Impact of body mass index in patients with tricuspid regurgitation after transcatheter edge-to-edge repair

This study was designed as a retrospective analysis of data from the Bonn registry, which is a prospective, consecutive collection of patient data from the Heart Center Bonn. We identified patients with symptomatic TR who underwent TEER interventions from June 2015 to May 2021. All included patients were considered as inoperable or at high-risk for surgery by the interdisciplinary heart team. After a standardized diagnostic workup, including transthoracic (TTE) echocardiography, left- and right-heart catheterization, the patient’s anatomical suitability for a TEER system was assessed using CT images and transesophageal echocardiography (TEE).

The indication for a tricuspid valve intervention was evaluated based on the current guidelines, and the decision to perform transcatheter tricuspid valve interventions as well as the device selection was made by the interdisciplinary heart team. The registry was approved by the local ethics committee. This study was conducted in accordance with the Declaration of Helsinki and its amendment, and all patients provided written informed consent.

The TEER procedure was performed with either the MitraClip System (Abbott Structural Heart, Santa Clara, CA, USA), TriClip System (Abbott Structural Heart, Santa Clara, CA, USA) or PASCAL Implant System (Edwards Lifesciences, Irvine, CA, USA). The details of each device system and procedure have previously been well described [3, 21‐24].

Comprehensive TTE and TEE were performed at baseline, and at discharge and 1-year follow-up a comprehensive TTE was performed according to the current guidelines. The severity of TR was graded by the current recommendations using the five-scaled grading scheme as described in detail before [25]. All measurements were reviewed by two independent cardiologists dedicated to echocardiographic evaluation.

Underweight was defined as BMI < 20 kg/m² based on the Academic Research Consortium and 2 [26] and 3 [27] definition of frailty. Four groups were defined, after calculating the preprocedural BMI for each patient: underweight (BMI < 20 kg/m²), normal weight (BMI 20 to < 25 kg/m², reference group), overweight (BMI 25 to < 30 kg/m²) and obesity (BMI ≥ 30 kg/m²).

The primary endpoint of this study was all-cause mortality within 1 year after the procedure. Secondary outcomes were the reduction of TR, New York Heart Association class (NYHA), major adverse cardiovascular and cerebrovascular events (MACCE) (myocardial infarction and stroke) and changes in right ventricular function, measured by TAPSE and right ventricular fractional area change (RVFAC) at 1-year follow-up.

All patients were followed through interviews at scheduled hospital visits, telephone, or documentation from the referring general practitioners.

All patients (n = 211) who underwent TEER were stratified according to their baseline BMI into four groups: underweight (n = 29), normal weight (n = 71), overweight (n = 64) and obesity (n = 47). Overall, the mean age was 78.3 ± 7.2, 55.5% of patients were female and the median EuroSCORE II was 9.6 ± 6.7% and the median TRI-SCORE was 5.2 ± 2.0. 182 (86.3%) of subjects were classified as NYHA functional class III and more. The prevalence of hyperlipidemia (p < 0.001) and diabetes (p = 0.01) was significantly higher in obese patients. All other comorbidities were without significant differences and in addition there were no statistically significant differences in echocardiographic and laboratory assessment between the four groups at baseline (Table 1).

Periprocedural findings are summarized in Table 2. Most cases of TEER were treated with the TriClip system (51.2%, Abbott Structural, Santa Clara, CA, USA), followed by the PASCAL Implant System (25.1%, Edwards Lifesciences, Irvine, CA, USA) and the MitraClip system (23.7%, Abbott Structural, Santa Clara, CA, USA). The mean number of implanted devices and implantation failure was comparable between the groups with 1.8 ± 0.8 devices per procedure (p = 0.06 and p = 0.07). Extensive bleeding events occurred in two patients (one patient with normal weight and one patient with overweight) without fatal outcome. These patients required blood transfusion. There was a tendency towards more failed intervention in the obese group, mainly driven through a lesser postprocedural reduction of TR grade (definition of failed procedure: reduction of TR grade < 2 grades or no device implantation)—without statistical significance (p = 0.06).

All patients had ≥ severe TR at baseline according to the five-scale grading scheme: 21.8% were graded torrential, 32.7% massive and 45.5% severe. There were no significant differences between the stratified BMI subgroups (p = 0.48). At discharge, TR severity was significantly reduced in all subgroups (Fig. 1a, b). In addition, there were no differences between the groups in TR severity at discharge and at 1 year-follow-up (p = 0.13 and p = 0.67, respectively) (Fig. 1c). 143 (67.8%) of subjects had a sustained TR reduction of ≥ 2 grades at discharge in comparison to baseline. Stratified according to the four BMI classes, underweight and obese patients showed significant lower rates of TR reduction ≥ 2 grades at discharge: 55.2% vs. 70.4% vs. 78.1%. vs. 57.4%; p = 0.05 (Fig. 1d).

Stable and significant improvements in NYHA functional class were found for all patients without significant differences among the subgroups at baseline (p = 0.17). At 12 months, 167 of 172 patients (97.1%) were in NYHA functional classes I or II, similar according to BMI subgroups (p = 0.93) (Fig. 2).

In the overall cohort, the primary endpoint (all-cause mortality at 1-year follow-up) was observed in 18.5% (39/211) of the patients. Kaplan–Meier curves showed that underweight and obese patients appeared to have a significantly higher all-cause mortality within 1 year after the procedure, compared to normal and overweight patients (p log rank < 0.001), 37.9% of the underweight group and 29.8% of the obese group died within 1 year (Fig. 3).

