Prognostic significance of reduced eGFR for kidney outcomes and all-cause mortality in older Japanese individuals with type 2 diabetes and normoalbuminuria: a single-center cohort study

Because glomerular filtration rate (GFR) physiologically declines with age, a fixed estimated GFR (eGFR) threshold of < 60 mL/min/1.73 m² to define chronic kidney disease may overestimate risk in older adults. We examined whether lower eGFR was associated with adverse outcomes in older Japanese adults with type 2 diabetes (T2D) and normoalbuminuria.

We studied 901 Japanese adults aged ≥ 65 years with T2D and normoalbuminuria who underwent baseline assessment between 2012 and 2017. Participants were categorized by Japanese Society of Nephrology (JSN) eGFR as ≥ 90 (n = 41), 60–89 (n = 538), 45–59 (n = 251), and 15–44 mL/min/1.73 m² (n = 71). Outcomes were a composite kidney outcome (≥ 30% eGFR decline, kidney replacement therapy, or kidney-related death), all-cause mortality, and eGFR slope. Analyses included Poisson regression, competing risk models, and restricted mean survival time. A sensitivity analysis recalculated eGFR using the 2021 CKD-EPI equation with a Japanese coefficient (CKD-EPI_J).

The mean age was 72.6 ± 5.5 years (SD) and 41.3% were women. Over median follow-up periods of 6.8 and 7.8 years, 245 kidney events and 48 deaths occurred, respectively. Baseline eGFR by the JSN equation was not independently associated with the composite kidney outcome, mortality, or eGFR slope. In contrast, the CKD-EPI_J equation reclassified 65.7% of participants with a JSN eGFR of 45–59 mL/min/1.73 m² into a higher category, yielding a clearer gradient in the risk of kidney outcomes, whereas mortality risk estimates remained similar across categories.

In this population, the association between lower eGFR and kidney outcomes differed by the equation used.