Amlodipine is one of the frequently used drugs in the treatment of arterial hypertension. Plasma levels of amlodipine can be measured to assess the treatment adherence. There is only limited data on plasma levels in patients treated with amlodipine and correlation of plasma levels with adverse events.Design and method:
We measured 642 plasma levels of amlodipine in 343 patients treated for resistant arterial hypertension. The minimum time gap between two measurements in one patient was three months. All patients used amlodipine chronically and should be in steady state of the drug. The amlodipine reference range between 6 and 18 μg/l has been regarded as therapeutic according to literature. Patients were divided into three groups - with sub-therapeutic or unmeasurable levels (group 0), with therapeutic levels (group 1) and levels above upper therapeutic limit (group 2). Clinical status, complaints, evidence of edemas, number of antihypertensive drugs and dose of amlodipine were recorded.Results:
Only 30.7% of the measurements were within the therapeutic range. Sub-therapeutic or unmeasurable level was detected in 19.9% of measurements (n = 128) and these patients were regarded as non-compliant with therapy. Level above the upper limit was detected in 49.4% (n = 317) of specimens. In patients adherent to amlodipine, 61.7% of measurements were above the therapeutic range. High levels of amlodipine and higher rate of overdose were more often detected in older patients (Table 1), which is in concordance with decreased clearance of amlodipine with age. Levels of amlodipine exceeding 3 times or more upper therapeutic limit were most commonly found in patients using higher daily doses of amlodipine than recommended, renal impairment or using drugs interfering with amlodipine metabolism on cytochrome CYP 3A4. 39% of these patients reported undesirable effects of the drug.Conclusions:
Overdose by amlodipine without clinically manifest side effects is frequent and probably undervalued in clinical practice. Risk of overdose can by predicted by simple clinical characteristics of patients.