Absolute level and rate of change of albuminuria over 1 year independently predict mortality and cardiovascular events in patients with diabetic nephropathy

Aims To determine the nature of the association between baseline albuminuria and risk of all-cause mortality and cardiovascular disease, and to determine if the rate of change of albuminuria from baseline over 1 year predicts these endpoints in patients with diabetic nephropathy. Methods Cohort study of 427 patients (161 Type 1 and 266 Type 2) with diabetic nephropathy defined as urinary albumin excretion (UAE) greater than or equal to 30 mg/24 h at baseline (mean age 53.4 years). Patients were recruited at the time of referral to a diabetic nephropathy clinic and followed up annually for an average of 5 years. UAE rate was re-measured at 1 year and the rate of change from baseline calculated. Results All-cause mortality and cardiovascular disease increased significantly and continuously across quintiles of baseline UAE (P for linear trend < 0.001 in both outcomes). The rate of change of albuminuria over 1 year (log(10) ) independently predicted all-cause mortality (hazard ratio (95% confidence interval) 1.76 (1.39, 3.11)) and cardiovascular mortality (1.57 (1.13, 5.22)). Taken as a categorical variable, a rate of change of albuminuria greater than or equal to 30% independently predicted mortality and cardiovascular events (2.07 (1.5, 4.30) and 1.89 (1.33, 4.06), respectively). Conclusions The rate of change of albuminuria over 1 year independently predicts mortality and cardiovascular disease in diabetic nephropathy and may have clinical utility as a risk marker in identifying a subgroup of patients at particular risk. The risk of these outcomes is continuous across the range of baseline albuminuria in patients with diabetic nephropathy.