Increased intima-media thickness of the common carotid artery predicts increased risk of myocardial infarction and stroke. Preliminary evidence suggests that a decrease in blood pressure (BP) is associated with diminished wall thickness. It is not known if all classes of anti-hypertensive agents have similar protective effects. In this double-blind parallel-group clinical trial, 69 previously untreated patients with hypertension were allocated randomly to 1 year of treatment with either amlodipine (5-10mg daily) or lisinopril (5-20mg daily). Doxazosin and bendrofluazide were added if required to achieve BP control. After 12 months of treatment, clinic BP, ambulatory BP and cardiac mass were reduced similarly by the two treatment regimens. Common carotid artery intima-media thickness decreased by 0.048mm (95% confidence intervals -0.066, -0.031mm) in the amlodipine-treated group, but decreased by only 0.027 mm (-0.046, -0.007mm) in the lisinopril-treated group ( P < 0.05 for difference between treatments). Common carotid artery lumen diameter declined significantly only in patients treated with lisinopril [amlodipine, -0.02mm (-0.14, 0.10mm); lisinopril, -0.21mm (-0.32, -0.11mm); P < 0.02], while intima-media area declined similarly in the two treatment groups [amlodipine -1.32mm 2 (-1.91, -0.74mm 2 ), lisinopril -1.26mm 2 (-1.80, -0.72mm 2 ); not significant]. The results confirm that a decrease in BP causes regression of structural changes in the carotid artery in hypertensive patients. The nature of the structural regression differed markedly between the two treatment regimens, in spite of similar decreases in BP. The calcium channel blocker induced greater regression of common carotid artery intima-media thickness than the angiotensin-converting enzyme inhibitor. However, carotid artery wall mass, as indicated by intima-media area, was reduced to a similar extent by the two treatments. It remains to be established whether such differences confer a prognostic advantage.