CKD prevalence based on real-world data: continuous age-dependent lower reference limits of eGFR with CKD–EPI, FAS and EKFC algorithms

Abstract

Purpose

Several recent articles discuss the need for a definition of chronic kidney disease (CKD) that embarks age-dependency and its impact on the prevalence of CKD. The relevance is derived from the common knowledge that renal function declines with age. The aim of this study was to calculate age-dependent eGFR lower reference limits and to consider their impact on the prevalence of CKD.

Methods

A real-world data set from patients with inconspicuous urinalysis was used to establish two quantile regression models which were used to calculate continuous age-dependent lower reference limits of CKD–EPI, FAS and EKFC–eGFR based on either single eGFR determinations or eGFR values that are stable over a period of at least 3 months (± 10% eGFR). The derived lower reference limits were used to calculate the prevalence of CKD in a validation data set. Prevalence calculation was done once without and once with application of the chronicity criterion.

Results

Both models yielded age-dependent lower reference limits of eGFR that are comparable to previously published data. The model using patients with stable eGFR resulted in higher eGFR reference limits. By applying the chronicity criterion, a lower prevalence of CKD was calculated when compared to one-time eGFR measurements (CKD–EPI: 9.8% vs. 8.3%, FAS: 8.0% vs. 7.2%, EKFC: 9.0% vs. 7.1%).

Conclusion

The application of age-dependent lower reference intervals of eGFR together with the chronicity criterion result in a lower prevalence of CKD which supports the estimates of recently published work and the idea of introducing age-dependency into the definition of CKD.