Patients with a TR reduction of ≤ 2 grades showed a tendency towards a higher 1-year mortality stratified by BMI groups (Fig. 4).

In a multivariable cox proportional hazard model, underweight patients (3.88, 95% CI 1.64–7.66; p < 0.01) and obese patients (HR 3.24, 95% CI 1.37–9.16; p < 0.01) significantly revealed to be associated with an increased risk of reaching the primary endpoint. COPD (HR 2.52, 95% CI 1.14–5.56; p = 0.02) and a depressed left-ventricular ejection fraction (HR 3.28, 95% CI 1.21–8.86; p = 0.02) were also identified as predicting factors for reaching the primary endpoint in our cohort (Table 3).

Obesity has become an increasingly common chronic condition in the western civilization which is associated with significant morbidity and mortality [32‐34]. It has also been characterized as a major and also modifiable risk factor for cardiovascular morbidity and mortality by the American Heart Association/American College of Cardiology and the Nutrition Council of the American Heart Association [35]. However, a considerable number of studies reported a beneficial effect of overweight and obesity on survival in patients with established heart disease in general (e.g. CAD, atrial fibrillation) and specifically in patients undergoing interventional procedures [36‐39], such as transcatheter aortic valve replacement [40], TEER for symptomatic mitral regurgitation [17] and in patients with heart failure [41, 42]—in these studies, obesity was characterized inconsistently but mainly as a BMI ≥ 30 kg/m². The phenomenon of a beneficial effect of overweight and obesity on survival was termed the obesity paradox, and its validity is still under discussion in the literature.

In our analysis, obesity—defined as a BMI ≥ 30 kg/m²—was associated with inferior survival (HR 3.24, 95% CI 1.37–9.16; p < 0.01) compared to normal weight and overweight patients. One reason in this regard might trace back to the fact that the underlying pathology mainly consists of a leading right heart disease and right heart failure whereas the studies showing a beneficial effect of obesity examined leading left heart diseases and valvulopathies, respectively. Thus, a different effect of overweight on outcome in patients with right heart failure and TR does not categorically contradict the proposed obesity paradox as it merely stresses the different disease entities of right and left sided valvulopathies and heart failure and their different clinical presentation and pathophysiology, respectively.

The investigators against the validity of the obesity paradox in the aforementioned cohorts argue that the obese population is younger, seeks medical care earlier, is treated medically more aggressively, and therefore, benefits more from medical and interventional treatment [36], thus obesity being merely a surrogate. Indeed, in our cohort obese patients were slightly but not statistically significant younger (mean age in: obesity 76.6 ± 7.4 years vs. overweight 78.1 ± 6.8 years vs. normal weight 79.5 ± 7.5 years vs. underweight age 78.3 ± 6.3 years; p = 0.18), but did not show a beneficial outcome. A possible explanation could be inferior rates of pronounced TR reduction (≥ 2 TR grades to baseline) of 42.6% in patients with the highest BMI. Additionally, rates of failed procedures (no device implantation or TR reduction ≤ 2 grades to baseline) appeared to be slightly higher in obese patients (17%, 8/47)—mainly driven by fewer cases with a pronounced TR reduction postprocedurally compared to baseline, but without statistical significance (p = 0.06).

A possible explanation might also be a potential difference in leaflet characteristics in obese patients compared to normal- and overweight patients making grasping more difficult, but by now this cannot easily be examined. Moreover, there were no significant differences regarding occurrence of SLDA (single leaflet device attachment) or number of implanted devices.

Additionally, we hypothesize that increasing BMI may have associated independent confounders not demonstrated in our study population for which it was potentially not powered enough and will be addressed in follow-up studies in further detail—such as leaflet characteristics, difficulty with catheter—or device guiding and navigation, challenging imaging, or different atrial and ventricular volumetries and TV annulus anatomies and dimensions.

Lastly, our analysis documented an increased 1-year all-cause mortality of underweight and obese patients receiving TEER for significant TR. Obese patients showed a comparable cardiovascular mortality after 1 year compared to the normal and overweight cohorts. As a main reason we postulate a potential underestimation of cardiovascular mortality in the obese cohort as in 5 deaths (10.6%) reasons for death remained unknown (in comparison to 4.7%, 10.3%, 1.4% and 1.5% for all, underweight, normal- and overweight patients).

Underweight has been previously documented as a risk factor for cardiovascular mortality [14]. However, there is no generally accepted definition of underweight—especially in the elderly, thus limiting comparability throughout the studies. Kalbacher et al. [17] reported underweight patients with mitral regurgitation undergoing TEER procedure with the MitraClip system to be exposed to increased short- and long-term mortality. In our cohort, underweight was associated with inferior 1-year survival (HR 3.88; 95% CI 1.64–7.66; p < 0.01) compared to normal weight and overweight patients. A significantly higher rate of cardiovascular death was documented in the underweight cohort compared to the other groups (24.1% vs. 7.0% vs. 6.3% vs. 6.4%; p < 0.01). Overall, only limited data exist about potential reasons as to why underweight is associated with increased mortality rates in cardiovascular patients. It cannot be excluded that underweight might act as a surrogate parameter for frailty as the most common definition used for frailty includes underweight and weight-loss, respectively—as well as serum-albumin levels, grip strengths, cognitive function and mobility (6 min-walking test) [26]. Moreover, underweight might effect a decreased capacity for reconvalescence and an overall reduced physiological reserve which might partly be an explanation for an increased short-term mortality. In addition, increased vulnerability to stressors has been discussed.