FFRct

FFRct v2.0 aims to non-invasively assess the hemodynamic significance of coronary artery stenoses to aid in the functional evaluation of coronary artery disease. It achieves this by applying computational fluid dynamics principles to a 3D model generated from coronary CT angiography data, simulating blood flow and pressure to derive FFRct values.

HeartFlow FFRct v2.0 is a software device designed for the post-processing analysis of previously acquired Computed Tomography (CT) DICOM data from clinically stable, symptomatic patients with coronary artery disease. Its purpose is to support the functional evaluation of coronary artery disease by providing a non-invasive assessment of hemodynamic significance. The methodology involves generating a 3D computer model from static coronary CT images and then applying computational fluid dynamics to simulate pressure, velocity, and blood flow information, from which a mathematically derived quantity, FFRct, is computed. This FFRct value is intended to aid qualified clinicians, in conjunction with other clinical information, in assessing coronary arteries. The technological characteristics of FFRct v2.0 are similar to its predicate, FFRct v1.4. Clinical validation for FFRct v2.0 was performed by re-processing data from the HeartFlowNXT study, a prospective, multicenter, non-randomized study. This study originally evaluated FFRct v1.4 against invasively measured Fractional Flow Reserve (FFR ≤0.80) as the reference standard in 254 subjects who were scheduled for invasive coronary angiography. For FFRct v2.0, the re-analysis demonstrated a per-vessel sensitivity of 84.2% and specificity of 84.9% in detecting hemodynamically significant lesions. On a per-subject basis, FFRct v2.0 showed a diagnostic accuracy of 80.0%, which was superior to site-read coronary CT angiography (cCTA) alone (51.9%) when compared to invasive FFR as the reference standard.

AI Regulatory Concerns

Powered by

Jev reads the full FDA letter and returns calibrated concern probabilities per regulatory category. These are AI estimates, not regulatory advice.

• Q1. How does the FFRct algorithm account for variations in patient-specific physiological parameters (e.g., heart rate, blood pressure at the time of CT scan) that might affect coronary flow dynamics?; What is the impact of CT image quality, including artifacts or suboptimal contrast, on the accuracy and reliability of FFRct values, and are there defined image quality thresholds for reliable analysis?; How does the performance of FFRct v2.0 compare in specific patient subgroups, such as those with severe calcification, prior stents, or diffuse coronary artery disease?; What is the computational time required to generate an FFRct analysis from a given CT dataset, and how does this integrate into typical clinical workflows for urgent or semi-urgent patient evaluations?; Beyond diagnostic accuracy, what is the evidence for FFRct's impact on clinical decision-making, patient outcomes (e.g., reduction in unnecessary invasive procedures, MACE rates), and cost-effectiveness compared to standard diagnostic pathways?