Beta blocker Use May Not Reduce Mortality, Myocardial Infarction, or Stroke In Patients Without Coronary Artery Disease
DynaMed Weekly Update - Volume 7, Issue 44
Beta blockers have long been a standard of care following myocardial infarction (MI), and their use has been generalized to other forms of coronary artery disease (CAD) for secondary prevention and even to patients with CAD risk factors for primary prevention. Most of the evidence supporting beta blocker use comes from randomized trials in patients with MI or heart failure that were conducted 20 years ago or more. In a 1999 systematic review of 82 randomized trials with 54,234 patients with previous MI, beta blocker use for at least 6 months was associated with a 23% reduction in death (BMJ 1999 Jun 26;318(7200):1730), but, the median publication date of the included trials was 1982 and the duration of follow-up was less than 18 months. In recent years, the more frequent use of reperfusion and revascularization therapies may have decreased the amount of scarring of the myocardium associated with an infarction and thus modified the amount of benefit of beta blockers in these populations.
A large, longitudinal study (median follow-up 44 months) assessed the current efficacy of beta blocker treatment in 3 cohorts: 14,043 patients with prior MI, 12,012 patients with CAD but no prior MI, and 18,653 patients with at least 3 cardiovascular risk factors. To compare the effects of beta blocker use vs. no use, a propensity score was calculated for each patient characterizing his or her likelihood for receiving beta blockers based on multiple demographic factors, medical history and use of other medications. About half of the patients in each cohort were included in the propensity-matched analyses.
There were no statistically significant differences comparing beta blocker use vs. no use in all-cause mortality or cardiovascular mortality in any of the 3 cohorts (level 2 [mid-level] evidence). For cardiovascular mortality, the hazard ratios were 0.91 (95% CI 0.76-1.09) in the prior MI cohort, 0.90 (95% CI 0.72-1.11) in the CAD without prior MI cohort, and 1.05 (95% CI 0.85-1.30) in the CAD risk factor cohort. The study was not powered for a mortality difference of only 9% so a benefit in the groups with known CAD cannot be excluded. In patients with risk factors for CAD, beta blocker use was associated with trends toward increased risk of both nonfatal MI (p = 0.08) and nonfatal stroke (p = 0.06). There were no significant differences in the rates of nonfatal stroke and nonfatal MI in either the prior MI cohort or CAD without prior MI cohort (JAMA 2012 Oct 3;308(13):1340).
American Heart Association guidelines, updated in 2011, took a more conservative approach to recommendations for beta blocker use. The previous guideline, from 2006, had strongly recommended indefinite beta blocker use in patients with MI, acute coronary syndrome, or left ventricular dysfunction. The current guideline calls for treatment for up to 3 years (class I recommendation [should be given]) with a weaker recommendation for longer treatment (class IIa recommendation [reasonable to give]). The new guideline also lowered its recommendation for beta blockers in patients with other vascular diseases from “reasonable to give” (class IIa) to “may be considered” (class IIb) (Circulation 2011 Nov 29;124(22):2458). Current European guidelines recommend long-term beta blocker use only for patients with left ventricular dysfunction (Eur Heart J2011;32(23):2999).
This study is limited by the inherent problems of cohort studies and the complexity involved with propensity matching, which may also be an inadvertent source of bias. While this study raises questions regarding the optimal use of beta blockers in patients with known CAD it does not seem compelling enough to warrant changes in current recommendations. However, the data would seem to caution against the use of beta blockers in patients with risk factors for CAD but no specific CAD history. In that group, which includes patients being treated for hypertension, alternative agents may be preferred.
For more information, see the Coronary artery disease (CAD) topic in DynaMed.