Non-contrast enhanced cardiac MRI using 3D whole heart acquisition protocol is non-inferior to contrast enhanced CT in the assessment of aortic annulus complex for pre-TAVR imaging.

1. Background/Rationale:

   Optimizing procedural and patient outcomes relies heavily on tomographic imaging data guidance for patients being evaluated for trans-catheter aortic valve replacement or implantation (TAVR) by providing accurate information of the aortic valve apparatus (aortic annulus and its pertinent neighboring structures). Though selection of the trans-catheter heart valve size was initially driven by echocardiographic data, recently aortic annulus sizing by electrocardiographic gated cardiac computed tomographic angiography (CCTA) is now widely accepted as the "gold-standard" for this purpose. 8 However, CCTA requires administration of iodinated contrast. Most of the patients being considered for (TAVR) have chronic kidney disease. In these patients, avoiding contrast administration and at the same time being able to accurately assess the aortic annulus apparatus (AVA) 1 2 will be of paramount importance. Also with CCTA, the presence of significant or exuberant aortic annulus calcification can make the annulus border delineation challenging, and in some instances practically impossible to demarcate the boundary of the annulus in some of its sectors. This is due to the blooming and shadowing artifacts caused by the calcification.

   In these patients, when not contraindicated, a non-contrast cardiac magnetic resonance imaging (CMRI) has been suggested as an alternate method to size the AVA. A free-breathing non-contrast navigator-gated 3D whole-heart acquisition can also be obtained to mimic the volumetric acquisition of a CT image. 1 4 However, in this study by Roos et al that is referenced for this recommendation, the authors used the 3D data set only to measure coronal and single oblique sagittal dimensions of the aortic valve annulus. This was adopted because the coronal view is similar to the anterior-posterior view on aortic root angiography, and the reconstructed sagittal view has the same orientation as the mid-esophageal long-axis view on trans-esophageal echocardiogram. It is now established well beyond doubt that this is not the most accurate way to size the aortic annulus, as the aortic valve annulus is a dynamic complex elliptical or ovoid 3D entity that is also subject to constant shape deformation across the cardiac cycle in addition to being influenced by the thoracic and intra-thoracic anatomy or variations. Ideally, this needs to be done creating a plane that precisely corresponds to the aortic annulus/basal ring which includes all 3 lowest insertion points of the aortic valve cusps (hinge points) by multi-planar reconstructions. A review of literature shows that this method of post-processing CMRI 3D data sets using multi-planar reformats has not so far been validated against the currently available "gold standard" CCTA.

   In CMRI, the conventional 2-dimensional (2D) breath-hold cine MRI has been used to assess cardiac anatomy and function. However, this technique requires cooperation from the patient, and in some cases the scan planning is complicated. However, similar to standard echocardiography, most MRI sequences are 2D (i.e. the plane of imaging has to be chosen at the time of the examination and cannot be changed by subsequent manipulation of the data set).2 On the contrary, 3-dimensional (3D) cardiac MRI can overcome some of these problems because it requires minimal planning and can be reformatted in any plane with post-processing. Cardiac MRI using 3D whole heart acquisition protocol may be a practical option for these patients, as this test is able to assess the aortic annulus complex without contrast administration, and without the blooming and shadowing artifacts caused by calcification in some instances.

2. Study Objectives

   2.1. Scope and Duration To assess agreement between cMRI and CCTA, 35 subjects with aortic valve stenosis who are referred for pre-TAVR imaging will be selected for the study. The investigators estimate it will take approximately 6 months to reach our target enrollment of 35 patients. There will not be any follow-up visits.

   2.2. Objective

   To assess if there are measurement differences between contrast-enhanced CT and non-contrast enhanced cardiac MRI using 3D whole heart acquisition protocol of the 3-pronged aortic annulus complex post-processed by a workstation capable of advanced image processing, manipulation and optimal multi-planar reformatting.

   2.3. Hypothesis

   Non-contrast enhanced cardiac MRI using 3D whole heart acquisition protocol is non-inferior to contrast enhanced CT in the assessment of aortic annulus complex for pre-TAVR imaging.

3. Study Design

   The study is a prospective study of 35 patients, men and/or women, with aortic valve stenosis who are referred for pre-TAVR imaging will be selected for the study.

4. Study Procedures

Patients seen at The Center for Advanced Cardiac Care for aortic stenosis will be enrolled into the study if they are eligible and consent. Upon enrollment the following will be collected:

1. Demographic information
2. Pre-CCTA vital signs, height, weight
3. Standard of care laboratory testing including BUN and Creatinine
4. All females of childbearing age will have a urine pregnancy test prior to
5. Clinical risk factor data
6. Current medications
7. Post CCTA laboratory testing including BUN and Creatinine
8. Non-contrast Cardiac MRI
9. Data to be abstracted from surgical procedure will include Valve size, deployment and positioning details and complications, if present.

