Letter to the Editor Regarding Kolodner et al.

Authors

  • Katayoun Mostafaie MD,

    1. Division of General Internal Medicine, David Geffen School of Medicine UCLA, Harbor-UCLA Medical Center, Los Angeles, CA, USA.
    Search for more papers by this author
  • Darrell W. Harrington MD

    1. Division of General Internal Medicine, David Geffen School of Medicine UCLA, Harbor-UCLA Medical Center, Los Angeles, CA, USA.
    Search for more papers by this author

To the Editor:—Kolodner et al.'s multicenter study1 concludes that preoperative β-blocker was significantly underutilized when compared with the current guideline recommendations. Despite the authors' acknowledgment that part of this underutilization is because of ongoing clinical questions regarding the appropriate selection of candidates for this therapy, the study design and the population described in this publication assumes that there is adequate evidence-based data supporting perioperative β-blocker use in patients with coronary artery disease (CAD) risk factors. We make objection to this conclusion.

The authors cite the results of 2 randomized clinical trials carried out by Mangano and Poldermans and their colleagues. Mangano's study of 200 patients has been criticized since its publication a decade ago. Many concerns have been appropriately raised with regard to the analysis of only postdischarge data and the exclusion of 6 inpatient deaths (while patients were receiving the study drug postrandomization). Inclusion of these deaths in the analysis contradicts the conclusion of the study. Poldermans' study of 112 patients was un-blinded and limited to a very small group of highly selected patients undergoing vascular surgery. Data from 3 recent large null studies totaling 1,521 subjects (MaVS, POBBLE and DIPOM) failed to show any benefit. Additionally, Devereaux's review of 22 trials (that randomized a total of 2,437 patients) concluded that the evidence is too unreliable to allow definitive conclusions to be drawn. This meta-analysis clearly demonstrates that preoperative β-blockade is not without risk. Devereaux concluded that the individual safety outcomes in patients treated with perioperative β-blockers showed a relative risk for bradycardia needing treatment of 2.27 (95% confidence interval [CI] 1.53 to 3.36) and nominally statistically significant relative risk for hypotension requiring treatment of 1.27 (95% CI 1.04 to 1.56).

In addition, the authors cite that many organizations have identified perioperative β-blockade as a quality measure.2 We acknowledge that pay for performance is inevitable. Yet, the public and medical community's perception is that quality measures are based on recommendations derived from data supported by body of evidence-based literature. However, it can be readily appreciated that many guidelines, including those recently published by ACC/American Heart Association,3 are not based on high quality evidence but instead on consensus opinion. Inevitably, this is further biased as many committee members have industry relationships, potentially influencing not only their personal practice but also their expert recommendations ultimately utilized for consensus building.

We agree with the authors that the results of POISE trial should provide us with a definitive answer. The evidence to date is not supportive of the routine use of β-blockers in the majority of patients at risk for CAD undergoing elective major noncardiac surgery. We need to demand that quality measures and guidelines should be based on evidence-based data and not driven by political or economic pressures.—Katayoun Mostafaie, MD, and Darrell W. Harrington, MD, Division of General Internal Medicine, David Geffen School of Medicine UCLA, Harbor-UCLA Medical Center, Los Angeles, CA, USA.

Ancillary