CMR-AI and Outcomes in Aortic Stenosis
Study Details
Study Description
Brief Summary
Background: Artificial Intelligence (AI) in cardiac imaging has previously been shown to provide highly reproducible and accurate results, outperforming clinical experts. Cardiac magnetic resonance (CMR) imaging represents the gold standard for assessment of myocardial structure and function. However, measurements of more sensitive markers of early left (LV) and right ventricular (RV) function, such as global longitudinal shortening (GLS), mitral annular plane systolic excursion (MAPSE), and tricuspid annular plane systolic excursion (TAPSE), are frequently not performed due to the lack of automated analysis.
Objectives: The investigators aim to evaluate whether AI-based measurements of ventricular structure and function convey important prognostic information in patients with severe aortic stenosis (AS) beyond LV and RV ejection fraction (EF) and represent early markers of adverse cardiac remodeling.
Materials & Methods: This large-scale international, multi-center, observational study will recruit ~1500 patients with severe AS scheduled for aortic valve replacement (AVR). Patients are invited to undergo CMR imaging prior to AVR and at 12-months post-AVR. An AI-based algorithm, developed in the UK, will be used for fully automated assessment of parameters of cardiac structure (end-diastolic volume, end-systolic volume, LV mass, maximum wall thickness) and function (EF, GLS, MAPSE, TAPSE). Application of the AI-model allows to capture these parameters for large patient cohorts within seconds (as opposed to the current practice of time-consuming manual post-processing). Association of AI-based CMR parameters with clinical outcomes post-AVR will be analyzed. The composite of all-cause mortality and heart failure hospitalization will serve as the primary endpoint. Trajectories of AI-based parameters from pre- to post-AVR will be assessed as a secondary endpoint.
Future Outlook: In severe AS, a novel AI-based algorithm allows immediate and precise measurements of ventricular structure and function on CMR imaging. Our goal is to identify early markers of cardiac dysfunction indicating adverse prognosis post-AVR. This has guideline-forming potential as the optimal timepoint for AVR in patients with AS is currently a matter of debate.
Condition or Disease | Intervention/Treatment | Phase |
---|---|---|
|
Detailed Description
Artificial Intelligence (AI) and Machine Learning are reshaping our daily clinical practice, which has the potential to be more efficient, precise, and personalized. Adopting these technologies in cardiac imaging does not only affect decision making by improved accuracy and risk stratification but also significantly reduces scan times and post-imaging workup.
Current guidelines acknowledge cardiac magnetic resonance (CMR) imaging as gold standard for assessment of myocardial structure and function. Although of fundamental importance in various cardiac diseases, measurements of size, mass, and ejection fraction (EF) are flawed by the inherent variability and subjectivity of human analysis. Recent developments in deep learning using convolutional neural networks (CNNs) allow for automated segmentation of the ventricular blood pool and myocardium using pre-existing CMR datasets. Importantly, these tools are integrated into CMR scanners generating real-time measurements without the need of time-consuming image post-processing. AI-based models have previously shown superior precision in ventricular contouring, volumetry, and maximum wall thickness measurements, outperforming clinical experts.
In patients with aortic stenosis (AS), changes in EF more often occur late in the disease process. Longitudinal shortening represents an earlier and more sensitive marker of left ventricular (LV) dysfunction. However, these measurements remain subjective, time-consuming, and are therefore not routinely performed due to the lack of automated analysis. Recently, AI-measured global longitudinal shortening (GLS) and mitral annular plane systolic excursion (MAPSE) have been demonstrated to provide more reproducible and accurate results, when compared to human experts. The investigators hypothesize that AI-based GLS and MAPSE could convey important prognostic information beyond LVEF in severe AS and represent early markers of adverse cardiac remodeling.
Furthermore, the investigators could previously demonstrate that right ventricular (RV) dysfunction on CMR, rather than conventional parameters assessed by echocardiography, was independently associated with outcome in individuals with AS undergoing transcatheter aortic valve implantation. The investigators aim to extend on our findings and investigate whether AI-based RV GLS and tricuspid annular plane systolic excursion (TAPSE) represent early markers of RV dysfunction indicating adverse prognosis.
Finally, the assessment of reverse cardiac remodeling by CMR requires high precision (=reproducibility). AI has been proven to outperform humans in both precision and accuracy, and therefore has great potential for the comprehensive evaluation of longitudinal structural changes in AS following valve replacement. The investigators aim to analyze reverse cardiac remodeling in patients with AS using novel AI technology.
Aims
With significant previous contributions in cardiac imaging and valvular heart disease being made in the past, the investigators aim to expand the knowledge in this field by exploring the following:
-
Association of AI-measured LV and RV structural and functional markers on CMR in patients with severe AS with clinical outcomes following aortic valve replacement (AVR).
