Affiliation:
1. Department of Nephrology, Blood Purification Research Center, The First Affiliated Hospital, Fujian Medical University, Fuzhou 350005, China
2. Fujian Clinical Research Center for Metabolic Chronic Kidney Disease, The First Affiliated Hospital, Fujian Medical University, Fuzhou 350005, China
Abstract
Objective. We aim to identify independent risk factors to predict CKD progression to end stage renal disease (ESRD) in patients with or without diabetes. Methods. In this retrospective study, we enrolled CKD stage 3-4 patients between January 2013 and December 2018 and followed them until December 2020 or the initiation of dialysis. We used Kaplan-Meier to plot the survival curve. Univariate and multivariable Cox proportional hazards model was used to explore risk factors affecting the progression of CKD. The final model was used to construct nomogram for predicting CKD progression. Calibration plots and concordance index (C-index) were used to evaluate the accuracy and discrimination of the risk model. Results. We enrolled 309 CKD patients, including 80 cases in G3a, 98 cases in G3b, and 131 cases in G4. Among them, 141 patients had diabetes and 168 did not. The mean age of patients at enrolled was 57.86 ± 15.10 years, and 67% were male. The median follow-up time was 25.6 months. There were 81 patients (26.2%) that started dialysis in the total CKD cohort, 52 cases (36.9%) in the CKD with diabetes group, and 29 cases (17.3%) in the CKD without diabetes group. Hypoalbuminemia (HR =2.655,
), proteinuria (HR =2.592,
), increased LDL (HR =2.494,
), diabetes (HR =2.759,
), hypertension (HR =3.471,
), and CKD stage (HR =2.001,
) were risk factors for CKD progression to ESRD in the overall population. For those without diabetes, only hypoalbuminemia (HR =2.938,
) was a risk factor for CKD progression to ESRD. For those with diabetes, both hypoalbuminemia (HR =2.758,
), the increased level of LDL (HR =3.982,
), and CKD stage (HR =3.781,
) were risk factors for CKD progression to ESRD. The C-index of the final nomograms was 0.760 (
). Conclusions. The results from our risk factor model suggest that CKD disease progression can be predicted and early strategic intervention is necessary for CKD patients to avoid renal function deterioration.
Funder
Fujian Provincial Health Technology Project
Subject
Endocrinology,Endocrinology, Diabetes and Metabolism