Left and Right Ventricular Hemodynamic Response After Transcatheter Mitral Valve Replacement

Background Transcatheter mitral valve replacement (TMVR) represents a novel treatment option for patients with mitral regurgitation (MR), but little is known about the hemodynamic impact of MR elimination following TMVR. We sought to investigate the hemodynamic impact of TMVR on left ventricular (LV) and right ventricular (RV) function using noninvasive pressure-volume loops. Methods All consecutive patients undergoing TMVR with dedicated devices between May 2016 and August 2022 were enrolled. The end-diastolic and end-systolic pressure-volume relationships were estimated from 26 patients using single-beat echocardiographic measurements at baseline and after TMVR at discharge. RV function was assessed by RV-pulmonary artery (PA) coupling and RV fractional area change. One-year follow-up was available for 19 patients. The prognostic impact of calculated end-diastolic volume at an end-diastolic pressure of 20 mmHg (VPed20) reduction was assessed by Cox regression. Results A total of 26 patients (77.0 years [interquartile range 73.9-80.1], N = 17 [65.4%] male) with successful TMVR were included (secondary MR [N = 21, 80.8%]; median LV ejection fraction was 37.0% [interquartile range 30.7-50.7]). At discharge, a decrease in VPed20 (p < 0.001) indicating leftward shift of end-diastolic pressure-volume relationship, and an increase of the end-systolic elastance slope (p = 0.007) were observed after TMVR. No changes were observed for RV-PA coupling (p = 0.19) and RV fractional area change (p = 0.22). At 1-year follow-up, LV contractility (end-systolic elastance) and RV-PA coupling remained stable. Vped20 reduction at discharge was significantly associated with 1-year all-cause mortality or heart failure hospitalization (hazard ratio 0.16, 95% CI 0.04-0.71, p = 0.016). Conclusions Noninvasive assessment of pressure-volume loops demonstrated early LV reverse remodeling and improved LV contractility, while RV performance was preserved. These results indicate the potential prognostic impact of complete MR elimination after TMVR.

Background: Transcatheter mitral valve replacement (TMVR) represents a novel treatment option for patients with mitral regurgitation (MR), but little is known about the hemodynamic impact of MR elimination following TMVR.We sought to investigate the hemodynamic impact of TMVR on left ventricular (LV) and right ventricular (RV) function using noninvasive pressure-volume loops.Methods: All consecutive patients undergoing TMVR with dedicated devices between May 2016 and August 2022 were enrolled.The end-diastolic and end-systolic pressure-volume relationships were estimated from 26 patients using single-beat echocardiographic measurements at baseline and after TMVR at discharge.RV function was assessed by RV-pulmonary artery (PA) coupling and RV fractional area change.One-year follow-up was available for 19 patients.The prognostic impact of calculated end-diastolic volume at an end-diastolic pressure of 20 mmHg (VPed20) reduction was assessed by Cox regression.Results: A total of 26 patients (77.0 years [interquartile range 73.9-80.1],N ¼ 17 [65.4%]male) with successful TMVR were included (secondary MR [N ¼ 21, 80.8%]; median LV ejection fraction was 37.0% [interquartile range 30.7-50.7]).At discharge, a decrease in VPed20 (p < 0.001) indicating leftward shift of end-diastolic pressure-volume relationship, and an increase of the end-systolic elastance slope (p ¼ 0.007) were observed after TMVR.No changes were observed for RV-PA coupling (p ¼ 0.19) and RV fractional area change (p ¼ 0.22).At 1-year follow-up, LV contractility (end-systolic elastance) and RV-PA coupling remained stable.Vped20 reduction at discharge was significantly associated with 1-year all-cause mortality or heart failure hospitalization (hazard ratio 0.16, 95% CI 0.04-0.71,p ¼ 0.016).Conclusions: Noninvasive assessment of pressure-volume loops demonstrated early LV reverse remodeling and improved LV contractility, while RV performance was preserved.These results indicate the potential prognostic impact of complete MR elimination after TMVR.Clinical Trial Registration: Hamburg TranscathEteR Mitral Valve REplacement RegiStry (HERMES, ClinicalTrials.govIdentifier NCT04914468). 1 Dr S. Ludwig and L. Strotmann contributed equally as first authors.

