Prognostic Value of Coronary CT Angiography-Derived Quantitative Flow Ratio in Suspected Coronary Artery Disease

Academic Background

Coronary artery disease (CAD) is one of the leading causes of death and disability worldwide. Coronary CT angiography (CTA) is recommended as a first-line noninvasive test for patients with suspected or known CAD. CTA allows direct visualization of coronary artery anatomy and helps identify the presence and extent of CAD. However, CTA has only moderate positive predictive value for identifying hemodynamically significant lesions, which may lead to unnecessary clinical resource utilization. Therefore, reliably identifying patients who may benefit from revascularization through noninvasive testing has become an important research topic.

Fractional flow reserve (FFR) is the gold standard for assessing the hemodynamic significance of coronary stenosis and has been proven to improve clinical outcomes. However, FFR measurement typically requires invasive coronary angiography (ICA). In recent years, CT-derived FFR algorithms based on computational fluid dynamics (CFD) have shown good diagnostic accuracy and prognostic value in clinical practice. To reduce the high computational resource requirements of CFD simulations, machine learning-based CT-derived FFR methods have been developed, demonstrating excellent computational efficiency and correlation with invasive FFR measurements.

The quantitative flow ratio (QFR) is an alternative method for rapidly calculating FFR from coronary images by deriving the pressure drop using fluid dynamics equations. The QFR algorithm has been widely validated for diagnosing the functional significance of coronary artery disease. By integrating the QFR algorithm into coronary CTA, researchers have developed a fast CTA-derived QFR (CT-QFR) method, which has shown good agreement with invasive FFR in clinical practice. However, the prognostic value of CT-QFR in predicting long-term outcomes in CAD patients has not been fully investigated.

Research Objectives

This study aimed to determine the prognostic value of CT-QFR in predicting long-term major adverse cardiovascular events (MACEs) in patients with suspected CAD and to compare its predictive ability with that of ICA/single-photon emission computed tomography (SPECT). Additionally, the study explored the impact of prior percutaneous coronary intervention (PCI) on the prognostic value of CT-QFR.

Research Methods

Study Cohort

This study is a secondary analysis of the prospective international diagnostic study CORE320 (Combined Non-invasive Coronary Angiography and Myocardial Perfusion Imaging Using 320 Detector Computed Tomography). The CORE320 study enrolled participants from 16 centers in eight countries between November 2009 and July 2011. All participants underwent coronary CTA and SPECT within 60 days before ICA and were followed up for 5 years to assess the occurrence of MACEs.

Image Acquisition and Interpretation

CTA and SPECT images were acquired within 60 days before ICA and interpreted by independent core laboratories. The methods for image acquisition and interpretation have been described in detail in previous studies. CT-QFR analysis was performed by an analyst with 4 years of experience in CT-QFR measurements using dedicated software (CTAPlus v1, Pulse Medical). CT-QFR was calculated based on fluid dynamics equations, with the lowest value recorded as the participant-level result.

Primary Endpoint

The primary endpoint was the time to the first MACE during the 5-year follow-up. MACEs included cardiac death, myocardial infarction, hospitalization for chest pain or congestive heart failure, late revascularization (beyond 30 days after ICA), cardiac arrhythmia requiring hospitalization, non-cardiac death, and cerebrovascular events.

Statistical Analysis

Statistical analysis was performed by an author with 15 years of experience in biostatistics. Continuous variables were expressed as medians and interquartile ranges, while categorical variables were expressed as frequencies and percentages. Kaplan-Meier curves, multivariable Cox regression models, and the area under the receiver operating characteristic curve (AUC) were used to evaluate and compare the predictive abilities of CT-QFR and ICA/SPECT.

Research Results

Participant Characteristics and Events

Among the 310 participants, 205 (66%) were male, with a median age of 62 years. During the 5-year follow-up, 82 participants (26.5%) experienced MACEs. The participant-level agreement between CT-QFR and ICA/SPECT in defining hemodynamically significant disease was 78%.

MACE Prediction Using CT-QFR

Of the 82 first events, 56 occurred in participants with CT-QFR ≤0.80 (36.8% of 152 participants), and 26 occurred in participants with CT-QFR >0.80 (16.5% of 158 participants). Participants with CT-QFR ≤0.80 had significantly lower MACE-free survival rates than those with CT-QFR >0.80 (60% vs 82%). In the multivariable Cox proportional hazards regression model, CT-QFR (hazard ratio 1.9) and prior myocardial infarction (hazard ratio 2.5) were independent predictors of MACEs.

Comparison of CT-QFR and ICA/SPECT Predictive Abilities

The AUCs for CT-QFR and ICA/SPECT in predicting MACEs were 0.64 and 0.67, respectively, with no significant difference. The MACE-free survival rates were similar between CT-QFR and ICA/SPECT in both normal and abnormal cases.

Impact of Prior PCI on the Prognostic Value of CT-QFR

In participants with prior PCI, the AUC for CT-QFR in predicting MACEs was significantly lower than in participants without prior PCI (0.44 vs 0.70). In participants without prior PCI, those with normal CT-QFR had significantly higher MACE-free survival rates than those with abnormal CT-QFR (86% vs 59%).

Conclusion

CT-QFR is an independent predictor of MACEs during the 5-year follow-up in patients with suspected CAD, with prognostic value similar to that of ICA/SPECT. However, prior PCI affected the ability of CT-QFR to predict MACEs.

Research Highlights

1. Independent Predictive Value of CT-QFR: CT-QFR and prior myocardial infarction were independent predictors of MACEs, indicating the important clinical value of CT-QFR in decision-making.
2. Similarity to ICA/SPECT: The prognostic value of CT-QFR was similar to that of ICA/SPECT, providing a new noninvasive alternative.
3. Impact of Prior PCI: Prior PCI reduced the predictive ability of CT-QFR, suggesting that patient surgical history should be considered in clinical applications.

Research Significance

This study is the first to explore the value of CT-QFR in predicting long-term outcomes in CAD patients, providing new evidence for the application of noninvasive testing. As a fast, low-radiation examination method, CT-QFR has the potential to replace some invasive tests in clinical practice, reducing patient radiation exposure and healthcare costs. However, the impact of prior PCI on the predictive ability of CT-QFR suggests that future research should further optimize algorithms to improve accuracy in complex cases.