850 million individuals have kidney diseases[J ] . Nephrol Dial Transplant, 2019, 34(11): 1803-1805. DOI: 10.1093/ndt/gfz174.

ZHANG L X, WANG F, WANG L, et al . Prevalence of chronic kidney disease in China: a cross-sectional survey[J ] . Lancet, 2012, 379(9818): 815-822. DOI: 10.1016/S0140-6736(12)60033-6.

LEVEY A S, ECKARDT K U, TSUKAMOTO Y, et al . Definition and classification of chronic kidney disease: a position statement from kidney disease: improving global outcomes (KDIGO)[J ] . Kidney Int, 2005, 67(6): 2089-2100. DOI: 10.1111/j.1523-1755.2005.00365.x.

ZOCCALI C, MALLAMACI F, ADAMCZAK M, et al . Cardiovascular complications in chronic kidney disease: a review from the European Renal and Cardiovascular Medicine Working Group of the European Renal Association[J ] . Cardiovasc Res, 2023, 119(11): 2017-2032. DOI: 10.1093/cvr/cvad083.

FOREMAN K J, MARQUEZ N, DOLGERT A, et al . Forecasting life expectancy, years of life lost, and all-cause and cause-specific mortality for 250 causes of death: reference and alternative scenarios for 2016-40 for 195 countriesand territories[J ] . Lancet, 2018, 392(10159): 2052-2090. DOI: 10.1016/S0140-6736(18)31694-5.

GANSEVOORT R T, CORREA-ROTTER R, HEMMELGARN B R, et al . Chronic kidney disease and cardiovascular risk: epidemiology, mechanisms , and prevention[J ] . Lancet, 2013, 382(9889): 339-352. DOI: 10.1016/S0140-6736(13)60595-4.

NORRIS K C, OLABISI O, BARNETT M E, et al . The role of vitamin D and oxidative stress in chronic kidney disease[J ] . Int J Environ Res Public Health, 2018, 15(12): 2701. DOI: 10.3390/ijerph15122701.

KOPPE L, FOUQUE D, KALANTAR-ZADEH K. Kidney cachexia or protein-energy wasting in chronic kidney disease: facts and numbers[J]. J Cachexia Sarcopenia Muscle, 2019, 10(3): 479-484. DOI: 10.1002/jcsm.12421.

EDMONSTON D, GRABNER A, WOLF M. FGF23 and klotho at the intersection of kidney and cardiovascular disease[J]. Nat Rev Cardiol, 2024, 21(1): 11-24. DOI: 10.1038/s41569-023-00903-0.

YANG K, WANG C, NIE L, et al . Klotho protects against indoxyl sulphate-induced myocardial hypertrophy[J ] . J Am Soc Nephrol, 2015, 26(10): 2434-2446. DOI: 10.1681/ASN.2014060543.

YANG K, YANG J X, BI X J, et al . Serum klotho, cardiovascular events, and mortality in nondiabetic chronic kidney disease[J ] . Cardiorenal Med, 2020, 10(3): 175-187. DOI: 10.1159/000506380.

THONGPRAYOON C, KAEWPUT W, CHOUDHURY A, et al . Is it time for machine learning algorithms to predict the risk of kidney failure in patients with chronic kidney disease?[J ] . J Clin Med, 2021, 10(5): 1121. DOI: 10.3390/jcm10051121.

NIEL O, BOUSSARD C, BASTARD P. Artificial intelligence can predict GFR decline during the course of ADPKD[J]. Am J Kidney Dis, 2018, 71(6): 911-912. DOI: 10.1053/j.ajkd.2018.01.051.

BARBIERI C, MOLINA M, PONCE P, et al . An international observational study suggests that artificial intelligence for clinical decision support optimizes anemia management in hemodialysis patients[J ] . Kidney Int, 2016, 90(2): 422-429. DOI: 10.1016/j.kint.2016.03.036.

ZHANG J J, FRIBERG I M, KIFT-MORGAN A, et al . Machine-learning algorithms define pathogen-specific local immune fingerprints in peritoneal dialysis patients with bacterial infections[J ] . Kidney Int, 2017, 92(1): 179-191. DOI: 10.1016/j.kint.2017.01.017.

