Thirty-five insulin-treated diabetics without overt proteinuria or hypertension, and taking no antihypertensive medications were screened at three clinical centers for the presence of microalbuminuria. In addition to the presence of albuminuria, patients were evaluated for duration and type of diabetes, retinopathy, blood pressure, and degree of diabetic control. In these patients, it was possible to examine the degree of microalbuminuria as a function of systolic and diastolic blood pressures, age and sex of the patient, site of recruitment, duration of diabetes, and glycemic control. On multivariate statistical analysis, systolic blood pressure was the only factor that contributed to microalbuminuria. An additional 37 patients had urinary albumin excretion measured, although biochemical and clinical characteristics were incompletely determined. Blood pressures were documented to be normal in 23 of these individuals, while the other fourteen were normal by history. The range of urinary albumin excretion was comparable in the patients with complete data bases and those without. Overall, 22.2% of the normotensive insulin-treated patients screened had microalbuminuria, 5.5% had gross albuminuria, while 72.2% had normal urinary albumin excretion. We agree with previous reports that microalbuminuria is relatively uncommon in the normotensive diabetic population, but further conclude that even in the context of "normal" blood pressure, systolic blood pressure should be carefully observed in diabetic patients. It is possible that these individuals should be considered for more aggressive monitoring programs, e.g., ambulatory blood pressure recording.