Patients with IDDM, both with normal albumin excretion and microalbuminuria, have no evidence of a defect of urinary dopamine excretion. This genetically determined association with essential hypertension does not, therefore, explain the link between early diabetic nephropathy and hypertension. In hypertensive patients with IDDM, captopril and nifedipine retard have a similar effect in reducing blood pressure and albumin excretion after 8 weeks treatment. Captopril, however, retains an acute effect on renal haemodynamics, resulting in a decreased filtration fraction, and this may be of specific importance in the management of patients with diabetic nephropathy. In newly-presenting patients with NIDDM, 26% had abnormal urinary albumin excretion, this correlating with age, glycaemic control, systolic blood pressure and generalised vascular disease. Over the year following diagnosis, 16% of patients had persistent microalbuminuria. Although tubular dysfunction is commonly found in patients with microalbuminuria, the relatively poor correlation of currently available markers would seem to limit their value in the detection of early nephropathy. In patients with IDDM and normal albumin excretion, acute hypoglycaemia causes a number of changes in renal function, notably a significant fall in both effective renal plasma flow and glomerular filtration rate (GFR). The fall in GFR is more marked than in non-diabetic subjects. It is possible that hypoglycaemia may have a role in the aetiology or progression of diabetic nephropathy.