Dalarna University's logo and link to the university's website

Abstract Background: The combined impact of reduced glomerular filtration rate (GFR) and microalbuminuria on the risk for cardiovascular disease is scarcely studied. Thus, we aimed to identify optimal cut-offs for albuminuria and GFR for the prediction of cardiovascular mortality in a community-based cohort of elderly men and to investigate whether the combined addition of these kidney markers adds independent prognostic information. Material and methods: In a sub-sample, without cardiovascular disease at baseline, of the community-based Uppsala Longitudinal Study of Adult Men (ULSAM, n=649, mean age 71 years, median follow-up 12.9 years; 86 cardiovascular deaths during follow-up), GFR (cystatin C-based) and urinary albumin excretion rate (UAER, overnight urine collection) were calculated. Results: The following cut-off points were identified in order to achieve optimal model discrimination based on the integrated discriminative improvement: UAER 6.25 µg/min and GFR 45 ml/min/1.73m2. In Cox-proportional hazard models adjusted for established risk factors (age, systolic blood pressure, antihypertensive treatment, total cholesterol, HDL cholesterol, lipid lowering treatment, diabetes, smoking, body-mass-index and previous cardiovascular disease), participants with low-grade albuminuria only (>6.25 µg/min, HR 1.75, 95 % CI 1.05-2.89), participants with reduced GFR only (<45 ml/min/1.73m2, HR 2.56, 95 % CI 1.05-6.28) and participants with both low-grade albuminuria and reduced GFR (HR 5.91, 95% CI 2.87-12.18) were at higher risk for cardiovascular mortality compared to participants with normoalbuminuria and normal GFR.