Timing of Complete Revascularization with Multivessel PCI for Myocardial Infarction (MULTISTARS AMI)

Authors, Journal, Affiliations, Type, DOI

• Stähli BE, Varbella F, Linke A, et al. for the MULTISTARS AMI Investigators
• N Engl J Med 2023;389:1368–79
• Lead institutions: University Heart Center, University Hospital Zurich; 37 sites across Europe
• Type: International, investigator-initiated, open-label, randomized noninferiority trial
• DOI: https://doi.org/10.1056/NEJMoa2307823

Overview

MULTISTARS AMI was a European multicenter RCT (n=840) in hemodynamically stable STEMI patients with multivessel CAD that compared immediate multivessel PCI (all lesions treated during the index procedure) versus staged multivessel PCI (nonculprit lesions treated 19–45 days later). The trial demonstrated that immediate multivessel PCI was not only noninferior but statistically superior to staged PCI for the composite primary endpoint at 1 year (8.5% vs 16.3%; RR 0.52; P<0.001 for both noninferiority and superiority). The benefit was driven primarily by fewer nonfatal MIs and unplanned ischemia-driven revascularizations in the first 45 days, supporting single-procedure complete revascularization as a viable and preferable strategy in appropriate STEMI patients.

Keywords

STEMI, multivessel coronary artery disease, complete revascularization, immediate PCI, staged PCI, nonculprit lesion, percutaneous coronary intervention

Key Takeaways

Background and Rationale

• Multivessel CAD is present in ~40–50% of STEMI patients and is associated with increased risk of recurrent MI and death
• Prior trials (COMPLETE, DANAMI-3-PRIMULTI, PRAMI, CvLPRIT) established that complete revascularization is superior to culprit-lesion-only PCI; however, the optimal timing (immediate vs staged) had not been established by an RCT
• The COMPLETE trial randomized patients to staged complete revascularization vs culprit-only — it did not evaluate immediate multivessel PCI during the index procedure

Trial Design

• International, open-label, randomized noninferiority trial; 37 European centres; October 2016–June 2022
• Eligibility: STEMI within 24 h of symptom onset + multivessel CAD (≥70% stenosis in ≥1 nonculprit artery ≥2.25 mm) + successful culprit PCI + hemodynamically stable
• Exclusions: cardiogenic shock, left main CAD, chronic total occlusion, prior CABG, stent thrombosis, in-stent restenosis
• Randomization 1:1 after successful culprit PCI: immediate multivessel PCI (same procedure) vs staged PCI (19–45 days post-index)
• Recommended stent: third-generation biodegradable-polymer everolimus-eluting stent (Synergy, Boston Scientific)
• FFR-guided and intravascular imaging-guided PCI at operator's discretion (low uptake in practice)
• Primary endpoint: composite of all-cause death, nonfatal MI, stroke, unplanned ischemia-driven revascularization, or HF hospitalization at 1 year
• Sample size: 840 patients (designed for 18% event rate, NI margin RR 1.46, one-sided α=0.05)

Results — Primary Endpoint

• Immediate group: 35/418 patients (8.5%) with primary endpoint event
• Staged group: 68/422 patients (16.3%) with primary endpoint event
• Risk ratio 0.52 (95% CI 0.38–0.72); P<0.001 for noninferiority; P<0.001 for superiority
• Per-protocol population confirmed: 8.3% vs 16.5% (RR 0.50, 95% CI 0.36–0.71)

Results — Individual Components

• Nonfatal MI: 2.0% (immediate) vs 5.3% (staged); HR 0.36 (95% CI 0.16–0.80)
  • 7 of 22 nonfatal MIs in staged group occurred before the planned staged procedure
  • Procedure-related MIs: 3 (immediate) vs 14 (staged); exploratory analysis excluding type 4 MI still supported primary result
• Unplanned ischemia-driven revascularization: 4.1% vs 9.3%; HR 0.42 (95% CI 0.24–0.74)
  • 23 of 39 unplanned revascularizations in staged group occurred before the planned staged procedure
• All-cause death: HR 1.10 (95% CI 0.48–2.48) — no significant difference
• Stroke: No significant difference
• HF hospitalization: No significant difference

Landmark Analysis

• Days 0–45: primary endpoint 3.6% (immediate) vs 10.7% (staged); HR 0.33 (95% CI 0.18–0.59)
• Day 45 to 1 year: no significant difference (HR 0.86; 95% CI 0.47–1.57)
• The between-group difference was entirely concentrated in the pre-staged intervention window

Safety

• Major bleeding (BARC type 3/5): 3.1% vs 4.8%; HR 0.65 (95% CI 0.32–1.31) — no significant difference
• Serious adverse events: 104 patients (immediate) vs 145 patients (staged)
• Crossover to immediate from staged: 0%; crossover from immediate to staged: 2.9% (12 patients)
• Quality of life (EQ-5D-5L): no significant between-group difference

Mechanistic Insights

• Unstable plaque features in nonculprit lesions in STEMI patients may predispose to plaque rupture during the waiting period before staged PCI (supported by COMPLETE OCT substudy and natural-history studies)
• Improved coronary blood flow after immediate complete revascularization may reduce the ischemic burden in the early post-STEMI phase
• Immediate strategy also reduces total contrast volume, radiation exposure, need for a second arterial puncture, and second hospitalization

Limitations of the Document

• Primary endpoint was expanded mid-trial (July 2019, after 217 patients enrolled) due to slow recruitment — stroke and HF hospitalization were added to the composite; these additions were relatively rare and balanced, but the protocol amendment introduces potential noninferiority bias
• Open-label design — knowledge of treatment assignment may have influenced clinical decisions (e.g., lower threshold for earlier ischemia-driven revascularization in staged group when coronary anatomy was already known)
• Small female representation — limits generalizability to women
• Selective eligibility: Excludes cardiogenic shock, left main CAD, CTO, prior CABG — these are clinically important subgroups where the strategy may differ
• Nonculprit lesion indication based on visual angiographic assessment only — no FFR or intravascular imaging mandate; low intravascular imaging uptake limits contemporary applicability to imaging-guided practice
• Procedure-related MI ascertainment bias — diagnosing periprocedural MI in the setting of STEMI (with elevated biomarkers and symptoms) is inherently challenging; immediate group may have had procedure-related MIs go undetected; exploratory analysis excluding type 4 MI was supportive but not definitive
• COVID-19 pandemic — may have affected enrollment and procedural timing in some patients
• Window for staged PCI (19–45 days): Does not address whether earlier staged intervention (e.g., 3–7 days) might achieve equivalent results with lower event risk
• Supported by Boston Scientific (manufacturer of Synergy stent), though investigators retained full control of design, analysis, and publication

Key Concepts Mentioned

• concepts/Multivessel-PCI-STEMI-Timing — primary subject: immediate vs staged nonculprit PCI in STEMI
• concepts/Intracoronary-Imaging-Guided-PCI — FFR and intravascular imaging were optional; low uptake in this trial

Key Entities Mentioned

• entities/Acute-Coronary-Syndrome — STEMI management and complete revascularization strategy

Wiki Pages Updated

• Created wiki/sources/complete-pci-multistars-ami-nejm-2023.md
• Created wiki/concepts/Multivessel-PCI-STEMI-Timing.md
• Updated wiki/entities/Acute-Coronary-Syndrome.md
• Updated wiki/sourceindex.md
• Updated wiki/wikiindex.md