Letting Go of Restraint: Tricuspid Valve Intervention in Heart Failure with Preserved Ejection Fraction

‘The Action is in the Interaction’ – Douglas Conant. Severe tricuspid valve regurgitation (TR) is identified during routine echocardiography with a prevalence ranging between 0.9% and 2.7%.1-3 Yet, newly diagnosed severe TR has exceedingly high 1-year risks for death (40%) and heart failure (HF) hospitalization or death (54%).3, 4 Not only is worse TR severity associated with poor prognosis, but its progression from mild to severe TR is associated with worst short-term outcomes.5 Despite the increasing recognition of this adverse valvular condition, few medical treatments exist, and surgical intervention for TR is infrequently performed with less than 5% over 5 years following diagnosis.3, 6 This observation is likely attributed to wide variations surgical practices, patient selection, and high-risk comorbidities.7 More recently, tricuspid transcatheter edge-to-edge repair (T-TEER) has shown promise in reducing TR severity and improving quality of life among patients with prohibitive surgical risk,8 although trial populations to date have included modest risk groups compared with real-world patients with severe TR.3, 4, 8 Recent data from a large, well-characterized population with serial echocardiography demonstrated remarkable heterogeneity among patients with newly diagnosed severe TR.3 Adverse phenotypic characteristics associated with nearly 80% 1-year risk of death included presence of severe mitral valve regurgitation, reduced left ventricular (LV) ejection fraction, liver dysfunction, and renal dysfunction.3 Phenotypic characteristics associated with better prognosis included HF with preserved ejection fraction (HFpEF) and absence of comorbid mitral valve disease — a population with potential to exhibit favourable therapeutic response to T-TEER. However, the question remains whether correction of TR (and resultant increase in forward flow and LV preload) may worsen HFpEF, a condition characterized by impaired diastolic function and exaggerated LV pressures to changes in end-diastolic volumes. While large outcomes data in patients with severe TR and HFpEF are lacking, characterizing the mechanistic responses of T-TEER in this subgroup may provide additional insight in determining whom T-TEER may best serve as well as suitable populations for larger trial investigation. In this issue of the Journal, Kresoja et al.9 examined physiologic mechanistic response to reduction in TR severity following T-TEER among patients with severe TR and HFpEF. The authors used single-beat LV pressure–volume analysis, cardiac magnetic resonance (CMR) imaging, and two-dimensional transthoracic echocardiography (TTE) to characterize right (RV) and LV pressure–volume relationships and interventricular dependence immediately before and 1 month after T-TEER. Using these elegant haemodynamic and imaging measures, the authors find T-TEER improved LV filling and RV-LV septal interactions without adverse changes to end-diastolic pressures. In fact, the LV end-diastolic pressure–volume relationship shifted down and rightward, suggestive of less diastolic impairment. This study provides an important and nuanced analysis of the physiologic properties of TR and the mechanistic response to T-TEER among patients with comorbid severe TR and HFpEF. Furthermore, it lends insight into RV and LV interactions following TR reduction.9 The findings build upon other early mechanistic descriptions of tricuspid valve (TV) repair and RV remodelling using transcatheter mitral repair devices,10 but further adds an important characterization of LV function particularly in a population with LV diastolic impairment on baseline haemodynamics. The authors should be commended for their efforts. In particular, the inclusion of multimodality imaging and pressure–volume analyses beautifully highlight the anatomic and haemodynamic responses to valvular repair and interventricular interactions, perhaps providing us with a signature for HFpEF patients likely to respond to such therapies. The authors also describe longitudinal physiologic changes which help clarify potential sustained treatment responses in this complex and heterogeneous valvular condition. Despite the study's clear strengths and its important observations, some notable questions remain. First and foremost, the patients enrolled in this study represent a very unique HFpEF phenotype, not typical of the general HFpEF population. With right atrial pressures that exceed LV end-diastolic pressure (LVEDP), the median pre-intervention LV transmural pressure (a surrogate of true LV preload which is estimated by LVEDP minus right atrial pressure) was negative 5 mmHg. This haemodynamic profile is consistent with significant pericardial restraint and suggests that the actual LV preload was low in most patients. One would therefore predict that a reduction in TR and improved right-sided volume loading conditions, if effective in reducing pericardial restraint, would be associated with improved LV filling and possibly even a reduction in LV filling pressures depending on the magnitude of reduction in pericardial restraint. Although immediate