Estimation of the glomerular filtration rate in children and young adults using the CKD-EPI equation with age-adjusted creatinine values.

Abstract The CKD-EPI creatinine-based estimation equation for glomerular filtration rate (GFR) cannot be used in children, overestimates GFR in young adults, and its combination with the KDIGO recommended pediatric CKiD (Schwartz bedside) equation causes implausible increases in estimated GFR when switching from pediatric to adult care. By establishing sex-specific creatinine growth curves for children and young adults, creatinine levels of children and young adults below age 40 years were adjusted with 40 as assigned age and applied in the CKD-EPI equation. Validation was performed in 4005 children (2-17 years) and 3309 young adults (18-39 years) using metrics based on bias, precision, and accuracy including percentage of estimates within 30% (P30) of measured GFR (mGFR). Comparisons were made with the CKiD and Schwartz-Lyon equations in children. CKD-EPI with age-adjusted creatinine instead of actual age and creatinine led to extensive improvements in bias, precision, and accuracy at all ages, in both sexes and at all levels of mGFR. At mGFR below and above 75 mL/min/1.73m2, the P30 increased from 12% to 75% and 33% to 88% in children, respectively, and from 56% to 73% and 83% to 92% in young adults, respectively. In children adjusted CKD-EPI was more accurate than CKiD, especially above mGFR 75 mL/min/1.73m2 (P30 88% vs. 82%), while Schwartz-Lyon was more accurate than adjusted CKD-EPI at mGFR below 75 mL/min/1.73m2 (P30 81% vs. 75%). Thus, the proposed strategy based on age-adjusted creatinine in children and young adults makes the CKD-EPI equation applicable across the full spectrum of age and kidney function.