According to international consensus, microalbuminuria is defined as an elevated urinary albumin excretion rate (UAER) of 20-200 micrograms/min, which is below the proteinuric range. Nephropathy is a major complication in IDDM, seen in about 30% of patients after many years of diabetes. Increasing microalbuminuria is an excellent marker of subsequent nephropathy in these patients. End-stage diabetic nephropathy is also important in NIDDM, but in most Western countries this serious complication eventually develops in only 5 to 10% of cases, whereas the majority of patients die before this from cardiovascular disease. In completely healthy individuals there is no clear correlation between age and UAER, at least up to about 70 years of age. The mean excretion rate is around 5 micrograms/min, with a considerable range, but excretion only rarely exceeds 15 micrograms/min. In population studies among middle-aged and elderly individuals, higher values are seen. In newly diagnosed NIDDM about 40% of patients show an excretion rate above 15-20 micrograms/min. There is a significant but not precise correlation between albumin excretion rate and glycemic control, and usually UAER is reduced by standard antidiabetic treatment. In a considerable number of patients, high values cannot be reduced. In the course of NIDDM about 20-30% of patients show microalbuminuria. In patients with known diabetes, microalbuminuria is related not only to subsequent diabetic proteinuria, but even more strongly to early death, mainly from cardiovascular disease. Even slight microalbuminuria (15-40 mg/l in early morning urines) is clearly associated with increased mortality. In subjects with newly detected elevated blood glucose (by screening) microalbuminuria also predicts early mortality. The mechanisms are not established, but several arteriosclerosis-related risk factors are seen more frequently in patients with microalbuminuria, e.g. lipid abnormalities, elevated systolic blood pressure (BP), hemostatic measures, as well other markers of cardiovascular disease. Usually there is a significant but not precise correlation between BP and UAER in groups of patients throughout the course of diabetes. New studies document that also in the elderly background population microalbuminuria is a significant risk factor for early death, maybe even stronger than the established risk markers, which thus may be confounded with the presence of microalbuminuria.