an international multicentre registry designed to answer 3 clinical questions:

1. Describing the characteristics and outcomes of current patients undergoing aortic + mitral transcatheter heart valve procedures.
2. Better understanding of the predictors for MR regression following isolated TAVI and consequently estimating the fraction of patients who will be suitable for TMVR/r post TAVI
3. Examining the clinical outcomes of patients with significant MR post TAVI who subsequently underwent TMVR/r compared to those left for medical management.

Transcatheter aortic valve implantation (TAVI) , has revolutionised the treatment for severe aortic stenosis (AS) over the past decade. The use of TAVI is growing exponentially, and the current trend of treating younger and lower risk patients is expected to continue [1], with TAVI eventually becoming the treatment of choice for the majority of severe AS patients. When treating younger, lower risk patients, whose life expectancy is longer than that of current typical TAVI patients, long term prognosis with good functional status is as significant as periprocedural morbidity when selecting the preferred treatment strategy. Consequently, treatment of comorbidities, especially cardiac comorbidities is of paramount importance.

Significant mitral regurgitation (MR) is present in approximately 25% of patients with severe AS [2], there are several pathophysiological mechanisms linking the two entities [3,4]. Significant MR is associated with worse prognosis in patients undergoing either surgical aortic valve replacement (SAVR) [5] or TAVI [6,7] . Using a surgical intervention, both AS and MR can be treated (albeit at a cost of increased perioperative mortality) [8], while treating both valves using a transcatheter technique in a single procedure is not customary. A sizeable fraction of patients from AS+MR will have some reduction in MR following TAVI [2], but unlike the case of surgical double valve replacement [6], the prognostic benefit of MR reduction following TAVI has not yet been shown [9].This issue may be of utmost importance in patients who are viable surgical candidates, in whom the feasibility of total surgical valvular treatment may be a "tiebreaker" when contemplating the pros and cons of transcatheter vs. surgical intervention.

The percutaneous treatment options for MR with either transcatheter mitral valve replacement (TMVR) or repair (TMVr) are expanding and clinical experience is growing [1], raising the option for a combined percutaneous treatment for patients with AS+MR - either a combined TAVI + TMVR/r procedure, or performing TAVI first followed by a TMVR/r for those patients whose MR does not improve sufficiently following an isolated TAVI. One multicentre study from Spain dedicated to patients with significant MR undergoing TAVI found two predictors for persistence of MR following TAVI (mitral annular diameter of >35.5 mm and the degree of calcification of the mitral apparatus by multidetector computed tomography), and also suggested that only a small fraction of such patients will be suitable for TMVR/r, but this was based on a small number of patients (n=177) [10].

To examine whether such approach is appropriate, several evidence gaps will need to be addressed:

1. describing the characteristics and outcomes of current patients undergoing aortic + mitral transcatheter heart valve procedures.
2. Better understanding of the predictors for MR regression following isolated TAVI and consequently estimating the fraction of patients who will be suitable for TMVR/r post TAVI
3. Examining the clinical outcomes of patients with significant MR post TAVI who subsequently underwent TMVR/r compared to those left for medical management.

To try and address these questions we propose to compile an international multicentre registry focusing on patients with significant MR undergoing TAVI - the AMTRAC registry.

Methods

A multicentre registry of interventional cardiology centres experienced in transcatheter valvular interventions. This registry will be established in 2 phases (each centre will choose whether it wishes to participate in both phases or just the initial phase):

Phase 1:

Collecting data on patients who underwent TAVI+TMVR/r (either as a single or staged procedure).

This phase will focus on assessing the feasibility and safety of performing TAVI+TMVR/r procedures and will adress 2 clinical question:

1. Describing the demographic, clinical and imaging characteristics of patients undergoing TAVI+TMVR/r in current clinical practice
2. Describing the clinical outcomes (with focus on 30 day outcomes) of patients undergoing TAVI+TMVR/r in current clinical practice

Phase 2:

Adding data on the overall TAVI population of the participating centres.

This phase will be directed at assessing the prevalence, clinical significance, and optimal treatment strategies of TAVI patients who also suffer from significant MR and will address 5 clinical questions:

1. The prevalence of significant MR in patients undergoing TAVI
2. The effect of isolated TAVI on MR in these patients
3. The prognostic effect of significant MR persistence following isolated TAVI
4. The characteristics of patients with persistence of significant MR following isolated TAVI (regarding pre-TAVI predictors of MR persistence and post TAVI suitability for TMVR/r)
5. The outcomes of patients undergoing TMVR/r following isolated TAVI

Patient population

Inclusion criteria:

For phase 1:

TAVI+TMVR/r (either as a single or staged procedure). Follow up of at least 30 days post TMVR/r (unless deceased) Available TTE and/or TOE data pre and post TAVI + TMVR/r

For phase 2:

Patients undergoing successful isolated TAVI Assessment of MR grade prior and post TAVI

Exclusion criteria:

Death within 30 days post TAVI

Patient's data will be collected through the TAVI registries of each participating centre and after anonymization of patient personal details will be sent to the coordinating centre (Rabin Medical Centre) using a streamlined uniform CRF sheet that will be circulated to the participating centres, this will include basic clinical, demographic, echocardiographic and procedural data as well as post procedural clinical outcomes and echocardiographic assessment.

