Cardiac Structural and Functional Remodeling After Transcatheter Mitral Valve in Valve Implantation: Early Changes and Prognostic Significance

Background Transcatheter mitral valve-in-valve (MViV) replacement has emerged as an alternative to redo mitral valve (MV) surgery for the management of failed bioprosthetic MVs. The degree of cardiac remodeling assessed by echocardiography has been shown to have prognostic implications in degenerative mitral regurgitation patients undergoing MV surgery. The impact of transcatheter MViV in patients with degenerative bioprosthetic MV failure on cardiac remodeling and its associated prognosis remains undescribed. Objectives The aim of this study is to describe the early anatomic and functional changes of the left-sided chambers and right ventricle by echocardiography posttranscatheter MViV intervention and their impact on mortality outcomes. Additionally, we sought to analyze the outcome of heart failure in bioprosthetic MV failure patients undergoing transcatheter MViV replacement. Methods We analyzed consecutive patients undergoing MViV intervention for symptomatic bioprosthetic MV failure. Echocardiograms before intervention and within 100 days postintervention were analyzed. A chart review was performed to obtain baseline characteristics, follow-up visits, 30-day heart failure and 1-year all-cause mortality outcomes. Results A total of 62 patients (mean age 69 ± 13 years, 61% male) were included in the study. Most patients were undergoing MViV intervention for prosthetic mitral stenosis n = 48 (77.4%) and the rest for mitral regurgitation or mixed disease. Compared with baseline, significant reductions were observed in median left atrial volume (LAV; 103 [81–129] ml vs. 95.2 [74.5–117.5] ml, p < 0.01) and mean (SD) left atrial conduit strain (9.1% ± 5.2% vs. 10.8% ± 4.8%, p = 0.039) within 100 days postintervention. Early reduction in right ventricular free wall global longitudinal strain and fractional area change also occurred postintervention. No significant change in left ventricular chamber dimensions or ejection fraction was observed. During the 1-year follow up period, 5 (8%) patients died. While baseline LAV was not associated with 1-year all-cause mortality (OR 0.98 CI 0.95–1.01; p = 0.27), a change in LAV in the follow up period was associated with all-cause mortality at 1 year (OR 1.06 CI 1.01–1.12; p = 0.023). At 30 days postintervention, 65% of patients had an improvement in their New York Heart Association functional class. Conclusion In this retrospective study of patients undergoing transcatheter MViV intervention for failed bioprosthetic MVs, early reverse remodeling of the left atrium occurs within 100 days postintervention and reduction in LAV is associated with reduced all-cause mortality at 1 year. In addition, there is significant improvement in heart failure symptoms at 30 days following intervention but further investigation into the longitudinal remodeling changes and long-term outcomes is needed.


A B B R E V I A T I O N S
MViV, mitral valve-in-valve; NYHA, New York Heart Association; FAC, fractional area change; LAV, left atrial volume; TAPSE, tricuspid annular plane systolic excursion; S 0 , right ventricular peak systolic velocity; PALS, peak atrial longitudinal strain; PACS, peak atrial contractile strain; CS, conduit strain.

Introduction
Bioprosthetic mitral valve (MV) degeneration that requires repeat intervention within 10 years postimplantation occurs in up to 35% of patients.][3][4] The procedural success and durability of transcatheter MV replacement has been reported in outcome studies from the Society of Thoracic Surgeons/American College of Cardiology Transcatheter Valve Therapy Registry (TVT), with higher procedural success and lower short-term mortality reported in MV-in-valve (MViV) intervention cohorts compared to valve-in-ring and valve-in-mitral annular calcification cohorts.Those undergoing MViV intervention showed sustained clinical improvement in heart failure with an overall all-cause mortality of approximately 5% at 30 days and 17% at 1 year. 5,6his success is likely driven by underlying anatomical and functional changes in heart chamber dynamics that are not yet completely understood.Changes in chamber size are a hallmark of structural remodeling due to changes in the function of the chambers.[9][10][11][12][13] Left atrial strain reflects the role of the left atrium in modulating left ventricular filling in different phases of the cardiac cycle.Peak atrial longitudinal strain (PALS) contributes about 40%-50% of total LA function and reflects the end of the reservoir phase when the LA fills and reaches its peak in systole.Following this, passive left atrial emptying, also known as the conduit phase, can be captured by conduit strain (CS) measurements and contributes to about 25% of atrial function.Finally, atrial contraction which can be measured as the peak atrial contractile strain contributes to about 25% of atrial function. 7,14Similarly, right ventricular remodeling and function can be understood through dimensions and pressures at various phases, as well as through echocardiographic measurements of tricuspid annular plane systolic excursion (TAPSE), lateral tricuspid annulus peak systolic velocity (S 0 ), fractional area change (FAC), and strain. 15,16n short, though the effect of MV surgery on left atrial and biventricular remodeling and prognosis has been extensively studied, 8,[10][11][12][13][17][18][19] there remains a paucity of data on the effect of MViV replacement on cardiac structural and functional remodeling and the prognostic impact of changes in chamber dimensions, function, and reverse cardiac remodeling in patients with degenerative bioprosthetic MV failure.
Therefore, the aim of this study was to describe the early anatomic and functional changes of the left-sided chambers and right ventricle by echocardiography due to transcatheter MViV intervention and to determine their association with the primary outcome of all-cause mortality at 1 year.Additionally, we sought to analyze the impact of MViV replacement on heart failure symptom improvement.

