CCTA for Coronary Artery Disease

(CarDIA-AI Trial)

Recruiting at 2 trial locations

Overseen By Jon-David Schwalm, MD, MSc

Age: 18+

Sex: Any

Trial Phase: Academic

Sponsor: Hamilton Health Sciences Corporation

Disqualifiers: Atrial fibrillation, Severe renal dysfunction, others

No Placebo Group All trial participants will receive the active study treatment (no placebo)

Compensation: Varies

Number of Visits: Unknown

252 Participants Needed

What You Need to Know Before You Apply

What is the purpose of this trial?

The trial aims to determine if a new AI triage tool can assist in deciding whether patients should undergo a less invasive heart test called CCTA or the usual ICA. This is crucial because ICA is costly and risky, and it sometimes fails to provide any treatment benefit. The trial will compare the usual care, where patients proceed directly to ICA, with a new method using AI to determine if CCTA might be a better initial option. Individuals referred for a heart test to check for coronary artery disease (CAD) and who haven't recently undergone a CCTA might be suitable candidates for this trial. As an unphased trial, this study offers patients the chance to contribute to innovative research that could enhance heart test decision-making.

Will I have to stop taking my current medications?

The trial information does not specify whether you need to stop taking your current medications. It's best to discuss this with the trial coordinators or your doctor.

What prior data suggests that this AI triage tool is safe for patients?

Research has shown that AI tools for coronary artery disease (CAD) are generally safe and easy to use. One study found that using an AI tool in patient care was safe and helpful, streamlining the patient process and supporting the healthcare system. Another study found that AI can improve the accuracy of CAD diagnosis, providing more precise results without additional risk. These studies suggest that the AI tool being tested should be safe for participants. ^{1 2 3 4 5}

Why are researchers excited about this trial?

Researchers are excited about the AI triage tool for coronary artery disease because it offers a new way to determine the most suitable diagnostic path for patients. While standard care often involves directly proceeding to invasive coronary angiography (ICA), this tool uses data from referral forms to predict the likelihood of obstructive coronary artery disease. By doing so, it can decide whether a patient should undergo a less invasive coronary computed tomography angiography (CCTA) instead. This approach could potentially reduce unnecessary invasive procedures, streamline patient care, and improve overall diagnostic accuracy.

What evidence suggests that this AI triage tool is effective for coronary artery disease?

This trial will compare two approaches for diagnosing coronary artery disease: usual care, where patients proceed directly to invasive coronary angiography (ICA), and a centralized triage approach using an AI triage tool. Research has shown that cardiac computed tomographic angiography (CCTA) can be as effective as ICA for diagnosing heart disease in certain patients, while being less risky and less expensive. In this trial, the AI triage tool will analyze patient information to determine whether CCTA or ICA is the better choice, potentially avoiding unnecessary procedures and guiding patients to the most appropriate test for their condition. This method aims to reduce risks for patients, lower costs, and make healthcare more efficient. ^{3 4 6 7 8}

Who Is on the Research Team?

Jon-David Schwalm, MD, MSc

Principal Investigator

Hamilton Health Sciences Corporation

Jeremy Petch, PhD

Principal Investigator

Hamilton Health Sciences Corporation

Natalia Pinilla-Echeverri, MD, PhD

Principal Investigator

Niagara Health

Are You a Good Fit for This Trial?

This trial is for individuals with suspected coronary artery disease. It aims to determine the best diagnostic approach by using a new AI model to decide if patients should get a non-invasive CCTA scan or an invasive angiography.

Inclusion Criteria

I am referred for a non-urgent heart artery check.

Patients able to provide informed consent in English

I need a heart artery check for chest pain, heart muscle disease, or stable heart artery disease.

Exclusion Criteria

I am scheduled for heart surgery that is not on the coronary arteries.

Patients with known severe coronary artery calcification (calcium score >1000)

I have had heart issues like blocked arteries or heart surgery.

Timeline for a Trial Participant

Screening

Participants are screened for eligibility to participate in the trial

2-4 weeks

Centralized Triage and Risk Score-based Screening

Patients originally referred for ICA will be screened for obstructive CAD with a decision support tool that uses data from their referral forms. Patients will receive either CCTA or ICA based on their predicted probability of obstructive CAD.