5. Risks/Benefit

5.1. Risks MRI does not use ionizing radiation (high-energy radiation that can potentially cause damage to DNA, like the x-rays used CT scans).

No contrast will be used for the MRI protocol.

There are no known harmful side-effects associated with temporary exposure to the strong magnetic field used by MRI scanners. However, there are important safety concerns to consider before performing or undergoing an MRI scan (U.S. Food and Drug Administration):

1. The magnet may cause pacemakers, artificial limbs, and other implanted medical devices that contain metal to malfunction or heat up during the exam.
2. Any loose metal object may cause damage or injury if it gets pulled toward the magnet.
3. Dyes from tattoos or tattooed eyeliner can cause skin or eye irritation.
4. Medication patches can cause a skin burn.
5. The wire leads used to monitor an electrocardiogram (ECG) trace or respiration during a scan must be placed carefully to avoid causing a skin burn.
6. Prolonged exposure to radio waves during the scan could lead to slight warming of the body.

The CCTA will be done as per the established standard practice and is currently the the standard of care in accurately sizing aortic valve annulus in patients when clinically indicated. The risks of CCTA include:

Coming from radiation department

5.2. Minimization of Risks Patient safety during this study will be the highest priority. All diagnostic tests/treatments provided are commonly accepted and FDA approved. The patients will not incur any psychological, social, legal or economic risks by participating in this study.

5.3. Benefits Subjects will not directly benefit from this research. The results of this study may provide important information for the medical imaging options for future valve replacement patients.

6. Statistical Methods and Analyses Assuming a 23.5mm mean difference of the aortic valve annulus: area between cardiac MRI and CT measurements, a standard deviation of the mean difference of 7.5mm, a lower equivalence bound of -30mm, a upper equivalence bound of 30mm, and alpha = 0.05 with 35 patients the study power is estimated to be over 80% (see Figure).

Statistical Analysis A generalized propensity score (to adjust for pre-operative clinical and non-clinical factors) approach will be used to account for possible confounding of the association between MRI and CT and annulus area, annulus perimeter, and maximum and minimum annulus diameter measurements. Specifically, a general linear (which will account for clustering of data within patients and will include the propensity score) will be developed to assess measurement differences between diagnostic tests.

Moreover, adjusted (by the propensity score) Kappa tests, specificity, and sensitivity will be used to assess agreement between MRI and CT.

7. Data Management The data collection form with the clinical and non-clinical risk factors and measurements that the investigators will collect for each patient is attached (see attachment).

Data Storage: An appropriate database will be created and managed in compliance with HIPAA regulations. Data will be stored safely in a secure environment, and maintenance for back-up purposes will be performed on a routine basis to avoid catastrophic data loss.

Research Subject Privacy: Patients will be approached discretely and privately by the study coordinator on location at The Center for Advanced Cardiac Care prior to their procedure. The discretion of the study coordinator will be used to judge if privacy (from both medical personnel and other persons in the hospital) is ensured during the consent process. The patient will be informed that they can stop the consent process at anytime if they feel their privacy is being compromised, and no information will be collected directly from the patient after that point. Patients will be informed that participating / not participating in this study will in no way affect their care and that they are in no way obligated or expected to participate. If the patient agrees to participate, they will be taken through the informed consent process. Upon data collection/acquisition, study identifiers will be encrypted to protect patients' privacy and meet HIPAA regulations, and data will be stored in a secured database while periodic back-up will protect against catastrophic data loss. All hard copy material (summary printouts) will be stored by the principal investigator in locked file cabinets and the disks, diskettes, and other removable storage media files will be safely stored as per HIPAA regulations. All patient identifiers will be removed from research related documents and a study specific identifier assigned to each subject. . Only de-identified data will be used for analysis and only aggregate results will be reported. Only those directly related to the research will be allowed access to subject information. Patients will not be identified as research subjects to those individuals they interact with except for those individuals who need to be made aware of the status for safe conduct of the study.

8. Data Monitoring An internal audit will occur within thirty days after the first subject is enrolled and at least annually until study closure. CRFs, source documents, informed consent forms, and study deviations will be included in the audits. Audit findings will be reported to the PI.

Regulatory documents will be audited by the Baylor Research Institute Department of Research Compliance upon request.

9. Publication The findings will be presented at a national/international conference and submitted for peer-review publication. Likely venues for publication include Circulation and Journal of the American College of Cardiology.

Inclusion Criteria

1. Patients aged ≥18 years
2. Patients with aortic stenosis.

Exclusion Criteria

1. Pregnancy-Women of childbearing potential should have negative urine pregnancy test prior to enrollment
2. Patients with implanted pacemaker
3. Patients with implantable cardiac defibrillator
4. Patients with contraindication to MRI
5. Medically unable to provide consent
6. Any surgical prosthesis
7. Moderate or severe mitral regurgitation
8. Central aortic valve leak

The study cohort will consist of 35 patients who are referred for pre-TAVR imaging with arotic valve stenosis who meet the following eligibility criteria and have provided written informed consent.