Serum uric acid and eGFR_CKDEPI differently predict long-term cardiovascular events and all causes of deaths in a residential cohort

ObjectivesSerum uric acid (SUA) and estimated glomerular filtration rate (eGFR) were separately assessed as risk factors for incident coronary hard (CHDH), cardiovascular disease (CVDH) or all-cause (ALL) deaths but never concomitantly in a residential cohort.Material and methodsMen and women aged 35-74 years, totaling 2888 subjects were followed 13.5-19.5 years for incident CVDH, CHDH and ALL deaths. Systematic comparisons among different end-points were based on: age, gender, systolic blood pressure (SBP), total and HDL cholesterol, cigarette consumption, body mass index, blood glucose, SUA, eGFR from the Chronic Kidney Disease Prognosis Consortium (eGFR_CKDEPI) and (eGFR_CKDEPI)2.ResultsSignificant (p < 0.00001) differences in SUA quintiles were seen for SBP, total and HDL cholesterol, body mass index and eGFR_CKDEPI whereas cigarettes and blood glucose were not statistically different. There were increasingly larger proportions of all events in SUA quintiles (0.05 > p < 0.0001). Among 4 major continuous variables, SUA was largely accurate (ROC > 0.610) to predict all end-points whereas eGFR_CKDEPI was the worse univariate predictor. Multivariately, age, gender, SBP and cigarettes were significant predictors for all end-points. Total cholesterol was a significant predictor only for CHDH events. Blood glucose and SUA were contributors for CVDH events (RR, for 1 mg/dl of SUA, 1.09, 95%CI 1.01-1.17), CVD deaths (RR 1.11, 95%CI 1.03-1.20) and ALL deaths (RR 1.08, 95%CI 1.03-1.14) whereas (eGFR_CKDEPI)2 was for ALL deaths only (RR 1.02, 95%CI 1.00-1.04).ConclusionSUA is a predictor of long-term incidence of cardiovascular events and deaths and all-cause mortality and should be considered for risk predictive purposes and instruments whereas eGFR_CKDEPI only predicts all-cause mortality by a U-shaped relation.