Clinical Relevance of Coronary Fractional Flow Reserve: Art-of-state

A deep learning-based fully automatic and clinical-ready framework for regional myocardial segmentation and myocardial ischemia evaluation

Myocardial ischemia diagnosis with CT perfusion imaging (CTP) is important in coronary artery disease management. Traditional analysis procedure is time-consuming and error-prone due to the semi-manual and operator-dependent nature. To improve the diagnostic performance, a deep learning-based, fully automatic, and clinical-ready framework was developed. Two collaborating deep learning networks including a 3D U-Net for left ventricle segmentation and a CNN for anatomical landmarks detection were trained on 276 subjects. With our processing framework, the 17-segment left ventricular model was automatically generated conformed to the clinical standard. Myocardial blood flow computed by commercial software was extracted within each segment and visualized against the bull's eye plot. The performance was validated on another 45 subjects. Coronary angiography and invasive fractional flow reserve measurements were also performed in these patients to serve as the gold standard for myocardial ischemia diagnosis. As a result, the diagnostic accuracy for our method was 81.08%, much higher than that for commercially available CTP analysis software (56.75%), and our method demonstrated a higher consistency (Kappa coefficient 0.759 vs. 0.585). Besides, the average processing time of our method was much lower (30 ± 10.5 s/subject vs. over 30 min/subject). In conclusion, the proposed deep learning-based framework could be a promising tool for assisting CTP analysis.

Is it the Time to Move Towards Coronary Computed Tomography Angiography-Derived Fractional Flow Reserve Guided Percutaneous Coronary Intervention? The Pros and Cons

Coronary artery disease is the leading cause of mortality worldwide. Diagnosis is conventionally performed by direct visualization of the arteries by invasive coronary angiography (ICA), which has inherent limitations and risks. Measurement of fractional flow reserve (FFR) has been suggested for a more accurate assessment of ischemia in the coronary artery with high accuracy for determining the severity and decision on the necessity of intervention. Nevertheless, invasive coronary angiography-derived fractional flow reserve (ICA-FFR) is currently used in less than one-third of clinical practices because of the invasive nature of ICA and the need for additional equipment and experience, as well as the cost and extra time needed for the procedure. Recent technical advances have moved towards non-invasive high-quality imaging modalities, such as magnetic resonance, single-photon emission computed tomography, and coronary computed tomography (CT) scan; however, none had a definitive modality to confirm hemodynamically significant coronary artery stenosis. Coronary computed tomography angiography (CCTA) can provide accurate anatomic and hemodynamic data about the coronary lesion, especially calculating fractional flow reserve derived from CCTA (CCTA-FFR). Although growing evidence has been published regarding CCTA-FFR results being comparable to ICA-FFR, CCTA-FFR has not yet replaced the invasive conventional angiography, pending additional studies to validate the advantages and disadvantages of each diagnostic method. Furthermore, it has to be identified whether revascularization of a stenotic lesion is plausible based on CCTA-FFR and if the therapeutic plan can be determined safely and accurately without confirmation from invasive methods. Therefore, in the present review, we will outline the pros and cons of using CCTA-FFR vs. ICA-FFR regarding diagnostic accuracy and treatment decision-making.

Calculation of Intracoronary Pressure-Based Indexes with JLabChart

The Fractional Flow Reserve (FFR) and instantaneous wave-Free Ratio (iFR) have been proposed and clinically validated to measure the pressure gradient across coronary stenoses. They provide quantitative information on stenosis severity. Both are used in coronary revascularization procedures to measure intracoronary pressure giving quantitative information to evaluate coronary diseases during angiographic procedures. We designed and implemented a tool able to acquire and measure iFR and FFR supporting the physicians studying and treating patients in interventional cardiology laboratories. We designed an extensive case study to assess the performance of the tool in (i) acquiring pressure signals from blood pressure measurement systems; (ii) calculating FFR and iFR; and (iii) filtering out extra-beats signals during realtime signal analysis phases. The tool, named JLabChart, is available online. We tested it on two sets of data for a total of 600 cycles from 201 pressure measurements performed on 65 patients, from the Interventional Cardiology Unit of Magna Graecia University. The recognition of cardiac cycles and keypoint of the pressure curve was effective in 100% of cases for proximal (aortic) pressure and in 99.2% for distal pressure. The FFR calculated by JLabChart had an excellent correlation (Rp=0.960; p<0.001) with the FFR values obtained through the commercial systems. Similar results were obtained with iFR (Rp=0.998; p<0.001). Finally, the tool measurement results were compared with a commercial tool proving JLabChart’s efficiency with real cases. It was also compared with measurements performed on synthetic vessels and stenosis designed using the Comsol commercial tool. JLabChart is able to provide reliable measurements of FFR and iFR indexes used to support decisions on interventional procedures. It represents a valuable open source support system that can be used in an interventional cardiology laboratory.

Assessing the association of appropriateness of coronary revascularization and 1-year clinical outcomes for patients with stable coronary artery disease in China

BACKGROUND

The Chinese appropriate use criteria (AUC) for coronary revascularization was released in 2016 to improve the use of coronary revascularization. This study aimed to evaluate the association between the appropriateness of coronary revascularization based on the Chinese AUC and 1-year outcomes in stable coronary artery disease (CAD) patients.

