ObjectiveTo compare the therapeutic coverage and safety of amlodipine and perindopril in patients with mild to moderate hypertension (diastolic blood pressure ≥ 90 mmHg and ≤ 109 mmHg).DesignA double-blind, randomized, parallel-group, multicentre study.MethodsFollowing a 2-week placebo wash-out period, the patients were randomly allocated to treatment with either amlodipine at 5–10 mg once a day or perindopril at 4–8 mg once a day, for 60 days. Trough : peak ratios were calculated by two different methods (global and individualized approaches) from 24 h ambulatory blood pressure recordings made after the placebo period and after the active treatment period. Residual lowering of blood pressure after single-blind, single-dose omission was also investigated with further 24 h ambulatory blood pressure monitoring. Safety assessments were made throughout the study.ResultsThe placebo-adjusted, global, diastolic blood pressure trough : peak ratio was 0.80 in the amlodipine group (n= 47) and 0.81 in the perindopril group (n= 49) in an intent-to-treat analysis. The corresponding global systolic blood pressure trough : peak ratio was 0.83 for amlodipine and 0.68 for perindopril. Individual trough : peak ratios were non-normally distributed. Mean (± SD) individual trough : peak ratios (intent-to-treat analysis) for diatolic blood pressure were 0.50 ± 0.69 for amlodipine (median 0.42) and 0.15 ± 3.27 for perindopril (median 0.33). In the per protocol analysis, the corresponding values were 0.50 ± 0.72 (median 0.34) for amlodipine and 0.01 ± 3.90 for perindopril (median 0.21). Both treatments produced comparable decreases in clinic systolic and diastolic blood pressure between days 0 and 60. Forty-eight hours after the last dose, both systolic and diastolic blood pressure were lower in amlodipine-treated patients than perindopril-treated patients. Amlodipine and perindopril were generally well tolerated. The most frequently reported adverse event was leg oedema in amlodipine-treated patients (19.1%), and coughing in perindopril-treated patients (14.3%).ConclusionsThese results showed no statistically significant difference in trough : peak ratios between amlodipine and perindopril. However, the ambulatory blood pressure trough : peak ratios showed very large variations. Determination of trough : peak ratios by the conventional approach or by an individual approach can yield disparate values. After omitting one dose, a condition imitating noncompliance, blood pressure was more effectively controlled with amlodipine than with perindopril.