OCT-Derived Virtual Fractional Flow Reserve Associated With 1-Year Outcomes After PCI in ACS Patients

Area of Science:

• Cardiovascular Medicine
• Interventional Cardiology
• Medical Imaging

Background:

• Percutaneous coronary intervention (PCI) in acute coronary syndrome (ACS) patients often fails to prevent major adverse cardiovascular events (MACE).
• Traditional angiography has limitations, and invasive fractional flow reserve (FFR) requires hyperemia and specialized resources.
• Optical coherence tomography (OCT)-derived FFR (OFR) offers a non-hyperemic functional assessment, but its prognostic implications post-PCI are not well-established.

Purpose of the Study:

• To evaluate the association between post-PCI OFR and 1-year target vessel-related MACE in ACS patients.
• To develop and assess machine learning models for risk stratification using OFR and imaging features.

Main Methods:

• A retrospective analysis of 719 ACS patients who underwent OCT-guided PCI.
• Primary endpoint: target vessel-related MACE (cardiac death, revascularization, myocardial infarction, angina rehospitalization).
• Feature selection via LASSO and Boruta; classification models (XGBoost, random forest, SVM, logistic regression, LGBoost) built and evaluated using ROC curves and GINI index.

Main Results:

• OFR demonstrated prognostic significance for 1-year MACE in ACS patients after PCI.
• Machine learning models, incorporating OFR and imaging data, showed robust performance in risk stratification.
• Feature importance analysis identified key predictors of MACE.

Conclusions:

• Post-PCI OFR is a valuable predictor of long-term outcomes in ACS patients.
• Machine learning approaches enhance risk stratification for MACE, aiding clinical decision-making.
• OFR represents a promising tool for functional assessment and risk prediction in interventional cardiology.