Perioperative Ischemic Evaluation - POISE

Contribution To Literature:

The POISE trial revealed that perioperative beta-blockade was effective at reducing adverse events, but at a cost of increased mortality and stroke.


The goal of the trial was to evaluate the effect of treatment with the beta-blocker metoprolol compared with placebo on major cardiovascular (CV) events in patients undergoing noncardiac surgery.

Study Design

  • Placebo Controlled
  • Randomized
  • Blinded

Patients Screened: 9,298
Patients Enrolled: 8,351
Mean Follow-Up: 30 days
Mean Patient Age: 69 years
Female: 37%

Patient Populations:

• Undergoing noncardiac surgery
• Age at least 45 years
• Expected length of stay at least 24 hours
• Any one of the following six criteria: coronary artery disease, peripheral vascular disease, history of stroke due to atherothrombotic disease, hospitalization for congestive heart failure within 3 years of randomization, undergoing major vascular surgery, or any three of seven other high-risk criteria


• Contraindication to metoprolol
• Clinical plan to use a beta-blocker preoperatively or during the first 30 postoperative days
• Prior adverse reaction to a beta-blocker
• Coronary artery bypass graft (CABG) surgery with complete revascularization in the preceding 5 years and no evidence of cardiac ischemia since the CABG surgery
• Patients undergoing low-risk surgical procedures
• Concurrent use of verapamil

Primary Endpoints:

Major CV events (defined as CV death, nonfatal MI, or nonfatal cardiac arrest)

Secondary Endpoints:

• Clinically significant atrial fibrillation
• Rehospitalization for cardiac reasons
• Components of the composite endpoint
• Total mortality
• Revascularization procedures
• Congestive heart failure
• Clinically significant bradycardia
• Clinically significant hypotension

Drug/Procedures Used:

Patients were randomized in a double-blind manner to treatment with extended-release metoprolol (n = 4,174) or placebo (n = 4,177). A dose of the study drug (metoprolol 100 mg controlled release or placebo) was given 2-4 hours prior to surgery and again 0-6 hours after surgery. Daily doses of study drug (metoprolol 200 mg or placebo) were then taken for the next 30 days.

Concomitant Medications:

At baseline, the use of aspirin was 36% versus 36%, heparin was 9.3% versus 9.2%, angiotensin-converting enzyme inhibitor or angiotensin-receptor blocker was 44% versus 45%, statin was 32% versus 32%, and diuretic was 22% versus 20%, respectively, for metoprolol versus placebo.

Principal Findings:

At study entry, 43% of patients had coronary artery disease, 41% had peripheral vascular disease, and 15% had a prior stroke. Type of surgery performed was vascular in 42% of patients, intraperitoneal in 22%, orthopedic in 21%, and other types in 15%.

The primary endpoint of CV death, myocardial infarction (MI), or cardiac arrest was reduced in the metoprolol group compared with placebo (5.8% vs. 6.9%, hazard ratio [HR] 0.84, 95% CI 0.70-0.99, p = 0.04), driven by a reduction in nonfatal MI (3.6% vs. 5.1%, HR 0.70, p = 0.0008). There were also reductions with metoprolol in revascularization (0.3% vs. 0.6%, p = 0.012) and new-onset atrial fibrillation (2.2% vs. 2.9%, p = 0.044). However, total mortality was increased in the metoprolol group (3.1% vs. 2.3%, HR 1.33, p = 0.032), as was stroke (1.0% vs. 0.5%, HR 2.17, p = 0.0053). The metoprolol group also had increased rates of significant hypotension (15.0% vs. 9.7%, p < 0.0001).

One-year follow-up:

  • All-cause mortality: 10% with metoprolol vs. 9% with placebo (p = 0.036)
  • MI: 5% with metoprolol vs. 6% with placebo (p = 0.008)
  • Stroke: 2% with metoprolol vs. 1% with placebo (p = 0.014)


Among patients undergoing noncardiac surgery, treatment with the beta-blocker metoprolol was associated with a reduction in the primary endpoint of CV death, MI, or cardiac arrest at 30 days compared with placebo, but total mortality and stroke were higher in the metoprolol group. Findings were materially unchanged at 1 year.

Prior studies have evaluated prophylactic beta-blocker use for patients undergoing vascular surgery and have shown mixed results, with the MaVS and DIPOM trials finding no benefit on clinical events, but an earlier trial using bisoprolol finding a reduction in cardiac death and MI. However, as with the present study, hypotension and bradycardia increased with metoprolol in MaVS and DIPOM.

Additionally, in the present trial, total deaths and strokes were more frequent in the metoprolol arm, with the reduction in the primary endpoint driven by nonfatal MIs. Although the post-surgical CV event rate was high, even in this population of patients undergoing noncardiac surgery, given the increased risk of death, stroke, and severe hypotension with metoprolol, routine prophylactic therapy does not appear to be a safe approach to reducing CV events in this population.


Presented by Dr. P.J. Devereaux at the American College of Cardiology Annual Scientific Session (ACC 2018), Orlando, FL, March 12, 2018.

POISE Study Group. Effects of extended-release metoprolol succinate in patients undergoing non-cardiac surgery (POISE trial): a randomised controlled trial. Lancet 2008;371:1839-47.

Presented by Dr. P.J. Devereaux at the American Heart Association Annual Scientific Sessions, Orlando, FL, November 2007.

Clinical Topics: Arrhythmias and Clinical EP, Heart Failure and Cardiomyopathies, Atherosclerotic Disease (CAD/PAD), Implantable Devices, SCD/Ventricular Arrhythmias, Atrial Fibrillation/Supraventricular Arrhythmias, Acute Heart Failure

Keywords: ACC18, ACC Annual Scientific Session, Coronary Artery Disease, Myocardial Infarction, Stroke, Heart Failure, Hypotension, Atrial Fibrillation, Bradycardia, Bisoprolol, Heart Arrest, Metoprolol, Myocardial Ischemia, Peripheral Vascular Diseases, Surgical Procedures, Operative

< Back to Listings