Use of estimated glomerular filtration rate to predict incident chronic kidney disease in patients at risk of cardiovascular disease: a retrospective study

Abstract

Background: Patients with cardiovascular disease are at an increased risk of chronic kidney disease (CKD). However, data on incident CKD in patients with multiple vascular comorbidities are insufficient. In this study, we identified the predictors of CKD stages 3-5 in patients at risk of cardiovascular disease and used their estimated glomerular filtration rate (eGFR) to construct a nomogram to predict the 5-year risk of incident CKD.

Methods: Ambulatory data on 622 adults with preserved kidney function and one or more cardiovascular disease risk factors who attended outpatient clinics at a tertiary care hospital in Al-Ain, United Arab Emirates were obtained retrospectively. eGFR was calculated using the Chronic Kidney Disease Epidemiology Collaboration equation and assessed every 3 months from baseline to December 12, 2017. Fine and Gray competing risk regression model was used to identify the independent variables and construct a nomogram to predict incident CKD at 5 years, which is defined as eGFR < 60 mL/min/1.73 m² for ≥3 months. Time-dependent area under the receiver operating characteristic curve (AUC) was used to evaluate the discrimination ability of the model. Calibration curves were applied to determine the calibration ability and adjusted for the competing risk of death. Internal validation of predictive accuracy was performed using K-fold cross-validation.

Results: Of the 622 patients, 71 had newly developed CKD stages 3-5 over a median follow-up of 96 months (interquartile range, 86-103 months). Baseline eGFR, hemoglobin A1c, total cholesterol, and history of diabetes mellitus were identified as significant predictors of CKD stages 3-5. The nomogram had good discrimination in predicting the disease stages, with a time-dependent AUC of 0.918 (95% confidence interval, 0.846-0.964) at 5 years, after internal validation by cross-validation.

Conclusions: This study demonstrated that incident CKD could be predicted with a simple and practical nomogram in patients at risk of cardiovascular disease and with preserved kidney function, which in turn could help clinicians make more informed decisions for CKD management in these patients.

Keywords: Cardiovascular disease; Chronic kidney disease; Estimated glomerular filtration rate; Nomogram; Prediction; Sub-distribution hazards model.

Fig. 1

Flow diagram of the patient population. CKD chronic kidney disease, eGFR estimated glomerular filtration rate

Fig. 2

Estimated cumulative incidence curves for CKD stages 3–5. Unadjusted estimated cumulative incidence curves (solid lines) for CKD stages 3–5 in the presence of death as a competing event according to eGFR groups with 95% pointwise CIs (broken lines). a eGFR, 60–89 mL/min/1.73 m². b eGFR, 90–99 mL/min/1.73 m². c eGFR, ≥100 mL/min/1.73 m². CKD chronic kidney disease, eGFR estimated glomerular filtration rate, CI confidence interval

Fig. 3

Time-dependent AUC for CKD stages 3–5 risk prediction models AUC area under the curve, CKD chronic kidney disease

Fig. 4

Calibration curves. Fine-Gray regression model after backward-stepwise selection and the full model with all variables included

Fig. 5

Nomogram to predict the development of CKD stages 3–5 at 5 years HbA1c glycosylated hemoglobin A1c, eGFR estimated glomerular filtration rate, CKD chronic kidney disease