CAFÉ: Central aortic pressure differences may explain better outcomes with amlodopine in ASCOT.

November 13, 2005—Lower central aortic systolic and pulse pressures with amlodipine-based treatment compared to atenolol-based treatment—although brachial pressures were similar—were found in CAFÉ, a substudy of ASCOT. This may explain the better outcomes with the amlodipine-based treatment in the main study. The Conduit Artery Function Evaluation (CAFÉ) Study was presented by Dr. Bryan Williams, University of Leicester, England, at the 2005 Scientific Sessions of the American Heart Association, held in Dallas, Texas from November 13-16.

“This study demonstrates for the first time in a large clinical outcomes trial that blood pressure-lowering drugs have profoundly different effects on central aortic pressures and hemodynamics, despite a similar impact on brachial blood pressure,” said Williams. Central aortic systolic blood pressure was 4.3 mmHg lower and central aortic pulse pressure was 3.0 mmHg lower in the amlodipine group, even though brachial pressures were similar in both groups.

A total of 2,073 patients (1031 patients in the atenolol group and 1042 in the amlodipine) group were recruited at 5 of the ASCOT study centers in the UK and Ireland. More than 70% of the ASCOT patients at each center were recruited into CAFÉ, from 1-year after ASCOT began when blood pressures were stable. The patients had a mean 3.4 tonometry measurements, and the average follow-up after the first tonometry was 3 years.

Central aortic pressures were measured non-invasively using Sphygomocor©, an FDA-approved device with a pressure sensitive probe that takes readings just above the radial artery of the wrist, which then are computed to pulse pressure by computer using a mathematical algorithm. These computed pressures have been validated in studies by other investigators using direct catheter measurement of central pressures.

Key results

Brachial blood pressures were well controlled in both groups. Systolic blood pressure was 133 mmHg in both groups, diastolic 78.6 mmHg with atenolol and 76.9 mmHg with amlodipine (p<0.0001), and pulse pressure was 55 mmHg on both groups. Heart rate was 58.6 with atenolol and 69.3 with amlodipine (p<0.001).

In contrast, central systolic blood pressure (SBP) was significantly lower in the amlodipine group (121.2 mmHg) compared to atenolol (125.9 mm Hg) throughout the entire study (p<0.0001). In the amlodipine arm, a significant difference between the brachial and central SBP was found (12.0 mmHg; p<0.0001). In the atenolol arm, the difference was 8.3 mm Hg.

Also, a significantly lower central aortic pulse pressure with amlodipine was found (43.4 mmHg vs 46.4 mmHg with amlodipine; p<0.0001). A significant difference between the brachial pulse pressure and the aortic central pulse pressure was found with amlodopine (12.8 mm Hg; p<0.0001), but none with atenolol (9.9 mmHg difference). The greater augmentation index in the atenolol group compared to the amlodipine group (31.9 vs 25.3) suggests that increased wave reflection increased the central pulse pressure.

A significant 16% reduction in the secondary composite endpoint of all cardiovascular events and procedures plus development of renal impairment was found (p<0.0001). Because of the small number of events (305), this result should be interpreted with caution, Williams noted. This pre-specified analysis sought to determine the relation between central aortic pressures and hemodynamics and clinical outcomes. Notably, only central pulse pressure was an independent predictor for the secondary composite endpoint (p=0.048; HR 1.11; 95%CI 1.00-1.21).

What do the results mean?

The results support the study hypotheses that different drugs have different effects on central aortic blood pressure and hemodynamics and that this pressure has more effect on clinical outcomes. Further, that “brachial blood pressure is not always a perfect surrogate for the effects of drug therapy on central aortic pressures,” Williams said. “In CAFÉ, brachial blood pressures overestimated the hemodynamic benefit of atenolol and underestimated the benefit of amlodipine treatment on central aortic pressures and hemodynamics.”

“The shape of the pulse wave is influenced by the treatments we use to lower blood pressure. The CAFÉ study suggests that treatment based on amlodipine had more favorable effects on the shape of the pulse wave and pressures in the main arteries, than treatment based on atenolol,” said Williams.

In comments to the press, Williams noted that with amlodipine, the reflected wave is smaller because of the drug’s vasodilatory effect, which shifts the wave away from the heart. Atenolol has an opposing effect, because of its vasoconstriction, with the reflected wave more likely to hit the outgoing wave and thereby increasing central aortic pressures. This may be particularly relevant in older patients with stiffer arteries.

He also commented that because patients taking amlodipine in ASCOT had better outcomes, patients who are over 55 years of age who fit the ASCOT patient profile probably should take amlodipine.

Notably, previous studies have shown a 25% reduction in stroke is associated with this difference between brachial and central aortic pressures. In the ASCOT trial, there was a 27% reduction in stroke. Therefore, it is likely that differences in central aortic pressures explain the difference in outcomes in the ASCOT study with amlodipine compared to atenolol.

What’s next?

It is likely that measuring central aortic pressures in clinical trials of hypertension drugs will become standard hereafter, Williams said, to better understand the physiology and the differential effects of specific drugs in monotherapy and combination therapy.

He also noted that the CAFÉ results have major implications for interpreting clinical trials, and that guidelines may need to be reviewed to reconsider this.

Further analyses include calculating central aortic pressures in the overall trial, based on the data from CAFÉ, for each time point and outcomes, and evaluating the effect of atorvastatin on central aortic pressures in the CAFÉ patients.