Introduction
Mitral regurgitation (MR) is the most common valvular heart disease in industrialized countries with an increasing incidence due to an aging society. 1 If left untreated, MR can lead to elevated left ventricular (LV) preload, volume overload, and eventually heart failure (HF) contributing to increased mortality rates. 2 Patients with severe symptomatic MR should be treated surgically (i.e., mitral valve repair or replacement) or by transcatheter edge-to-edge repair (TEER) according to international guidelines. 3,4However, there is a significant subset of patients who are either ineligible for surgery because of comorbidities or advanced age or who present unfavorable anatomy for TEER.For these patients, transcatheter mitral valve replacement (TMVR) represents a potential alternative.Early clinical experience with different TMVR devices has demonstrated effective and durable MR elimination in the vast majority of patients treated. 5,6While residual and recurrent MR occur frequently in patients undergoing TEER, MR elimination is considered the central therapeutic advantage of TMVR over TEER. 7,8[10][11][12] Therefore, we sought to investigate the effect of MR elimination through TMVR on both LV hemodynamic and volumetric changes, as well as right ventricular (RV) function, in patients with severe MR using noninvasive LV pressure-volume loops derived from single-beat echocardiographic measurements.By analyzing LV hemodynamic and volumetric changes, as well as RV function, we aim to gain a better understanding of the implications of MR elimination through TMVR on ventricular performance and prognosis in this patient population.

Study Design
Data for the conduct of this study were derived from the HERMES prospective, observational, single-center registry (Hamburg Trans-cathEteR Mitral Valve Replacement RegiStry; ClinicalTrials.gov:NCT04914468).In total, this registry included 188 patients with clinically significant mitral valve disease undergoing screening for TMVR devices from May 2016 to August 2022, according to an interdisciplinary screening process described before. 13,14All patients were considered to have complex, nonfavorable TEER anatomy and were at high or prohibitive surgical risk.Among these, a total of 47 patients underwent TMVR using dedicated transapical or transfemoral/transseptal devices.Implanted TMVR devices included Tiara [Neovasc], Tendyne [Abbott Vascular], HighLife [HighLife Medical], or CardiAQ [Edwards Lifesciences] valves.Each case was evaluated by an interdisciplinary heart team, and decisions were based on good clinical standards.For this study, 26 consecutive patients undergoing TMVR with available single-beat echocardiographic data to derive noninvasive pressure-volume loops at baseline and discharge were included.Discharge echocardiography was performed at a mean of 14 days after the intervention.One-year follow-up was available in 19 patients and was obtained after 12 AE 4 months.One patient was treated via transfemoral/transseptal approach, and 25 patients were treated via transapical approach (Figure 1).

Echocardiographic Measurements
All echocardiographic measurements were performed by experienced physicians based on current recommendations as part of the standard clinical work-up in the echocardiography laboratory of the structural heart disease department. 15Forward stroke volume (SV for- ward ) was determined from the product of the LV outflow tract diameter and the LV outflow tract velocity time integral (measured by pulsed-wave Doppler).Left ventricular ejection fraction (LVEF) was calculated using the Simpson method, and forward LVEF (LVEF forward ) was measured as SV forward divided by the end-diastolic volume (EDV).Early diastolic tissue velocity was determined by pulsed-wave tissue Doppler imaging, assessing the septal and lateral mitral annular plane velocity and taking the mean of both measurements.Early trans-mitral diastolic peak velocity was measured by pulsed-waved Doppler imaging by placing the sample volume at the tip of the mitral leaflets at baseline and at the valve plane after valve implantation.In patients with atrial fibrillation, sequential measurements were performed and mean values were calculated.
RV function was determined by measurement of tricuspid annular plane systolic excursion (TAPSE), pulmonary artery systolic pressure (PASP), RV-pulmonary artery (RV-PA) coupling, as well as RV fractional area change (RVFAC). 16Based on apical four-chamber views, TAPSE was measured by placing an M-mode line at the lateral tricuspid valve annulus.The tricuspid regurgitant jet was assessed by a continuous-wave Doppler placed to obtain the maximum velocity.Based on the Bernoulli equation, the maximum velocity was transferred into pressure values.The peak value plus the estimated right atrial pressure was taken as the PASP.The RV-PA coupling was defined as the TAPSE/PASP ratio.RVFAC was obtained by tracing the RV endocardial border at end-diastole as well as end-systole in apical four-chamber views.

