TCT-161 AI-Assisted Versus Interventional Cardiologist-Guided Intravascular Imaging for PCI in Calcified Coronary Artery Disease: A Comparative Analysis

Document Type

Abstract

Publication Date

10-28-2025

Abstract

Background: Heavily calcified coronary lesions present a major challenge during percutaneous coronary intervention (PCI), where optimal stent expansion is critical for long-term success. Intravascular imaging modalities such as intravascular ultrasound (IVUS) and optical coherence tomography (OCT) assist in procedural planning, but their interpretation is often subjective. Artificial intelligence (AI)-enhanced image analysis may improve decision-making consistency and outcomes. Out study objective was to compare procedural and clinical outcomes between AI-assisted versus interventional cardiologist-guided intravascular imaging for PCI in patients with severely calcified coronary artery disease. Methods: We retrospectively analyzed 800 patients undergoing PCI for calcified coronary lesions using IVUS or OCT, with 400 managed via AI-guided imaging interpretation and 400 via manual interpretation by experienced interventional cardiologists. Primary endpoint was optimal stent expansion (>90%). Secondary endpoints included procedural success, 12-month major adverse cardiovascular events (MACE), target lesion revascularization (TLR), contrast volume, and radiation exposure. Results: Baseline characteristics were similar between groups. Optimal stent expansion was significantly higher in the AI-guided group (87.0% vs. 77.0%, p=0.0018), with greater procedural success (97.5% vs. 93.5%, p=0.018) and lower contrast volume (146.3 ± 28.7 mL vs. 158.5 ± 30.2 mL, p< 0.001). TLR at 12 months was reduced in the AI group (3.0% vs. 5.5%, p=0.045), while MACE rates were comparable (7.0% vs. 9.0%, p=0.29). Radiation exposure was also lower with AI guidance (p< 0.001). Conclusion: AI-assisted intravascular imaging guidance during PCI for calcified coronary lesions is associated with improved stent optimization, higher procedural success, and lower procedural burden compared to cardiologist-guided interpretation. These results support the integration of AI in intraprocedural imaging workflows to enhance PCI outcomes. Categories: IMAGING AND PHYSIOLOGY: Artificial Intelligence: Imaging and Physiology

Publication Title

Journal of the American College of Cardiology

Volume

86

Issue

17 Supplement

First Page

B73

Last Page

B73

Comments

Transcatheter Cardiovascular Therapeutics Conference held 2025-10-25 to 2025-10-28 in San Francisco, CA

This document is currently not available here.

Open Access

Share

COinS