The Depths of the Heart
By Keith Loria
Radiology Today
Vol. 25 No. 6 P. 14

Advances in Heart Imaging

Coronary CT angiography (CCTA) is a well-validated, highly sensitive test for the identification and risk assessment of coronary artery disease (CAD) in appropriately selected symptomatic patients. It is the first-line noninvasive test for patients with acute and chronic chest pain syndromes, according to guidelines from several international medical societies.

Outpatient CCTA exams increased by 355% in the past decade, and the volume projections for CCTA are higher than any other cardiac imaging test, with expected increases of approximately 24% over the next five years, according to Siemens Healthineers.

“Coronary CTA is unique among available noninvasive testing as the only imaging test able to visualize both obstructive and nonobstructive coronary artery disease, which has been shown in multiple studies to be the most important marker of a patient’s risk for future cardiovascular events (eg, heart attacks, strokes),” says Blake Richards, CEO of Elucid. “As we know, it is atherosclerosis (plaque) that is the root cause of the majority of heart attacks and a large proportion of strokes.”

However, there is essential information contained within these images that is unable to be utilized today. Current CT software evaluates plaque by leveraging Hounsfield unit (HU) thresholding, a radiodensity measure. This approach does not discriminate between overlapping dense tissues, such as lipid-rich necrotic core (LRNC) and intraplaque hemorrhage, and is prone to numerous limitations and biases. Elucid’s PlaqueIQ is the only histologyderived plaque quantitation software with the potential to provide greater insights into plaque quantification and classification when compared with HU thresholding alone.

“By applying Elucid’s software to CT images to analyze plaque, physicians have the potential to accurately identify high-risk plaques in patients with cardiovascular disease (CVD), track disease progression or regression over time, assess urgency of treatment to reduce the risk of adverse events, and identify blockages to help determine the need for revascularization,” Richards says.

The software mitigates calcium blooming and blurring effects to enable better visualization definition of the signature of different tissues within the imaging data, Richards explains. The software corrects the HU values for the partialvolume effect, often experienced as blooming artifacts from calcified plaque and enhanced arterial lumen, as well as reduced ability to discriminate LRNC plaque. The software then runs proprietary, histology- validated plaque characterization algorithms to categorize and quantify plaque.

“Plaque is the key driver of adverse cardiovascular events,” Richards says. “By accurately assessing all plaque types, including lipid-rich necrotic core, physicians are equipped with additional information on a patient’s level of risk and can treat accordingly.”

Quantitative Analysis
Quantitative analysis of plaque helps physicians understand the total plaque volume as well as high-risk plaques, such as LRNC. Physicians can incorporate this information into their clinical assessment to help guide medical therapy selection, recommended lifestyle changes, and frequency of follow-up visits.

“Being able to provide patients with clear information and visualization of their plaque helps them take a more active role in their health, including increased medication adherence and conformity to lifestyle changes,” Richards says. “Additionally, for patients going to the cath lab, this plaque information can help inform physician device selection, size, and placement. A vessel with significant bulky calcified plaque would be treated differently than one with more noncalcified plaque, etc.”

Elucid’s PlaqueIQ plaque analysis is trained and validated against groundtruth histopathology, allowing physicians to quantify and characterize plaque tissue types, inclusive of LRNC.

“Broadly speaking, earlier identification and classification of plaques can help care providers avoid logjams and condense the care pathway by more quickly and accurately aligning them with the treatment best suited for their disease,” Richards says. “By prescribing a patient one of the abundant pharmaceutical therapies that exist today, we can head off cardiovascular disease earlier, dramatically improving outcomes. And, the quicker a provider can get each patient into appropriate treatment, the more time they will have to spend with others.”

There are potential economic benefits, as well. The United States spends $400 billion each year on CVD, but studies show that approximately 55% of the more than 2 million invasive cardiac catheterizations performed in the United States each year are unnecessary. Also, an estimated 20% to 30% of patients with undetected CAD are sent home due to false negative tests without receiving treatment, which can lead to more expensive treatments down the road.

“Effective screening that identifies patients who need timely, invasive intervention—and excludes those who don’t—could pare billions in costs from the health care system,” Richards says.

Fractional Flow Reserve
Fractional flow reserve derived from CT (FFR-CT) is a new, advanced technology that provides detailed information about the heart and how blood flows through the arteries. The technique identifies patients with coronary artery blockage or narrowing who could benefit from revascularization. FFR-CT enhances diagnosis for patients with or at risk for CAD and offers personalized care for CAD by identifying CAD earlier in the cycle, delivering high visibility of blockages in the arteries, and offering high-quality diagnostic data that helps dictate personalized treatment plans for patients.

Keya Medical, an international medical technology company, recently released its flagship AI-based FFR-CT product, DEEPVESSEL FFR (DVFFR). This FDA-cleared, CE-marked, and National Medical Products Association-approved product is in clinical use around the world with customers in the United States, European Union, Middle East, and Asia. DVFFR utilizes deep learning technology to perform a noninvasive physiological functional assessment of the coronary arteries using CCTA.

“This software offers an accurate, efficient, and easily accessible evaluation of FFR, significantly improving the triage of patients with stenosis between 30% and 90% and avoiding unnecessary invasive tests,” says Mary-Pierre Waiss, vice president of market development for Keya Medical. “The DVFFR analysis is conducted by Keya Medicalcertified analysts and delivered electronically to clinicians through the Intelerad/Ambra imaging platform using US-based [Amazon Web Services] servers.”

The analysis includes a 3D interactive model of the coronary artery tree with DVFFR values and a PDF report detailing computed DVFFR values and branch-by-branch visualizations.

“This achieves diagnostic accuracy using invasive FFR as the reference standard, with a low rejection rate—per-vessel and per-patient accuracy are 86.9% and 85.2%, respectively,” Waiss says. “The deep learning technology powering DVFFR enables rapid performance, minimizes server power requirements, and continuously enhances analysis accuracy through expanding data inputs.”

John A. Rumberger, PhD, MD, FACC, MSCCT, medical director and director of cardiac imaging for Corazon Imaging in Southfield, Michigan, has been doing cardiac CT for more than 40 years. He says FFR-CT and DVFFR have transformed his practice, as they allow him to provide superior patient care and achieve better outcomes.

“The Achilles’ heel of cardiac CT is the calcification, which makes it difficult to determine if that is a significant narrowing that may be causing the patient’s symptoms,” he says. “With the advent of FFR, we now have the ability to look at the entire anatomic situation, looking at heart disease from the beginning, as opposed to the end.”

For example, if a patient has a stenosis of between 40% and 90%, mistakes often happen based on visual estimation. By using FFR and looking at the severity of the disease, Rumberger can make a clearer determination much more efficiently. Rumberger believes the next steps for improving CT in heart imaging will come from doing more with plaque burden, noting Elucid’s efforts.

“The next step is going to be looking at inflammation,” he says. Inflammation around coronary arteries can be a sign of disease, “and densities can give you an excellent idea about the inflammation driving plaque development. You will have the ability to look with cardiac CT well beyond the coronary arteries.”

— Keith Loria is a freelance writer based in Oakton, Virginia. He is a frequent contributor to Radiology Today.