An updated clinical prediction model of protein-energy wasting for hemodialysis patients

Si Chen; Xiaoyan Ma; Xun Zhou; Yi Wang; WeiWei Liang; Liang Zheng; Xiujuan Zang; Xiaobin Mei; Yinghui Qi; Yan Jiang; Shanbao Zhang; Jinqing Li; Hui Chen; Yingfeng Shi; Yan Hu; Min Tao; Shougang Zhuang; Na Liu

doi:10.3389/fnut.2022.933745

An updated clinical prediction model of protein-energy wasting for hemodialysis patients

Front Nutr. 2022 Dec 6:9:933745. doi: 10.3389/fnut.2022.933745. eCollection 2022.

Authors

Si Chen¹, Xiaoyan Ma¹, Xun Zhou¹, Yi Wang¹, WeiWei Liang¹, Liang Zheng², Xiujuan Zang³, Xiaobin Mei⁴, Yinghui Qi⁵, Yan Jiang³, Shanbao Zhang⁵, Jinqing Li¹, Hui Chen¹, Yingfeng Shi¹, Yan Hu¹, Min Tao¹, Shougang Zhuang^{1

6}, Na Liu¹

Affiliations

¹ Department of Nephrology, Shanghai East Hospital, Tongji University School of Medicine, Shanghai, China.
² Key Laboratory of Arrhythmias of the Ministry of Education of China, Research Center for Translational Medicine, Shanghai East Hospital, Tongji University School of Medicine, Shanghai, China.
³ Department of Nephrology, Shanghai Songjiang District Central Hospital, Shanghai, China.
⁴ Department of Nephrology, Shanghai Gongli Hospital, Shanghai, China.
⁵ Department of Nephrology, Shanghai Punan Hospital, Shanghai, China.
⁶ Department of Medicine, Rhode Island Hospital and Alpert Medical School, Brown University, Providence, RI, United States.

Abstract

Background and aim: Protein-energy wasting (PEW) is critically associated with the reduced quality of life and poor prognosis of hemodialysis patients. However, the diagnosis criteria of PEW are complex, characterized by difficulty in estimating dietary intake and assessing muscle mass loss objectively. We performed a cross-sectional study in hemodialysis patients to propose a novel PEW prediction model.

Materials and methods: A total of 380 patients who underwent maintenance hemodialysis were enrolled in this cross-sectional study. The data were analyzed with univariate and multivariable logistic regression to identify influencing factors of PEW. The PEW prediction model was presented as a nomogram by using the results of logistic regression. Furthermore, receiver operating characteristic (ROC) and decision curve analysis (DCA) were used to test the prediction and discrimination ability of the novel model.

Results: Binary logistic regression was used to identify four independent influencing factors, namely, sex (P = 0.03), triglycerides (P = 0.009), vitamin D (P = 0.029), and NT-proBNP (P = 0.029). The nomogram was applied to display the value of each influencing factor contributed to PEW. Then, we built a novel prediction model of PEW (model 3) by combining these four independent variables with part of the International Society of Renal Nutrition and Metabolism (ISRNM) diagnostic criteria including albumin, total cholesterol, and BMI, while the ISRNM diagnostic criteria served as model 1 and model 2. ROC analysis of model 3 showed that the area under the curve was 0.851 (95%CI: 0.799-0.904), and there was no significant difference between model 3 and model 1 or model 2 (all P > 0.05). DCA revealed that the novel prediction model resulted in clinical net benefit as well as the other two models.

Conclusion: In this research, we proposed a novel PEW prediction model, which could effectively identify PEW in hemodialysis patients and was more convenient and objective than traditional diagnostic criteria.

Keywords: hemodialysis; impact factors; nomogram; prediction model; protein-energy wasting.