TANGRI N, KITSIOS G D, INKER L A, et al . Risk prediction models for patients with chronic kidney disease: a systematic review[J ] . Ann Intern Med, 2013, 158(8): 596-603. DOI: 10.7326/0003-4819-158-8-201304160-00004.

LEVEY A S, STEVENS L A, SCHMID C H, et al . A new equation to estimate glomerular filtration rate[J ] . Ann Intern Med, 2009, 150(9): 604-612. DOI: 10.7326/0003-4819-150-9-200905050-00006.

KOBAYASHI I, SHIDARA K, OKUNO S, et al . Higher serum bone alkaline phosphatase as a predictor of mortality in male hemodialysis patients[J ] . Life Sci, 2012, 90(5-6): 212-218. DOI: 10.1016/j.lfs.2011.11.006.

MENON V, GREENE T, WANG X, et al . C-reactive protein and albumin as predictors of all-cause and cardiovascular mortality in chronic kidney disease[J ] . Kidney Int. 2005, 68(2): 766-772. DOI: 10.1111/j.1523-1755.2005.00455.x.

YANG K, DU C H, WANG X M, et al . Indoxyl sulfate induces platelet hyperactivity and contributes to chronic kidney disease-associated thrombosis in mice[J ] . Blood, 2017, 129(19): 2667-2679. DOI: 10.1182/blood-2016-10-744060.

LIU L, LIU Y, ZHANG Y, et al . High phosphate-induced downregulation of PPARγ contributes to CKD-associated vascular calcification[J ] . J Mol Cell Cardiol, 2018, 114: 264-275. DOI: 10.1016/j.yjmcc.2017.11.021.

TRAN N T D, BALEZEAUX M, GRANAL M, et al . Prediction of all-cause mortality for chronic kidney disease patients using four models of machine learning[J ] . Nephrol Dial Transplant, 2023, 38(7): 1691-1699. DOI: 10.1093/ndt/gfac316.

LANDRAY M J, EMBERSON J R, BLACKWELL L, et al . Prediction of ESRD and death among people with CKD: the Chronic Renal Impairment in Birmingham (CRIB) prospective cohort study[J ] . Am J Kidney Dis, 2010, 56(6): 1082-1094. DOI: 10.1053/j.ajkd.2010.07.016.

DEO R, DUBIN R F, REN Y, et al . Proteomic cardiovascular risk assessment in chronic kidney disease[J ] . Eur Heart J, 2023, 44(23): 2095-2110. DOI: 10.1093/eurheartj/ehad115.

ZELNICK L R, SHLIPAK M G, SOLIMAN E Z, et al . Prediction of incident atrial fibrillation in chronic kidney disease: the chronic renal insufficiency cohort study[J ] . Clin J Am Soc Nephrol, 2021, 16(7): 1015-1024. DOI: 10.2215/CJN.01060121.

SEGAL Z, KALIFA D, RADINSKY K, et al . Machine learning algorithm for early detection of end-stage renal disease[ J ] . BMC Nephrol, 2020, 21(1): 518. DOI: 10.1186/s12882-020-02093-0.

Views

293

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Risk factors for postoperative respiratory failure in patients undergoing cardiovascular surgery and construction of a prediction model

Construction and validation of a prediction model for central lymph node metastasis of papillary thyroid carcinoma based on contrast-enhanced venous phase CT radiomics

Pregnancy probability prediction models based on 5 machine learning algorithms and comparison of their performance

Construction and validation of a risk prediction model for high altitude de-acclimatization syndrome

Construction of a prediction model of histologic chorioamnionitis based on machine learning

Related Author

ZHOU Fuchao

LIU Xiang

LU Kaizhi

SHAO Jianliang

HE Xingyun

LIU Chen

DU Junze

LI Qian

Related Institution

Department of Anesthesiology， First Affiliated Hospital， Army Medical University （Third Military Medical University）

Department of Radiology

Department of Breast and Thyroid Surgery， First Affiliated Hospital， Army Medical University （Third Military Medical University）， Chongqing

Department of Radiology， No. 9

Hospital of PLA Army

⁰