OCT-based Machine Learning FFR for Predicting Post-PCI FFR

Tomography, Optical Coherence Clinical Trial

Official title:

Optical Coherence Tomography-based Machine Learning for Predicting Fractional Flow Reserve After Coronary Artery Stenting

Clinical Trial Details — Status: Not yet recruiting

Administrative data

• NCT number: NCT06341361
• Other study ID #: OCT-FFR
• Status: Not yet recruiting
• Start date: April 15, 2024
• Est. completion date: October 15, 2025

Study information

• Verified date: March 2024
• Source: Yonsei University
• Contact: Oh-Hyun Lee, MD
• Phone: +82-31-5189-8786
• Email: Decenthyun[@]yuhs.ac
• Is FDA regulated: No
• Study type: Observational

Clinical Trial Summary

This study aims to compare the diagnostic accuracy of the fractional flow reserve (FFR) model derived by machine learning based on optical coherence tomography (OCT) exam after coronary artery stent implantation with the wire-based FFR.

Description:

FFR and OCT exam are used for different purposes during percutaneous coronary intervention (PCI). The FFR is a decision-making tool to determine if additional procedures are necessary, while the OCT exam is used to optimize the stent procedure. The use of both tests provides additional information to help perform a excellent procedure, but it is more expensive and time-consuming. Therefore, an OCT-derived machine learning FFR test may be helpful. Previous studies have demonstrated that OCT-based machine learning FFR before the procedure has shown good diagnostic performance in predicting FFR, irrespective of the coronary territory. Despite the rapid development of technologies and tools for PCI, a significant number of patients experienced adverse events, such as recurrence of angina and silent ischemia despite angiographically successful PCI. Suboptimal PCI is a well-known independent prognostic factor for major cardiovascular accidents. Therefore, measuring post-PCI FFR immediately after stent implantation is crucial to optimize the procedure outcome and improve the patient's prognosis. Although the importance of measuring post-PCI FFR is gradually emerging, there is currently no model for OCT-based machine learning FFR that predicts FFR after stent insertion. In patients who underwent percutaneous coronary intervention using stents for ischemic heart disease, we will compare the diagnostic accuracy of the fractional flow reserve (FFR) model derived by machine learning based on optical coherence tomography (OCT) exam after coronary artery stent implantation with the wire-based FFR.

Recruitment information / eligibility

• Status: Not yet recruiting
• Enrollment: 82
• Est. completion date: October 15, 2025
• Est. primary completion date: April 30, 2024
• Accepts healthy volunteers: No
• Gender: All
• Age group: 19 Years and older

Eligibility

Inclusion Criteria:

1. Patients who underwent stent implantation for ischemic heart disease

2. Patients who underwent both OCT examination and FFR using a pressure wire after PCI

Exclusion Criteria:

1. Poor OCT imaging quality

2. Patients with severe left ventricular dysfunction (<30%)

3. Patients with severe valvular heart disease

4. Patients with a life expectancy of less than 1 year

Related Conditions & MeSH terms

• Fractional Flow Reserve, Myocardial
• Tomography, Optical Coherence

Intervention

Diagnostic Test:
• OCT-based machine learning FFR
OCT-based machine learning FFR and wire-based FFR

Locations

Country	Name	City	State
n/a

Sponsors (3)

Lead Sponsor	Collaborator
Yonsei University	Gangnam Severance Hospital, Severance Hospital

Outcome

Type	Measure	Description	Time frame	Safety issue
Primary	Correlation of OCT-based machine learning FFR compared to wire-based FFR	Determining the diagnostic accuracy of CT-FFR values obtained by the new method compared with invasive coronary angiography with fractional flow reserve	4 weeks
Secondary	Diagnostic performance of OCT-based machine learning FFR compared to wire-based FFR	Accuracy, sensitivity, specificity, positive predictive value, negative predictive value	4 weeks
Secondary	Diagnostic performance of OCT-based machine learning FFR according to the coronary artery (LAD, LCx or RCA) compared to wire-based FFR	Accuracy, sensitivity, specificity, positive predictive value, negative predictive value	4 weeks