For patients who have subsequently undergone TMVR/r - data on the availability of pre-procedural imaging assessment (TOE, MDCT) and follow up TTE data will be collected as well. When possible (according to the decision of each centre), we would request for the imaging files for patients with significant MR post isolated TAVI to be sent to the coordinating centre to be assessed by an imaging CoreLab

Study endpoints

Phase 1:

30 day mortality (from the date of TMVR/r) Periprocedural outcomes (according to the VARC II criteria) NYHA class 30 days post TMVR/r Grade of MR 30 days post TMVR/r

Phase 2:

For all patients:

Grade of MR prior and following isolated TAVI (at least 30 days post TAVI) Periprocedural outcomes (according to the VARC II criteria) Overall mortality at 1,6,12 months and longest available follow up post TAVI NYHA class at 1,6,12, months and longest available follow up (when available)

For patients with significant MR pre-TAVI:

Regression of MR grade post isolated TAVI Suitability for TMVR/r

For patients undergoing TMVR/r:

Grade of MR following TMVR/r

Definitions

Significant MR:

MR grade 3/4 (moderate+) as defined by each participating centre

TMVR/r:

Transcatheter procedure for the treatment of MR using any available percutaneous device.

Statistical methods

Phase 1:

Summary statistics of the baseline characteristics using mean +/- SD / median + IQR as appropriate for continuous variables and counts (%) for categorical variables.

Graphic display of the cumulative risk for 30 day mortality using Kaplan-Meier curves Graphic display of change in MR grade from baseline to 30 day using bar charts (using the Chi-square test to determine statistical significance).

Identifying factors associated with 30 day mortality and 30 day persistence of MR using a multivariate adjusted logistic regression model.

Phase 2:

Baseline characteristics will be compared between patients with/without significant MR prior to TAVI. categorical variables will be reported as counts or percentages, continuous data as means ±SD or median (interquartile range) as appropriate and compared using t-test or the Wilcoxon sum rank test for continuous variables and Chi Square test for categorical variables as appropriate. The probabilities 1 year and long term mortality according to MR category will be plotted by Kaplan-Meier curves, with comparison of cumulative events by the log-rank test. A multivariate adjusted cox regression model will be fitted to examine the association of post TAVI mortality with baseline MR category.

The same analysis, restricted to patients with significant MR at baseline will be repeated to compare the characteristics and post TAVI mortality between those with significant MR at baseline whose MR has/has not post isolated TAVI.

To identify independent predictors of MR persistence post TAVI we shall fit a multivariate logistic regression model with MR persistence as the outcome variable and all relevant baseline clinical and imaging characteristics as covariates.

For all patients who remained with significant MR post TAVI, suitability for TMVR/r will be assessed after CoreLab review of all available imaging data (TTE/TOE/MDCT).

Baseline characteristics of patients with significant MR post TAVI will be compared between those suitable/unsuitable for TMVR/r as described above.

The preprocedural outcomes (in hospital complications and 30 day mortality) will be presented for those patients undergoing TMVR/r.

Finally, the long term survival (starting from the date of TAVI) will be compared between 5 groups of patients stratified according to their baseline MR, it's response to TAVI and their subsequent treatment strategy (see Figure 1):

1. No significant MR at baseline
2. Significant MR at baseline with regression post isolated TAVR
3. Persistence of significant MR post TAVI, unsuitable for TMVR/r
4. Persistence of significant MR post TAVI, suitable for TMVR/r, treated medically
5. Persistence of significant MR post TAVI, suitable for TMVR/r, treated with TMVR/r

Sample size For phase 1, our goal is to collect data on 50-100 patients who underwent TAVI + TMVR/r (either as a single or staged procedure).

Based on the data by Cortes et al [10], assuming a baseline prevalence of 20% for severe MR, with 40% persistence post TAVI and 13% of those cases suitable for TMVR/r - we would need a sample size of just under 10,000 TAVI cases. We aim to include at least 15-20 high volume TAVI centres in this registry.

Inclusion Criteria:

For phase 1:

• TAVI+TMVR/r (either as a single or staged procedure).
• Follow up of at least 30 days post TMVR/r (unless deceased) Available TTE and/or TOE data pre and post TAVI + TMVR/r

For phase 2:

• Patients undergoing successful isolated TAVI
• Assessment of MR grade prior and post TAVI

Exclusion Criteria:

-Death within 30 days post TAVI

Patients undergoing TAVI who also have data regarding MR grade pre TAVI and at least 30 days post TAVI.