Development and validation of a prognostic model for survival in peritoneal dialysis patients / 中华肾脏病杂志

ABSTRACT

Objective:To develop and validate a nomogram for predicting the 1-and 3-year survival rates of patients receiving peritoneal dialysis.Methods:Patients who underwent peritoneal dialysis for the first time in Zhujiang hospital from January 1, 2010 to December 31, 2017 were enrolled. The patients from January 1, 2014 to December 31, 2017 were enrolled in a training dataset. Baseline clinical data were collected and the primary endpoint was all-cause death. Cox proportional hazard regression models were used to analyze risk factors affecting the survival rates. Nomograms were generated using the R rms package. The Harrell' concordance index (C-index), receiver operating characteristic curve and calibration curve were used to verify the performance of the model. Patients who underwent peritoneal dialysis from January 1, 2010 to December 31, 2013 were then selected to validate the external predictive accuracy of the prediction models.Results:The prediction cohort enrolled 457 patients, with a median follow-up time of 27.67(18.37, 39.22) months, and 64 patients (14.00%) died during follow-up. The 1-and 3-year cumulative survival rates were 96.4% and 83.0%. Multivariate analysis showed that aging (every 1 year old increase, HR=1.07, 95% CI 1.04-1.09, P<0.001), stroke ( HR=3.63, 95% CI 1.93-6.85, P<0.001), higher cholesterol (every 1 mmol/L increase, HR=1.51, 95% CI 1.20-1.89, P<0.001), higher neutrophil-to-lymphocyte ratio (every 1 increase, HR=1.12, 95% CI 1.05-1.20, P=0.001), and lower albumin ( HR=0.89, 95% CI 0.82-0.95, P=0.001) were independent risk factors affecting the survival rates of PD patients. The C-index of the prediction cohort and the validation cohort were 0.815(95% CI 0.765-0.865) and 0.804(95% CI 0.744-0.864, respectively). Both internally and externally verified calibration curves showed that the predicted results were close to the actual survival rates. Conclusion:Based on age, blood total cholesterol level, stroke history, and NLR, the prognosis prediction model of peritoneal dialysis patients established with nomogram can help predict the 1-year and 3-year survival rates of peritoneal dialysis patients.