Controversies in Hypertension From ESC 2008

August 30, 2008 - September 3, 2008, Munich, Germany

Linda Brookes, MSc


January 05, 2009

In This Article


When the results of the Blood Pressure Lowering Arm of the Anglo-Scandinavian Cardiac Outcomes Trial (ASCOT-BPLA), one of the largest hypertension trials ever conducted, were announced at the American College of Cardiology annual meeting in the spring of 2004, they brought about a serious reassessment of the use of beta-blockers as first-line therapy in uncomplicated hypertension patients. A pre-planned substudy of ASCOT, the Conduit Artery Function Evaluation (CAFE), suggested that there was a difference between blood pressure measured with the traditional brachial sphygmomanometer and pressure assessed by calculated pulse wave derivation method, and that this could explain the differential effect (or lack of effect) of the beta blocker. Now, however, a study reported at this year's European Society of Cardiology (ESC) Congress in Munich, Germany, raises doubts about this explanation.

Another of the largest-ever hypertension trials, the Losartan Intervention for Endpoint Reduction in Hypertension (LIFE) trial, was similarly bad news for beta blockers (for prevention of stroke, at least) -- but again, a study presented at ESC 2008 has suggested that brachially measured blood pressure values do not reveal the true hemodynamic effects of the alternative therapies.

Finally, the results of a new meta-analysis have confirmed the benefits traditionally attributed to calcium channel blockers (CCBs), and also suggested a possible benefit against all-cause death and (compared with placebo) against risk for heart failure and coronary revascularization.

ASCOT was designed to assess the benefits of 2 different blood pressuring lowering strategies, a "traditional" approached using a beta-blocker plus or minus a diuretic, vs a "modern" approach with a CCB plus or minus an angiotensin-converting enzyme (ACE) inhibitor. However ASCOT also included a substudy, CAFE, which looked at the differential effects of these 2 strategies on blood pressure assessed by 2 different methods, the standard brachial pressure (as measured in all doctors' offices) and "central aortic blood pressure," measured by a calculated derivation of the pulse pressure wave.

The results of CAFE, first announced in 2005[1] and published in 2006,[2] showed that, despite having similar effects on brachial blood pressure in hypertensive patients at moderate risk of developing cardiovascular events, a beta-blocker-based regimen was less effective in lowering central aortic pressure and had different effects on hemodynamics compared with a CCB-based regimen. These results were consistent with data from earlier smaller-scale studies of shorter duration, and the CAFE investigators concluded that their findings could provide a plausible mechanistic explanation for the superior clinical outcome associated with the CCB-based regimen in ASCOT, in support of the explanation originally suggested by the ASCOT investigators[3] and commentators.[4] It might also explain the differential effects of blood pressure-lowering drugs on cardiovascular structure and clinical outcomes in other outcome trials, such as LIFE[5] and the Heart Outcomes Prevention Evaluation (HOPE).[6]

An editorial published in conjunction with the CAFE results, however, proposed alternative explanations for the differences between the 2 treatments in CAFE, such as differences in 24-hour blood pressure control or "tissue mechanisms yet to be described."[7] Nonetheless, the CAFE investigators pointed out that the results of their study supported the results of meta-analyses that challenged the use of beta-blockers as first-line treatment for uncomplicated hypertension,[8,9] and several national blood pressure management guidelines have since removed beta-blockers, as a class, from first-line treatment of hypertension in the absence of compelling indications, either explicitly[10] or implicitly.[11]

The CAFE Substudy

The CAFE substudy of the blood pressure-lowering arm of ASCOT (ASCOT-BPLA), supported by Pfizer, involved 2037 patients randomized in the main trial to an atenolol-based regimen (with or without bendroflumethiazide) or an amlodipine-based regimen (with or without perindopril).[12] ASCOT-BPLA patients were aged 40 to 79 years and had hypertension plus 3 or more other cardiovascular risk factors. Over a mean follow-up of 5.5 years, those randomized to the amlodipine-based regimen had fewer major cardiovascular events and less incident diabetes than those randomized to the atenolol-based regimen.

