The antihypertensive efficacy and tolerability of combination therapy with candesartan cilexetil, 16 mg plus hydrochlorothiazide (CC/HCTZ), 12.5 mg was compared with that of amlodipine, in a multicentre, double-blind, randomised, parallel-group study in patients with mild-to-moderate essential hypertension inadequately controlled by monotherapy. After a two week run-in period on existing therapy, patients with a sitting diastolic blood pressure (DBP) of 90-110 mmHg and a sitting systolic blood pressure (SBP) <or= 180 mmHg were switched to either CC/HCTZ (n=101) or amlodipine (n=102), once-daily by mouth. After eight weeks of treatment, both regimens reduced mean trough blood pressure (BP) by a similar amount: mean sitting SBP/DBP reductions were -15.4/-11.9 mmHg for CC/HCTZ, and -15.7/-12.0 mmHg for amlodipine (group differences, p=0.835/0.963). The BP of 84.2% of patients on CC/HCTZ and 84.5% on amlodipine was controlled (sitting DBP < 90 mmHg and sitting SBP < 140 mmHg) (p=1.00). Six (5.9%) patients on CC/HCTZ and 18 (17.6%) on amlodipine discontinued treatment, including one (1%) and 13 (12.7%) owing to adverse events (p<0.001). The most common adverse event was peripheral oedema, which occurred in two patients on CC/HCTZ and 19 on amlodipine. In conclusion, CC/HCTZ and amlodipine were equally effective in reducing BP in hypertensive patients not controlled by monotherapy, but CC/HCTZ was much better tolerated. Tolerance is an important clinical consideration in the chronic treatment of an asymptomatic disease.