Noninvasive Hemodynamic Measurements
[19] According to this method, the equation end-diastolic pressure [EDP] ¼ α EDV β describes the EDPVR, with EDP defined as the end-diastolic pressure and EDV as the end-diastolic volume.The variables α and β are calculated as curve-fit parameters.With the assumption of a common underlying shape for volume-normalized EDPVRs, the curve-fit parameters α and β can be obtained for each individual patient from a single EDV/EDP data set. 18EDP was assessed using the equation EDP ¼ 11.96þ (0.596 x early mitral inflow velocity/early diastolic tissue velocity). 20To obtain an estimate of the entire EDPVR and to compare its change after intervention, we used the calculated EDV at an EDP of 20 mmHg, known as VPed20. 18,21or the estimation of ESPVR and end-systolic elastance (Ees), respectively, we used a method assuming a linear relation of ESPVR as published by Chen et al. 19 To describe the Ees noninvasively, we assessed systolic and diastolic brachial artery cuff pressures, forward stroke volume, and an estimated normalized ventricular elastance at arterial end-diastole.To estimate end-diastole, group-averaged values were used and adjusted using ejection fraction, systolic and diastolic arterial blood pressure, pre-ejection time, and total ejection time.Pre-ejection and total ejection time were derived from the LV outflow tract pulsed-wave Doppler.Ees, echocardiography derived end-systolic volume (ESV), and systolic blood cuff pressures were used for calculation of the x-axis intersection of the linear ESPVR (V0).In order to compare the position of the ESPVR with integration of V0 and Ees, we calculated ESV at an end-systolic pressure (ESP) of 120 mmHg, known as the systolic parameter VPes120. 19Ea describes the effective arterial elastance, which is derived from a 3-element Windkessel model described by Kelly et al. 22,23 Ea can be determined approximately by using the arterial ESP and stroke volume. 24The ratio of Ea/Ees indicates the interaction between cardiac contractility and the arterial system. 25Systemic vascular resistance was assessed through pulsed-wave Doppler imaging as the ratio of peak mitral regurgitant velocity and the LV outflow time-velocity integral. 26These methods were then used to compare the hemodynamic status between baseline and discharge, and between discharge and 1-year follow-up (in patients with available data).

Statistical Analysis
Continuous variables are shown as medians (25th percentile and 75th percentile) or as means AE standard deviation.Binary variables are described as counts (frequencies).To assess differences before and after TMVR, paired Student's t-test is applied.A p value of <0.05 was considered statistically significant.Cox regression for the combined 1year endpoints of all-cause mortality or HF hospitalization and cardiovascular mortality or HF hospitalization were performed.Combining several potential risk factors the adjustment model included only Euro-SCORE II due to the small sample size of this study.All analyses were performed with R statistical software version 4.0.3(R Foundation for Statistical Computing, Vienna, Austria).

Clinical and Echocardiographic Baseline Characteristics
A detailed description of the baseline variables is given in Table 1.A total of 26 patients including 17 male and 9 female patients were enrolled in the present study.Median age of the group was 77 years (interquartile range [IQR] 73.9-80.1),with a median body mass index of 27 kg/m 2 (IQR 24.9-30.9).All patients were symptomatic according to the New York Heart Association functional class (New York Heart Association II: N ¼ 2