post-intervention haemodynamics were not available, at 1 month LV transmural pressure was more positive and LV filling pressures were reduced. In fact, with a median post-intervention LVEDP of 16 mmHg, nearly half of the cohort would no longer have met haemodynamic criteria for HFpEF after intervention. This would suggest that a substantial portion of the study population had only mild LV diastolic impairment despite inclusion criteria of increased pulmonary artery (PA) wedge pressure or LVEDP ≥15 mmHg at baseline. Although the cohort at large had high H2FPEF scores and comorbid conditions associated with HFpEF, we are left to wonder if some patients may not have had significant HFpEF physiology (i.e. myocardial diastolic abnormalities). We also do not know if a similar response would occur in HFpEF patients without marked pericardial restraint or more typical HFpEF haemodynamics (Figure 1). Secondly, while the authors note that 60% of patients enrolled at baseline had at least mild pulmonary hypertension with a pulmonary vascular resistance (PVR) >2 Wood units (WU),9 the median PVR was <3 WU. This suggests that most patients did not have significant RV pressure overload as a cause of their functional TR. It remains unknown if repairing the TV in this setting would yield a similar response to a predominant volume overloaded state as in this cohort. Consistent with the study's aim to understand how increased volume return to the left ventricle following TV repair adversely affects LV function in HFpEF, the authors performed pressure–volume analysis in the left ventricle to best characterize haemodynamic changes following T-TEER. While they described favourable physiologic response in the left ventricle, changes in RV function were not directly measured, rather estimated by characterizing interventricular relationships via CMR/TTE-derived septal wall characteristics, biventricular end-diastolic volumes, estimated PA pressures, and RV wall stress. However, a definitive understanding of changes to RV function and RV-PA interaction is lacking. In fairness, assessing RV function in the setting of TR is challenging. Even pressure–volume assessment may be misleading since RV stroke volume is ejected into both the right atrium (backwards flow–low afterload) and PA (forward flow–higher afterload). A recent study by Yoshida and colleagues in patients with TR has suggested partitioning of RV afterload (e.g. elastances) into right atrial and pulmonary arterial components may be useful.11 Assessment of RV reserve with provocative manoeuvres may ultimately add value.12 Considering the aforementioned limitations of RV assessment in the setting of TR, RV function appeared relatively well-preserved in the current study (CMR-based RV ejection fraction 47 ± 7%). Thus, it remains unknown the threshold of RV dysfunction at which favourable RV-LV interactions would still occur with T-TEER. This remains a major unmet need in our field. Finally, although the present study describes improved LV diastolic function from baseline to 1 month following T-TEER and other mechanistic studies have described additional RV volume overload reduction between 1 month and 6 months following transcatheter valve repair,10 we must acknowledge that adverse risk following severe TR diagnosis increases drastically within 1 year.3 Thus it remains to be seen if the present study's observed changes from baseline to 1 month is a long-enough duration for sustained remodelling of RV and LV function and whether we can extrapolate these observed mechanistic changes to downstream outcomes. A longitudinal characterization of biventricular interaction and RV-PA function over several months or even years may elucidate the time horizon for treatment efficacy or identify periods of reverse biventricular remodelling or relapse in valvular severity. With the development of novel and minimally invasive TV interventions, larger outcome trials of T-TEER and TV replacement will allow for characterization of the mechanistic and clinical benefits of TR reduction in patients with severe TR. This study is an important first step in identifying a group of HFpEF patients with severe TR that are likely to benefit from such therapies. It would seem that those with haemodynamic and imaging evidence of pericardial restraint and altered RV-LV interactions may be such a group. We again congratulate Kresoja and co-authors on this study and appreciate the reminder that the ‘action is in the interaction’. Conflict of interest: R.J.T. is the co-chair of the PH due to left heart disease task force for Seventh World Symposium on Pulmonary Hypertension and Deputy Editor for the Journal of Heart and Lung Transplantation. He reports general disclosures to include consulting relationships with and receiving honorarium from Abbott, Acorai, Aria CV Inc., Acceleron/Merck, Alleviant, Boston Scientific, Cytokinetics, Edwards Life Sciences, Gradient, Medtronic, Morphic Therapeutics, Restore Medical, and United Therapeutics. He serves on steering committee for Abbott, Edwards, Endotronix, and Merck as well as a research advisory board for Abiomed. He also does haemodynamic core lab work for Merck. All other authors have nothing to disclose.