How AI and CCTA help heart teams plan ahead before PCI
The use of artificial intelligence (AI) in cardiology is making significant strides, particularly with automating coronary CT angiography (CCTA) analyses and helping pre-plan percutaneous coronary intervention (PCI) procedures. To learn more, Cardiovascular Business spoke with Evan Scott Shlofmitz, DO, director of intravascular imaging at St. Francis Hospital in Roslyn, New York, who leverages HeartFlow’s AI technology to get a complete picture of the cardiovascular health of his patients before stepping into the cath lab.
Historically, CCTA scans were used primarily to either rule out the need for an angiogram or flag abnormalities for further investigation. However, HeartFlow’s technology is transforming this approach by extracting crucial data from CT scans and using AI to provide an in-depth analysis of the coronary anatomy before the patient reaches the cath lab.
"It's such a change in the way that we're using CT scans, because historically it was just a gatekeeper. People would just read it as abnormal and they'd go for an angiogram, and no one even looked at the CT scan, but you have this wealth of data there that actually can drive most of your procedural planning. Before you even get to the cath lab and do your angiogram, you have an idea of what you need to treat, where to treat and what the best tools are to treat this," Shlofmitz explained.
CT can enhance pre-procedure planning
Heartflow’s technology automates the post-processing off CCTA exams and offers the cardiologist CT fractional flow reserve (CT-FFR) values for all coronary vessel segments to identify areas of ischemia, an interactive 3D coronary tree showing FFR pressure drops, curved multi-planer imaging of each vessel to show detailed plaque morphology and vessel remodeling, and a color-coded assessment of the types of soft plaques with automated quantification. This enables interventional cardiologists to determine the best course of action before performing an invasive angiogram or PCI to determine if the patient would be best served by cardiac surgery.
Shlofmitz said many studies have shown how intravascular imaging and pressure-wire based FFR can improve PCI outcomes. But the Heartflow CT-FFR and Plaque Analysis AI now allows users to see all of this before the procedure, saving time and avoiding surprises.
The AI-driven analysis provides insights into:
• The location and extent of ischemia
• Plaque burden and composition (calcific or lipid-rich plaques)
• Vessel diameter and optimal stent size
• Side branch involvement and potential complications
"We used to say, 'let's go to the cath lab and let's figure it out.' We were used to just making decisions on the fly where you see the angiogram and we make ad hoc decisions for implanting a permanent prosthetic device in these patients," Shlofmitz explained.
Now, he adds, this advanced AI software is helping heart teams do all of that planning in advance.
Improving procedural efficiency and patient outcomes
He said Heartflow shows the optimal angiographic views before entering the cath lab to save procedure time. The AI also aids in decision-making regarding treatment strategies. For instance, by evaluating the density and location of calcified plaques, physicians can determine in advance whether they need intravascular lithotripsy or atherectomy.
Another advantage of AI-driven pre-procedural planning is improved patient communication. With detailed visualizations and risk assessments available before the procedure, physicians can engage in more informed discussions with patients about their treatment options, including whether PCI or coronary artery bypass grafting (CABG) would be the most appropriate intervention.
“In some cases, we’ve had the CABG discussion before even performing a diagnostic angiogram because the AI-enhanced CT scan provided such clear insights into the severity of the disease,” Shlofmitz shared.
The future of AI in interventional cardiology
A decade ago, there were few CCTA scans shown in interventional cardiology conference sessions or patient cases, but that has changed dramatically in recent years. CCTA has also seen an acceleration in use since it was included as a class 1A recommendation in the 2021 ACC chest pain evaluation guidelines. The guidelines also included recommended use of Heartflow's algorithm, making it one of the first AI technologies to be included in any medical society clinical guidance.
CCTA advocates are now pushing for the modality to be used as a screening tool for cardiovascular disease; the addition of AI can evaluate the patient's risks and potentially improve outcomes. These same CCTA images would then likely be used as a key reference if interventions are needed.
Dave Fornell has covered healthcare for more than 17 years, with a focus in cardiology and radiology. Fornell is a 5-time winner of a Jesse H. Neal Award, the most prestigious editorial honors in the field of specialized journalism. The wins included best technical content, best use of social media and best COVID-19 coverage. Fornell was also a three-time Neal finalist for best range of work by a single author. He produces more than 100 editorial videos each year, most of them interviews with key opinion leaders in medicine. He also writes technical articles, covers key trends, conducts video hospital site visits, and is very involved with social media. E-mail: dfornell@innovatehealthcare.com