IVUS-Guided versus Angiography-Guided PCI in Unprotected Left Main Coronary Disease (OPTIMAL)

Authors, Journal, Affiliations, Type, DOI

• Luca Testa, Ernest Spitzer (co-first authors), Adrian P. Banning (last author), and 23 co-authors for the OPTIMAL Investigators
• 28 European centres across Italy, Spain, and the United Kingdom; lead site: IRCCS Policlinico San Donato, Milan, Italy
• The New England Journal of Medicine, published March 30, 2026
• Type: Investigator-initiated, international, multicentre, open-label, randomised superiority trial
• DOI: 10.1056/NEJMoa2600440
• Funded by Philips Image Guided Therapy Devices and Boston Scientific (unrestricted grants; no role in design, conduct, or publication)
• ClinicalTrials.gov: NCT04111770

Overview

The OPTIMAL trial randomised 806 patients with unprotected left main coronary artery disease (LMCA) at 28 European centres to IVUS-guided PCI versus angiography-guided PCI. At a median follow-up of 2.9 years, the patient-oriented composite primary endpoint (any stroke, MI, revascularisation, or death from any cause) occurred in 33.7% of the IVUS group versus 30.9% of the angiography group (HR 1.11; 95% CI 0.87–1.42; P=0.40), showing no superiority of IVUS guidance. No significant differences were seen across device-oriented, vessel-oriented, or any individual secondary endpoints. Stroke was numerically higher in the IVUS arm (3.0% vs 1.0%; HR 3.11) but considered an unexpected, unexplained finding. These results challenge the Class IA guideline recommendation for routine IVUS use in LMCA PCI, at least in expert high-volume centres.

Keywords

IVUS guidance, unprotected left main coronary artery disease, PCI, angiography-guided PCI, OPTIMAL trial, patient-oriented composite, SYNTAX score, bifurcation, stent optimisation

Key Takeaways

Background and Rationale

• Unprotected left main coronary artery (LMCA) disease is found in 4–6% of diagnostic coronary angiographies and carries substantial ischaemic risk
• PCI is an accepted alternative to CABG for low-to-intermediate complexity LMCA disease (established by EXCEL, NOBLE, PRECOMBAT, SYNTAX trials)
• IVUS guidance used in >50% of LMCA PCI in contemporary practice; recommended Class IA by 2024 ESC chronic coronary syndromes guidelines and 2025 ACC/AHA ACS guidelines
• Prior observational studies and meta-analyses suggested IVUS use associated with lower death, MI, and stent thrombosis in LMCA PCI; but large-scale randomised data were lacking specifically for LMCA

Methods

• Randomisation: 1:1 IVUS-guided vs angiography-guided PCI; stratified by centre
• Eligibility: LMCA stenosis ≥50% (visual estimation), deemed suitable for PCI by heart team, with evidence of ischaemia or symptoms on GDMT; included ostial LAD/LCx extensions
• All patients received everolimus-eluting platinum-chromium stents (Synergy/Synergy Megatron)
• IVUS group: pre-procedural IVUS strongly recommended; post-procedural IVUS mandatory; prespecified minimum lumen area targets (LMCA body ≥8 mm², bifurcation ≥7 mm², proximal LAD/LCx ≥6 mm², proximal LCx ≥5 mm²); additional dilation/stenting if criteria not met
• Angiography group: IVUS discouraged unless safety concerns; 14 patients (3.5%) received IVUS for safety reasons
• Primary endpoint: patient-oriented composite of any stroke, any MI, any revascularisation, or death from any cause at longest follow-up
• Observation period extended by protocol amendment to allow sufficient event accrual

Patient Characteristics

• 806 patients randomised July 2020–June 2023; median follow-up 2.9 years (IQR 2.2–3.7)
• Mean age 71.4±10.7 years; 78.4% male; 34.7% diabetes
• Clinical presentations: NSTEMI 39.1%, unstable angina 10.1%, chronic coronary syndrome 50.8%
• Mean SYNTAX score: 29.7±12.6 — intermediate-to-high complexity (higher than EXCEL mean 26.5, NOBLE mean 22.5)
• Population: older and more comorbid than prior left main PCI trials; many ineligible for CABG
• Previous heart failure: 25.8% (IVUS) vs 19.8% (angiography) — slight imbalance at baseline