4-6 weeks

Follow-up

Participants are monitored for safety and effectiveness after treatment

12 weeks

What Are the Treatments Tested in This Trial?

Interventions

• AI Triage Tool

Trial Overview The study tests usual care against a new triage pathway that uses an AI risk score to screen patients. The goal is to see if this method can more accurately direct patients either towards CCTA or ICA, potentially reducing unnecessary procedures and costs.

How Is the Trial Designed?

Group I: Centralized triage with risk score-based screening for obstructive CAD Experimental Treatment 1 Intervention

Patients originally referred for ICA will be screened for obstructive CAD with a decision support tool that uses data from their referral forms. Patients will receive either CCTA or ICA based on their predicted probability of obstructive CAD.

Group II: Usual Care Active Control 1 Intervention

Patients will proceed directly to ICA as originally referred.

Who Is Running the Clinical Trial?

Hamilton Health Sciences Corporation

Lead Sponsor

Trials

380

Recruited

345,000+

Hamilton Academic Health Sciences Organization

Collaborator

Trials

22

Recruited

5,200+

Population Health Research Institute

Collaborator

Trials

165

Recruited

717,000+

Published Research Related to This Trial

In a study of 172 patients with moderate coronary artery disease, those who experienced coronary adverse events had a significantly higher coronary Fatty Acid Index (FAI) compared to those who did not, indicating a potential link between FAI levels and adverse cardiac events.

Even after accounting for age differences, the coronary FAI remained significantly different between the adverse and non-adverse groups, suggesting that FAI could be a useful biomarker for assessing risk in patients with coronary artery disease.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

The Association of Coronary Fat Attenuation Index Quantified by Automated Software on Coronary Computed Tomography Angiography with Adverse Events in Patients with Less than Moderate Coronary Artery Stenosis. Zhang, W., Li, P., Chen, X., et al.[2023]

Using Computer Aided Diagnosis (CAD) with Coronary Computed Tomographic Angiography (CCTA) can effectively triage 60% of low to intermediate risk chest pain patients in the Emergency Department, potentially reducing patient wait times.

The CAD demonstrated a sensitivity of approximately 85% and a high Negative Predictive Value (NPV) of 97.8% for the 64-slice scanner and 97.1% for the 320-slice scanner, indicating it is reliable for ruling out serious conditions, although it should not completely replace human interpretation.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Computer-aided analysis of 64- and 320-slice coronary computed tomography angiography: a comparison with expert human interpretation. Abd Alamir, M., Noack, P., Jang, KH., et al.[2019]

The study will evaluate a low-dose CCTA protocol for diagnosing acute chest pain in 681 younger adults, aiming to demonstrate that it is safe and effective, with a target of less than 1% chance of hard events within 30 days after a negative result.

If successful, this low-dose approach could reduce radiation exposure and costs without needing to invest in new technology, making it a practical option for emergency departments.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

SEALONE (Safety and Efficacy of Coronary Computed Tomography Angiography with Low Dose in Patients Visiting Emergency Room) trial: study protocol for a randomized controlled trial. Kim, J., Kang, JW., Kim, K., et al.[2020]

In a study of 8030 patients who underwent coronary computed tomography angiography (CCTA), only 12.03% required further coronary angiography (CAG), indicating that CCTA can effectively triage patients and reduce unnecessary procedures.

Among those who did undergo CAG after CCTA, there was a significantly higher revascularization rate (63.4% underwent PCI) compared to patients who had direct CAG, suggesting that CCTA improves the accuracy of identifying patients who need further intervention.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

[Impact of coronary computed tomography angiography on patient triage strategies]. Gai, J., Zhai, X., Bai, Q., et al.[2016]

Coronary computed tomography angiography (CCTA) is a safe and effective method for triaging patients with low-to-intermediate likelihood of acute coronary syndrome, allowing for quick discharge if no coronary artery disease (CAD) is present.

