The aim of this study was to analyse the respective values of clinic and ambulatory blood pressure measurements in patients with treated hypertension. Twenty-eight patients with mild to moderate hypertension were studied (clinic blood pressure > or = 160/95 mmHg), treated with benazepril alone (n = 7) or combined with hydrochlorothiazide (n = 21). A 24 h ambulatory blood pressure measurement was performed before (d0) and after 8 weeks of treatment (d56). The treatment was effective on the ambulatory blood pressure in 19 patients (Staessen criteria) and effective on clinic measurements in 20 patients (WHO criteria). The results were concordant in 25 of the 28 patients. The decrease of the ambulatory blood pressure values during the day between d0 and d56 was not significantly related to the decrease of the clinic values. The comparison of the ambulatory blood pressure values at each hour between d0 and d56 showed that the treatment was effective during 24 h. The 'white coat' effect has been evaluated at d0 and d56 by the difference between mean daytime ambulatory blood pressure and clinic values. Clinic values were higher than mean daytime ambulatory values in benazepril (14 +/- 12 mmHg and 7 +/- 7 mmHg for the systolic and diastolic pressure) and benazepril-hydrochlorothiazide group (12 +/- 12 mmHg and 0.2 +/- 9 mmHg for the systolic and diastolic pressure). This 'white coat' effect was not modified during the treatment. Previous studies have emphasized a better assessment of the cardiovascular risk with ambulatory than with clinic measurements. The differences in the assessment of the efficiency of the treatment between clinic and ambulatory measurements shown in our study prompt us to use ambulatory measurements in patients with apparently uncontrolled hypertension.