Noninvasive Pressure-Volume Loops
A detailed description of hemodynamic changes following TMVR (baseline vs. discharge, discharge vs. 1-year follow-up) is given in Table 2. Comparing noninvasive pressure-volume loop results before and after TMVR, a significant decrease of VPed20 (165.3AE 75.8 mL vs. 134.0AE 65.9 mL, p < 0.001) was observed between baseline and discharge, indicating a leftward shift of the EDPVR toward overall lower volumes in most patients (Figure 2).Accordingly, the systolic parameter VPes120 (129.9AE 94.4 mL vs. 91.4AE 84.0 mL, p < 0.001) also decreased.At 1year follow-up, VPed20 showed a statistically not nonsignificant trend toward an increase (142.9AE 65.9 mL vs. 166.6AE 88.1 mL, p ¼ 0.052), whereas VPes120 remained stable (106.0AE 89.1 mL vs. 135.5AE 116.0 mL, p ¼ 0.10) (Supplementary Figure 1).End-systolic elastance (Ees) showed a significant increase at discharge following TMVR (1.4 AE 0.9 vs. 2.5 AE 2.4, p ¼ 0.007) and remained stable between discharge and followup (2.2 AE 2.4 vs. 2.6 AE 3.5, p ¼ 0.60).This reflects an increase in the slope of ESPVR, which is displayed in the pressure-volume diagram (Figure 3).V0 did not change significantly from baseline to discharge (-3.No significant changes in Ea and systemic vascular resistance were observed following TMVR.The Ea/Ees ratio declined from 1.64 at baseline to 0.84 at discharge and remained stable (Ea/Ees 0.85) between discharge and 1year follow-up (Figure 4).

Right Ventricular Hemodynamic Changes
PASP significantly decreased between baseline and discharge (50.5 AE 16.5 mmHg vs. 35.7 AE 10.8 mmHg, p < 0.001) which was maintained from discharge to 1-year follow-up (35.4

Cox Regression
To assess the prognostic impact of hemodynamic changes following TMVR (measured by VPed20), a Cox regression analysis adjusting for EuroSCORE II was performed.The reduction of VPed20 from baseline to discharge was inversely associated with all-cause mortality or HF hospitalization at 1 year (hazard ratio 0.16, 95% CI 0.04-0.71,p ¼ 0.016) as well as with cardiovascular mortality or HF hospitalization at 1 year (hazard ratio 0.18, 95% CI 0.04-0.77,p ¼ 0.021).

Discussion
This is the first study to analyze hemodynamic changes after TMVR using noninvasive pressure-volume loops.In 26 patients with MR treated with dedicated TMVR devices, successful MR elimination resulted in the following changes: reduced end-diastolic and end-systolic LV volumes at discharge, improved LV contractility, improved LVEF forward , improved energetic efficiency (conceptually depicted by Ea/Ees ratio), increased EDP, and preserved RV performance (Graphical Abstract).In patients with available 1-year follow-up data, LV volumes increased, but LV contractility and EDP remained stable.Reduction of VPed20 at discharge was associated with beneficial clinical outcomes at 1 year.

