CCTA to Optimize Diagnostic Yield of Invasive Angiography With AI

NCT06648239 | Active Not Recruiting DMC

• 1 other identifier: 17103
• Study Type: interventional
• Target: 251
• Locations: 1 country
• Sites: 3

Brief Summary

Coronary artery disease (CAD) is a leading cause of death. The gold-standard test used to diagnose CAD is invasive coronary angiography (ICA). However, nearly half the patients who receive ICA are found to have no disease or non-significant disease. This means that while they receive a diagnosis, they do not receive any therapeutic benefit. This is concerning because ICA is expensive and it carries a risk to patients. A non-invasive diagnostic test, cardiac computed tomographic angiography (CCTA), has been shown to be as effective as ICA at diagnosing CAD in the right patient population, while being less expensive and less risky for patients. An optimal solution would involve screening to identify which patients are good candidates for CCTA vs. which should receive ICA. This screening tool could be used in a triage pathway to ensure that every patient gets the test that is best for them. The investigators have used Artificial Intelligence (AI) to develop a model for determining which patients should receive ICA vs. which should receive CCTA. The investigators have also developed a triage pathway to direct patients to the most appropriate test. The investigators now plan to evaluate the AI tool combined with the triage pathway through a clinical trial at Hamilton Health Sciences and Niagara Health. This model of care will reduce risk to patients, reduce wait times for ICA and reduce costs to the health care system.

Conditions

Coronary Artery Disease

Keywords

Artificial Intelligence (AI); Coronary Angiography; Coronary Computed Tomography Angiography

Outcome Measures

Primary Outcomes (1)

• Rate of normal/non-obstructive CAD diagnosed through ICA

  The rate of normal or non-obstructive CAD diagnosed through ICA in patients referred for cardiac investigation. The rate for an arm (control vs experimental) is calculated by dividing the number of patients diagnosed with normal/non-obstructive CAD through ICA by the total patients allocated to the arm.

  90 days (after randomization)

Secondary Outcomes (7)

• Quantitative assessment of number of angiograms avoided

  90 days (after randomization)

• Deviation from management recommendations following CCTA (i.e. angiograms performed when not recommended)

  90 days (after randomization)

• Diagnostic yield of invasive angiography

  90 days (after randomization)

• Sex differences in rate of normal/non-obstructive CAD diagnosed through ICA

  90 days (after randomization)

• Site differences in rate of normal/non-obstructive CAD diagnosed through ICA

  90 days (after randomization)

Study Arms (2)

Usual Care

ACTIVE COMPARATOR

Patients will proceed directly to ICA as originally referred.

Other: Usual Care

Centralized triage with risk score-based screening for obstructive CAD

EXPERIMENTAL

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.

Other: Centralized triage with risk score-based screening for obstructive CAD

Interventions

Usual Care OTHER

In the usual care group, patients will proceed directly to ICA following referral from community cardiology, as is the current standard of care. Research staff will screen participants in this group for significant CAD using the decision support tool; however, the tool's recommendations will not affect their care, as all patients in this group will invariably receive ICA.

Usual Care

Centralized triage with risk score-based screening for obstructive CAD OTHER

Patients randomized to the intervention will have selected features of their medical history, recorded on their referral form, entered into a decision support tool by research personnel to generate a recommendation of whether they should proceed directly to ICA or whether they should receive CCTA. Patients with recommendations for ICA will proceed directly to ICA. Patients with recommendations for CCTA will be referred to CCTA. Based on the results of the CCTA, recommendations for medical management versus referral for ICA will be made.

Centralized triage with risk score-based screening for obstructive CAD

Eligibility Criteria

• Age: 18 Years+
• Sex: all
• Healthy Volunteers: Yes
• Age Groups: Adult (18-64), Older Adult (65+)

You may qualify if:

• Patients are eligible if they:
• are ≥18 years of age;
• are referred for non-urgent (elective) outpatient ICA;
• have an indication for ICA that includes 'Rule out CAD', 'Cardiomyopathy', or 'Stable CAD'; and
• are able to provide informed consent in English.

