Xu You1, Baohuan Gu2, Tianlu Chen3, Xiangyong Li4, Guoxiang Xie5, Chao Sang6, Hequn Zou7. 1. Department of Nephrology, the Third Affiliated Hospital, Southern Medical University, Guangzhou, China; Department of Clinical Laboratory, the Third Affiliated Hospital, Southern Medical University, Guangzhou, China. 2. Department of Infectious Diseases, Yuedong Hospital, the Third Affiliated Hospital of Sun Yat-Sen University, Meizhou, China. 3. Center for Translational Medicine, Shanghai Jiao Tong University Affiliated Sixth People's Hospital, Shanghai, China. 4. Department of Infectious Disease, the Third Affiliated Hospital of Sun Yat-sen University, Guangzhou, China. 5. Human Metabolomics Institute, Inc., Shenzhen, China. 6. Center for Translational Medicine, Shanghai Jiao Tong University Affiliated Sixth People's Hospital, Shanghai, China. sang_chao@sjtu.edu.cn. 7. Department of Nephrology, the Third Affiliated Hospital, Southern Medical University, Guangzhou, China; Department of Nephrology, Pinghu Hospital, Health Science Center, Shenzhen University, Shenzhen, China. hequnzou@hotmail.com.
Abstract
BACKGROUND: Chronic kidney disease (CKD) is a leading public health problem worldwide. Cardiovascular diseases are the primary cause of death in hemodialysis patients with CKD. Therefore, it is necessary to develop a simple risk assessment tool for cardiovascular events in hemodialysis patients with CKD. METHODS: A cohort of 370 hemodialysis patients, who were recruited between January 2015 to September 2019 in south China, were involved in the present study. On the basis of routine blood test indicators and ultrasonic cardiogram parameters, the optimal parameter set was determined and a Cox proportional hazards model coupled with a nomogram was used to predict cardiovascular risk over 3, 5, and 10 years. Predictive performance was evaluated using Harrell's concordance index (C-index) and the area under the receiver-operating characteristic curve (AUROC). The results were validated using both 10-fold cross-validation and hold-out validation (70% training and 30% validation, repeated 100 times). RESULTS: The optimal parameter set consisted of hypertension, diabetes mellitus, age, phosphate, triglyceride, C-reactive protein, white blood cells, and interventricular septum thickness. The time-dependent AUROCs for predicting 3-, 5-, and 10-year cardiovascular event occurrence risk were 0.836, 0.845, and 0.869, respectively. The nomogram showed satisfactory prediction performance (C-index: 0.808, 95% confidence interval: 0.773-0.844) and was well-calibrated. The results were further confirmed by 10-fold cross-validation and hold-out validation (C-index: 0.794 and 0.798, respectively). CONCLUSIONS: On the basis of several easy-to-detect clinical parameters, we developed a simple and useful nomogram for predicting cardiovascular risk in long-term hemodialysis patients that is of potential value for clinical application.
BACKGROUND:Chronic kidney disease (CKD) is a leading public health problem worldwide. Cardiovascular diseases are the primary cause of death in hemodialysis patients with CKD. Therefore, it is necessary to develop a simple risk assessment tool for cardiovascular events in hemodialysis patients with CKD. METHODS: A cohort of 370 hemodialysis patients, who were recruited between January 2015 to September 2019 in south China, were involved in the present study. On the basis of routine blood test indicators and ultrasonic cardiogram parameters, the optimal parameter set was determined and a Cox proportional hazards model coupled with a nomogram was used to predict cardiovascular risk over 3, 5, and 10 years. Predictive performance was evaluated using Harrell's concordance index (C-index) and the area under the receiver-operating characteristic curve (AUROC). The results were validated using both 10-fold cross-validation and hold-out validation (70% training and 30% validation, repeated 100 times). RESULTS: The optimal parameter set consisted of hypertension, diabetes mellitus, age, phosphate, triglyceride, C-reactive protein, white blood cells, and interventricular septum thickness. The time-dependent AUROCs for predicting 3-, 5-, and 10-year cardiovascular event occurrence risk were 0.836, 0.845, and 0.869, respectively. The nomogram showed satisfactory prediction performance (C-index: 0.808, 95% confidence interval: 0.773-0.844) and was well-calibrated. The results were further confirmed by 10-fold cross-validation and hold-out validation (C-index: 0.794 and 0.798, respectively). CONCLUSIONS: On the basis of several easy-to-detect clinical parameters, we developed a simple and useful nomogram for predicting cardiovascular risk in long-term hemodialysis patients that is of potential value for clinical application.