LV Reverse Remodeling After TEER
Currently, available data on the immediate hemodynamic impact of TMVR are scarce.8][29] Evidence suggests that MR reduction by TEER leads to early LV reverse remodeling with acute reductions in end-systolic and end-diastolic volumes. 17Additionally, other studies have demonstrated that a sustainable reduction in LV volumes can be achieved in case of durable resolution of MR.Even in patients with only mild residual MR at 12 months after TEER, the LV volume reduction was still significant. 30,31In contrast, the COAPT trial could not document a sustained reduction in LV volumes after TEER.Nevertheless, the progression of LV remodeling was mitigated after TEER compared to the medical control group. 28In a study by Schrage et al. that utilized a similar noninvasive approach as the present study, patients undergoing TEER showed postprocedural improvement in hemodynamic parameters such as VPed20.However, this beneficial impact was only observed for patients with preserved LVEF. 173][34] A similar impact of TMVR on LV remodeling at 1-month follow-up has been described by Fukui et al. 35 using cardiac computed tomography scans.In terms of EDV, these changes seem to occur only as acute postprocedural changes detectable at discharge, whereas this effect was reversed in patients with a 1-year follow-up.Of note, when comparing the hemodynamic impact at discharge and follow-up, an incomplete 1-year follow-up must be taken into consideration.In our patient cohort, the decrease in LV volumes immediately after TMVR was paralleled by a leftward shift of the LV pressure-volume relationship, which was valid for both ESPVR and EDPVR, respectively.This was documented by a significant overall decrease in the VPed20 and VPes120.While VPed20 reduction hints at the restoration of hemodynamic LV diastolic function, a decrease in VPes120 is considered to correlate with improvements in LV contractility.This is confirmed by an increase in LVEF forward and, more importantly, by an increase in the Ees slope after TMVR.In addition, we observed energetically more efficient LV performance after TMVR, as indicated by a reduced Ea/Ees ratio.It could be shown that the ratio of LV stroke work and potential energy were optimized at Ea/Ees ratios between 0.3 and 1.3, which corresponds to enhanced Ea/Ees ratios in our study cohort indicating an improvement of ventriculo-arterial coupling. 24,25s mentioned before, the results show an apparent increase in contractility despite lower LV preload, which does not seem consistent with Frank-Starling Law.As per Frank-Starling Law, contractility increases with preload up to a certain point referred to as the descending limb. 36Better contractility in the light of less preload could mean that (1) the ventricles operated on the descending limb before TMVR and were optimized after, or (2) more complex LV geometry changes caused by the intervention lead to a better contractile function, resulting in a different Frank-Starling curve then.
Despite improvements in hemodynamic parameters, we observed a nonsignificant decrease of global LVEF.This is a well-known phenomenon following MR treatment, which has been described for both TEER and TMVR 28,30,32,37 and may be explained by overestimation of LVEF in the presence of MR 38 as well as some degree of afterload mismatch after eradication of the regurgitant volume following a sudden increase of end-diastolic filling pressures. 39Dupuis et al. 40 were able to show that LVEF forward is superior to global LVEF in the assessment of postoperative LV systolic function in patients undergoing mitral valve surgery.Our study confirms increased filling pressures by MR resolution with an increased EDP following TMVR.Increased LVEF forward further illustrates the fact that proportionally more blood ejection occurs toward the high-pressure circulatory system, leading to increased afterload.Nevertheless, complete restoration of LV function, including diastolic function, would include normalization of EDP, which we could not observe in 1-year follow-up data.This finding requires confirmation by larger prospective studies.
The early hemodynamic changes described in this study are unique to TMVR and have not been reported in similar studies including TEER patients.Complete elimination of MR, as opposed to some degree of residual MR, could be a plausible explanation for this finding and may represent an advantage of TMVR over TEER with potentially prognostic implications.
In general, the beneficial impact of LV reverse remodeling on outcomes after MR treatment has been well described. 30,41,42Although the sample size of our study was small, our data suggest a significant prognostic impact of early VPed20 reduction on clinical outcomes.Therefore, early LV reverse remodeling could have prognostic implications despite attenuation at follow-up.

Right Ventricular Changes After TMVR
The impact of TMVR on PA pressures has been consistently described by several studies involving TMVR devices. 33,37Severe MR is frequently associated with some degree of postcapillary pulmonary hypertension, which results from the backward transmission of increased left atrial pressures. 43Hence, by eliminating MR, TMVR generally results in effective recovery of PA pressures.Our study supports previous findings that TMVR significantly reduces PASP.However, in our patient population, the decrease in PASP was accompanied by a decline in TAPSE, which might indicate a decline in RV contractility.In line with an early decrease in TAPSE, we observed a significant decrease in RVFAC from discharge to 1-year follow-up.Reduced RV function following mitral valve intervention might be attributed to geometric changes in RV contraction, which may result from the disruption of pericardial integrity following transapical access. 446][47] Balancing the reduction of both PASP and TAPSE, RV-PA coupling remained stable after TMVR.Given the fact that TAPSE should generally be interpreted in the light of PASP, RV-PA coupling is considered superior to TAPSE alone for defining hemodynamic RV function.Therefore, we conclude that RV function after TMVR was indeed preserved.Nevertheless, the absence of recovery of TAPSE and a decrease in RVFAC at follow-up warrant further investigation of RV performance following TMVR.

