Inappropriate GFR equations misrepresent CKD epidemiology and risk interpretation in scientific reports

The accuracy of chronic kidney disease (CKD) prevalence and its associations with metabolic risk factors depends heavily on the formula used to estimate glomerular filtration rate (eGFR). In Iran, many epidemiologic studies still rely on outdated equations, which may distort epidemiologic findings and misinform public health estimates.

We analyzed 9,911 adults from the Fasa Adult Cohort Study (FACS) at baseline and 2,977 at 5-year follow-up. eGFR was calculated using Cockcroft-Gault, MDRD, and CKD-EPI 2021 equations. CKD was defined as eGFR < 60 mL/min/1.73 m². Associations of metabolic dysfunction-associated steatotic liver disease (MASLD) and the fatty liver index (FLI) with CKD prevalence and incidence were examined using logistic regression, with polynomial regression applied for continuous eGFR outcomes.

At baseline, mean eGFR varied substantially by equation (Cockcroft-Gault: 80.7 ± 20.6; MDRD: 71.0 ± 11.1; CKD-EPI: 82.7 ± 13.3). CKD prevalence was estimated at 15.1% with Cockcroft-Gault, 14.5% with MDRD, but only 3.5% with CKD-EPI (χ² p < 0.001). Similar discrepancies were observed for 5-year incidence (23.0%, 29.5%, and 9.9%, respectively; χ² p < 0.001). In longitudinal analyses, MASLD was associated with higher CKD risk when using CKD-EPI (OR 1.35, 95%CI 1.01-1.79) and MDRD (OR 1.32, 95%CI 1.08-1.62), but lower CKD risk with Cockcroft-Gault (OR:0.10,95%CI:0.08-0.13,p < 0.001). Continuous analyses showed the same directionally inconsistent associations between FLI and eGFR depending on the equation used.

The choice of GFR estimating equation substantially alters CKD prevalence estimates and the observed relationships with MASLD. CKD-EPI 2021 produced results consistent with biological plausibility and guideline recommendations. For accurate epidemiology, public health planning, and international comparability, researchers in Iran and globally should adopt the most updated eGFR equations.

Not applicable.