POISE-1: Perioperative Metoprolol in Non-Cardiac Surgery

Metoprolol cut MI 27% but caused more strokes + deaths — net harm.

Population

8,351 patients with or at risk for atherosclerotic disease undergoing non-cardiac surgery; 190 hospitals, 23 countries.

Intervention

Metoprolol succinate 100 mg PO 2-4 hr pre-op + 100 mg q12 × 30 days.

Comparison

Matching placebo.

Outcome

Composite of CV death, non-fatal MI, non-fatal cardiac arrest at 30 days.

Double-blind RCT. NO TITRATION — fixed dose started day-of-surgery. This methodologic choice (no β-blocker exposure history before high-dose start) was central to the harm signal.

• Primary composite: 5.8% metoprolol vs 6.9% placebo (HR 0.84, P=0.04) — modest benefit.
• MI: 4.2% vs 5.7% (P=0.0008) — clear benefit.
• All-cause mortality: 3.1% vs 2.3% (HR 1.33, P=0.03) — increased.
• Stroke: 1.0% vs 0.5% (HR 2.17, P=0.005) — DOUBLED.
• Hypotension: 15.0% vs 9.7% (P<0.0001) — major mediator of strokes.

Do NOT START a β-blocker on day of surgery in β-blocker-naïve patients. If a patient is ALREADY on a chronic β-blocker, continue it (abrupt withdrawal is also harmful). Start preoperative β-blockade only in carefully selected high-risk patients (RCRI ≥3 + active ischemia) with ≥1 week of titration, monitoring HR + BP. POISE-1 reshaped ACC/AHA guidelines from 'consider β-blocker for everyone with CAD' to 'continue if on, otherwise individualize.'

• Fixed-dose extended-release → severe hypotension was somewhat baked-in.
• Did not test gradual titration over weeks (which DECREASE trial supports).
• Underpowered for some subgroups; vascular surgery subgroup analysis hypothesis-generating.

1. How do you handle the patient on chronic atenolol the morning of surgery? Continue? Hold?
2. Should POISE-1 close the door entirely on prophylactic perioperative β-blockade, or just on day-of-surgery initiation?
3. What about ESC guidelines that still recommend titrated β-blockade in selected vascular surgery patients?