Long-Term Effects of Left Ventricular Assist Device Therapy on Pulmonary Vascular Resistance in Patients Bridged to Heart Transplant

Jason J Han1 Salman Zaheer 1 Rahul Kanade1 Jennifer Chung1 Carol W Chen1 Ann C Gaffey1 Christyna Justice1 Alyse E Ameer1 J Eduardo Rame2 Michael A Acker1 Pavan Atluri1* 1Division of Cardiovascular Surgery, Department of Surgery, University of Pennsylvania, Philadelphia, Pennsylvania, United States 2Division of Cardiovascular Medicine, Department of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, United States


Introduction
Severely elevated pulmonary vascular resistance (PVR) is contraindicated in orthotopic heart transplant (OHT) due to increased risk of right heart failure [1][2][3][4]. While institutions vary in their policies, most centers consider a PVR greater than three on maximal medical therapy an absolute contraindication [5]. Historically, no surgical intervention was available for these patients, whose advanced heart failure and chronic pulmonary congestion led to adverse cardiac remodeling and pulmonary hypertension (PH). The maturation of mechanical circulatory support technology hassled to the development of new management and optimization strategies for these challenging patients. In addition to destination therapy (DT), short-term use of ventricular assist devices (VAD) as bridge-to-transplant (BTT) has been shown to be effective in normalizing PVR for many patients with advanced heart failure and PH, rendering them eligible for OHT [6,7].
Despite these benefits in bridging patients to transplantation, the long-term efficacy and durability of reducing PVR with mechanical circulatory support are not yet fully understood. Therefore, this study aimed to better characterize patients' pulmonary vascular profiles as they progressed from end-stage heart failure to VAD therapy then ultimately to OHT. We stratified patients based on their PVR prior to VAD implantation and evaluated the effects of mechanical circulatory support on PVR and incidence of adverse events in the long-term following successful transplantation.

Patient Selection
The Institutional Review Board (IRB) at the Hospital of the University of Pennsylvania approved this study. Patients who received HeartMate II (HMII) (Thoratec Corp., Pleasanton, CA, USA) or HeartWare (HVAD) (HeartWare International Inc., Framingham, MA, USA) as bridge-to-transplant (BTT) indications between the time period May 2008 to November 2016 were identified from the institutional MCS database. Patients who required bi-ventricular support during their index operations were excluded. Based on their PVR values immediately prior to VAD implantation, patients were divided into two groups: "Elevated PVR," denoting PVR ≥ 3, and "Normal PVR," denoting PVR <3. LVAD settings at time of implantation were titrated to adequate ventricular unloading based on echocardiographic findings and hemodynamic profiles, as per standard practice protocol. A subset of patients who successfully underwent OHT during the study period were identified.

Variable Selection
Primary outcome was defined as survival at both 30 days and most recent follow-up date after OHT. Secondary outcomes were defined as incidence of right heart failure and PVR post-OHT. Patient demographics and implant characteristics, including hemodynamic and laboratory profiles, were compared between groups. Patients underwent formal right heart catheterization studies (RHC) at following time points: prior to VAD implantation, within 30 days after OHT, and at the most recent follow-up time point after OHT. These studies recorded mean arterial pressure (MAP), mean pulmonary arterial pressures (mPAP), pulmonary capillary wedge pressure (PCWP), cardiac output (CO), index (CI) and central venous pressure (CVP), which were used to calculate PVR. Pulmonary arterial pressure recordings while on VAD support were derived from post-operative Swan-Ganz measurements within 7 days after VAD implantation.
Adverse events post-VAD implantation and post-OHT, including right ventricular (RV) failure, gastrointestinal bleeding, stroke, transient ischemic attacks, renal failure requiring hemodialysis, concern for pump thrombosis, and major infection requiring oral or IV antibiotic therapy, were also assessed. RV failure at any time point during VAD support were included, defined as requiring right ventricular assist device (RVAD) support, severe RV dysfunction on echocardiography or inotropic therapy for longer than 14 days postoperatively as per INTERMACS.

Statistical Analysis
Continuous variables were reported as mean ± standard deviation and non-parametric variables were reported as median and interquartile range. All statistical analyses were conducted using GraphPad Prism (La Jolla, CA) and Stata software 14 (College Station, Tx). Patient demographics and baseline characteristics were compared using univariable analysis. Differences in hemodynamic profiles at varied time points between the two groups were compared using T-tests. Longitudinal progression of PVR within the cohort at pre-VAD, pre-OHT, and post-OHT time points were compared using paired-t-test. Categorical variables were compared by chi-square analysis. Survival at 30 days and 1 year were compared using Kaplan-Meier curves and log-rank tests. For all analyses, values of p greater than 0.05 were considered not significant (NS).

Results
During the study period, 235 patients received LVADs with a mean mechanical support duration of 470 ± 570 days. Baseline cohort statistics are outlined in Table 1. Of the total, 83 patients received HeartMate II (n=51) or HVAD (n=32) for BTT indications. Prior to VAD implantation, the median PVR in the overall cohort was 2.7 (1.5-4.0) Woods units (W.U.). Thirty-five patients (42%) had PVR ≥ 3 with a median of 4.6 (4.0-5.0) W.U, while 48 (58%) patients had normal PVR with a median of 1  Table 2). Other hemodynamic parameters including MAP and CVP were comparable. There were no differences between the groups in terms of demographic and laboratory variables.

Discussion
In this study, our goal was to investigate the efficacy and durability of using mechanical circulatory support to reduce PVR among patients listed for heart transplant. Our principal findings are as follows.

