Estimating glomerular filtration rate at the transition from pediatric to adult care

The current Kidney Disease Improving Global Outcomes (KDIGO) guidelines recommend the use of the bedside creatinine-based Chronic Kidney Disease in Children (CKiD) equation to estimate glomerular filtration rate (GFR) in children and the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation in adults. However, this approach causes implausible changes in estimated GFR (eGFR) at the transition from pediatric to adult care. We investigated the performance of the KDIGO strategy and various creatinine-based eGFR equations in a cross-sectional dataset of 5,764 subjects (age 10-30 years), using directly measured GFR (mGFR) as reference. We also evaluated longitudinal GFR slopes in 136 subjects who transitioned to adult care. Implausible changes in eGFR resulted from the large overestimation (bias=+21 mL/min/1.73m²) and poor precision of the CKD-EPI equation in the 18-20 year age group, compared to CKiD in the 16-18 year age group (bias=-2.7 mL/min/1.73m²), resulting in a mean change of 23 mL/min/1.73m² at the transition to adult care. Averaging the CKiD and CKD-EPI estimates in young adults only partially mitigated this issue. The Full Age Spectrum equation (with and without height), the Lund-Malmö Revised equation, and an age-dependent weighted average of CKiD and CKD-EPI resulted in much smaller changes in eGFR at the transition (change of 0.6, -2.1, -0.9 and -1.8 mL/min/1.73m², respectively). The longitudinal analysis revealed a significant difference in average GFR slope between mGFR and the KDIGO strategy (-2.2 vs. +2.9 mL/min/1.73 m²/year), which was not observed with the other approaches. These results suggest that the KDIGO recommendation for GFR estimation at the pediatric-adult care transition should be revisited.