Epidemiology and prognostic significance of chronic kidney disease in the elderly—the Three-City prospective cohort study

Abstract

Background. Little is known about normal kidney function level and the prognostic significance of low estimated glomerular filtration rate (eGFR) in the elderly. Methods. We determined age and sex distribution of eGFR with both the Modification of Diet in Renal Disease (MDRD) study and the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equations in 8705 community-dwelling elderly aged ≥65 years and studied its relation to 6-year mortality. In a subsample of 1298 subjects examined at 4 years, we assessed annual eGFR decline and clinically relevant markers including microalbuminuria (3–30 mg/mmol creatinine) with diabetes, proteinuria ≥50 mg/mmol, haemoglobin Results. Median (interquartile range) MDRD eGFR was 78 (68–89) mL/min/1.73m² in men and 74 (65–83) in women; there were 79 (68–87) and 77 (67–85) for CKD-EPI eGFR, respectively. Prevalence of MDRD eGFR 2 was 13.7% and of CKD-EPI eGFR was 12.9%. After adjustment for several confounders, only those with an eGFR 2 had significantly higher all-cause and cardiovascular mortality than those with an eGFR of 75–89 mL/min/1.73m² whatever the equation. In the subsample men and women with an MDRD eGFR of 45–59 mL/min/1.73m², 15 and 13% had at least one clinical marker and 15 and 3% had microalbuminuria without diabetes, respectively; these percentages were 41 and 21% and 23 and 10% in men and women with eGFR 2/year. Conclusions. Moderately decreased eGFR is more often associated with clinical markers in men than in women. In both sexes, eGFR 2 is related to poor outcomes. The CKD-EPI and the MDRD equations provide very similar prevalence and long-term risk estimates in this elderly population.