Using CCTA can significantly reduce hospital length of stay and may lead to cost-effective care compared to traditional triage methods.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Cardiac computed tomography for the evaluation of the acute chest pain syndrome: state of the art. Schlett, CL., Hoffmann, U., Geisler, T., et al.[2022]

An artificial intelligence (AI) algorithm developed for coronary computed tomography angiography (CCTA) showed strong performance in identifying the absence of coronary atherosclerosis, achieving an impressive area under the receiver operating characteristic curve (AUC-ROC) of 0.96 in a study of 100 patients.

The AI-assisted workflow demonstrated high negative predictive values of 95%, indicating that it can effectively support physicians in safely discharging patients with chest pain when no coronary disease is detected.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Artificial Intelligence to Assist in Exclusion of Coronary Atherosclerosis During CCTA Evaluation of Chest Pain in the Emergency Department: Preparing an Application for Real-world Use. White, RD., Erdal, BS., Demirer, M., et al.[2022]

Cardiac computed tomography angiography (CCTA) demonstrated high diagnostic accuracy in detecting significant coronary artery stenosis, with a sensitivity of 99% and specificity of 81% in a study of 400 patients with non-ST segment elevation myocardial infarction.

CCTA effectively triaged patients into appropriate treatment groups 86% of the time, indicating its potential utility in managing high-risk coronary artery disease patients, although logistical and safety challenges limit its current use as a routine diagnostic tool before invasive coronary angiography.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Coronary CT angiography in clinical triage of patients at high risk of coronary artery disease. Kühl, JT., Hove, JD., Kristensen, TS., et al.[2019]

In a study of 2023 patients with acute chest discomfort, 11.2% were diagnosed with acute coronary syndrome (ACS), highlighting the importance of accurate triage in potentially life-threatening situations.

The semi-automatic triage system had a sensitivity of 73% for identifying urgent cases, but it underestimated urgency in 27% of patients with ACS or other life-threatening events, indicating a need for improved accuracy despite some benefits from nurse overrides.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Accuracy of telephone triage in primary care patients with chest discomfort: a cross-sectional study. Wouters, LT., Rutten, FH., Erkelens, DC., et al.[2020]

Citations

1. pmc.ncbi.nlm.nih.gov pmc.ncbi.nlm.nih.gov/articles/PMC11307411/

Using Artificial Intelligence to Optimize the Use of Cardiac ... Plans to evaluate this centralized triage and AI-supported decision support for elective patients with suspected CAD in a randomized controlled trial are in the ...

2. sciencedirect.com sciencedirect.com/science/article/pii/S1386505625000553

AI-driven triage in emergency departments: A review of ... AI-driven triage systems present a promising solution, automating patient prioritization by analyzing real-time data, such as vital signs, medical history, and ...

3. pmc.ncbi.nlm.nih.gov pmc.ncbi.nlm.nih.gov/articles/PMC12138305/

Protocol for the CarDIA-AI Randomized Controlled Trial - PMC Coronary computed tomographic angiography (CCTA) is a noninvasive option that provides similar information with less risk and is recommended as a first-line ...

4. ahajournals.org ahajournals.org/doi/10.1161/CIR.0000000000001201

Use of Artificial Intelligence in Improving Outcomes in Heart ... This scientific statement outlines the current state of the art on the use of artificial intelligence algorithms and data science in the diagnosis, ...

5. jscai.org jscai.org/article/S2772-9303(24)02208-7/fulltext

Artificial Intelligence in Coronary Artery Interventions AI has enhanced diagnostic accuracy in coronary artery disease through advanced invasive and noninvasive imaging modalities, facilitating more precise diagnosis ...

6. pmc.ncbi.nlm.nih.gov pmc.ncbi.nlm.nih.gov/articles/PMC12511812/

Clinical Performance Evaluation of an Artificial Intelligence ... The main objective of this study is to evaluate the clinical performance of an artificial intelligence (AI)-based tool in predicting significant ...

7. nature.com nature.com/articles/s41591-025-03620-y

Implementation of a national AI technology program on ... Implementation of an AI-diagnostic tool as part of a health intervention program was safe and beneficial to the patient pathway and health system with fewer ...

8. pubs.rsna.org pubs.rsna.org/doi/10.1148/radiol.242087

Diagnostic Accuracy of On-Premise Automated Coronary ... On-premises artificial intelligence was accurate in ruling out obstructive coronary artery disease at coronary CT angiography and achieved substantial to near- ...