Which glomerular filtration rate estimation equations should be used in youth with type 1 diabetes?

Diabetic kidney disease (DKD) is the primary cause of kidney failure in type 1 diabetes (T1D). Early identification of subclinical DKD is based on albuminuria and glomerular filtration rate (GFR). However, no GFR equations have been validated in youth with T1D. This study evaluates the performance of GFR equations in youth with T1D.

We compared a direct measure of GFR (mGFR) via plasma clearance of exogenous inulin, with estimated GFR (eGFR). eGFR was calculated with the following equations: Full-Age Spectrum (FAS-height and FAS-age); revised Lund-Malmö (LM-Rev); European Kidney Function Consortium (EKFC); Chronic Kidney Disease Epidemiology Collaboration, 2009 (CKD-EPI 2009); CKD-EPI 2021 (revised CKD-EPI 2009 without race adjustment); CKD-EPI40 (revised CKD-EPI 2021 with age-adjusted creatinine); Chronic Kidney Disease in Children (CKiD); CKiD Under 25 (sex-dependent); CKiD Under 25 (age- and sex-dependent); and Improving Renal Complications in Adolescents with Type 2 Diabetes through Research (iCARE). Bland-Altman analysis estimated performance (bias and accuracy (P10 and P30)).

In total, 141 children and adolescents at a mean age of 13.5 ± 3.3 years (range 6.2-18.5) and diabetes duration 5.7 ± 1.6 years were included. Their mean serum creatinine was 45.6 ± 12.1 μmol/L (0.52 ± 0.14 mg/dL) and mGFR 142.3 ± 25.0 mL/min/1.73 m². For the whole population, FAS-height (bias = 8.37 mL/min/1.73 m²; P30 = 85.1%) and CKiD (bias = 11.9 mL/min/1.73 m²; P30 = 81.6%) performed best. These equations also outperformed others in subgroups (females, age groups < 11 years and 11-15.9 years). In contrast, CKiD in males (bias = 5.23 mL/min/1.73 m²; P30 = 81.0%) and CKD-EPI 2009 in the 16-18.5 years subgroup (bias = -0.04 mL/min/1.73 m²; P30 = 84.6%) were top performers.

Subgroups of children and adolescents with T1D may benefit from different eGFR equations for accurate subclinical DKD evaluation.