There have been substantial changes in the management of diabetes since the 1960' s.We studied all 213 patients in our catchment area in whom Type I diabetes was diagnosed before the age of 15 years between 1961 and 1980, 92% were followed from the onset of diabetes to 1991 or to the time of deathThe cumulative incidence of diabetic nephropathy has decreased substantially in recent decades from 30% to 9% after 25 year's diabetes duration, probably as a result of improved glycaemic control. Neither the cumulative incidence of severe retinopathy nor hypertension (140/90mmHg) changed during the last decades.The risk to develop severe retinopathy or nephropathy was higher in patients with Very Poor glycaemic control (HbA1c;::.8.4%) vs. patients with Poor control (HbA1c ~ 7 .2<8.4% )(p<0.001). Patients with Poor control had an increased risk to develop severe retinopathy vs. patients with Good control (HbA1c<7 .2% )(p<0.008) but there was no difference in the risk for nephropathy. No patients with Good control developed nephropathy and only one patient developed severe retinopathy during 25 years of diabetes. Up to a diabetes duration of 25 years both diabetic nephropathy and severe retinopathy can be avoided but the degree of glycaemic control needed for prevention differs.The course of microalbuminuria during the 1980's was studied with a 10-year follow up of 109 Type 1 diabetes patients. Only 5 (19%) of the initially micro-albuminuric patients developed macro-albuminuria during the 10 year follow up period and 15 (58%) patients decreased their AER to normal. The initially micro-albuminuric patients, who normalised their AERimproved their glycemic control. In the majority of patients with micro-albuminuria in whom it is possible to obtain a good glycemic control, micro-albuminuria will disappear and the risk of developing nephropathy is markedly reduced.PRA and All concentrations were significantly lower in Type 1 diabetic patients (n=80) than in matched healthy controls (n=75). ANP levels were higher in patients than in controls. In the patients PRA correlated negatively to the mean value of HbA1c during the previous five years. Patients with Type 1 diabetes, specially those with very poor glycaemic control, have a suppressed RAS and increased ANP levels.