PCI in Patients Undergoing Transcatheter Aortic-Valve Implantation (NOTION-3)

Authors, Journal, Affiliations, Type, DOI

• Authors: Jacob Lønborg, Reza Jabbari, Muhammad Sabbah, et al. for the NOTION-3 Study Group
• Journal: New England Journal of Medicine, 2024;391:2189–2200
• Affiliations: Copenhagen University Hospital–Rigshospitalet (lead centre); 12 hospitals across Denmark, Finland, Sweden, and Latvia
• Type: International, open-label, randomized superiority trial (investigator-initiated)
• DOI: 10.1056/NEJMoa2401513
• Funding: Unrestricted grants from Boston Scientific and the Danish Heart Foundation (Boston Scientific had no role in conduct, writing, or approval)

Overview

NOTION-3 is the largest RCT to test whether FFR-guided or angiographically significant PCI in addition to TAVI improves outcomes versus TAVI alone (conservative treatment) in patients with stable CAD and severe symptomatic aortic stenosis. Enrolling 455 patients across 12 Nordic–Baltic sites (September 2017–October 2022), the trial demonstrated that PCI significantly reduced the primary composite of death/MI/urgent revascularisation at 2 years (HR 0.71; P=0.04) compared to conservative treatment. However, this came at the cost of higher bleeding (HR 1.51). All-cause mortality was not significantly different (HR 0.85). The trial superseded the prior neutral ACTIVATION trial by using FFR guidance, a larger sample, and longer follow-up.

Keywords

Percutaneous coronary intervention, transcatheter aortic valve implantation, aortic stenosis, coronary artery disease, fractional flow reserve, major adverse cardiac events, revascularization, NOTION-3

Key Takeaways

Background

• Coronary artery disease is present in approximately 50% of patients undergoing TAVI
• Between 10–20% of TAVI patients also receive PCI, yet no clear guideline recommendation existed prior to this trial
• The prior ACTIVATION trial (n=235; 1-year follow-up) did not meet non-inferiority for conservative treatment and was stopped early for futility — leaving the question unanswered

Study Design

• International, open-label, 1:1 randomized superiority trial; 12 sites in Denmark, Finland, Sweden, and Latvia
• Enrollment: September 2017–October 2022; n=455 randomized (227 PCI, 228 conservative)
• Inclusion criteria: Severe symptomatic aortic stenosis selected for TAVI by local heart team; ≥1 physiologically significant coronary stenosis (FFR ≤0.80, or angiographic diameter stenosis ≥90%) in vessel ≥2.5 mm
• Exclusion criteria: Life expectancy <1 year; eGFR <20 mL/min/1.73 m²; ACS within 14 days; left main coronary artery stenosis; valve-in-valve TAVI indication
• Primary endpoint: MACE = composite of death from any cause, myocardial infarction, or urgent revascularisation at median 2 years
• Analysis: Intention-to-treat principle; Cox regression; Kaplan–Meier time-to-event plots

Patient Population

• Median age 82 years (IQR 78–85); 67% male
• Median STS-PROM 3% (IQR 2–4) — low-intermediate surgical risk
• Median SYNTAX score 9 (IQR 6–14) — low coronary complexity
• ~2/3 of patients enrolled between 2020–2022 (contemporary practice)
• Groups well-balanced except: lower current/former smokers and COPD in PCI group; PCI group had more lesions meeting FFR/angiographic criteria

Intervention Details

• PCI group: Complete revascularisation of all eligible lesions in vessels ≥2.5 mm meeting FFR ≤0.80 or ≥90% diameter stenosis; completed before TAVI (strongly recommended, protocol); concomitant or ≤2 days post-TAVI allowed; 8 patients did not undergo PCI; complete revascularisation achieved in 89%
• Antithrombotic (PCI group): Lifelong aspirin 75 mg OD + clopidogrel 75 mg OD × 6 months post-PCI; OAC patients: aspirin 7 days (post-AUGUSTUS) + clopidogrel 6 months + lifelong OAC
• Conservative group: No PCI before TAVI; no planned PCI after TAVI; aspirin monotherapy post-TAVI (modified per POPular TAVI trial)
• Median time from randomisation to TAVI: 34 days (PCI group) vs 25 days (conservative)
• PCI performed concomitantly with or shortly after TAVI in 26% of PCI patients