Study Limitations
This study has several limitations that should be acknowledged.First, the absence of a medical or TEER control group limits our ability to assess the natural course of the disease and draw conclusions regarding differences compared to TEER.Second, the retrospective nature of this study warrants verification of the results in prospective studies.Third, the accuracy of echocardiography measurements partly relies on approximation formulas, where small measurement differences may have a significant impact on the overall data.
Especially, parameters calculated with measurements regarding the LV outflow tract are typically error prone.Additionally, assessment of both diastolic and systolic function is flawed when considering valve replacement and transapical access.In particular, the assessment of ejection time in the presence (baseline) or absence of MR (following TMVR) remains a potential limitation of the noninvasive approach used in this study.Furthermore, SV forward was used for calculation of Ees at baseline and discharge, thus not considering additional ejection to the left atrium at baseline.However, we are convinced of the reliability of our results given the somewhat expected hemodynamic changes following MR elimination as demonstrated by noninvasive measurements. 31Nevertheless, these results should be validated by studies utilizing invasive hemodynamic measurements.Further studies including a larger patient population with longer echocardiographic follow-up time, are necessary to assess the durability of the short-term hemodynamic and volumetric changes we observed.

Conclusion
In this single-center study, noninvasive assessment of pressurevolume loops before and after TMVR demonstrated acute LV reverse remodeling and improved LV contractility, while RV performance was preserved.At 1-year follow-up, the LV reverse remodeling was discharge and 1-year follow-up.The PVA can be divided into two parts: 1.The stroke work, which includes the area within the pressure-volume loop (orange).2. The potential energy, which describes the triangular area formed by the ESPVR and EDPVR slopes (yellow) Abbreviations: 1-year FU, 1-year follow-up; Ea, effective arterial elastance; EDPVR, end-diastolic pressure-volume relationship; Ees, end-systolic elastance; ESPVR, endsystolic pressure-volume relationship; PVA, pressurevolume area.
attenuated, but LV contractility remained improved.Considering available data with other treatment options, these findings suggest that complete MR elimination by TMVR may have a greater impact on LV function compared to other MR therapies.In addition, complete abolishment of MR may still be compensated by the ventricle even in the presence of LV dysfunction, irrespective of transapical access.However, it remains to be determined whether the potential benefit of complete MR elimination by TMVR on LV function and reverse remodeling can translate into favorable long-term functional and clinical outcomes.

Impact on Daily Practice
This study is the first to assess the hemodynamic changes in patients undergoing TMVR by deriving noninvasive assessment of pressurevolume loops.Our results highlight the potential beneficial impact of complete MR elimination on LV contractility and support the need for further investigation of hemodynamics after TMVR.With the transition from transapical to transfemoral/transseptal access in TMVR, procedural safety is likely to improve significantly.As procedural risk improves and TMVR becomes available to a broader patient population, the distinct LV and RV hemodynamic consequences of complete MR elimination and TMVR become more important.

Figure 2 .
Figure 2. End-diastolic pressure-volume relationship following TMVR.Patients are compared based on their VPed20 as a marker of the EDPVR.The yaxis displays the change in VPed20 from baseline to discharge, and the x-axis displays the value of VPed20 at baseline Abbreviations: EDPVR, end-diastolic pressure-volume relationship; TMVR, transcatheter mitral valve replacement.;VPed20, calculated end-diastolic volume at an end-diastolic pressure of 20 mm Hg.

Figure 3 .
Figure 3. Schematic left ventricular pressure-volume visualization between (a) baseline and discharge and (b) discharge and 1-year follow-up.The ESPVR and EDPVR both shift leftward from baseline to follow-up and remain statistically unchanged from discharge to 1-year follow-up, despite an overall increase in LV volumes.VPed20 indicates an end-diastolic volume at a pressure of 20 mmHg, VPes120 indicates an end-systolic volume at a pressure of 120 mmHg, respectively.Ees indicates the slope of ESPVR Abbreviations: 1-year FU, 1-year follow-up; EDPVR, enddiastolic pressure-volume relationship; Ees, end-systolic elastance;ESPVR, end-systolic pressure-volume relationship.

Figure 4 .
Figure 4.The PVA of a single beat at (a) baseline and (b) discharge and 1-year follow-up.The PVA can be divided into two parts: 1.The stroke work, which includes the area within the pressure-volume loop (orange).2. The potential energy, which describes the triangular area formed by the ESPVR and EDPVR slopes (yellow) Abbreviations: 1-year FU, 1-year follow-up; Ea, effective arterial elastance; EDPVR, end-diastolic pressure-volume relationship; Ees, end-systolic elastance; ESPVR, endsystolic pressure-volume relationship; PVA, pressurevolume area.