The Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation incorporating both cystatin C and creatinine best predicts individual risk: a cohort study in 444 patients with chronic kidney disease.

BACKGROUND The recently introduced CKD-EPIcreat-cys equation surpassed creatinine-based equations for GFR estimation in a large cross-sectional analysis. However, its performance to predict individual risk of CKD progression and death in patients with various underlying CKD etiologies is unknown. METHODS We recruited 444 patients with CKD GFR categories 2-4 (eGFR 15-89 mL/min/1.73 m2); baseline eGFR was estimated by the established MDRD and CKD-EPIcreat equations and by the novel CKD-EPIcreat-cys equation. RESULTS Patients were followed for 2.7±1.2 years for the occurrence of the combined predefined endpoint event: death, need for renal replacement therapy or halving of eGFR. The endpoint occurred in 62 patients. Reclassification from MDRD determined categories to CKD-EPIcreat-cys categories yielded net reclassification improvements for those with the endpoint event (NRIevents) of 27.4% (95% CI: 16.7-40.0%) and for those without the event (NRInon-events) of -3.1% (-8.2 to 1.6%). Similarly, reclassification from CKD-EPIcreat categories to CKD-EPIcreat-cys categories yielded an NRIevents of 22.6% (10.2-34.3%) and NRInon-events of -11.3% (-15.9 to -6.5%). Addition of albuminuria to each eGFR equation increased the calculated risk of the outcome for a net 26-32% of those who subsequently reached the endpoint, and reduced the calculated risk in a net 21-23% in non-event patients, but only minimally. CONCLUSIONS The CKD-EPIcreat-cys equation assigned patients who went on to have the event to more appropriate CKD risk categories than MDRD and CKD-EPIcreat, but patients without the event to less appropriate categories than CKD-EPIcreat. Addition of albuminuria marginally improved risk classification for those who had the event.