Author + information
Bangalore et al. (1) suggest that beta-blocker–induced reduction of heart rate increases the risk for cardiovascular events and death for hypertensive patients. Results of meta-analyses should be viewed with circumspection, especially when clinical practice might be influenced (2). With one-fifth of the general population susceptible to migraine (and other primary vascular headaches) and with nonvasodilating beta-blockers still regarded as first-line preventive agents, the conclusions of this study (1) raise concern. The authors admittedly ignore the dose of atenolol (1), a critical variable that limits this analysis. Besides, biologically, in no 2 patients can the evolution of atherosclerosis or its complications be strictly comparable. Statistics do permit such mathematical comparisons in randomized controlled trials (RCTs), but they can extract a hidden biological price (3).
First, contrary to the assertion that the “slower the heart rate, the greater the benefit” (1), heart rates below 50 beats/min cannot generally be claimed to promote overall cardiovascular integrity. Second, the negative inotropic action of beta-blockers has been ignored in this study (1) as well as in the CAFE (Conduit Artery Functional End Point) study (4). Any rise in central aortic pressure/pulse pressure by nonvasodilating beta-blockers would be intrinsically countered by their negative inotropic action. A pharmacologically reduced stroke volume would maximally affect central conduit vessels to reduce central vessel wall stress. Not surprisingly, atenolol reduces the elevated augmentation index in hypertensive patients compared with that in normotensive subjects (5). Third, the investigators (1) did not stratify their results according to age. An aging cohort is likely to have stiffer conduit arteries that, in turn, would exacerbate any differential drug effects on central aortic pressure (4). Fourth, the concept of dyssynchrony or uncoupling between outgoing and reflected aortic waves consequent to pharmacologically induced bradycardia (1) is purely speculative.
Cardiovascular morbidity and mortality are too complex to be resolved through multiple mathematics-based comparisons of diverse pharmacologic agents, particularly when polytherapy with several drugs might be involved. Moreover, all vasodilators usually worsen migraine headache; among antihypertensive agents, beta-blockers, however, do not generally aggravate headache. Regardless of age or race, hypertension is commonly associated with headache that has several features of migraine (6). Proscription of beta-blockers for initial or primary management of hypertension will increase the incidence of associated vascular headaches including migraine and make their management more complex. Next, beta-blockers hold center stage in management of predominantly systolic hypertension as well as the anxiety-related white coat hypertension/effect. Weight gain and precipitation of diabetes mellitus by beta-blockers is a relative risk unrelated to pancreatic islet cell damage.
In essence, this study (1) and the accompanying editorial (7) seek to convert a small-to-moderate statistical relative risk into an absolute biological risk with important practical implications. RCTs allow scientists to carry out credible research without having to discern crucial clinical phenomena or diminishing the need for the same (3). While using research tools such as RCTs or meta-analysis involving RCTs, we must remain cognizant of the intrinsic biological limitations of mathematical data mining. To seek or force a clinical/therapeutic consensus in the face of biological uncertainty cannot be commended as the best scientific/research practice.
- American College of Cardiology Foundation
- Bangalore S.,
- Sawhney S.,
- Messerli F.H.
- Williams B.,
- Lacy P.S.,
- Thom S.M.,
- et al.
- Kaplan N.M.