Primary Endpoint Results

• MACE (death/MI/urgent revascularisation): PCI 26% vs conservative 36%; HR 0.71 (95% CI 0.51–0.99; P=0.04)
• Modified ITT (excluding screening failures): HR 0.69 (95% CI 0.49–0.97) — consistent
• Benefit consistent across prespecified subgroups (age, sex, diabetes, stenosis diameter, LVEF, STS-PROM, SYNTAX score, balloon-expandable valve)

Secondary Endpoint Results

• All-cause death: PCI 27% vs conservative 32% at 2 years; HR 0.85 (95% CI 0.59–1.23) — not significant
• Myocardial infarction: HR 0.54 (95% CI 0.30–0.97) — significant
• Urgent revascularisation: HR 0.20 (95% CI 0.08–0.51) — significant
• Benefit for MI and urgent revascularisation appeared greatest in lesions with ≥90% diameter stenosis (descriptive finding, no formal interaction)

Safety Endpoints

• Bleeding (any — minor, major, life-threatening/disabling): PCI 28% vs conservative 20%; HR 1.51 (95% CI 1.03–2.22) — significantly higher with PCI
• Acute kidney failure: PCI 5% vs conservative 11%; HR 0.45 (95% CI 0.23–0.89) — significantly lower with PCI
• PCI procedure-related complications: 7 patients (3%) in PCI group
• Stent thrombosis: Similar in both groups (numerically low)

Comparison with ACTIVATION Trial

Mechanistic Hypotheses for Benefit

1. Post-TAVI change in coronary physiology → altered vessel-wall shear stress → plaque destabilisation
2. Patients become more physically active post-TAVI → unmasking of previously asymptomatic coronary disease
3. Events in conservative arm represent natural history of stable but physiologically significant coronary stenosis

Limitations of the Document

• Left main coronary stenosis and recent ACS (within 14 days) excluded — findings not generalisable to these populations
• Antithrombotic regimens changed during the 5-year enrollment period (AUGUSTUS 2019, POPular TAVI 2020 both modified treatment protocols during trial)
• PCI timing (pre- vs peri- vs post-TAVI) not fully standardised: 26% had PCI concomitantly or post-TAVI — optimal timing remains unknown
• Knowledge of untreated coronary lesion in conservative arm may have increased crossover and unplanned revascularisation (a driver of the primary endpoint)
• Low SYNTAX score population (median 9) — results may not apply to more complex multivessel disease
• Open-label design; adjudicated endpoints by blinded committee mitigates but does not eliminate bias
• Urgent revascularisation driven component (HR 0.20) may partly reflect open-label knowledge bias in the conservative arm

Key Concepts Mentioned

• concepts/PCI-Before-TAVI — primary topic: FFR-guided or angiographic PCI strategy in TAVI candidates
• concepts/TAVI — the valve intervention context for all patients in this trial
• concepts/Aortic-Stenosis — the valve disease indication
• concepts/Fractional-Flow-Reserve — the physiological lesion-selection tool

Key Entities Mentioned

• entities/Chronic-Coronary-Disease — stable CAD as comorbidity in TAVI population

Wiki Pages Updated

• Created wiki/sources/PCI-TAVI-NOTION3-NEJM-2024.md
• Created wiki/concepts/PCI-Before-TAVI.md
• Updated wiki/concepts/TAVI.md — expanded NOTION-3 evidence in CAD Management Before TAVI section
• Updated wiki/sourceindex.md
• Updated wiki/wikiindex.md