Perioperative Ischemic Evaluation-3 - POISE-3

Contribution To Literature:

Among patients undergoing noncardiac surgery at risk of vascular events in the POISE-3 trial, tranexamic acid resulted in lower bleeding compared with placebo but did not meet criteria for noninferiority for cardiovascular safety. Further, a perioperative hypotension-avoidance strategy did not differ from hypertension-avoidance strategy with regard to 30-day major vascular complications.


The goals of the trial were to compare the effects of intraoperative tranexamic acid vs. placebo on bleeding, as well as perioperative hypotension-avoidance strategy vs. hypertension-avoidance strategy on major adverse complications.

Study Design

Patients were randomized in a 2x2 factorial design: 9,535 to tranexamic acid (n = 4,757) vs. placebo (n = 4,778), and 7,490 patients randomized to a hypotension-avoidance strategy (n = 3,742) vs. a hypertension-avoidance strategy (n = 3,748).

Tranexamic acid vs. placebo: Tranexamic acid was given intravenously as a 1 g bolus at the start and end of surgery.

Hypertension- vs. hypotension-avoidance strategy: Patients were given chronic antihypertensive medications and mean arterial pressure (MAP) ≥60 was targeted intraoperatively vs. a hypotension-avoidance strategy where antihypertensives were only given if hypertensive prior to surgery and MAP ≥80 was targeted intraoperatively.

  • Duration of follow-up: 30 days
  • Mean patient age: 70 years
  • Percentage female: 44%

Inclusion criteria:

  • Age ≥45 years undergoing inpatient noncardiac surgery at risk of perioperative bleeding and cardiovascular events
  • Taking ≥1 antihypertensive medications

Exclusion criteria:

  • New York Heart Association class III-IV heart failure, or left ventricular ejection fraction (LVEF) ≤30%; hemodynamic instability
  • Cranial neurosurgery or cardiac surgery
  • Hypertension-related cerebral hemorrhage, thyrotoxicosis, pheochromocytoma
  • Planned use of systemic tranexamic acid during surgery (for tranexamic acid vs. placebo comparison)
  • Creatinine clearance of <30 ml/min or on hemodynamic (for tranexamic acid vs. placebo comparison)

Other salient features/characteristics:

  • Mean two chronic antihypertensive medications with approximately 25% taking ≥3 medications
  • Urgent or emergent surgery: 79%
  • Type of surgery: general (37%), orthopedic (22%), vascular (15%)
  • History of chronic heart failure: 14%

Principal Findings:

The primary efficacy outcome, composite bleeding event, for tranexamic acid vs. placebo at 30 days, was 9.1% vs. 11.7% (hazard ratio [HR] 0.76, 95% confidence interval [CI] 0.67-0.87, p < 0.001).

  • Life-threatening bleeding: 1.6% vs. 1.7%
  • Major bleeding: 7.6% vs. 10.4%
  • Bleeding into a critical organ: 0.3% vs. 0.4%

The primary safety outcome, composite cardiovascular outcome, for tranexamic acid vs. placebo, was 14.2% vs. 13.9% (HR 1.02, 95% CI 0.92-1.14; p for noninferiority = 0.04).

  • Myocardial injury after noncardiac surgery (MINS): 12.8% vs. 12.6%
  • Nonhemorrhagic stroke: 0.5% vs. 0.3%
  • Symptomatic proximal venous thromboembolism: 0.7% vs. 0.6%

Secondary outcomes for tranexamic acid vs. placebo:

  • Myocardial infarction: 1.4% vs. 1.1%

The primary outcome of major vascular complication, a composite of vascular death, nonfatal MINS, stroke, or cardiac arrest, for hypotension-avoidance vs. hypertension-avoidance, was 13.9% vs. 14.0% (p = 0.92).

Results were not modified by status of randomization to tranexamic acid or placebo group (p for interaction = 0.49).

Secondary outcomes for hypotension-avoidance vs. hypertension-avoidance:

  • MINS 12.7% vs. 12.8% (p = 0.84)
  • Myocardial infarction: 1.4% vs. 1.2% (p = 0.41)
  • Stroke: 0.5% vs. 0.5% (p > 0.99)
  • Vascular mortality: 0.7% vs. 0.6% (p = 0.88)
  • All-cause mortality: 1.3% vs. 1.1% (p = 0.46)

Tertiary outcomes for hypotension-avoidance vs. hypertension-avoidance:

  • Length of stay: 4.0 days vs. 4.0 days (p = 0.34)
  • Days alive and at home: 25.0 days vs. 25.0 days (p = 1.00)


The results of this trial indicate that:

a) Routine administration of tranexamic acid during noncardiac surgery resulted in less bleeding compared with placebo, but it did not meet criteria for noninferiority for the primary cardiovascular safety endpoint. Providers will need to weigh these risk-benefits when making decisions about use of tranexamic acid for noncardiac surgery. Cost-effectiveness analyses will also be important.

b) Among patients at risk of vascular events, a hypotension-avoidance strategy did not differ from hypertension-avoidance strategy with regard to 30-day major vascular complications. Chronic hypertension is common among patients undergoing noncardiac surgeries, and practice varies with regard to taking antihypertensive medications prior to surgery. Patients with chronic hypertension undergo frequent hemodynamic abnormalities, with intraoperative or postoperative hypotension being linked to complications.

There are questions regarding the safety of withholding beta-blockers and what blood pressure to target during surgery. The results of the POISE-3 trial inform these questions, with no difference observed between targeting a MAP of 60 vs. 80 intraoperatively. Additionally, there were no differences observed between administering chronic antihypertensives to avoid hypertension or continuing medications based on blood pressure prior to surgery. Future research should focus on strategies to reduced vascular events in the peri- and postoperative setting.


Highlighted text has been updated as of May 24, 2023.

Marcucci M, Painter TW, Conen D, et al., on behalf of the POISE-3 Trial Investigators and Study Groups. Hypotension-Avoidance Versus Hypertension-Avoidance Strategies in Noncardiac Surgery: An International Randomized Controlled Trial. Ann Intern Med 2023;176:605-14.

Devereaux PJ, Marcucci M, Painter TW, et al., on behalf of the POISE-3 Investigators. Tranexamic Acid in Patients Undergoing Noncardiac Surgery. N Engl J Med 2022;386:1986-97.

Editorial: Berger JS. Bleeding Outcomes After Noncardiac Surgery — Are We POISEd to Do Better? N Engl J Med 2022;386:2052-3.

Presented by Dr. Maura Marcucci at the American College of Cardiology Annual Scientific Session (ACC 2022), Washington, DC, April 4, 2022.

Presented by Dr. P.J. Devereaux at the American College of Cardiology Annual Scientific Session (ACC 2022), Washington, DC, April 2, 2022.

Clinical Topics: Heart Failure and Cardiomyopathies, Prevention, Pulmonary Hypertension and Venous Thromboembolism, Vascular Medicine, Acute Heart Failure, Hypertension

Keywords: ACC22, ACC Annual Scientific Session, Antihypertensive Agents, Antifibrinolytic Agents, Heart Failure, Hemorrhage, Hypertension, Hypotension, Inpatients, Myocardial Infarction, Perioperative Care, Primary Prevention, Risk, Stroke, Tranexamic Acid, Vascular Diseases, Venous Thromboembolism

< Back to Listings