METHODS

We conducted a prospective, multi-center cohort study of stable CAD patients with coronary lesion stenosis ≥50%. After the classification of appropriateness based on Chinese AUC, patients were categorized into the coronary revascularization group or the medical therapy group based on treatment received. The primary outcome was a composite of death, myocardial infarction, stroke, repeated revascularization, and ischemic symptoms with hospital admission.

RESULTS

From August 2016 to August 2017, 6085 patients were consecutively enrolled. Coronary revascularization was associated with a lower adjusted hazard of 1-year major adverse cardiovascular and cerebrovascular events (MACCEs; hazard ratio [HR]: 0.62; 95% confidence interval [CI]: 0.45-0.86; P = 0.004) than medical therapy in patients with appropriate indications (n = 1617). No significant benefit in 1-year MACCEs was found after revascularization compared to after medical therapy in patients with uncertain indications (n = 2658, HR: 0.81; 95% CI: 0.52-1.25; P = 0.338) and inappropriate indications (n = 1810, HR: 0.80; 95% CI: 0.51-1.23; P = 0.308).

CONCLUSIONS

In patients with appropriate indications according to Chinese AUC, coronary revascularization was associated with significantly lower risk of MACCEs at 1 year. No benefit was found in coronary revascularization in patients with inappropriate indications. Our findings provide evidence for using Chinese AUC to guide clinical decision-making.

CLINICAL TRIAL REGISTRATION

NCT02880605. https://www.clinicaltrials.gov.

Thrombus growth modelling and stenosis prediction in the cerebral microvasculature

Cerebral microvascular occlusions cause restriction of blood supply to the brain, thus potentially severely impacting cognitive abilities. Thus, accurate prediction of thrombus growth in realistic geometries is important. Thrombi growth in an existing 13-generation cerebral microvasculature network is simulated here to study the haemodynamic effects of single and multiple blockages on the occlusion of the network. Compared to a single vessel, in a network, the occlusion probability is found to be different. It is the downstream/smaller arterioles (i.e. the 3rd, 4th, 5th, 6th generation arterioles in this study) that tend to reach occlusion first in a network and thus are the critical vessels. Simulations of simultaneous growth of two independent thrombi in the network (referred to here as the two-block case) show a close coupling between the locations of the various blocks in the network, each influencing the other's growth. The presence of the lead block (LB) slows the growth of the trailing block (TB). In some cases, it stops the TB's growth thereby preventing it from occluding the vessel. Findings in this work thus indicate that, to prevent ischaemia, blocks in the smaller arterioles need to be identified and treated first, and that this is more critical if the number of simultaneous blocks is higher.

Numerical analysis of the impact of flow rate, heart rate, vessel geometry, and degree of stenosis on coronary hemodynamic indices

Background

The stenosis of the coronary arteries is usually caused by atherosclerosis. Hemodynamic significance of patient-specific coronary stenoses and the risk of its progression may be assessed by comparing the hemodynamic effects induced by flow disorders. The present study shows how stenosis degree and variable flow conditions in coronary artery affect the oscillating shear index, residence time index, pressure drop coefficient and fractional flow reserve. We assume that changes in the hemodynamic indices in relation to variable flow conditions and geometries evaluated using the computational fluid dynamics may be an additional factor for a non-invasive assessment of the coronary stenosis detected on multi-slice computed tomography.

Methods

The local-parametrised models of basic shapes of the vessels, such as straight section, bend, and bifurcation as well as the global-patient-specific models of left coronary artery were used for numerical simulation of flow in virtually reconstructed stenotic vessels. Calculations were carried out for vessels both without stenosis, and vessels of 10 to 95% stenosis. The flow rate varied within the range of 20 to 1000 ml/min, and heart rate frequency within the range of 30 to 210 cycles/min.

The computational fluid dynamics based on the finite elements method verified by the experimental measurements of the velocity profiles was used to analyse blood flow in the coronary arteries.

Results

The results confirm our preliminary assumptions. There is significant variation in the coronary hemodynamic indices value caused by disturbed flow through stenosis in relation to variable flow conditions and geometry of vessels.

Conclusion

Variations of selected hemodynamic indexes induced by change of flow rate, heart rate and vessel geometry, obtained during a non-invasive study, may assist in evaluating the risk of stenosis progression and in carrying out the assessment of the hemodynamic significance of coronary stenosis. However, for a more accurate assessment of the variability of indices and coronary stenosis severity both local (near the narrowing) and global (in side branches) studies should be used.

Physiome approach for the analysis of vascular flow reserve in the heart and brain

This work reviews the key aspects of coronary and neurovascular flow reserves with an emphasis on physiomic modeling characteristics by the use of a variety of numerical approaches. First, we explain the definition of fractional flow reserve (FFR) in coronary artery and introduce its clinical significance. Then, computational researches for obtaining FFR are reviewed, and their clinical outcomes are compared. In the case of cerebrovascular reserve (CVR), in spite of substantial progress in the simulation of cerebral hemodynamics, only a few computational studies exist. Thus, we discuss the limitations of CVR simulation study and suggest the challenging issue to overcome these. Also, the future direction of physiomic researches for the flow reserves in coronary arteries and cerebral arteries is described. Also, we introduce a machine learning algorithm trained by the existing physiomic simulation data of flow reserve and suggest a prospective research direction related to this.