April 23rd, 2013
Study Suggests Benefit for Beta Blockers During Noncardiac Surgery
Larry Husten, PHD
The use of perioperative beta-blockade for noncardiac surgery has been declining as a result of the controversial POISE study, which turned up evidence for harm associated with extended-release metoprolol in this setting. Now a large new observational study published in JAMA offers a contrary perspective by suggesting that perioperative beta-blockade may be beneficial in low- to intermediate-risk patients. But without better evidence the debate about this topic is unlikely to be resolved.
Martin London and colleagues performed a retrospective analysis of 136,745 patients who underwent noncardiac surgery at VA hospitals, 40% of whom received beta-blockade. Beta-blockers were used more frequently in vascular surgery patients and in patients who were at higher cardiovascular risk, as indicated by a greater number of Revised Cardiac Risk Index factors. Over the course of the study period, from January 2005 through August 2010, the rate of beta-blockade declined from 43.5% to 36.2%, a finding likely related to the publication of the POISE study in 2008.
Overall, the mortality rate was 1.1% and the cardiac morbidity rate was 0.9%. After propensity matching, beta-blockade was associated with lower mortality (RR 0.73, CI 0.65-0.83, p<.001) and cardiac complications (RR 0.67, CI 0.57-0.79, p<.001). The difference was significant only in patients with 2 or more Revised Cardiac Risk Index factors and only in the group of nonvascular surgery patients. The authors speculated that the lack of effect in the nonvascular surgery patients may have been due to the smaller sample size and that some beta-blocker usage in this group may not have been captured in the database. The risk of stroke — which was elevated in the beta-blocker group in POISE — was not significantly different between the groups in the VA analysis.
The authors concluded that their results “highlight a need for a randomized multicenter trial of perioperative beta-blockade in low- to intermediate-risk patients scheduled for noncardiac surgery.”
Franz Messerli and Sripal Bangalore sent the following response to this study:
London et al in a large observational study, found an association of perioperative beta blockade with lower morbidity and mortality in 137,000 patients (~75,000 in the matched cohort) with low to intermediate risk. The strength of the observational study is the large number of patients and the use of the VA database which perhaps reduces the number of missing data and out of network referral. The role of perioperative beta blockade in patients undergoing non cardiac surgery is controversial, and has become all the more controversial in the recent past given the ongoing investigations into the research conducted by Poldermans. The Poldermans issue notwithstanding, there are significant limitations to the present analysis including inability to account for unmeasured confounders, selection and ascertainment bias. We also are puzzled that a similar beneficial association was not seen in patients undergoing vascular surgery, the highest risk group where some RCTs had indicated a possible benefit. Additionally, it appears that most of the patients in the beta blocker group were on it at 90 days indicating chronic exposure. Conceivably therefore beta blockers were given for a possible compelling indication and not necessarily only for cardioprotection for noncardiac surgery such as in the POISE trial. The present data may therefore further strengthen the ACC/AHA recommendation of continuing beta blockers in patients who are already on it. However, as far as to the question of starting beta blocker for non cardiac surgery, this study does not provide any new insights.