PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
January 22, 2026Catheterization and Cardiovascular Interventions0 citations

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

View Full Paper
QXQianhang XiaSYShuangya YangPLPan Li

Key Points

  • This research aims to explore how OCT-derived fractional flow reserve (OFR) impacts one-year outcomes in acute coronary syndrome (ACS) patients after PCI.
  • Single-center retrospective study of 719 ACS patients undergoing OCT-guided PCI.
  • Primary endpoint focused on target vessel-related major adverse cardiovascular events (MACE).
  • Imaging features selected using LASSO and Boruta methods.
  • Machine learning models developed using various algorithms including XGBoost and logistic regression.
  • Performance assessed via ROC curves and feature importance calculated with the GINI index.
  • Significant association found between post-PCI OFR and one-year MACE in ACS patients.
  • Machine learning models effectively predicted MACE risk, demonstrating robust classification capabilities.
  • Feature selection highlighted key imaging parameters relevant to patient outcomes.

Abstract

ABSTRACT Background Despite PCI, many acute coronary syndrome (ACS) patients experience major adverse cardiovascular events (MACE). Angiography is limited, and fractional flow reserve (FFR) is restricted by cost and specialized resource requirements. Optical coherence tomography (OCT)‐derived FFR (OFR) allows functional assessment without hyperemia, but its prognostic role after PCI is unclear. Aims To investigate the association between post‐PCI OFR and 1‐year target vessel‐related MACE in ACS patients and to develop machine learning‐based models to explore risk stratification. Methods This single‐center retrospective study included 719 ACS patients undergoing OCT‐guided PCI at the Affiliated Hospital of Zunyi Medical University (May 2022 to December 2023). The primary endpoint was target vessel‐related MACE (cardiac death, revascularization, myocardial infarction, or angina rehospitalization). Imaging features were selected by LASSO and Boruta, and machine learning–based classification models were built with XGBoost, random forest, support vector machine, logistic regression, and light gradient boosting. Performance was assessed using ROC curves and feature importance via GINI index.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Xia et al. (2026) studied this question.

synapsesocial.com/papers/6971be8d642b1836717e3334https://doi.org/10.1002/ccd.70458
Ask AI
Helpful
Bookmark
Share
View Full Paper