Ischemia, Myocardial
Conditions
Keywords
Myocardial velocity imaging, Strain, Strain rate, Speckle tracking, Deformation, Quantitative, Echocardiography, Machine learning
Brief summary
Patients with chest pain on exertion need a reliable non-invasive test to identify if they have inducible myocardial ischaemia. This would reduce the use of diagnostic coronary arteriography, avoid its risks and costs, and guide clinical decisions. Conventional stress echocardiography has poor reproducibility because it relies on qualitative and subjective interpretation. Quantitative approaches based on precise and reliable measurements of myocardial velocity, strain, strain rate and global longitudinal strain have been shown to be able to accurately diagnose myocardial ischaemia. A more accurate test using myocardial velocity imaging was not implemented by ultrasound vendors although it provided an objective measurement of myocardial functional reserve on a continuous scale from normality to severe ischaemia. The investigators propose an original approach to create a diagnostic software tool that can be used in routine clinical practice. The investigators will extract and compare quantitative data obtained through myocardial velocity imaging and speckle tracking in subjects who undergo dobutamine stress echocardiography. The data will be analysed using advanced computational mathematics including multiple kernel learning and joint statistics applied to multivariate data across multiple dimensions (including velocity, strain and strain rate traces). This approach will be validated against quantitative coronary arteriography and fractional flow reserve. The results will be displayed as parametric images and placed into a reporting tool. The output will determine the presence and severity of myocardial ischaemia. These new tools will have the capacity for iterative learning so that the precision of the diagnostic conclusions can be continuously refined.
Interventions
DIAGNOSTIC_TESTDeformation imaging
Deformation parameters derived using myocardial velocity imaging or speckle tracking
Sponsors
Universitaire Ziekenhuizen KU Leuven
Universitat Pompeu Fabra
Hull University Teaching Hospitals NHS Trust
Danderyd Hospital
University Hospital of Wales
Study design
Observational model
COHORT
Time perspective
PROSPECTIVE
Eligibility
Sex/Gender
ALL
Age
20 Years to 89 Years
Healthy volunteers
Yes
Inclusion criteria
* Chest pain, chest pain equivalent
Exclusion criteria
* acute coronary syndrome with elevated troponin, severe heart valve disease, uncontrolled hypertension (resting SBP \>200mmHg), cardiomyopathy, contraindication to dobutamine, pregnancy
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Diagnostic accuracy of quantitative measures of dobutamine stress echocardiography | 18 months | Echocardiographic measurements of segmental myocardial velocity, strain, strain rate and wall motion scoring referenced against measurements derived from coronary angiography. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Lowest dose of dobutamine to provoke measurable marker of inducible myocardial ischaemia | 18 months | Using modelling techniques applied predict lowest dose of dobutamine to maintain diagnostic accuracy |
| Diagnostic accuracy of using machine learning to interpret multiparametric and multidimensional datasets to diagnose myocardial ischaemia | 18 months | Use modelling to combine pre-test probabilities (based on risk factors such as age), physiological factors (e.g., heart rate) that are associated with longitudinal function and data derived throughout the cardiac cycle (i.e., based on analysis of velocity or strain curves and not just a single value like peak velocity or strain). |
Countries
Belgium, Sweden, United Kingdom
Contacts
Primary ContactImran D Sunderji
Backup ContactAlan G Fraser
Outcome results
None listed