We wondered whether, in an elderly hypertensive population in a primary prevention setting, free from diabetes mellitus and clinical atherosclerosis, differences between end organ damage and microalbuminuria (MA) could be found using a lower level of urinary albumin excretion than that of classically defined MA. From a population survey of 173 previously untreated hypertensive patients (4x blood pressure systolic > or = 160 and < or = 220 mmHg, and/or diastolic > or = 95 and < or = 115 mmHg), mean age 67 +/- 4 years, were screened for MA (defined as albumin excretion between 20 and 300 mg/24 h). End organ damage was determined by B-mode ultrasound scanning of carotid and femoral arteries and echocardiography. Out of 173 hypertensives, 14 showed MA (8%). These hypertensives had a significantly higher intima media thickness (IMT; 1.01 +/- 0.21 vs 0.88 +/- 0.6 mm, p < 0.05) and increased left ventricular mass index (118 +/- 31 vs 103 +/- 22 g/m2, p < 0.05) than hypertensives without MA. Linear regression analysis showed that MA, age, male gender and diastolic blood pressure were independently related to IMT, while systolic blood pressure, male gender and body mass index were independently related to left ventricular mass. Even using lower levels of urinary albumin excretion rate, patients with MA had significantly higher IMT and increased left ventricular mass. Moreover, MA was independently related to IMT in these elderly hypertensives. These results suggest that the threshold value for MA should be reconsidered in hypertension.