Perioperative beta-blockers in noncardiac surgery: Evolution of the evidence
ABSTRACTAfter studies in the 1990s suggested that beta-blockers offer substantial benefits when given before surgery, several national organizations endorsed the perioperative use of these drugs as a best practice in certain patients. However, subsequent research has cast doubt on whether it is appropriate to use these drugs as widely as suggested by those early studies.
KEY POINTS
- Beta-blockers reduce perioperative ischemia, but the benefit may be only in high-risk patients undergoing high-risk surgery. Currently, the best evidence supports their use in two groups: patients undergoing vascular surgery who have known ischemic heart disease or multiple risk factors for it, and patients who are already on beta-blockers.
- The Perioperative Ischemic Evaluation (POISE) findings suggest that beta-blockers should be used in the immediate preoperative period only with great caution, after ensuring that the patient is clinically stable and without evidence of infection, hypovolemia, anemia, or other conditions that could make heart-rate titration misleading or use of the drug dangerous.
- When feasible, beta-blockers should be started a month before surgery, titrated to a heart rate of 60 beats per minute, and continued for approximately a month. If the drug is then to be discontinued, it should be tapered slowly.
The Revised Cardiac Risk Index. Although the Boersma risk-factor index is not used in general practice, numerous experts27,20–32 recommend a similar one, the Revised Cardiac Risk Index, devised by Lee et al.8 This index consists of six risk factors, each of which is worth one point:
- Congestive heart failure, based on history or examination
- Myocardial infarction, symptomatic ischemic heart disease, or a positive stress test
- Renal insufficiency (ie, serum creatinine level > 2 mg/dL)
- History of stroke or transient ischemic attack
- Diabetes requiring insulin
- High-risk surgery (defined as intrathoracic, intra-abdominal, or suprainguinal vascular surgery).
Patients with three or more points are considered to be at high risk, and those with one or two points are considered to be at intermediate risk. The ACC/AHA 2007 guidelines6 use a modified version of this index that considers the issue of surgical risk separately from the other five clinical conditions.
Devereaux et al33 performed a meta-analysis, published in 2005, of 22 studies of perioperative beta-blockade. They concluded that beta-blockers had no discernable benefit in any outcome measured, including deaths from any cause, deaths from cardiovascular causes, other cardiac events, hypotension, bradycardia, and bronchospasm. However, they based this conclusion on the use of a 99% confidence interval for each relative risk, which they believed was justified because the trials were small and the numbers of events were only moderate. When the outcomes are assessed using the more common 95% confidence interval, benefit was detected in the combined end point of cardiovascular death, nonfatal myocardial infarction, and nonfatal cardiac arrest.
Yang et al,34 Brady et al,35 and Juul et al36 performed three subsequent randomized trials that added to the controversy. Most of the patients in these trials were at intermediate or low risk, and none of the trials found a significant benefit with perioperative beta-blocker use. However, the protocols in these studies were different from the one in the study by Poldermans et al,3 which had found perioperative beta-blockade to be beneficial. Whereas patients in that earlier study started taking a beta-blocker at least 1 week before surgery (and on average much earlier), had their dose aggressively titrated to a target heart rate, and continued taking it for 30 days afterward, the protocols in the later trials called for the drug to be started within 24 hours before surgery and continued for only a short time afterward.
Lindenauer et al,37 in a retrospective study published in 2005, found that fewer surgical patients who received beta-blockers in the hospital died in the hospital. The researchers used an administrative database of more than 780,000 patients who underwent noncardiac surgery, and they used propensity-score matching to compare the postoperative mortality rates of patients who received beta-blockers and a matched group in the same large cohort who did not. Beta-blockers were associated with a lower morality rate in patients in whom the Revised Cardiac Risk Index score was 3 or greater. However, although there was a trend toward a lower rate with beta-blocker use in patients whose score was 2 (ie, at intermediate risk), the difference was not statistically significant, and patients with a score of 0 or 1 saw no benefit and were possibly harmed.
The authors admitted that their study had a number of limitations, including a retrospective design and the use of an administrative database for information regarding risk index conditions and comorbidities. In addition, because they assumed that any patient who received a beta-blocker on the first 2 hospital days was receiving appropriate perioperative treatment, they may have incorrectly estimated the number of patients who actually received these drugs as a risk-reduction strategy. For instance, some patients at low risk could have received beta-blockers for treatment of a specific event, which would be reflected as an increase in event rates for this group. They also had no data on what medications the patients received before they were hospitalized or whether the dose was titrated effectively. The study excluded all patients with congestive heart failure and chronic obstructive pulmonary disease, who may be candidates for beta-blockers in actual practice. In fact, a recent observational study in patients with severe left ventricular dysfunction suggested that these drugs substantially reduced the incidence of death in the short term and the long term.38 Finally, half the surgeries were nonelective, which makes extrapolation of their risk profile by the Revised Cardiac Risk Index difficult, since Lee et al excluded patients undergoing emergency surgery from the cohorts from which they derived and validated their index criteria.
Nevertheless, the authors concluded that patients at intermediate risk derive no benefit from perioperative beta-blocker use, and that the odds ratio for death was actually higher in patients with no risk factors who received a beta-blocker.
