DALLAS, TX - Results of a randomized trial suggest that different blood-pressure-lowering medications can have quite different effects on central aortic BP, despite similar effects on brachial BP.

The findings are from Conduit Artery Function Evaluation (CAFE) [1], a substudy of the Anglo-Scandinavian Cardiac Outcomes Trial (ASCOT), which compared the beta blocker atenolol with or without a diuretic with a regimen based on amlodipine with or without the ACE inhibitor perindopril [2]. CAFE findings showed substantial reductions in central aortic BP with amlodipine+perindopril over atenolol+diuretic, despite very similar brachial BPs between groups.

The researchers, with first author Dr Bryan Williams (University of Leicester, Leicester Royal Infirmary, UK), suggest that these findings may explain at least in part why patients in the amlodipine/perindopril arm appeared to have better clinical outcomes in the overall ASCOT trial. "Moreover, we speculate that the central-blood-pressure hypothesis might also explain differential effects of blood-pressure-lowering drugs on cardiovascular structure and clinical outcomes in other recent trials," such as LIFE and HOPE, they write.

The final CAFE results—first presented last November at the American Heart Association Scientific Sessions 2005 and reported by heartwire at that time—were published online February 14, 2006 in Circulation and will appear in the March 7, 2006 issue.

The CAFE trial recruited 2073 patients from five ASCOT centers about a year after their randomization into the overall trial, so that their treatment regimens were stable. They used radial artery applanation tonometry and pulse-wave analysis to derive central aortic pressures and hemodynamic indexes on study visits for up to four years.

In ASCOT, a total of 19 257 patients with hypertension and three additional risk factors were randomized in a 2x2 factorial design to receive one of two antihypertensive regimens: amlodipine (5/10 mg)+perindopril (4/8 mg) or atenolol (50/100 mg)+a diuretic.

CAFE results showed that despite similar brachial systolic BPs in the two antihypertensive regimens, there were significant reductions in both central aortic systolic BP and central aortic pulse pressures in the amlodipine+perindopril arm compared with the atenolol+diuretic regimen.

Measure	amlodipine-based vs atenolol-based regimen (mm Hg)	95% CI	p
Brachial systolic BP	0.7	-0.4 to 1.7	0.2
Central aortic systolic BP	4.3	3.3 to 5.4	<0.0001
Central aortic pulse pressure	3.0	2.1 to 3.9	<0.0001

Cox proportional hazards modeling suggested that central aortic pulse pressure was significantly associated with a post-hoc-defined composite outcome of total cardiovascular events and procedures and the development of renal impairment in the CAFE study (unadjusted p<0.0001; p<0.05 after adjustment for baseline variables).

"I think this type of technology is going to be increasingly used in clinical trials because it gives us information that we haven't had before," Williams told a press conference at the AHA Scientific Sessions where the study was first presented. "I think that it should be incorporated into blood-pressure-lowering trials so that we gain as much information as possible about what the drugs are doing."

The greater vasodilation seen with amlodipine-based treatment might translate into a reduction in the strength of the reflected wave velocity from the periphery, thereby reducing central arterial pressures. Williams pointed out that a 3- to 4-mm-Hg difference in BP seen between groups in central aortic pressures translates into roughly a 25% difference in stroke risk—similar to the 27% reduction in stroke risk seen in ASCOT in the amlodipine/perindopril arm, supporting the possibility that this difference in central pressures may explain the differences seen in outcomes between groups.

An editorial and clinical perspective related to this article will appear in the March 7, 2006 issue of Circulation.