VAD therapy successfully reverses elevated PVR in patients who
are bridged-to-transplant 2. Benefits of reduced PVR are equal among those with elevated and normal PVR values prior to VAD implantation Patients with advanced heart failure with long-standing pulmonary congestion, vasoconstriction and adverse remodeling often have concomitant pulmonary hypertension (PH). The severity as well as the reversibility of PH has been shown to have prognostic implications [8]. Due to increased risk of right heart failure after OHT, These 44 patients also demonstrated lasting improvements in rightheart hemodynamic parameters: Progressive reductions in mean pulmonary arterial pressures (pre-LVAD 34.6 ± 8.3 vs. LVAD 23.7 ± 7.5 vs.. OHT 18.7 ± 6 mmHg) and improvements in cardiac index (pre-LVAD 1.9 ± 0.5 vs. LVAD 2.8 ± 0.9 vs.OHT 3.1 ± 0.8) were observed ( Table 4).

Regardless of having normal (PVR<3) or elevated (PVR ≥ 3)
values prior to VAD implantation, both groups had comparable survival outcomes post-OHT at 30 days (88% vs. 96%, p=0.9) and 5 years (62% vs .74%, p=0.2) (Figure 1). There was no difference in the incidence of moderate or greater right ventricular failure post-OHT   The table delineates various complications associated with LVAD therapy based on patients' pre-implantation PVR profiles. RV failure was defined as inotropic requirement greater than 2 weeks or RVAD requirement.

4/6
irreversible elevated PVR is a well-established contraindication with varying degrees of stringency across institutions [9].
At our institution, PVR ≥ 4 is an absolute contraindication for transplant and PVR between 3 and 4 on optimal medical therapy is a relative contraindication. Therefore, hemodynamic optimization prior to OHT, especially mitigating PH using inotropic support, pulmonary vasodilators or mechanical circulatory support, remains crucial. In our study, approximately 40% of all BTT patients had PVR above 3 at the time of implantation, indicating the high prevalence of and the importance in further understanding this patient population's long-term outcomes.
Landmark clinical trials have demonstrated the efficacy of VAD therapy in restoring hemodynamic stability in patients with endstage heart failure with excellent long-term outcomes [10][11][12]. In addition to restoring cardiac output, various studies have shown the efficacy of using mechanical circulatory support to mitigate PH and optimize right heart function [6,7,13]. Particularly for potential OHT candidates who are contraindicated based on their fixed PH diagnoses, VAD therapy as BTT indication may render them eligible upon repeat right heart catheterization in as short as 3 to 6 months. According to the 2016 International Society for Heart Lung Transplantation (ISHLT) Listing Criteria, this strategy to assess the reversibility of PH is currently listed as a Class IIA recommendation [14].
In our study, the benefits of VAD support in optimizing pulmonary circulation among BTT patients were reaffirmed. While right heart catheterizations were not routinely performed in all BTT patients while on LVAD support, significant improvements in pulmonary arterial pressures, cardiac output and index were observed, consistent with previously reported findings [15]. Reassuringly, significantly elevated PVR values (≥3) prior to VAD implantation had effectively normalized by the time of transplant across the entire cohort. Moreover, these hemodynamic improvements were sustained in the long-term following OHT, suggesting stable reverse remodeling in the pulmonary vasculature. While right heart failure is a welldescribed adverse event after VAD implantation and OHT, especially among patients with increased PVR, its incidence was equivalent between both groups in our study [16,17].
Lastly, studies have noted the differential efficacy of axial versus centrifugal flow VADs in unloading the left ventricle [18]. Inherent differences in device capability may have implications for how effective each device type is in mitigating pulmonary hypertension and favoring reverse remodeling among BTT patients. In our study, there were no differences in survival or other adverse outcomes based on device type. Future studies that aim to understand the relationship between the type of intervention, the degree of unloading, and longterm post-OHT outcomes are warranted.
Our study has several limitations. First, it is a single institutional retrospective study with a limited cohort size. Moreover, although many patients are designated as BTT prior to VAD implant, patients have varying severity of PH that may still preclude their eligibility for OHT. While all patients had bedside Swan-Ganz catheter measurements, not all patients underwent formal right heart catheterization assessments while on VAD during the study period, which limited our overall analysis. Recent evidence by Schumer et al. [19] pointed to the possibility of increased risk of adverse outcomes post-OHT for patients with persistently elevated PVR on VAD. This subset of patients with irreversible PH may need to be described separately in future studies, as right heart catheterization studies become more commonly utilized in evaluating, optimizing as well as prognosticating BTT patients. This would help avoid selection bias for patients whose PVR values were able to be reversed prior to undergoing OHT.
Furthermore, a growing body of evidence supports an equally vital role of compliance, in addition to resistance, in RVF, especially in the setting of PH [20][21][22]. These two parameters are inversely correlated as evidenced by the formula describe the arterial time constant, =RC [23]. As described by Lankhaar et al. [24], the clinical implication is that the same degree of improvement in R can lead to different degree of improvement in C depending on how severely R is elevated at baseline. As such, it points to interesting questions about therapeutic potential of LVADs based on the severity of PVR, and subsequently, the optimal timing of implantation.

5/6 Conclusions
In conclusion, our study reaffirms the role of mechanical circulatory support among patients with end-stage heart failure who are bridged-to-transplant. Sustained and clinically meaningful reductions in PVR were observed regardless of the degree of pulmonary hypertension prior to VAD implantation and the type of device used.