POISE

From Wiki Journal Club
Jump to navigation Jump to search
Devereaux PJ, et al. "Effects of Extended-Release Metoprolol Succinate in Patients Undergoing Non-Cardiac Surgery: a Randomised Controlled Trial". The Lancet. 2008. 371(9627):1839-1847.
PubMedFull textPDF

Clinical Question

In patients undergoing non-cardiac surgery, does the use of perioperative beta-blockers reduce the risk of CV events or death?

Bottom Line

Although metoprolol reduced the risk of perioperative MI among patients undergoing non-cardiac surgery, this benefit was more than nullified by a higher rate of mortality and strokes.

Major Points

Studies investigating the role of perioperative beta-blockade have yielded conflicting results. Some trials have shown a reduction in perioperative MI, while others have demonstrated an increased risk of cardiogenic shock and death. Much of the evidence supporting the use of perioperative beta blockers came from the DECREASE trials, which were discredited in 2011 for lack of appropriate consent, inappropriate collection of data, and data fabrication.[1] Before they were discredited, DECREASE-I (1999) and DECREASE-IV (2009) supported the initiation of bisoprolol 7-21 days before surgery. No large trial had evaluated the use of beta blockers on the day of surgery.

The 2008 Perioperative Ischemic Evaluation (POISE) trial randomized 8,351 patients to metoprolol on the day of surgery. It found that beta-blockers reduced the risk of perioperative MI compared to placebo (5.8% vs. 6.9%), but this benefit was more than nullified by a higher rate of death (3.1% vs. 2.3%) and stroke (1% vs. 0.5%). A major drawback of the study was its blanket use of high-dose metoprolol succinate (100mg started 2-4 hours before surgery and then continued at 200mg afterwards) which may have resulted in more strokes and deaths due to hypotension.

A 2014 meta-analysis[2] excluding the DECREASE trials concluded that perioperative beta blockers started ≤1 day of noncardiac surgery prevented nonfatal MI but increases stroke, death, hypotension, and bradycardia. Without the results from the DECREASE studies, there was not enough information to inform the use of beta blockers initiated ≥2 days before surgery.

Guidelines

ACC/AHA Perioperative CV evaluation and management for non-cardiac surgery (2014, adapted)[3]

  • Patients on chronic beta blockers who are undergoing surgery should continue taking beta blockers (class I, level B)
  • It is reasonable to manage beta blockers after surgery with guidance by clinical circumstances, regardless of when the beta blocker was initiated (class IIa, level B)
  • If preoperative risk stratification indicates intermediate- or high-risk myocardial ischemia, it may be reasonable to initiate perioperative beta blockers (class IIb, level C)
  • If a compelling reason to be on long-term beta blockers but no other RCRI risk factors, initiation of beta blockers periperatively to reduce perioperative risk is of unknown benefit (class IIb, level B)
  • If starting beta blockers, it may be reasonable to initiate them long enough in advance (>1 day before surgery) to assess their safety and how well they are tolerated (class IIb, level B)
  • Do not start beta blockers on the day of surgery (class III, level B)

Design

  • Multicenter, parallel-group, randomized, placebo-controlled trial
  • N=8,351
    • Metoprolol (n=4,174)
    • Placebo (n=4,177)
  • Setting: 190 centers in 23 countries
  • Enrollment: 2002-2007
  • Analysis: Intention-to-treat

Population

Inclusion Criteria

  • Age ≥45 years
  • Non-cardiac surgery
  • Expected hospital length of stay ≥24h
  • Any one of the following:
    • History of CAD, PAD, or stroke
    • Hospitalization for CHF within previous 3 years
    • Undergoing major vascular surgery
    • Any 3 of:
      • Undergoing intrathoracic or intraperitoneal surgery
      • History of CHF, TIA, DM
      • Serum creatinine >175 µmol/L (1.98 mg/dL)
      • Age >70 years
      • Undergoing urgent or emergent surgery

Exclusion Criteria

  • Heart rate <50 bpm
  • Second or third degree heart block
  • Asthma
  • Receiving a beta-blocker or their physician planned to start one perioperatively
  • Prior adverse reaction to a beta-blocker
  • CABG surgery in the preceding 5 years and no cardiac ischemia since
  • Low-risk surgical procedure (based on individual physician's judgment)
  • Receiving verapamil
  • Prior enrollment in POISE

Baseline characteristics

  • Demographics: Age 69 years, female 37%
  • Baseline health data: HR 78 bpm, BP 139/78 mmHg
  • PMH: Current smoker 19%, HTN 63%, CAD 43%, PAD 41%, stroke 15%, TIA 10.5%, HF 6%
  • Receiving oral hypoglycemic agent or insulin: 29%
  • HF hospitalization within 3 years: 2.7%
  • Preoperative serum creatinine >175 µmol/L (1.98 mg/dl): 5%
  • Preoperative cardiac medications:
    • Aspirin: 36%
    • LMWH or UFH: 9.3%
    • ACE inhibitor or ARB: 44.5%
    • Statin: 32.0%
    • Diuretic: 21%
    • Calcium channel blocker: 22%
  • Emergent/urgent surgery: 10.5%
  • Surgery type: Vascular 42%, intraperitoneal 22%, orthopedic 21%
  • Anesthesia/analgesia: General 47.5%, spinal 17%, lumbar epidural 11%

Interventions

  • Randomized to metoprolol succinate or placebo
  • First dose was 100mg, given 2-4 hours before surgery
  • Second dose was 100mg, given 0-6 hours after surgery
  • Thereafter patients were given 200mg daily for 30 days
  • Study drug administration was given if HR ≥50 bpm and SBP ≥100 mmHg

Outcomes

Comparisons are metoprolol vs. placebo.

Primary Outcomes

Composite of CV death, non-fatal MI, and non-fatal cardiac arrest at 30 days
5.8% vs. 6.9% (HR 0.84; 95% CI 0.70-0.99; P=0.0399)

Secondary Outcomes

CV death
1.8% vs. 1.4% (HR 1.30; 95% CI 0.92-1.83; P=0.1368)
Non-fatal MI
3.6% vs. 5.1% (HR 0.70; 95% CI 0.57-0.86; P=0.0008)
Non-fatal cardiac arrest
0.5% in both groups (p=NS)
Any MI
4.2% vs. 5.7% (HR 0.73; 95% CI 0.60-0.89; P=0.0017)
Death
3.1% vs. 2.3% (HR 1.33; 95% CI 1.03-1.74; P=0.0317)
Stroke
1.0% vs. 0.5% (HR 2.17; 95% CI 1.26-3.74; P=0.0053)

Subgroup Analysis

  • Primary subgroup analysis based on the revised cardiac risk index (RCRI) scoring system
  • Secondary subgroup analysis based on sex, type of surgery, and use of epidural or spinal anaesthetic
  • None of these analyses showed a subgroup effect (data not provided in paper), but analyses were underpowered to detect modest differences in subgroup effects that one might expect to detect if there was a true subgroup effect.

Adverse Events

Clinically significant hypotension
15.0% vs. 9.7% (HR 1.55; 95% CI 1.38-1.74; P<0.0001)
Clinically significant bradycardia
6.6% vs 2.4% (HR 2.74; 95% CI 2.19-3.43; P<0.0001)

Criticisms

  • Blanket use of high-dose metoprolol succinate (100mg started 2-4 hours before surgery and then continued at 200mg afterwards) may have resulted in more strokes and deaths due to hypotension.

Funding

Received public funding and private support through AstraZeneca.

Further Reading