You may not qualify if:

• received a prior high-quality coronary computed tomographic angiography (CCTA) within the last 5 years;
• atrial fibrillation;
• known severe renal dysfunction (GFR \<35);
• planned non-coronary cardiac surgery;
• any prior obstructive CAD, acute coronary syndrome, percutaneous coronary intervention, or coronary artery bypass graft; or
• known severe coronary artery calcification (calcium score \>1000).

Sponsors & Collaborators

• Hamilton Health Sciences Corporation: lead
• Population Health Research Institute: collaborator
• Hamilton Academic Health Sciences Organization: collaborator

Study Sites (3)

Hamilton General Hospital

Hamilton, Ontario, L8L 2X2, Canada

Location

McMaster University Medical Centre

Hamilton, Ontario, L8N 3Z5, Canada

Location

St. Catharines Hospital

St. Catharines, Ontario, L2S 0A9, Canada

Location

Related Publications (4)

• Schwalm JD, Di S, Sheth T, Natarajan MK, O'Brien E, McCready T, Petch J. A machine learning-based clinical decision support algorithm for reducing unnecessary coronary angiograms. Cardiovasc Digit Health J. 2021 Dec 24;3(1):21-30. doi: 10.1016/j.cvdhj.2021.12.001. eCollection 2022 Feb.

  PMID: 35265932 BACKGROUND

• Schwalm JD, Bouck Z, Natarajan MK, Pinilla N, Walker D, Syed N, Landry D, Sabri A, Tandon V, Nkurunziza J, Taljaard M, Sheth T. Centralized Triage of Suspected Coronary Artery Disease Using Coronary Computed Tomographic Angiography to Optimize the Diagnostic Yield of Invasive Angiography. CJC Open. 2022 Nov 19;5(2):148-157. doi: 10.1016/j.cjco.2022.10.009. eCollection 2023 Feb.

  PMID: 36880068 BACKGROUND

• Schwalm JD, Sheth T, Pinilla-Echeverri N, Petch J. Using Artificial Intelligence to Optimize the Use of Cardiac Investigations in Patients With Suspected Coronary Artery Disease. J Soc Cardiovasc Angiogr Interv. 2024 Mar 26;3(3Part B):101305. doi: 10.1016/j.jscai.2024.101305. eCollection 2024 Mar. No abstract available.

  PMID: 39131228 BACKGROUND

• Petch J, Tabja Bortesi JP, Sheth T, Natarajan M, Pinilla-Echeverri N, Di S, Bangdiwala SI, Mosleh K, Ibrahim O, Bainey KR, Dobranowski J, Becerra MP, Sonier K, Schwalm JD. Coronary Computed Tomographic Angiography to Optimize the Diagnostic Yield of Invasive Angiography for Low-Risk Patients Screened With Artificial Intelligence: Protocol for the CarDIA-AI Randomized Controlled Trial. JMIR Res Protoc. 2025 May 21;14:e71726. doi: 10.2196/71726.

  PMID: 40397500 DERIVED

MeSH Terms

Conditions

Coronary Artery Disease

Condition Hierarchy (Ancestors)

Coronary Disease; Myocardial Ischemia; Heart Diseases; Cardiovascular Diseases; Arteriosclerosis; Arterial Occlusive Diseases; Vascular Diseases

Study Officials

• Jon-David Schwalm, MD, MSc

  Hamilton Health Sciences Corporation

  PRINCIPAL INVESTIGATOR

• Jeremy Petch, PhD

  Hamilton Health Sciences Corporation

  PRINCIPAL INVESTIGATOR

• Natalia Pinilla-Echeverri, MD, PhD

  Niagara Health

  PRINCIPAL INVESTIGATOR

Study Design

• Study Type: interventional
• Phase: not applicable
• Allocation: RANDOMIZED
• Masking: NONE
• Purpose: SCREENING
• Intervention Model: PARALLEL
• Sponsor Type: OTHER
• Responsible Party: SPONSOR

Study Record Dates

• First Submitted: October 16, 2024
• First Posted: October 18, 2024
• Study Start: January 9, 2025
• Primary Completion: February 1, 2026
• Study Completion: February 1, 2026
• Last Updated: December 12, 2025

Record last verified: 2025-01

Data Sharing

• IPD Sharing: Will not share

Locations

CA (3)