BETA-BLOCKERS MAY BE HARMFUL IN SATBLE CAD

NEW YORK, NY — There are more signs that beta-blockers are of no use—and may potentially be harmful—in patients with stable coronary artery disease (CAD) or with CAD risk factors only, this time from a post hoc analysis of the CHARISMAtrial^[1].

The new analysis, published online September 30, 2014 in Circulation: Cardiovascular Quality and Outcomes, adds to a growing body of evidence that is chipping away at the patient group in whom beta-blockers are clearly beneficial.

"The take-home message is that we can't just lump all CAD patients together as one big entity," lead author on the new analysis,  Dr Sripal Bangalore, told heartwire . "We are more and more realizing that the benefit of beta-blockade is different and we have to look specifically at the patient subsets."

Bangalore and colleagues looked at nonfatal MI, stroke, or cardiovascular mortality among patients who participated in the CHARISMA trial, grouped according to whether they'd had a prior MI (4772 patients), known atherothrombotic disease (7804 patients), or risk factors only, but no heart failure (2101 patients) at the study outset. Each group was then further analyzed according to whether or not they were taking beta-blockers at baseline.

Over 28 months of follow-up, baseline beta-blocker use was associated with a 31% reduced risk of the primary composite outcome among patients with prior MI—a finding driven by a roughly 40% reduced risk of new MI. No mortality difference, however, was seen between prior MI patients in the beta-blocker vs no beta-blocker group.

In both the known-atherothrombotic-disease group and the risk-factors-only group, no advantage was seen among those taking beta-blockers as compared with those not taking beta-blockers.

Signal of Stroke Risk

Of note, beta-blockers were associated with a trend toward an increased risk of stroke in the risk-factors-only group, a signal also seen among patients receiving perioperative beta-blockers in POISE , as well as in the 2012 analysis published by Bangalore et al, evaluating beta-blockers in the REACH registry.

The stroke signal in this CHARISMA analysis supports "a very consistent message we are seeing from a host of randomized-trial data, registry data, and observational studies, all pointing to this potential increased risk of stroke with beta-blockade," Bangalore said.

A mechanism previously proposed for the stroke findings is the "inefficient reduction" in central aortic pressures with beta-blockers, potentially related to heart-rate lowering, the authors note.

Important Caveats

There are important limitations to the CHARISMA analysis, Bangalore acknowledged, including the use of older beta-blockers at the time the study was conducted, as well as the key fact that investigators were unable to address whether patients continued to take their beta-blockers over the entire follow-up or, indeed, whether patients in the no-beta-blocker group started taking them during this same period.

As well, the standard of care for patients like those in CHARISMA has changed over the past 10 years. "The way we do medicine today is much different," Bangalore said. But he also noted that the REACH analysis included much more contemporary patients as well as a time-based analysis that took duration of therapy into account. In REACH, there were no differences in the findings based on duration of beta-blocker use, and "we showed something very similar [in terms of benefit/lack of benefit of beta-blockers] in a large, real-world registry of patients."

Bangalore also pointed to an important parallel—namely, that the randomized controlled trial evidence supporting the use of beta-blockers in these types of patients also comes from outdated studies that used earlier-generation drugs and were conducted prior to the widespread use of PCI, statins, and other drugs.

"People have been just extrapolating from those results and assuming that because beta-blockers worked back in the day they will work today, too," and it isn't known whether that's the case.

Another consideration is the other side effects of beta-blockers, including weight gain, new-onset diabetes, worsening of cholesterol control, noncompliance due to sexual dysfunction, and fatigue. "If the data strongly suggest substantial benefits [of beta-blockers] we can take all of the potential risks into consideration, but if the risk/benefit ratio is not favorable, we need to really consider all of these adverse effects."

Bangalore says there remain compelling indications to prescribe beta-blockers in patients with reduced systolic function and in patients having an MI who don't have a high risk of cardiogenic shock for a short period of time. "For post-MI patients, I use them for a longer period of time if they have a reduced ejection fraction. If systolic function is normal, however, I have a low threshold for taking them off [beta-blockade]."

The problem, he says, is that physicians today tend to put patients on a beta-blocker, then leave them on the drug for life, despite the lack of evidence supporting long-term use. Even in the specific setting of post-MI care, optimal duration of beta-blockers is a controversial area, with US STEMI guidelines recommending that patients stay on beta-blockers for three years, while European guidelines stipulate beta-blockers for one year.

The signal from observational studies and post hoc analyses from randomized studies is consistent, Bangalore told heartwire , and supports a growing recognition of the weaknesses in the evidence supporting broad use of beta-blockers, not only in the post-MI setting, but also perioperatively in noncardiac surgery, among heart-failure patients with  atrial fibrillation, for stable angina, and so on. Yet doctors typically default to using these drugs. How to get new randomized controlled trial evidence supporting a role for newer beta-blockers in specific patient groups is a key barrier to answering the growing number of questions, he said.

"As long as people continue using beta-blockers, there is no incentive for pharmaceutical companies to do new randomized controlled trials that will require thousands of patients and be very expensive. Unfortunately, that's the reality we are faced with. So somewhere, somehow we need to say there is really no good evidence [for using these agents], and we need to do a randomized trial."

Medical research tends to focus on finding new drugs that need to be added to patient regimens and less commonly on drugs that can be cut, despite the fact that this could actually improve patient care. "We know that the more medications a patient is on, the more compliance dramatically declines," Bangalore observed.

Bangalore disclosed serving on advisory boards for Daiichi Sankyo, Boehringer Ingelheim, and Pfizer; being an ad hoc consultant/speaker for Abbott Vascular, Abbott, and Gilead; and receiving research grants from Abbott Vascular and the National Heart, Lung, and Blood Institute. Disclosures for the coauthors are listed in the paper.