Anderson, Todd J.

Introduction Angina with no obstructive coronary artery disease (ANOCA) affects millions and is frequently under-recognised because diagnostic pathways and risk tools predominantly target obstructive coronary artery disease (CAD). This protocol describes shared methods for two machine-learning (ML) studies: (1) differentiating ANOCA from stable angina with obstructive CAD and (2) predicting long-term mortality among patients with ANOCA and obstructive CAD. Methods and analysis We will develop and cross-site validate ML classification models using a multicentre retrospective cohort drawn from the Alberta Provincial Project for Outcome Assessment in Coronary Heart Disease registry and institutional datasets from the University of Ottawa Heart Institute and the University Health Network. Eligible participants are adults (≥18 years) undergoing initial cardiac catheterisation for chest pain/anginal equivalents since 1995, excluding prior revascularisation, major structural heart disease and predefined non-anginal indications. Outcomes are (1) ANOCA (0% to <50% stenosis) versus obstructive CAD (≥50% stenosis) and (2) 1, 3 and 5-year mortality, modelled separately for ANOCA and obstructive CAD. Model development will use nested cross-validation with stratified k-fold inner-loop tuning and leave-one-site-out cross-validation for repeated external validation. Candidate predictors will be harmonised across sites, filtered for missingness and refined using expert/directed acyclic graph-guided selection plus Boruta and Least Absolute Shrinkage and Selection Operator. Preprocessing includes appropriate encoding, missing-data imputation (multivariate imputation by chained equations) and feature scaling. Algorithms will include elastic-net logistic regression, random forest, LightGBM and multilayer perceptron models; hyperparameters will be optimised via Bayesian optimisation. Performance and threshold tuning will be reported. Explainability and subgroup fairness will be assessed using SHapley Additive exPlanations. Final models will be deployed as a web-based clinical risk calculator. Ethics and dissemination Ethics approval has been obtained from the University of Calgary and the University Health Network (#24-5916). Analyses will use deidentified data in secure environments; only aggregate results will be reported. Findings will be disseminated via peer-reviewed publications, conferences and a web-based calculator.

Background: Intraplaque hemorrhage (IPH) is associated with plaque progression and ischemic events, and plaque lipid content (% lipid core) predicts the residual atherosclerotic cardiovascular disease risk. This study examined the impact of IPH on lipid content change in the setting of intensive lipid-lowering therapy. Methods: In total, 214 AIM-HIGH (Atherothrombosis Intervention in Metabolic Syndrome with Low High-Density Lipoprotein/High Triglycerides: Impact on Global Health Outcomes) participants with clinically established ASCVD and low high-density lipoprotein cholesterol received cartoid MRI at baseline and 2 years to assess changes in carotid morphology and composition. Patients were randomized to extended-release niacin or placebo, and all received simvastatin with optional ezetimibe as necessary to lower low-density lipoprotein cholesterol to 40 to 80 mg/dL. Changes in lipid content and carotid morphology were tested using the Wilcoxon signed-rank test. Differences between subjects with and without IPH and between subjects assigned extended-release niacin or placebo were tested using the Wilcoxon rank-sum test. Linear regression was used to test the association of IPH and lipid content changes after adjusting for clinical risk factors. Results: Among 156 patients (61±9 years; 81% men) with complete MRI, prior statin use: <1 year, 26%; 1 to 5 years, 37%; >5 years, 37%. Triglycerides and ApoB decreased significantly, whereas high-density lipoprotein cholesterol and ApoA1 increased significantly over time. Plaque lipid content was significantly reduced (−0.5±2.4 %/year, P = 0.017) without a significant difference between the 2 treatment groups. However, the lipid content increased in plaques with IPH but regressed in plaques without IPH (1.2±2.5 %/year versus −1.0±2.2, P = 0.006). Additionally, IPH was associated with a decrease in lumen area (−0.4±0.9 mm2/year versus 0.3±1.4, P = 0.033). IPH remained significantly associated with increase in lipid content in multivariable analysis (54.4%, 95% CI: 26.8, 88.0, P < 0.001). Conclusions: Carotid plaques under continued intensive lipid-lowering therapy moved toward stabilization. However, plaques with IPH showed greater increases in lipid content and greater decreases in lumen area than plaques without IPH. Registration: URL: https://www.clinicaltrials.gov ; Unique identifier: NCT01178320.

Background: Although vulnerable plaques are considered the pathological substrate for acute cardiovascular events, the pursuit of imaging surrogate markers for systemic atherothrombotic risk is currently focused on plaque burden. The prognostic value of vulnerable plaque characteristics as novel markers for systemic cardiovascular outcomes remains elusive. Advances in magnetic resonance imaging (MRI) have enabled characterization of the vulnerable carotid plaque. In a prospective study, we tested whether MRI-detected carotid plaque characteristics predict subsequent cardiovascular events. Methods: As part of an event-driven clinical trial, subjects with clinically established atherosclerotic disease were recruited. A multi-sequence protocol was used to measure the volumes of calcification and necrotic core (NC), and to identify the presence of intraplaque hemorrhage (IPH) and thin/ruptured fibrous cap (FC) using published criteria. The primary endpoint included fatal and non-fatal myocardial infarction or ischemic stroke, hospitalization for acute coronary syndrome, and symptom-driven revascularization. Cox regression analysis was used to present results as hazard ratio (HR) with 95% confidence interval (CI). Results: Of 232 subjects recruited, 214 (92.2%) with diagnostic image quality constituted the study population (mean age: 61±9 years; male: 82%; statin use: 94%). Calcification, NC, IPH, and thin/ruptured FC were detected in 48%, 52%, 8% and 14% of subjects, respectively. During a median follow-up of 35.1 months, 18 (8.4%) subjects reached the primary endpoint. MRI-detected plaque characteristics associated with the primary endpoint included larger NC (HR per 1 standard deviation increase in volume [1-SD]: 1.43 [1.16, 1.75], p<0.001), thin/ruptured FC (HR: 4.31 [1.67, 11.12], p=0.003), and marginally IPH (HR: 3.00 [0.99, 9.13], p=0.053). Although calcium deposition is traditionally pursued as a marker for atherosclerosis, subjects with larger calcification (HR per 1-SD: 0.60 [0.26, 1.41], p=0.20) did not show increased risk. Conclusions: Vulnerable plaque characteristics are prevalent in carotid arteries of patients with established atherosclerosis and are predictive of systemic cardiovascular outcomes.