Background It remains uncertain whether cystatin C is a superior marker of renal function than creatinine in older adults. We have investigated the association between estimated glomerular filtration rate (eGFR) using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equations based on creatinine (CKD-EPIcr) and cystatin C (CKD-EPIcys), and cardiovascular risk markers and mortality in older adults.

Methods This is a cross-sectional and prospective study of 1639 British men aged 71–92 years followed up for an average of 5 years for mortality. Cox survival model and receiving operating characteristic analysis were used to assess the associations.

Results The prevalence of chronic kidney disease (CKD) was similar using the two CKD-EPI equations, although cystatin C reclassified 43.9% of those with stage 3a CKD (eGFR 45–59 mL/min/1.73², moderate damage) to no CKD. However, CKD stages assessed using both CKD-EPIcr and CKD-EPIcys were significantly associated with vascular risk markers and with all-cause and cardiovascular disease mortality. In all men with CKD (eGFR <60 mL/min/1.73²), the HRs (95% CI) for all-cause mortality after adjustment for cardiovascular risk factors compared with those with no CKD were 1.53 (1.20 to 1.96) and 1.74 (1.35 to 2.23) using CKD-EPIcr and CKD-EPIcys, respectively. Comparisons of the two CKD equations showed no significant difference in their predictive ability for mortality (difference in area under the curve p=0.46).

Conclusion Despite reclassification of CKD stages, assessment of CKD using CKD-EPIcys did not improve prediction of mortality in older British men >70 years. Our data do not support the routine use of CKD-EPIcys for identifying CKD in the elderly British male population.