Prediction of cardiovascular death and non-fatal cardiovascular events by the Kidney age–Chronological age Difference (KCD) score in men and women of different ages in a community-based cohort

Abstract

Objective We examined the utility of the Kidney age–Chronological age Difference (KCD) score, an age-adapted measure of kidney function, to identify increased cardiovascular (CV) death or non-fatal CV event risk in participants of the Australian Diabetes, Obesity, and Lifestyle Study (AusDiab), a community-based cohort aged 23–95 years.Design Cohort study.Setting Community.Participants 11205 randomly selected participants from urban and nonurban areas across Australia.Outcome measures Mortality status and underlying and contributory causes of death obtained from the Australian National Death Index, and non-fatal CV events from adjudicated hospital records. The association of CV death or non-fatal CV event risk with KCD score was examined using penalised spline curve analysis.Results Of 11 180 participants with serum creatinine measurement at baseline and 5-year outcome data, there were 308 CV deaths or non-fatal CV events after 5 years. Penalised spline curve analysis showed similar progressive increase in CV death or non-fatal CV event risk with increasing KCD score in men and women, and participants aged <50 years to ≥80 years. Receiver operating characteristic curve analysis showed optimal discrimination at a KCD score ≥20 years (KCD20) for all participants. Among 148 participants aged<70 years with CV death or non-fatal CV event, KCD20 identified 24 (16%) participants, whereas estimated glomerular filtration rate (eGFR) <60 mL/min/1.73 m2 identified 8 (5%) participants (p=0.0001), with specificities of 95% and 99%, respectively (p<0.0001).Conclusion KCD20 predicted CV death or non-fatal CV event risk similarly in men and women of different ages in this population-based cohort. The higher sensitivity for prediction of CV death or non-fatal CV event risk in participants aged <70 years by KCD20 than by eGFR <60 mL/min/1.73 m2 offers opportunity for earlier renoprotective therapy in individuals with eGFR-associated increased CV death or non-fatal CV event risk.