For CAFE, radial artery applanation tonometry and pulse wave analysis were used to derive central aortic pressures and hemodynamic indexes on repeated visits for up to 4 years. This revealed that although brachial systolic blood pressure (SBP) levels were similar in the 2 treatment groups (area under the curve [AUC] difference: 0.7 mm Hg; 95% CI, -0.4 to 1.7; P = .2), there were substantial reductions in central aortic pressures with the amlodipine regimen (AUC difference in central aortic SBP: 4.3 mm Hg; 95% CI, 3.3 to 5.4; P < .0001; AUC difference in central aortic pulse pressure, 3.0 mm Hg; 95% CI, 2.1 to 3.9; P < .0001). The investigators also reported that the central pulse pressure was significantly associated with a composite outcome of all cardiovascular events and development of renal impairment in the CAFE cohort defined post hoc and validated by the ASCOT endpoints committee.

The New Study

It was against this background that the CAFE investigators' findings were challenged at the 2008 ESC meeting in Munich by the results of a study that, unlike CAFE, suggested that beta-blockers lower central blood pressure to the same extent as other antihypertensive drug classes. Presenting the results of the study, Frederico Soria-Arcos, MD (Hospital Universitario Virgen de La Arrixaca, Murcia, Spain), reported that there was no significant difference in hemodynamic patterns of adaptation to different blood pressure-lowering regimens in a nonselected, real-life population sent to the catheterization laboratory for diagnostic and/or therapeutic coronary angiography.[13]

Study Design. The study population consisted of 161 consecutive patients (age 68 ± 13 years, 71% male, 80% hypertensive) who underwent coronary angiography (elective or emergency) via the radial artery at 2 cardiology sites in Southeastern Spain. They were studied as 4 groups, analyzed according to their blood pressure-lowering medication: beta-blocker only (beta-blocker group, 30 patients); beta-blocker plus any other antihypertensive drug (combined group, 73 patients); any antihypertensive drug other than beta-blockers (no beta-blocker group, 46 patients); and no antihypertensive drugs (controls, 12 patients).

There were no differences among the 4 groups with respect to age, gender, diabetes, smoking status, hypercholesterolemia, height, or use of intravenous nitroglycerin, statins, or antiplatelets, although there were significant differences in weight (P = .027) and use of antidiabetic drugs (P = .03). Most of the beta-blockers used by these patients were the newer drugs in the class, including bisoprolol (36%), carvedilol (29.5%), and nebivolol (4%), as well as atenolol (29.5%), the beta-blocker used in the ASCOT trial, Dr Soria-Arcos noted.

Direct invasive blood pressure measurement was carried out in these patients, who were all pretreated with heparin and verapamil 2.5 mg administered intra-arterially to prevent arterial spasm. At the end of the procedure, and before all the catheters were withdrawn, 2 consecutive measurements of blood pressure were obtained, at the aortic arch (central) and at the radial artery (peripheral). A minimum of 10 beats were averaged for analysis. Patients who showed hemodynamic instability at the time of recording, required ≤ 5F catheters, or in whom there was a ≥ 1-minute time delay between tracings were excluded from the study.

Results and Implications. Significant differences were seen in central and radial SBP among all 4 groups in favor of the combined and no-beta-blocker groups ( Table 1 ). However, no differences emerged when comparing central-to-radial differences among groups. There were no significant differences in heart rate.

Dr. Soria-Arcos noted limitations of the study that might have affected the results, including the small numbers of patients and controls, the presence of 19.5% patients with acute coronary syndrome who started beta-blockers only hours before the study started, a lack of simultaneous tracings, and the fact that 75% of the patients taking beta-blockers were on doses considered low to medium, which could account for the lack of a difference in heart rate between the groups. Despite these limitations, however, he and his colleagues believe that their study suggests that mechanisms other than different hemodynamic profiles are responsible for the diminished cardiovascular protective effect demonstrated with beta-blockers. Dr Soria-Arcos noted that newer beta-blockers with direct vasodilating properties may have different patterns of hemodynamic adaptation and consequently different vascular protection from the older drugs in the class. They should be studied before they are excluded from first-line antihypertensive therapy and other mechanisms should also be investigated, he suggested.

During discussion, Dr. Soria-Arcos admitted that he and his colleagues had expected to see rather different, even opposite results in their study, since they had believed that central blood pressure lowering was important. However, they believe that the reduction in cardiovascular protection cannot be attributed solely to this effect. While agreeing that definitive conclusions cannot be drawn from the study, Dr. Soria-Arcos said that it was important to draw attention to this problem.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: