Usefulness of the Slope of the Preload Recruitable Stroke Work Relationship for the Prediction of Rehospitalization After Transcatheter Mitral Edge-to-Edge Repair

Abstract

Background: Among patients with functional mitral regurgitation (FMR), responders to transcatheter mitral edge-to-edge repair (TEER) remain unclear. We investigated whether the slope of the preload recruitable stroke work relationship (M_w; calculated as stroke work / [EDV − k × EDV + {1 − k} × LV wall], where EDV is end-diastolic volume, k is a constant, and LV wall is the volume of the left ventricular wall) could predict rehospitalization in FMR patients after TEER.

Methods and Results: M_w was calculated for 24 FMR patients using echocardiography. The median left ventricular ejection fraction was 27% and the median M_w was 32. Over a 498-day median follow-up period, 38% of patients were rehospitalized for heart failure, and only M_w had a high area under the curve in time-dependent receiver operating characteristic analysis.

Conclusions: M_w is an effective predictor for rehospitalization in FMR patients after TEER.

For surgical repair of the mitral valve, transcatheter mitral edge-to-edge repair (TEER) has recently emerged as a new therapeutic approach for both functional (FMR) and degenerative (DMR) mitral regurgitation. However, how to identify FMR patients who are likely to respond well to TEER remains uncertain. Percutaneous Repair with MitraClip Device for Severe Functional Mitral Regurgitation (MITRA-FR) trial, a large randomized trial, did not confirm the benefits of TEER up to 2 years after treatment.¹ Conversely, the Cardiovascular Outcomes Assessment of the MitralClip Percutaneous Therapy for Heart Failure Patients with Functional Mitral Regurgitation (COAPT) trial showed significantly better outcomes, including reduced rehospitalization for heart failure, in the TEER group compared with the group receiving medication alone.² The differences between these 2 studies have been discussed in detail in a joint position statement issued by several European heart/cardiovascular societies.³ The ongoing Randomized Study of the MitraClip Device in Heart Failure Patients with Clinically Significant Functional Mitral Regurgitation (RESHAPE-HF2) trial, which has a much lower expected crossover rate than the former trials, will provide further evidence on the efficacy of TEER.⁴

In a reanalysis of the COAPT trial, the rates of rehospitalization for heart failure following TEER treatment were comparable between patients with a baseline left ventricular ejection fraction (LVEF) ≤40% and >40%.⁵ Thus far, patient suitability for TEER cannot be adequately determined using conventional parameters, such as LVEF, and there is an urgent need for a new tool that can be easily assessed through routine noninvasive examinations. In a previous study we showed that the slope of the preload recruitable stroke work (PRSW) relationship (M_w), which can be estimated noninvasively by echocardiography, can predict outcomes after surgical mitral valve interventions in FMR patients.⁶ However, it has not been established whether M_w is also useful in determining outcomes after TEER treatment. The aim of this study was to explore the usefulness of M_w in predicting rehospitalizations due to heart failure after TEER treatment.

Methods

Study Design and Setting

This retrospective study included consecutive 24 patients with moderately severe or severe FMR and low LVEF who were treated with the MitraClip^® (Abbott) between December 2018 and December 2022 at Hokkaido University Hospital. Patients with a preoperative LVEF >40% were excluded to enable us to examine heart failure patients with reduced LVEF.⁷

The primary endpoint of the study was the time until rehospitalization due to heart failure after the MitraClip^® intervention.

The study protocol was approved by the Hokkaido University Hospital Clinical Research Review Board Office (No. 022-0266; date of approval: March 14, 2023). Due to the retrospective nature of the study, the requirement for written informed consent was waived. The study was conducted in accordance with the Declaration of Helsinki.

Assessment of Cardiac Function

Transthoracic echocardiography was performed a median of 12 days (interquartile range [IQR] 6–24 days) before TEER, with the following measurements record: left ventricular (LV) end-diastolic and systolic dimensions (LVDd and LVDs, respectively), LV posterior wall thickness at end-diastole (LVPWTd), interventricular septal thickness at end-diastole (IVSTd), end-diastolic and systolic volume indices (EDVI and ESVI, respectively), and LVEF. Systolic pulmonary artery pressure was estimated through analysis of regurgitant flow at the tricuspid valve. E/e′ was assessed by combining transmitral inflow velocities (E wave) and mitral annular motion (e′ wave) using tissue Doppler imaging. The mean of the lateral and septal e′ values was used for the calculation. Post-TEER echocardiography was performed prior to discharge.

M_w is recognized as a load-insensitive index of contractile function and was estimated in this study using a single-beat technique, with some modifications, as reported by Lee et al.⁸ as follows:

  

$${\text{M}}_{\text{w}}\left(\text{erg}\cdot {\text{cm}}^{-3}\cdot {10}^3\right)=\frac{Totalstrokework}{[EDV-k\times EDV+(1-k)\times LVwall]}$$

where EDV is end-diastolic volume and was estimated by the modified Simpson’s method. LV wall refers to the volume of the LV wall and was estimated from echocardiography-derived LV mass using the following formula:⁹

  

$$\text{LV mass}=1.1\times ({\left[LVDd+LVPWTd+IVSTd\right]}^3-LVD{d}^3)$$

The constant k was determined as follows:

  

$$k=0.0004\times \text{LV mass}+0.6408$$

Stroke work (mL·mmHg) in the heart was calculated by multiplying total stroke volume by mean arterial blood pressure (mABP). mABP was calculated as follows:

  

$$\text{mABP}=\left(SBP+2\times DBP\right)/3$$

where SBP and DBP are systolic and diastolic blood pressure, respectively, and were measured using a sphygmomanometer at the time of echocardiography. Total stroke volume (mL) was calculated by subtracting ESV from EDV. This single-beat method for determining M_w has been reported to correlate well with the invasive catheter method across different LV sizes, LV masses, and the presence of regional wall motion abnormalities.⁸ For patients with atrial fibrillation, M_w was measured by assessment of a single representative beat.

During the index hospitalization for TEER, right heart catheterization was performed to measure mean pulmonary artery pressure (mPAP), pulmonary capillary wedge pressure (PCWP), and the cardiac index. In some patients, right heart catheterization was repeated before discharge.

Statistical Analysis

All values are presented as the median with (IQR), and were compared using the Mann-Whitney U test. The Hodges-Lehman method was used to calculate 95% confidence intervals (CIs) of the median difference. Areas under the curve (AUCs) were derived from receiver operating characteristic (ROC) curves from logistic regression models, and were compared using Delong’s test. The patients were divided into 2 groups by Youden index in ROC analysis. Rehospitalization rates were estimated using Kaplan-Meier curves and compared by the log rank test. To evaluate the predictive performance of LVEF and M_w before TEER, we performed a comparative analysis using AUCs with time-dependent ROC curves up to 3 years. A bootstrap method with 1,000 iterations was used to estimate 95% CIs for differences in AUCs between LVEF and M_w.

Data were analyzed using SPSS (version 26; IBM) and R (version 4.4.0; R Foundation for Statistical Computing), with the “timeROC” package used for time-dependent ROC analysis.

Results

Patient Characteristics

Patient characteristics are presented in Table 1. The study population have various underlying etiologies, including dilated (46%) and ischemic (38%) cardiomyopathies. Three patients with persistent atrial fibrillation were diagnosed with atrial FMR, defined as mitral annular enlargement associated with left atrial dilatation, resulting in poor coaptation of the mitral leaflets.¹⁰ Most patients were classified as New York Heart Association Class III or higher, and the median N-terminal pro B-type natriuretic peptide (NT-proBNP) concentration was 2,539 pg/mL.

Table 1.

Patient Characteristics Before TEER

Age (years)	73 [66–79]
Male sex	14 (58)
Comorbidities
Hypertension	9 (38)
Diabetes	6 (25)
Dyslipidemia	6 (25)
COPD	0 (0)
CRT implantation	7 (29)
Etiology
Non-ischemic dilated cardiomyopathy	11 (46)
Ischemic dilated cardiomyopathy	9 (38)
Atrial functional mitral regurgitation	3 (13)
Drug-induced cardiomyopathy	1 (4)
NYHA functional class
II	6 (25)
III	12 (50)
IV	6 (25)
Serum NT-proBNP (pg/mL)	2,539 [1,785–4,981]
eGFR (mL/min/1.73 m2)	33 [22–43]
SBP (mmHg)	103 [92–118]
DBP (mmHg)	63 [50–71]
Medications
ACEi/ARB	23 (96)
β-blocker	23 (96)
SGLT2i	7 (29)
Catecholamine use during hospitalization	7 (29)

Data are given as the median [interquartile range] or n (%). ACEi, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; BNP, B-type natriuretic peptide; COPD, chronic obstructive pulmonary disease; CRT, cardiac resynchronization therapy; DBP, diastolic blood pressure; eGFR, estimated glomerular filtration rate; NYHA, New York Heart Association; SBP, systolic blood pressure; SGLT2i, sodium-glucose cotransporter 2 inhibitors; TEER, transcatheter mitral edge-to-edge repair.

Cardiac Function Before TEER

Cardiac functional parameters are presented in Table 2. The median LVEF was 27%, and the median M_w was 32 erg·cm⁻³·10³. EDVI was enlarged, with a median value of 107 mL/m².

Table 2.

Cardiac Functional Parameters Before and After TEER

	Before TEER	After TEER
Transthoracic echocardiography
LVDd (mm)	67 [62–70]	66 [60–71]
LVDs (mm)	57 [53–64]	57 [52–66]
LVPWTd (mm)	8.0 [7.0–8.0]	7.5 [7.0–8.0]
IVSTd (mm)	8.0 [8.0–9.0]	8.0 [7.3–9.0]
EDVI (mL/m2)	107 [93–141]	112 [90–157]
ESVI (mL/m2)	76 [62–107]	84 [63–119]
LVEF (%)	27 [21–34]	25 [21–32]
E/e′	14 [10–20]	19 [15–28]
Estimated sPAP	33 [27–48]	31 [26–41]
TR > moderate	11 (46)	9 (38)
LVSW (×103 mmHg·mL)	3.3 [2.7–4.7]	3.3 [2.5–4.5]
LVSWI (×103 mmHg·mL/m2)	2.1 [1.8–3.0]	2.1 [1.8–2.6]
Mw (erg·cm−3·103)	32 [26–44]	32 [25–35]
Right heart catheterization
Mean PAP (mmHg)	23 [20–37]	19 [16–26] (n=15)
PCWP (mmHg)	15 [10–27]	13 [10–16] (n=15)
Cardiac index (L/min/m2)	2.1 [1.8–2.5]	2.1 [1.8–2.5] (n=15)

Data are presented as the median [interquartile range] or n (%). EDVI, end-diastolic volume index; ESVI, end-systolic volume index; IVSTd, interventricular septal thickness at end-diastole; LVDd, left ventricular end-diastolic dimension; LVDs, left ventricular end-systolic dimension; LVEF, left ventricular ejection fraction; LVPWTd, left ventricular posterior wall thickness at end-diastole; LVSW, left ventricular stroke work; LVSWI, LVSW index; M_w, slope of the preload recruitable stroke work relationship; PAP, pulmonary artery pressure; PCWP, pulmonary capillary wedge pressure; sPAP, systolic pulmonary artery pressure; TEER, transcatheter mitral edge-to-edge repair; TR, tricuspid regurgitation.

Rehospitalization Due to Heart Failure

TEER was successful according to Mitral Valve Academic Research Consortium Part 2 definitions, and the reduction in mitral regurgitation was “optimal” in 20 (83%) patients and “acceptable” in 4 (17%).¹¹ Throughout the 498-day median follow-up period, 38% (9/24) of patients were rehospitalized due to heart failure. Table 3 presents results of univariate analysis of predictive factors for rehospitalization due to HF. M_w values were significantly lower in rehospitalized individuals compared to those without rehospitalization (median [IQR] 28 [23–31] vs. 41 [30–44] erg·cm⁻³·10³); however, LVEF did not differ between those who were rehospitalized and those who were not (median [IQR] 30% [24–34%] vs. 24% [21–33%], respectively). LV stroke work was lower in patients rehospitalized for heart failure than in those who were not. NT-proBNP, LVDd, LVDs, EDVI, and ESVI were comparable between the rehospitalization and non-rehospitalization groups. Of the parameters assessed by right heart catheterization, mPAP and PCWP were significantly higher in the rehospitalization than non-rehospitalization group, whereas there was no significant difference in cardiac index between the 2 groups. There was no correlation between M_w and any of the preoperative parameters except for mPAP and PCWP. The 3 patients with atrial FMR experienced no events during the follow-up period.

Table 3.

Univariate Analysis of Factors Predictive for Heart Failure Rehospitalization

	95% CI of median difference	P value
Pre-TEER transthoracic echocardiography
LVDd	−1.0, 14.0	0.12
LVDs	−4.0, 12.0	0.52
LVPWTd	−1.0, 1.0	0.77
IVSTd	−1.0, 1.0	0.52
EDVI	−18, 46	0.45
ESVI	−16, 39	0.45
LVEF	−9.0, 2.0	0.24
E/e′	−4.5, 8.1	0.76
Estimated sPAP	−5.0, 19.0	0.25
TR > moderate	NA	0.67
LVSW	224, 2,176	0.010
LVSWI	−119, 1,251	0.16
Mw	2.0, 18.0	0.012
Pre-TEER right heart catheterization
Mean PAP	−28.0, −4.0	0.002
PCWP	−23.0, −3.0	0.004
Cardiac index	−0.39, 0.68	0.52

CI, confidence interval. Other abbreviations as in Table 2.

The AUC of M_w for the prediction of rehospitalization was 0.806 (95% CI 0.63–0.98; P=0.015). There was no significant difference between M_w and LVEF for the prediction of rehospitalization (95% CI −0.20, 0.51; P=0.39). However, patients with an M_w <40 erg·cm⁻³·10³ had significantly higher rehospitalization rates (P=0.022), unlike those categorized using a cut-off value of LVEF 25% (P=0.27; Figure A,B). None of the patients with a pre-TEER M_w of ≥40 erg·cm⁻³·10³ were rehospitalized due to heart failure. Furthermore, time-dependent ROC analysis revealed that the AUCs for M_w and LVEF at 365 days were 0.72 and 0.37, respectively, indicating that M_w was a superior predictor of rehospitalization due to heart failure (95% CI −0.66, −0.041). No significant differences between M_w and LVEF were observed at 730 days (95% CI −0.36, 0.23) or 1,095 days (95% CI −0.48, 0.00; Figure C).

Figure.

Estimation of rehospitalization and time-dependent receiver operating characteristic (ROC) analysis. (A,B) Rehospitalization rates for heart failure, based on cut-off values derived from the Youden index, for left ventricular ejection fraction (LVEF; A) and the slope of the preload recruitable stroke work relationship (M_w; B). (C) Time-dependent ROC analysis. AUC, area under the curve; pts, patients.

Discussion

We have demonstrated that M_w predicts heart failure rehospitalization in FMR patients who undergo TEER, particularly when M_w is <40 erg·cm⁻³·10³. In contrast, LVEF did not predict heart failure rehospitalization in these patients. M_w shows superior predictive performance to LVEF, especially in the early postoperative period.

Heart Failure Rehospitalization After TEER

The major benefit of TEER is reportedly the prevention of heart failure in patients with FMR.^1,2 However, rehospitalizations after TEER are not uncommon, with a rate of 35% at 24 months after treatment in the COAPT trial, similar to the rate in our study (38%). Although 46% of the patients in our study had a “COAPT-like” profile,¹² the EDVI was similar between our study and the COAPT trial (107 and 101 mL/m², respectively). Although there is an urgent need for a tool to predict which patients will respond to TEER treatment, as yet no reliable cardiac functional parameters have been identified.

Prediction of Outcomes After TEER

The COAPT trial demonstrated more favorable outcomes than those observed in the MITRA-FR trial, and the proportionate–disproportionate FMR hypothesis became a topic of debate.^1,2 However, using the effective regurgitant orifice area and EDV, Lindenfeld et al. suggested that this hypothesis may be insufficient to explain the divergent results between the MITRA-FR and COAPT trials.¹³ Although LVEF remains the gold standard for assessing cardiac function, its utility in patients with reduced LVEF and FMR is contentious. In a subanalysis of the COAPT trial, the 2-year rehospitalization rates for patients with severe (LVEF ≤30%) and moderate (LVEF 30–40%) LV dysfunction were 39% and 28%, respectively, but no statistical analysis was performed between the 2 groups.⁵ In patients with diminished cardiac function, a more precise assessment of intrinsic cardiac power is essential.

In the present study, the rehospitalization rates were 55% for patients with lower (<40) M_w and 0% for those with higher (≥40) M_w, with no similar trend seen for LVEF. Previously, we reported that M_w, but not LVEF, can predict outcomes following surgical mitral valve procedures for FMR.⁶ Unlike LVEF, which does not account for LV pressure, M_w considers stroke work, encompassing both pressure and volume changes. Kashiyama et al. reported that the stroke work index was the sole significant predictor of composite outcomes, including all-cause mortality, admission for heart failure, and LV assist device implantation, after mitral valve surgery.¹⁴ That study strongly supports the findings of the present study, because stroke work is a major component in the calculation of M_w; stroke work also predicted rehospitalization in our study (Table 3). The COAPT risk score¹⁵ did not predict rehospitalization in our cohort (data not shown).

The estimated PA and E/e′ were not factors predictive of rehospitalization due to heart failure (Table 3). Tricuspid regurgitation was less than mild in half the patients, which may have led to an underestimation of pressure gradients and estimated pulmonary artery pressure. E/e′ reflects LV filling pressure. LV stroke work or M_w may be more useful for assessing overall cardiac function than a single pressure estimate.

Study Limitations

This study has several limitations. First, the number of cases was small, so a multivariate analysis for predicting rehospitalization was not possible. Second, we have modified the single-beat method for M_w as proposed by Lee et al.⁹ in which stroke volume is derived from pulse Doppler measurements at the LV outflow tract. Based on the concept that stroke work in FMR should be calculated using total stroke volume, not merely LV outflow,¹⁶ we used stroke volume calculated as EDV–ESV. Further studies are needed to confirm whether this modification is correlated with M_w derived from LV catheterization. M_w should be cautiously used in patients with mitral regurgitation until the single-beat method is validated through invasive assessment.

Conclusions

M_w may be a more reliable predictor of heart failure rehospitalization than LVEF in patients with FMR who undergo TEER. This is particularly the case when M_w is <40 erg·cm⁻³·10³. Given that M_w can be assessed non-invasively, it could serve as an effective measure for predicting rehospitalization in this patient group. Further studies involving more subjects and extended follow-up periods are necessary to standardize M_w as a tool for functional assessment in FMR patients.

Acknowledgment

During the preparation of this paper, the authors used ChatGPT for language editing.

Sources of Funding

This study did not receive any specific funding.

Disclosures

T.A. has received a clinical research grant from Abbott Medical Japan LLC. T.A. and T.N. are members of the Circulation Journal’s Editorial Team. The remaining authors have no conflicts of interest to declare.

IRB Information

This study was approved by the Hokkaido University Hospital Clinical Research Review Board Office (No. 022-0266; date of approval: March 14, 2023).

Data Availability

The data supporting the findings of this study are available from the corresponding author upon reasonable request. Specifically, preprocedural data will be shared via email for up to 5 years after publication. Correlation analysis can be conducted using the outcome data. The data will be shared with any physician involved in mitral valve therapy who requests data sharing.

References

• 1.   Obadia JF, Messika-Zeitoun D, Leurent G, Iung B, Bonnet G, Piriou N, et al. Percutaneous repair or medical treatment for secondary mitral regurgitation. N Engl J Med 2018; 379: 2297–2306.
• 2.   Mack MJ, Lindenfeld J, Abraham WT, Kar S, Lim DS, Mishell JM, et al. 3-Year outcomes of transcatheter mitral valve repair in patients with heart failure. J Am Coll Cardiol 2021; 77: 1029–1040.
• 3.   Coats AJS, Anker SD, Baumbach A, Alfieri O, von Bardeleben RS, Bauersachs J, et al. The management of secondary mitral regurgitation in patients with heart failure: A joint position statement from the Heart Failure Association (HFA), European Association of Cardiovascular Imaging (EACVI), European Heart Rhythm Association (EHRA), and European Association of Percutaneous Cardiovascular Interventions (EAPCI) of the ESC. Eur Heart J 2021; 42: 1254–1269.
• 4.   Anker SD, Friede T, von Bardeleben RS, Butler J, Fatima K, Diek M, et al. Randomized investigation of the MitraClip device in heart failure: Design and rationale of the RESHAPE-HF2 trial design. Eur J Heart Fail 2024; 26: 984–993.
• 5.   Lerakis S, Kini AS, Asch FM, Kar S, Lim DS, Mishell JM, et al. Outcomes of transcatheter mitral valve repair for secondary mitral regurgitation by severity of left ventricular dysfunction. EuroIntervention 2021; 17: e335–e342.
• 6.   Shingu Y, Kubota S, Wakasa S, Ooka T, Kato H, Tachibana T, et al. Slope in preload recruitable stroke work relationship predicts survival after left ventriculoplasty and mitral repair in patients with idiopathic cardiomyopathy. J Cardiol 2015; 65: 157–163.
• 7.   Heidenreich PA, Bozkurt B, Aguilar D, Allen LA, Byun JJ, Colvin MM, et al. 2022 AHA/ACC/HFSA guideline for the management of heart failure: A report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation 2022; 145: e895–e1032.
• 8.   Lee WS, Huang WP, Yu WC, Chiou KR, Ding PY, Chen CH. Estimation of preload recruitable stroke work relationship by a single-beat technique in humans. Am J Physiol Heart Circ Physiol 2003; 284: H744–H750.
• 9.   Devereux RB, Reichek N. Echocardiographic determination of left ventricular mass in man: Anatomic validation of the method. Circulation 1977; 55: 613–618.
• 10.   Zoghbi WA, Levine RA, Flachskampf F, Grayburn P, Gillam L, Leipsic J, et al. Atrial functional mitral regurgitation: A JACC: Cardiovascular Imaging expert panel viewpoint. JACC Cardiovasc Imaging 2022; 15: 1870–1882.
• 11.   Stone GW, Adams DH, Abraham WT, Kappetein AP, Genereux P, Vranckx P, et al. Clinical trial design principles and endpoint definitions for transcatheter mitral valve repair and replacement: Part 2: Endpoint definitions: A consensus document from the Mitral Valve Academic Research Consortium. J Am Coll Cardiol 2015; 66: 308–321.
• 12.   Adamo M, Rubbio AP, Zaccone G, Pighi M, Massussi M, Tomasoni D, et al. Prediction of mortality and heart failure hospitalisations in patients undergoing M-TEER: External validation of the COAPT risk score. EuroIntervention 2023; 18: 1408–1417.
• 13.   Lindenfeld J, Abraham WT, Grayburn PA, Kar S, Asch FM, Lim DS, et al. Association of effective regurgitation orifice area to left ventricular end-diastolic volume ratio with transcatheter mitral valve repair outcomes: A secondary analysis of the COAPT trial. JAMA Cardiol 2021; 6: 427–436.
• 14.   Kashiyama N, Toda K, Miyagawa S, Yoshikawa Y, Hata H, Yoshioka D, et al. Left ventricular stroke work index associated with outcome after mitral valve surgery for functional regurgitation in nonischemic dilated cardiomyopathy. Semin Thorac Cardiovasc Surg 2020; 32: 698–709.
• 15.   Shah N, Madhavan MV, Gray WA, Brener SJ, Ahmad Y, Lindenfeld J, et al. Prediction of death or HF hospitalization in patients with severe FMR: The COAPT risk score. JACC Cardiovasc Interv 2022; 15: 1893–1905.
• 16.   Shingu Y, Matsui Y. Did we misunderstand how to calculate total stroke work in mitral regurgitation by echocardiography? Circ J 2012; 76: 1533–1534.