Background: This study compares glomerular filtration rate (GFR) equations in children based on standardized cystatin C (CYSC) and creatinine (CREA) and their combinations with renal clearance of inulin (C-inulin).
Methods: A total of 220 children with different renal disorders were referred for C-inulin (median 84 ml/min/1.73 m(2)). Bias, precision (interquartile range, IQR), and accuracy (percentage of estimates ±30 % of C-inulin; P30) were evaluated for two cystatin C equations, CAPACYSC and BergCYSC, for creatinine equations, SchwartzCREA and GaoCREA, the arithmetic mean of CAPACYSC and SchwartzCREA (MEANCAPA+Schwartz), BergCYSC and SchwartzCREA (MEANBERG+SCHWARTZ) and the composite equation ChehadeCYSC+CREA.
Results: Overall results of CAPACYSC, BergCYSC, SchwartzCREA, GaoCREA, MEANCAPA+Schwartz, MEANBERG+SCHWARTZ and ChehadeCYSC+CREA were: median bias -7.6/-4.9/-3.7/-2.3/-4.6/-4.0/-10.1 %, IQR 20.0/19.9/21.7/22.4/21.0/20.9/23.3 ml/min/1.73 m(2) and P30 86/86/80/83/89/91/83 %. The cystatin C equations, MEANCAPA+Schwartz and MEANBERG+SCHWARTZ had a more stable performance across subgroups compared with SchwartzCREA, GaoCREA and ChehadeCYSC+CREA.
Conclusions: Cystatin C was the preferred filtration marker for GFR estimation in children, while the benefit of combining cystatin C and creatinine deserves further investigations.