The New England Journal of Medicine MARCH 31, 2014

Aspirin and Clonidine in Patients Undergoing Noncardiac Surgery

The POISE-2 Investigators

Bottom Line

The POISE-2 trial demonstrated that neither low-dose aspirin nor low-dose clonidine reduced the 30-day risk of death or nonfatal myocardial infarction in patients undergoing noncardiac surgery, while both interventions were associated with significant safety risks.

Key Findings

1. Aspirin did not reduce the primary outcome of death or nonfatal myocardial infarction at 30 days compared with placebo (7.0% vs. 7.1%; hazard ratio, 0.99; 95% CI, 0.86 to 1.15; P=0.92).
2. Aspirin significantly increased the risk of major bleeding compared with placebo (4.6% vs. 3.8%; hazard ratio, 1.23; 95% CI, 1.01 to 1.49; P=0.04).
3. Clonidine did not reduce the primary outcome of death or nonfatal myocardial infarction at 30 days compared with placebo (7.3% vs. 6.8%; hazard ratio, 1.08; 95% CI, 0.93 to 1.26; P=0.29).
4. Clonidine significantly increased the risk of clinically important hypotension (47.6% vs. 37.1%; hazard ratio, 1.32; 95% CI, 1.24 to 1.40; P<0.001) and nonfatal cardiac arrest (0.3% vs. 0.1%; hazard ratio, 3.20; 95% CI, 1.17 to 8.73; P=0.02).

Study Design

Design
RCT
Double-Blind
Sample
10,010
Patients
Duration
30 days
Median
Setting
Multicenter, International
Population Patients aged 45 years or older undergoing noncardiac surgery who were at risk for perioperative cardiovascular events.
Intervention Low-dose aspirin (initiated or continued) or low-dose clonidine (preoperative and 72-hour postoperative transdermal patch).
Comparator Matching placebo for either intervention.
Outcome Composite of death or nonfatal myocardial infarction at 30 days.

Study Limitations

The 2x2 factorial design necessitated strict monitoring for independent effects, though potential interactions were investigated.
The trial focused on 30-day outcomes, potentially missing longer-term cardiovascular or mortality benefits or harms.
The study population was heterogeneous regarding baseline cardiovascular risk and surgical procedures, which may limit the generalizability to specific sub-populations.
The results do not support the routine prophylactic use of these agents in broad noncardiac surgical populations, challenging prior observational hypotheses.

Clinical Significance

The trial provides robust evidence against the routine perioperative administration of aspirin or clonidine to prevent major adverse cardiovascular events in noncardiac surgery. The findings emphasize that the potential for harm, particularly major bleeding with aspirin and hypotension with clonidine, outweighs the lack of observed benefit in these perioperative settings.

Historical Context

The POISE-2 trial followed the initial POISE trial, which had previously discredited the routine use of perioperative beta-blockers due to an increased risk of stroke and mortality despite a reduction in myocardial infarction. POISE-2 aimed to investigate other commonly considered perioperative pharmacological interventions for cardioprotection, ultimately leading to a shift toward more cautious, evidence-based perioperative management.

Guided Discussion

High-yield insights from every perspective

Med Student
Medical Student

Based on the physiological mechanism of alpha-2 adrenergic agonists like clonidine, why was it hypothesized to reduce perioperative myocardial infarction, and what adverse hemodynamic effects observed in POISE-2 explain its failure to improve outcomes?

Key Response

Clonidine reduces sympathetic outflow, which theoretically prevents tachycardia and hypertension—common triggers for perioperative myocardial oxygen supply-demand mismatch. However, POISE-2 showed that clonidine significantly increased the risk of clinically important hypotension and bradycardia. These adverse effects likely offset any cardioprotective benefits by reducing coronary perfusion pressure, leading to the observed lack of benefit in preventing death or myocardial infarction.

Resident
Resident

A 65-year-old patient with hypertension and stable coronary artery disease is scheduled for a total hip arthroplasty. According to the POISE-2 results, what is the most appropriate management of their aspirin therapy in the immediate perioperative period?

Key Response

POISE-2 demonstrated that neither starting nor continuing low-dose aspirin reduced the risk of death or nonfatal MI, but it did significantly increase the risk of major bleeding (HR 1.23). For patients without a recent coronary stent (where the risk of stent thrombosis is paramount), the trial suggests that aspirin should generally be held perioperatively to minimize bleeding risk, as there is no evidence of a cardioprotective benefit in the noncardiac surgical setting.

Fellow
Fellow

How do the results of POISE-2 regarding aspirin efficacy inform our understanding of the pathophysiology of Perioperative Myocardial Infarction (PMI) compared to non-surgical spontaneous Myocardial Infarction (MI)?

