Predicting Kidney Failure, Cardiovascular Disease and Death in Advanced CKD Patients

Abstract

Introduction: Predicting the timing and occurrence of kidney replacement therapy (KRT), cardiovascular events, and death among patients with advanced chronic kidney disease (CKD) is clinically useful and relevant. We aimed to externally validate a recently developed CKD G4+ risk calculator for these outcomes and to assess its potential clinical impact in guiding vascular access placement.

Methods: We included 1517 patients from the European Quality (EQUAL) study, a European multicentre prospective cohort study of nephrology-referred advanced CKD patients aged ≥65 years. Model performance was assessed based on discrimination and calibration. Potential clinical utility for timing of referral for vascular access placement was studied with diagnostic measures and decision curve analysis (DCA).

Results: The model showed a good discrimination for KRT and "death after KRT," with 2-year concordance (C) statistics of 0.74 and 0.76, respectively. Discrimination for cardiovascular events (2-year C-statistic: 0.70) and overall death (2-year C-statistic: 0.61) was poorer. Calibration was fairly accurate. Decision curves illustrated that using the model to guide vascular access referral would generally lead to less unused arteriovenous fistulas (AVFs) than following estimated glomerular filtration rate (eGFR) thresholds.

Conclusion: This study shows moderate to good predictive performance of the model in an older cohort of nephrology-referred patients with advanced CKD. Using the model to guide referral for vascular access placement has potential in combating unnecessary vascular surgeries.

Keywords: CKD; cardiovascular disease; death; external validation; kidney failure; prognostic model.

Graphical abstract

Figure 1

Stacked cumulative incidence plot. The observed incidence of experiencing KRT, CVD or death as first event is shown. Censoring is accounted for and all outcomes are competing events. After 4 years 74.9% of patients experienced a first event of which 28.8% experienced KRT first, 25% CVD first and 21.1% died without experiencing another event. The number of patients remaining in the study at each time-point are shown below the x-axis. CVD, cardiovascular disease; KRT, kidney replacement therapy.

Figure 2

Scatterplot depicting predicted probabilities of KRT against predicted probabilities of death without KRT. Each patients actual observed outcome (death first, KRT or neither) is illustrated by the color and shape of the point. The dotted 45° line indicates where the predicted risks of KRT and death without KRT are equal: above this line patients had a higher predicted risk of KRT compared to death without KRT. KRT, kidney replacement therapy.

Figure 3

Calibration plots for each main outcome. The predicted probability is shown on the x-axis and the observed outcome rate (calculated with cumulative incidence functions) is given on the y-axis. The dotted 45 degree line represents perfect agreement between predicted and observed probability. The points represent a decile of the validation population (10%), ranked by predicted probability. CVD, cardiovascular disease; KRT,kidney replacement therapy.

Figure 4

Decision curves showing the clinical utility of the Grams model predictions, eGFR guidelines (a) and KFRE predictions (b) for KRT preparation. KRT preparation (including vascular access referral) is considered appropriate if patients initiate KRT within 1 year. These graphs should be read vertically; for any given harm-benefit ratio the guideline with the highest net benefit would result in the most beneficial ratio of correct referrals and incorrect referrals (given the weight that is given to a false positive compared to a true positive based on the harm-benefit ratio). For most harm-benefit ranges 2-year KRT risks predicted by the Grams prediction model have a higher net benefit than eGFR-based risks (a), the net benefit of the Grams and KFRE predictions are very similar, though the Grams model seems to be slightly more beneficial (b). eGFR, estimated glomerular filteration rate; KRT, kidney replacement therapy.