Methods
This study was based on the experience at Northwestern University with transcatheter MViV replacement between January 2016 and October 2021.Patients were identified and data was collected from comprehensive 2-dimensional echocardiograms performed before intervention (baseline) and within 100 days postintervention (follow-up).Outcome data for all patients undergoing percutaneous transcatheter MViV replacement with balloonexpandable SAPIEN 3 (Edwards Lifesciences, Irvine, California) transcatheter heart valves via transeptal approach 20 for symptomatic bioprosthetic MV failure was obtained from the TVT registry.The data were collected by trained abstractors who are unaffiliated with the author team.
The study protocol was approved and granted a waiver of informed consent by the Northwestern University institutional review board.

Echocardiographic Analysis
All two-dimensional echocardiograms included in the study were analyzed by an advanced cardiovascular imaging trained cardiologist following standardized American Society of Echocardiography guidelines 21,22 with assessment of variables including left ventricular ejection fraction by biplane Simpson's method of disks, left ventricular end systolic dimension, left ventricular end diastolic dimension, left atrial volume (LAV) by biplane method, right ventricular fractional area change, TAPSE, right ventricular peak systolic velocity, and MV gradient.All strain assessment was performed via speckle tracking analysis using TomTec Arena 2.41 (Philips); 2D echocardiograms with inadequate images for myocardial deformation analysis (defined as inadequate quality of tracking in !2 segments) were not utilized.

Statistical Analysis
For descriptive statistics, continuous variables were presented as mean AE standard deviation (SD) or median (IQR) as appropriate.Categorical variables were presented as frequency and percentages.Baseline characteristics were compared between patients with and without 1-year mortality using two sample t-test, Wilcoxon test, Chi-squared test or Fisher exact test as appropriate.
Pre-and post-MViV intervention echocardiographic parameters of interest were compared using the paired t-test and Wilcoxon sign-rank test as appropriate.Cox proportional hazards regression models adjusted for age and gender were used to evaluate the association between baseline echocardiogram parameters and early change in echo parameters with 1-year mortality.
A subgroup analysis was also conducted to evaluate differences of echocardiographic parameters pre-and post-MViV intervention in patients with atrial fibrillation/flutter. Comparisons between groups were performed using two sample t-test, or Wilcoxon test as appropriate.
Improvement in New York Heart Association (NYHA) class distribution 1 month after MViV procedure was assessed graphically using an alluvial flow plot.In exploratory analyses, echocardiographic parameters were further compared between valve lesion and postprocedural residual MR groups using Analysis of Variance or Kruskal-Wallis test.
All statistical analyses were performed using SAS (Base 9.4) software, with p-values of <0.05 considered statistically significant.

