Authors' Reply: The Potential Outcome-Modification Influences Introduced by ESKD Life Plan on eGFR Slopes

We thank Hung and Chao1 for their insightful comments about the potential impact of life plan during the transition to kidney failure on the slowing of eGFR decline in patients with advanced CKD. Below, we provide additional hypotheses to explain why eGFR decline would start decelerating before preemptive arteriovenous (AV) access creation as shown in our study published in JASN.2 First, as highlighted in the letter to the editor, treatment adherence and lifestyle behavior change are key for slowing kidney function decline. During the Chronic Kidney Disease—Renal Epidemiology and Information Network (CKD-REIN) cohort follow-up (2013–2020), best practices for CKD management were summarized by the Kidney Disease Improving Global Outcomes 2012 guidelines, updated in 2024.3 In advanced CKD, recommendations included preparation to KRT but also measures aimed at delaying progression, including management of cardiovascular and CKD-associated comorbidities, patient education, and lifestyle modification including diet, exercise, and smoking cessation. The international CKD Outcomes and Practice Patterns Study (including France) previously showed that adherence to these recommendations was low in typical clinical settings.4 However, more frequent nephrology visits, as recommend by the Kidney Disease Improving Global Outcomes and the French Health Authority guidelines—a minimum of 2–4 visits per year in patients with CKD stages 4–5—provide an opportunity for improving symptom management and reducing adverse drug reactions. In a previous study, we showed that patients with more frequent nephrologist visits had lower likelihood of urgent dialysis start.5 Second, GFR estimated from serum creatinine (as in our study) may be overestimated in advanced CKD because of muscle mass loss and reduced protein intake, affecting eGFR trajectory estimation. Third, regression toward the mean—a statistical phenomenon related to random error in repeated measurements—cannot be excluded because the decision to create an AV access often follows a period of rapid decline. Finally, our results may simply be due to the natural history of CKD in advanced stages, which stabilizes after reaching a certain level of GFR. Preserving kidney function is paramount in all phases of CKD care, and further research is needed to clarify the potential contribution of the aforementioned factors. For decades, part of the nephrology community included AV access creation as one of the factors that could slow eGFR decline and delay KRT initiation. By showing a smooth eGFR trajectory around the moment of preemptive AV access creation, our study provides evidence against a causal role of AV access on CKD progression.