-
Reverse cardiac remodeling at baseline to 12-months after AVR as determined by AI-based CMR methods.
-
The ultimate goal is to provide automated, precise, and time-saving algorithms to identify patients at risk undergoing AVR.
Methods
This is a large-scale international, multi-center, multi-cohort, observational study. Patients with severe degenerative AS scheduled for valvular treatment discussion in an interdisciplinary Heart Team will be invited to participate. Enrollment will be performed prospectively at seven university-affiliated tertiary care centers in Continental Europe, the UK, and Asia.
Baseline evaluation consists of medical history, physical examination, routine blood tests, electrocardiogram, echocardiography, and CMR imaging. Patients will be prospectively followed up by an ambulatory visit at 12 months. In addition, CMR will be repeated at 1 year.
Study Design
Outcome Measures
Primary Outcome Measures
- Number of patients with AI-measured parameters of impaired left and right ventricular structure and function on cardiac magnetic resonance imaging and association with the composite of all-cause death and heart failure hospitalization. [2 years]
Association of AI-based parameters (end-diastolic volume [ml], end-systolic volume [ml], left ventricular mass [gram], maximum wall thickness [mm], ejection fraction [%], global longitudinal shortening [%], mitral/tricuspid annular plane systolic excursion [mm]) on cardiac magnetic resonance imaging with the composite of all-cause death and heart failure hospitalization. Captured clinical endpoints will include all-cause death, cardiovascular mortality, and heart failure hospitalization. Data will be ascertained by follow-up visits, state-wide electronic hospital charts, and phone calls. In addition, mortality data will be obtained via National Death Registries of the participating countries.
Secondary Outcome Measures
- Number of patients with AI-measured parameters of impaired left and right ventricular structure and function on cardiac magnetic resonance imaging and association with components of the primary endpoint analyzed individually. [2 years]
Association of AI-based parameters (end-diastolic volume [ml], end-systolic volume [ml], left ventricular mass [gram], maximum wall thickness [mm], ejection fraction [%], global longitudinal shortening [%], mitral/tricuspid annular plane systolic excursion [mm]) on cardiac magnetic resonance imaging with components of the primary endpoint (all-cause death and heart failure hospitalization) analyzed individually. Captured clinical endpoints will include all-cause death, cardiovascular mortality, and heart failure hospitalization. Data will be ascertained by follow-up visits, state-wide electronic hospital charts, and phone calls. In addition, mortality data will be obtained via National Death Registries of the participating countries.
- Number of patients with AI-measured parameters of impaired left and right ventricular structure and function on cardiac magnetic resonance imaging and association with cardiovascular mortality. [2 years]
Association of AI-based parameters (end-diastolic volume [ml], end-systolic volume [ml], left ventricular mass [gram], maximum wall thickness [mm], ejection fraction [%], global longitudinal shortening [%], mitral/tricuspid annular plane systolic excursion [mm]) on cardiac magnetic resonance imaging with cardiovascular mortality. Captured clinical endpoints will include all-cause death, cardiovascular mortality, and heart failure hospitalization. Data will be ascertained by follow-up visits, state-wide electronic hospital charts, and phone calls. In addition, mortality data will be obtained via National Death Registries of the participating countries.
Other Outcome Measures
- Number of patients with AI-measured parameters of impaired left and right ventricular structure and function on cardiac magnetic resonance imaging at baseline and changes at 1-year follow-up after aortic valve replacement (AVR). [1 year]
Longitudinal trajectories of AI-based parameters (end-diastolic volume [ml], end-systolic volume [ml], left ventricular mass [gram], maximum wall thickness [mm], ejection fraction [%], global longitudinal shortening [%], mitral/tricuspid annular plane systolic excursion [mm]) on cardiac magnetic resonance imaging from pre- to 1-year post-AVR. Captured clinical endpoints will include all-cause death, cardiovascular mortality, and heart failure hospitalization. Data will be ascertained by follow-up visits, state-wide electronic hospital charts, and phone calls. In addition, mortality data will be obtained via National Death Registries of the participating countries. CMR will be repeated at 1 year.
Eligibility Criteria
Criteria
Inclusion Criteria:
-
Written informed consent
-
Severe AS scheduled for Heart Team decision
Exclusion Criteria:
-
Inability or unwillingness to perform any of the diagnostic tests
-
Inability or unwillingness to participate in follow-up visits
-
Metal implants, e.g. cochlear implants and pacemakers
-
Chronic kidney failure (GFR < 30 ml/min/1.73m2)
Contacts and Locations
Locations
No locations specified.Sponsors and Collaborators
- Medical University of Vienna
Investigators
None specified.Study Documents (Full-Text)
None provided.More Information
Publications
None provided.- VIE_CMR-AI