Results
During the study period, 62 patients had transcatheter MViV intervention, with a mean age of 69 AE 13 years, and 61% were male (Table 1).There were no significant differences in baseline characteristics between those who survived at 1 year and those who did not.
Fifty-seven (92%) patients had preintervention and follow-up (within 100 days after intervention) echocardiography that allowed measurement of at least the LAV.
Left atrial CS (p ¼ 0.039) was increased significantly after intervention.No change from baseline was found in left atrial PALS (p ¼ 0.08), or peak atrial contractile strain (p ¼ 0.32) after intervention (Table 2).
We additionally observed no structural or functional remodeling of the left ventricle as determined by left ventricular chamber dimension We observed significant differences in baseline MV gradients between valve lesion groups with higher gradients noted in patients with prosthetic stenosis (14.4mmHgAE4.6) compared to those with regurgitation (7.6mmHg AE2.7) or mixed valve disorders (11.8mmHgAE4.6) (p < 0.01) but no significant difference postintervention (p ¼ 0.05) (Supplemental Table 1).
Although 65% of patients had trace to mild residual mitral regurgitation following intervention, no echocardiographic parameter was significantly associated with postprocedure residual mitral regurgitation (Supplemental Table 2).

Association of Baseline Echocardiography Parameters and Early Cardiac Remodeling with 1-year all-cause Mortality
During the 1-year follow up period postintervention, 5 (8%) patients died.
All-cause mortality at 1 year was not associated with baseline echo derived left-sided chamber size or function parameters evaluated, nor with right ventricular systolic pressure or right ventricular function as assessed by FAC, TAPSE, S 0 or right ventricular free wall strain (Table 3).We further evaluated the association of CS and mortality, adjusted for left ventricular ejection fraction and left ventricular global longitudinal strain, showing that the results appear to be similar compared to the model without adjustment (Supplementary Table 3).
However, the change in LAV was associated with 1-year all-cause mortality (OR 1.06; CI 1.01-1.12;p ¼ 0.02) (Table 4, Model 1, Figure 1).When adjusted for small differences in postoperative examination time, given that data were sourced from the TVT registry where there is an allowed tolerance window, these findings do not significantly change (Table 4

Early cardiac remodeling in patients with atrial fibrillation/flutter following transcatheter MViV intervention
In a subgroup analysis conducted to evaluate differences in reverse remodeling in patients with atrial fibrillation/flutter compared to those without, 40 patients (65%) who underwent MViV intervention for failed mitral bioprosthetic valves had a history of atrial fibrillation/flutter.
No significant difference in baseline LAV (p ¼ 0.80) or postprocedural LAV (p ¼ 0.55) was observed in patients with atrial fibrillation/flutter compared to those without (Table 5).
We also observed no significant difference in left ventricular chamber dimensions and function and right ventricular function parameters pre and postprocedurally in patients with or without atrial fibrillation/flutter (Table 5).

Impact of MViV Intervention on NYHA Classification Improvement
Among the 57 patients in whom LAVs were measured pre and post MViV intervention, 37 (65%) patients had improvement in NYHA symptom classification at 1-month follow-up visit (Figure 2).

