The aim of this study was to evaluate the antihypertensive effect of drugs according to the initial ambulatory blood pressure (BP) level. After a 15-day placebo run-in period, 105 patients with moderate essential hypertension (mean age, 52 years) underwent 24-h BP monitoring (spacelabs: 1 measure/15 min). Patients were subdivided into two groups: the "High" group, with 24-h mean values of systolic BP (SBP) > 137 or diastolic BP (DBP) > 87 mm Hg, and the "Low" group, with SBP < or = 137 and DBP < or = 87 mm Hg. All patients received, in a random and double-blind design, either bisoprolol (10 mg q.d.) or lisinopril (20 mg q.d.) for 8 weeks. At the end of this active treatment period, office and ambulatory BP measurements were performed. Casual measurements revealed similar BP decreases in all subgroups receiving bisoprolol and lisinopril; BP monitoring showed that the antihypertensive effect depended on the baseline mean 24-h value; -15/-12 mm Hg for bisoprolol and -18/-13 mm Hg for lisinopril in the High group; -7/-6 mm Hg for bisoprolol and -6/-6 mm Hg for lisinopril in the Low group. This study shows that the antihypertensive effect depended on initial ambulatory BP values, with a lower BP decrease in the Low group. Assessment of the antihypertensive effect on ambulatory BP is useful in clinical trials.