Procedural Details

• IVUS performed in 97.5% of IVUS group (pre-stent 65.1%, post-stent 96.0%)
• Post-stent dilation: 97.0% (IVUS) vs 96.0% (angiography); proximal optimisation technique 89.6% vs 85.1%
• Mean stent number per patient: 2.48±1.57 vs 2.40±1.42
• Mean total stent length: 58.68±40.53 mm vs 57.64±37.37 mm
• Procedure duration: 24.7 minutes longer (95% CI 18.6–30.9) with IVUS guidance
• Periprocedural complications: similar (7.4% vs 7.2%)

Primary Outcome

• Patient-oriented composite at 2.9 years: 33.7% (IVUS) vs 30.9% (angiography); HR 1.11 (95% CI 0.87–1.42); P=0.40
• IVUS-guided PCI was NOT superior to angiography-guided PCI

Secondary Outcomes

• Device-oriented composite (CV death/target-vessel MI/target-lesion revascularisation): 22.4% vs 20.5%; HR 1.10 (NS)
• Vessel-oriented composite: 24.2% vs 21.5%; HR 1.14 (NS)
• Death from any cause: 15.7% vs 15.1%; HR 1.06 (NS)
• Myocardial infarction: 11.2% vs 10.9%; HR 1.04 (NS)
• Repeat revascularisation: 12.0% vs 11.1%; HR 1.10 (NS)
• Definite stent thrombosis: 0.7% vs 0.2%; HR 3.07 (95% CI 0.32–29.53) — numerically higher in IVUS arm (low event numbers; wide CI; NS)
• Stroke: 3.0% (IVUS) vs 1.0% (angiography); HR 3.11 (95% CI 1.00–9.65) — unexpected, unexplained finding; median time to stroke 19 months (not immediately periprocedural); considered possibly chance occurrence

Discussion — Why Results Were Neutral

1. Enrolment of high-risk patients with higher SYNTAX scores and more comorbidities than prior left main PCI trials; these patients may have higher background event risk that IVUS cannot mitigate
2. Participating centres had extensive IVUS expertise (50–100% IVUS use in LMCA PCI before trial start); experienced operators may achieve near-equivalent angiographic results with established left main stenting algorithms originally derived from IVUS
3. Rigorous procedural standards and contemporary second-generation DES platform in both arms reduced the gap that IVUS guidance may otherwise close
4. Trial powered for 35% relative risk reduction; observed event rate higher than assumed, but no difference persisted in sensitivity analyses

Limitations of the Document

• Open-label design; potential performance bias in both arms
• Enrolled exclusively in Europe (Italy, Spain, UK); limited generalisability to other regions or risk subgroups (e.g., acute presentations, lower SYNTAX scores)
• Recruitment occurred during COVID-19 pandemic, potentially affecting recruitment pace and follow-up completeness
• Powered for 35% RRR; underpowered to exclude smaller but clinically meaningful differences
• No independent Data and Safety Monitoring Board appointed — steering committee monitored aggregate event data
• Stroke finding (HR 3.11) requires cautious interpretation: low absolute event numbers, CI spanning 1.00 to 9.65, unexplained mechanism, median onset 19 months post-procedure

Key Concepts Mentioned

• concepts/Intracoronary-Imaging-Guided-PCI — central intervention; trial specifically evaluates IVUS guidance in left main PCI

Key Entities Mentioned

• entities/Left-Main-Coronary-Disease — the target disease; OPTIMAL is the pivotal trial for this entity
• entities/Chronic-Coronary-Disease — CCS patients comprised 50.8% of enrolment

Wiki Pages Updated

• wiki/sources/ivus-optimal-nejm-2026.md — created (this file)
• wiki/concepts/Intracoronary-Imaging-Guided-PCI.md — created
• wiki/entities/Left-Main-Coronary-Disease.md — created
• wiki/entities/Chronic-Coronary-Disease.md — updated
• wiki/wikiindex.md — updated
• wiki/sourceindex.md — updated