Discussion
The results of this retrospective study of patients who underwent transcatheter MViV intervention for failed bioprosthetic MV confirm the following main findings: (1) Early structural reverse remodeling of the left atrium as assessed by LAV occurs in patients following MViV intervention.(2) Improvement of left atrial CS following MViV also occurs.
(3) Change in left atrial chamber remodeling is associated with allcause mortality at 1 year.MViV intervention which is consistent with findings noted in larger registry data reports. 5,6oprosthetic valve stenosis and regurgitation is often caused by structural deterioration of the valve with associated leaflet thickening and calcification, pannus, or thrombus.
Severe prosthetic stenosis and regurgitation is often accompanied by secondary findings of left atrial dilatation due to increased chamber pressure and volume leading to downstream effects of elevated pulmonary capillary wedge pressures, and elevated right sided pressures that contribute to the development of heart failure symptoms.In this setting, echocardiography plays an important role in evaluating the mechanism and quantifying the severity of valvular dysfunction as described by the American Society of Echocardiography. 23sing echocardiography, our study provides, for the first time, data on cardiac structural and functional remodeling after MViV intervention which may offer more insight into the mechanisms driving the clinical improvements experienced by these patients and offer some understanding into their prognostication.
][29] Our study showed that the improvement in atrial function by strain in the MViV cohort was due to improvements in contractility during the passive filling phase.After MV surgery, Orde et al showed significant early RV dysfunction, likely due to a variety of factors including perioperative hypoperfusion, hypothermia and pericardial opening.In the transcatheter population however, a select group of patients can show acute improvement in RV function after MV repair. 30We show that even after MViV, there is improvement in right ventricular function as captured by FAC, S 0 , TAPSE, and strain in our cohort, likely due to improvement in right ventricular systolic pressure and right-sided remodeling.
In cardiac surgery, left atrial enlargement and left atrial strain have been shown to have significant prognostic implications in patients with severe mitral regurgitation undergoing MV surgery with baseline LAV and left atrial global longitudinal strain noted as independent predictors of postoperative cardiovascular events. 10,31Similarly, a study by Ledwoch et al 32 investigating transcatheter mitral interventions in native valves identified changes in atrial function predict mortality and long-term outcomes.Our study showed no association between baseline LAV or LA strain with 1-year all-cause mortality in patients with failed bioprosthetic valves undergoing transcatheter MViV intervention.Instead, change in LAV was associated with 1-year all-cause mortality.Further research identifying the factors associated with changes in atrial volume and reverse remodeling as MViV interventions increase in volume across the country may provide opportunities to guide clinical decision-making with regards to patient selection and optimal timing for intervention to improve outcomes.Notably, we observed through subgroup analysis lower baseline PALS in patients with atrial fibrillation/flutter compared to those without undergoing MViV intervention and a persistently lower but insignificant postprocedural PALS in the atrial fibrillation group.
These findings are consistent with a previously published pilot study by Albini et al. studying interventions in native valves. 33We hypothesize that significant baseline atrial cardiopathy is reflected in lower preprocedural PALS which could have an impact on reverse remodeling in atrial fibrillation patients after MViV but are limited by our small sample size.This is particularly pertinent in the MViV cohort, who presumably have been experiencing atrial fibrillation and associated fibrosis for a longer period and are more likely to have undergone irreversible deleterious remodeling.Ultimately, we demonstrate that echocardiography provides a readily available tool for the assessment of prosthetic valve dysfunction and the evaluation of early structural and functional remodeling in patients who have undergone MViV intervention.We describe, for the first time, early left atrial and ventricular remodeling after MViV intervention, which may serve as a useful tool in predicting the risk of poor outcomes; however, further investigation into longitudinal changes and their long-term prognostic implication is needed.

Study limitations
The limitations to this study are mostly derived from its retrospective, single-center nature with a relatively small sample size and particularly small number of deaths.Additionally, there was some missing echo data due to postprocedural image degradation related to prosthetic related artifacts.However, this study is the first and largest report to date exploring cardiac remodeling following MViV intervention in patients with bioprosthetic MV failure and its impact on outcomes.

Conclusion
Following MViV intervention, there is significant early improvement in heart failure symptoms.Early reverse remodeling results in reduction in LAV, improvement in left atrial CS and improvement in right ventricle FAC, S 0 , TAPSE, and free wall longitudinal strain.Reverse remodeling following MViV intervention provides more insight into potential mechanisms driving clinical improvement and offer some understanding into prognosis of these patients.

( 4 )
Early RV function improvement as assessed by FAC, S 0 , TAPSE, and strain occurs in patients following MViV.(5) No significant left ventricular remodeling was observed in our study following MViV intervention.(6) Improvement in heart failure symptoms occurs early following

Figure 1 .
Figure 1.Change in left atrial volume by mortality at 1 year in patients undergoing transcatheter mitral valve-in-valve implantation.

Figure 2 .
Figure 2. New York Heart Association (NYHA) functional class at baseline and 1-month among 57 patients undergoing transcatheter mitral valve-in-valve implantation.

Table 1
Baseline characteristics of 62 patients undergoing transcatheter mitral valve-in-valve implantation Abbreviations: IQR, interquartile range; NYHA, New York Heart Association.*Two sample t test, Wilcoxon test, Chi-squared test or Fisher exact test as appropriate.

Table 3
Association of baseline echo parameters and 1-year mortality of 62 patients undergoing transcatheter mitral valve-in-valve implantation * Cox regression modeling hazard ratio of 1-year mortality adjusting for age and gender.

Table 4
Association of change in pre and postprocedural echo parameters and 1-year mortality of 57 patients undergoing transcatheter mitral valve-in-valve implantation * Model 1 describes a Cox regression model of 1-year mortality adjusting for age, gender and baseline echo.Model 2 further adjusts for length of time to follow-up echo in days.
* Two sample t test, or Wilcoxon test as appropriate.