Key points are not available for this paper at this time.
Abstract Background and Aims The optimal timing for dialysis vascular access creation remains a complex clinical decision, balancing the risk of early intervention against the prospect of starting dialysis with a central venous catheter. This study sought to compare the predictive capabilities of the Kidney Failure Risk Equation (KFRE) and estimated Glomerular Filtration Rate (eGFR) in determining the appropriate timing for arteriovenous fistula (AVF) creation in patients with advanced chronic kidney disease (CKD). Method A retrospective cohort of 238 patients referred for ultrasound vascular mapping from April 2013 to June 2023 was analyzed. Data on patient characteristics, laboratory results, and clinical outcomes were collected. The 4-variable and 8-variable KFRE and eGFR were assessed for their ability to predict end-stage kidney disease (ESKD) within a 2-year period. Discrimination was evaluated using receiver operating characteristic (ROC) curves, and calibration was assessed graphically. Net benefit decision curves were utilized to compare the net benefit of the KFRE models at various threshold probabilities. Results The cohort comprised 238 patients with an average age of 65.2 years and a mean eGFR of 13.3 mL/min/m²73. During the median 10.7-month follow-up, 178 patients (74.8%) progressed to ESKD, and 21 (9.2%) died prior to reaching ESKD. At the 1-year mark, the probability of progressing to ESKD was 77.6% (95% CI, 69.9-85.3%) when a 40% threshold for the 4-variable KFRE was applied, compared to 65.8% (95% CI, 58.3-73.3%) with an eGFR threshold of 15 mL/min/m²73. Discrimination analysis demonstrated a superior predictive performance for the 8-variable KFRE at 6 months, with a C-statistic of 0.82 (95% CI: 0.76-0.88). At 1 year, both KFRE models showed strong predictive abilities, with C-statistics of 0.835 (95% CI: 0.78-0.89) for the 4-variable KFRE and 0.82 (95% CI: 0.76-0.875) for the 8-variable KFRE. Sensitivity and specificity analyses underscored the benefits of utilizing the 40% KFRE threshold for predicting ESKD over the eGFR threshold. Conclusion The study demonstrates that both the 4-variable and 8-variable KFRE are valuable tools for predicting the need for vascular access in pre-dialysis patients, outperforming eGFR alone. The 8-variable KFRE, in particular, provides a significant improvement in 6-months ESKD prediction, with potential implications for clinical decision-making and patient management in advanced CKD. These findings advocate for the incorporation of KFRE in the clinical evaluation process for timely and appropriate vascular access planning.
Ingwiller et al. (Wed,) studied this question.