Hidden risk in normal myocardial perfusion scans: AI-detected proximal coronary calcium on CT attenuation maps improves prognosis.
Authors
Affiliations (23)
Affiliations (23)
- Departments of Medicine (Division of Artificial Intelligence in Medicine), Biomedical Sciences, and Cardiology, Cedars-Sinai Medical Center, 6500 Wilshire Blvd, Los Angeles, CA, 90048, USA.
- Department of Cardiac Sciences, University of Calgary, Calgary, AB, Canada.
- Signal and Image Processing Institute, Ming Hsieh Department of Electrical and Computer Engineering, University of Southern California, Los Angeles, CA, USA.
- Departments of Medicine (Cardiology) and Population Health Science and Policy, Icahn School of Medicine at Mount Sinai, New York, NY, USA.
- Department of Interventional Cardiology and Cardiac Surgery, University of Zielona Góra, Zielona Góra, Poland.
- Division of Cardiology, Department of Medicine, Department of Radiology, Columbia University Irving Medical Center and New York- Presbyterian Hospital, New York, NY, USA.
- Section of Cardiovascular Medicine, Department of Internal Medicine, Yale University School of Medicine, New Haven, CT, USA.
- Division of Cardiology, University of Ottawa Heart Institute, Ottawa, ON, Canada.
- Intermountain Medical Center Heart Institute, Intermountain Healthcare, Murray, UT, USA.
- Department of Internal Medicine, University of Utah, Salt Lake City, UT, USA.
- Department of Cardiovascular Medicine, Mayo Clinic, Rochester, MN, USA.
- Department of Nuclear Cardiology, National Institute of Cardiology Ignacio Chávez, México City, México.
- Faculty of Medicine, National Autonomous University of Mexico, México City, México.
- Department of Cardiovascular Medicine, The University of Kansas Medical Center, Kansas City, KS, USA.
- Cardiology Division, Montefiore Health System/Albert Einstein College of Medicine, Bronx, NY, USA.
- Department of Radiology (Nuclear Medicine), Montefiore Medical Center, Albert Einstein College of Medicine, Bronx, NY, USA.
- Division of Cardiology, Department of Medicine, David Geffen School of Medicine, University of California, Los Angeles, CA, USA.
- Department of Advanced Biomedical Sciences, University of Naples Federico II, Naples, Campania, Italy.
- Houston Methodist DeBakey Heart & Vascular Center, Houston Methodist Academic Institute, Houston, TX, USA.
- Division of Cardiology, Department of Medicine, University of Ottawa Heart Institute, Ottawa, ON, Canada.
- Department of Nuclear Medicine, Cardiac Imaging, University Hospital Zurich, Zurich, Switzerland.
- Cardiovascular Imaging Program, Departments of Radiology and Medicine; Division of Nuclear Medicine and Molecular Imaging, Department of Radiology; and Cardiovascular Division, Department of Medicine, Brigham and Women's Hospital, Boston, MA, USA.
- Departments of Medicine (Division of Artificial Intelligence in Medicine), Biomedical Sciences, and Cardiology, Cedars-Sinai Medical Center, 6500 Wilshire Blvd, Los Angeles, CA, 90048, USA. [email protected].
Abstract
Spatial distribution of coronary artery calcium (CAC) may provide additional prognostic value in patients undergoing SPECT and PET myocardial perfusion imaging (MPI). We aimed to automatically identify CAC in proximal segments from attenuation correction CT (CTAC) scans using artificial intelligence (AI) and to evaluate prognostic significance in two large international multicenter registries. From hybrid MPI/CT imaging (N = 43,099) across 15 sites, we included 4,552 most relevant patients with (1) no prior coronary artery disease; (2) AI-derived mild CAC scores (1-99); and (3) normal perfusion (stress total perfusion deficit < 5%). The independent associations between AI-identified proximal CAC and major adverse cardiovascular events (MACE) and all-cause mortality (ACM) were evaluated using multivariable Cox regression, likelihood ratio test (LRT), and continuous net reclassification index (NRI). Among the patients with mild CAC and normal perfusion (mean age 65 ± 12 years, 51% male), 1,730 (38%) had proximal CAC. Over 3.6 (inter-quartile interval 2.1, 5.2) years follow-up, 599 (13%) and 444 (10%) patients had MACE or ACM, respectively. Proximal CAC was associated with an increased risk of MACE (adjusted hazard ratio [HR] 1.24, 95% CI 1.03-1.48, P = 0.02) and ACM (adjusted HR 1.25, 95% CI 1.01-1.53, P = 0.04) after the adjustment of CAC score and density, clinical risk factors, and perfusion deficit. Proximal CAC improved the risk stratification of MACE (LRT P = 0.02; NRI 12%) and ACM (LRT P = 0.04; NRI 12%). In patients with mild CAC and normal myocardial perfusion, AI-based proximal CAC detection identified a subgroup at increased risk of adverse outcomes. Automated identification of proximal CAC may provide incremental prognostic information beyond perfusion findings and CAC scoring and could improve risk stratification in patients who might otherwise be considered low risk.