The revised Lund-Malmö GFR estimating equation outperforms MDRD and CKD-EPI across GFR, age and BMI intervals in a large Swedish population.

BACKGROUND The performance of creatinine-based glomerular filtration rate (GFR) estimating equations may vary in subgroups defined by GFR, age and body mass index (BMI). This study compares the performance of the Modification of Diet in Renal Disease (MDRD) study and Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equations with the revised Lund-Malmö equation (LM Revised), a new equation that can be expected to handle changes in GFR across the life span more accurately. METHODS The study included 3495 examinations in 2847 adult Swedish patients referred for measurement of GFR (mGFR) 2008-2010 by plasma clearance of iohexol (median 52 mL/min/1.73 m²). Bias, precision [interquartile range (IQR)] and accuracy [percentage of estimates ±10% (P₁₀) and ±30% (P₃₀) of mGFR] were compared. RESULTS The overall results of LM Revised/MDRD/CKD-EPI were: median bias 2%/8%/11%, IQR 12/14/14 mL/min/1.73 m², P₁₀ 40%/35%/35% and P₃₀ 84%/75%/76%. LM Revised was the most stable equation in terms of bias, precision and accuracy across mGFR, age and BMI intervals irrespective of gender. MDRD and CKD-EPI overestimated mGFR in patients with decreased kidney function, young adults and elderly. All three equations overestimated mGFR and had low accuracy in patients with BMI <20 kg/m², most pronounced among men. CONCLUSIONS In settings similar to the investigated cohort LM Revised should be preferred to MDRD and CKD-EPI due to its higher accuracy and more stable performance across GFR, age and BMI intervals.