Key Response

In non-surgical settings, aspirin is highly effective because most MIs are Type 1 (plaque rupture and thrombosis). The lack of benefit for aspirin in POISE-2 suggests that a large proportion of PMIs are likely Type 2 MIs (supply-demand mismatch) or involve different triggers than typical plaque rupture. This shifts the focus toward hemodynamic stability and oxygen balance rather than potent antiplatelet therapy for the prevention of Myocardial Injury after Noncardiac Surgery (MINS).

Attending
Attending

The POISE-2 trial results were definitive for a broad population, but what specific high-risk subgroups must we be careful not to de-escalate based on this study, and how do you teach trainees to navigate this 'negative' trial evidence?

Key Response

Practitioners must distinguish between 'perioperative aspirin initiation/continuation' and 'maintenance of essential antiplatelet therapy.' Patients with recent coronary stents (within 6 weeks for BMS or 1 year for DES) were largely excluded or represented a different risk profile. Teaching should emphasize that POISE-2 applies to preventing MI in general noncardiac surgery, but does not override the standard of care for secondary prevention in patients with high-risk vascular hardware where the risk of catastrophic thrombosis outweighs bleeding.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD
PhD

Assess the statistical power and validity of the 2x2 factorial design used in POISE-2; specifically, how should the researchers have accounted for the potential biological interaction between aspirin-induced bleeding and clonidine-induced hypotension on the primary endpoint?

Key Response

Factorial designs assume the absence of interaction between interventions. However, hypotension (clonidine) and bleeding (aspirin) can both lead to reduced organ perfusion and MI. If a synergistic negative effect existed, it could mask individual benefits. A rigorous analysis would require a formal test for interaction (multiplicative vs. additive) and sufficient power to detect such an interaction, which is often lacking in trials designed primarily to assess the main effects of two distinct therapies.

Journal Editor
Journal Editor

If a manuscript were submitted claiming clonidine's failure in POISE-2 was due to sub-therapeutic dosing or improper timing of the transdermal patch, what internal validity metrics would you require the authors to demonstrate to support a rebuttal?

Key Response

A reviewer would look for 'dose-response' evidence or pharmacokinetic data within the trial. In POISE-2, the use of a 0.2mg oral dose followed by a patch was intended to provide rapid and sustained levels. An editor would check if the achieved heart rate and blood pressure reductions in the clonidine group were significant enough to confirm a biological effect; if hemodynamic changes occurred without clinical benefit, it suggests the failure was due to the drug's profile rather than inadequate dosing.

Guideline Committee
Guideline Committee

How does the POISE-2 evidence necessitate a revision of the ACC/AHA guidelines regarding perioperative beta-blocker and alpha-2 agonist use, and what is the current strength of recommendation for aspirin in light of these findings?

Key Response

Prior to POISE-2, some guidelines were neutral or slightly favored aspirin. Currently, based on this Level A evidence, guidelines (like the ESC or ACC/AHA) generally recommend against the routine initiation of aspirin perioperatively (Class III: No Benefit/Harm) unless there is a specific indication like a recent stent. For clonidine, the recommendation is similarly a Class III, as it mirrors the POISE-1 findings with metoprolol: while some ischemia may be reduced, the increase in hypotension and mortality/stroke risk makes it unsuitable for routine prophylaxis.

Clinical Landscape

Noteworthy Related Trials

2008

POISE Trial

n = 8,351 · NEJM

Tested

Metoprolol succinate extended-release

Population

Patients undergoing noncardiac surgery at risk for perioperative cardiovascular events

Comparator

Placebo

Endpoint

Composite of cardiovascular death, nonfatal myocardial infarction, or nonfatal cardiac arrest

Key result: Perioperative administration of metoprolol reduced the risk of myocardial infarction but increased the risk of death and stroke.
2010

DECREASE-IV Trial

n = 1,066 · JAMA

Tested

Beta-blocker and statin therapy

Population

Patients undergoing noncardiac surgery

Comparator

Usual care

Endpoint

All-cause mortality and nonfatal myocardial infarction

Key result: Neither beta-blocker nor statin therapy significantly reduced the primary endpoint compared to usual care.
2014

DASH Trial

n = 753 · JAMA

Tested

Perioperative aspirin

Population

Patients undergoing noncardiac surgery with preoperative aspirin use

Comparator

Placebo

Endpoint

Major adverse cardiovascular events

Key result: The trial found no significant difference in MACE or major bleeding when continuing aspirin compared to placebo in high-risk patients.

Tailored to your role

Want this tailored to you?

Add your specialty or training stage to get role-specific takeaways and more questions.

Personalize this analysis