Length of Follow-Up Time Needed for Stable eGFR Slope Estimation in Glomerular Disease

Slope of estimated glomerular filtration rate (eGFR) is an important measurement of kidney disease progression and clinical outcome in glomerular disease. However, the length of time required to reliably estimate long-term eGFR slopes is unknown. Furthermore, the impact of missing data on eGFR slope estimation is unclear.

Data from the Cure Glomerulonephropathy study were used to answer two questions. A Total Duration question compared eGFR slope estimates based on shorter intervals of follow-up to true long-term slopes using descriptive statistics, overall and within patient subgroups based on age, disease diagnosis, disease activity at study enrollment and during follow-up, non-linearity of the eGFR trajectory, and eGFR measurement frequency. A Duration After Missingness question simulated missing data to examine different lengths of follow-up time after initial gaps between biopsy and study enrollment to compare eGFR slope estimates to true complete-data eGFR slopes, overall and within patient subgroups.

For the Total Duration question, eGFR slope estimates gradually became more accurate with longer follow-up, with 86% of slopes based on 3.5-4 years of follow-up within 5 mL/min/1.73m2/year of true long-term slopes. For the Duration After Missingness question, 84% of eGFR slopes based on over 2 years of follow-up were within 5 mL/min/1.73m2/year of true complete-data slopes after initial missing data gaps between 6 and 24 months. For both questions, eGFR slopes stabilized faster for adults compared to children and for those diagnosed with membranous nephropathy or immunoglobulin A nephropathy compared with minimal change disease or focal segmental glomerulosclerosis. For the Total Duration question, participants with non-linear trajectories required much longer time to achieve stable eGFR slopes compared to those with linear trajectories.

Several years of follow-up are necessary to accurately estimate long-term eGFR slopes in patients with glomerular disease, but moderately sized gaps between eGFR measurements are acceptable.