Maurice Laville, MD, Pavilion P., Hôpital Edouard-Herriot, F-69437 Lyon Cedex 03 (France) Dear Sir, The excretion of small amounts of albumin, called microalbuminuria, is well known as an early feature of glomerular impairment in diabetics. It has also been observed in patients with essential hypertension [1-3], as a consequence of altered renal hemodynamics with increased glomerular capillary pressure. In diabetics, an-giotensin converting enzyme inhibitors (ACEI) have proved their ability to

more »

... educe albumin excretion rate (AER), a fact attributed to an elective vasodilation of postglomerular arteries [4]. In hypertensives however, a decrease in AER has been reported after normalization of blood pressure by antihypertensive drugs with different hemodynamic effects [1]. Thus, we designed a prospective study of the effect of an 8-week ß-blocker treatment on microalbuminuria in essential hypertensives. Fourteen patients (10 men and 4 women, mean age 46.8 ± 11.3 years, range 27-65) entered the study. They had mild essential hypertension (inclusion DBP 97.5 ± 7.1 mm Hg) for 5.2 ± 4.1 years. All of them had normal renal function measured by creatinine clearance, without macroalbuminuria or hematuria on dipstick uri-nalysis. None was yet treated by antihypertensive drugs at inclusion. In previously treated patients, all antihypertensive drugs were progressively withdrawn, and a 2-week wash-out period was performed before testing. All patients received a cardioselective ß-blocker, bis-oprolol, which lacks ß-agonist activity. The drug was taken at a dosage of 10 mg once a day, in the morning, even the day of visits, during 8 weeks as the sole therapy. The following measurements were performed the day before and after 2 months of treatment. Patients were requested to collect 24-hour urine the day before the test, then came in the morning as outpatients to the Nephro-logy Department where blood and urine samples were taken after 1 h of rest in supine position. During this period, blood pressure was measured every 6 min using an automated oscillometric device (Dinamap®). The mean of the 10 blood pressure readings was used for comparisons. Creatinine was measured on blood and urine samples by an automated Jaffé method and its clearance calculated. Albumin was measured on 24-hour and 1hour urine using immunoturbidimetry (Turbitimer, Behringwerke AG, Marburg, FRG). This method provides a sensitivity treshold of 6 mg/l with an intra-assay variation coefficient of 3% in