Biomarkers of kidney function and prediction of death from cardiovascular and other causes in the elderly: A 9-year follow-up study

Abstract

Background: Cystatin C is claimed to be superior to creatinine-based estimates of glomerular filtration rate (eGFRcr). The purpose of the study is to analyze whether cystatin C, creatinine, and/or estimated glomerular filtration rates (eGFR) predicted cardiovascular and/or non-cardiovascular deaths among Finnish elderly.

Methods: Hazard ratios (HR) of cystatin C, creatinine and eGFRs for cardiovascular and non-cardiovascular deaths.

Results: During a 9-year follow-up, 275 died, 192 deaths were a result of cardiovascular disease. In age-adjusted analyses, cystatin C predicted the risk of non-cardiovascular and cardiovascular death in men (HR for 0.1-unit increase 1.12 [95% CI, 1.04-1.19] for non-CVD deaths and 1.18 [1.09-1.28] for CVD deaths) and women (1.14 [1.07-1.21] and 1.14 [1.06-1.22], respectively). CKD-EPIcr-cyc predicted the risk of CVD deaths in men (HR for 5-unit decrease 1.17 [1.09-1.25]) and women (1.09 [1.02-1.17]) and non-CVD deaths in women (1.07 [1.01-1.14]). Also, MDRD (HR for 5-unit decrease 1.16 [1.05-1.27]) and CKD-EPI (HR for 5-unit decrease 1.15 [1.05-1.25]) predicted CVD deaths among men. After additional adjustments, predictive value of cystatin C remained significant. Also, the predictive value of CKD-EPIcr-cys remained significant in non-CVD deaths among women.

Conclusion: Cystatin C was clearly the best predictor for cardiovascular and non-cardiovascular deaths among Finnish elderly. Serum cystatin C is more accurate for clinical decision making than creatinine-based eGFR equations or the combined CKD-EPIcr-cys equation in persons older than 64years.

Keywords: Creatinine; Cystatin C; Elderly; Glomerular filtration rate; Mortality.