Machine leArning Based CT angiograpHy derIved FFR: a Multi-ceNtEr, Registry

Completed

• Conditions: Coronary Heart Disease; Coronary Artery Disease

• Registration Number: NCT02805621
• Lead Sponsor: Erasmus Medical Center

Brief Summary

Demonstrate in a large multicenter population the diagnostic performance of a pre-commercial on-site, local, CT angiography derived FFR algorithm in comparison to invasive FFR.

Detailed Description

To retrospectively evaluate the diagnostic accuracy of FFRCT, in patients with known or suspected CAD. the investigators propose to do technical assessment of the software and evaluate how different parameters effect the outcome. Validate the FFTCT outcome by comparing the FFRCT values with invasive FFR values from retrospective patient data. To analyze the potential of FFRCT on decision making and prognosis.

Study Timeline

• Study Start: 2016-04
• Study First Submitted: 2016-06-15
• Study First Submitted QC: 2016-06-15
• Study First Posted: 2016-06-20
• Status Verified: 2017-01
• Primary Completion: 2017-01
• Study Completion: 2017-01
• Last Update Submitted: 2017-01-03
• Last Update Posted: 2017-01-04

Recruitment & Eligibility

• Status: COMPLETED
• Sex: All
• Target Recruitment: 352

Inclusion Criteria

• Know or suspect coronary artery disease followed within 6 months by an invasive FFR measurement.

Exclusion Criteria

• Cardiac event between coronary CT angiography and the invasive FFR procedure, noninterpretable coronary CT angiography image quality, or incomplete coronary CT angiography coverage.

Study & Design

• Study Type: OBSERVATIONAL
• Study Design: Not specified

Primary Outcome Measures

Name	Time	Method
Diagnostic accuracy of local reduced order CFD and machine learning based CT angiography derived FFR, both validated against invasive FFR. Measured at both vessel and patient level.	6 months

Secondary Outcome Measures

Name	Time	Method
Influence of calcium on diagnostic accuracy of CT angiography derived FFR.	6 months
Direct vessel based comparison between CT angiography derived FFR and QCT stenosis measurements	6 months
Confidence intervals of CT angiography derived FFR	6 months
Analysis of anatomically mild stenosis (<50% lumen diameter reduction) but functionally significant (invasive FFR ≤ 0.80)	6 months
Long term clinical outcome of CT angiography derived FFR	12 months

Trial Locations

Locations (1)

ErasmusMC; 🇳🇱 Rotterdam, Zuid Holland, Netherlands