Adult Cardiac Scientific Session: Fads with VADs

Objective: Right heart failure (RHF) remains a significant limitation of left ventricular assist device (LVAD) therapy. Studies have observed that many of these patients have tricuspid valve regurgitation (TR) at the time of LVAD implantation. It remains unclear whether concurrent tricuspid valve surgery (TVS) may reduce post-operative RHF. This study is a randomized clinical trial to define the clinical impact of concurrent TVS in patients undergoing LVAD implantation. The interim analysis demonstrated that TVS did not appear to lower RHF incidence. The impact of TVS on late RHF, TR severity, and functional outcomes is presented here.
Methods: Patients with moderate or severe TR on pre-operative echocardiography (n=71) were randomized to either LVAD implantation alone (No TVS, n=34) or with concurrent TVS (TVS, n=37). Randomization was stratified by pre-operative right ventricular dysfunction (none/mild vs moderate vs severe). TVS consisted of either annuloplasty (n=31) or replacement (n=6) and was chosen at the surgeon's discretion. Patients were followed for 12+ months after surgery. The rate of moderate or severe RHF was determined by an adjudication committee using INTERMACS definitions.
Results: Demographic characteristics were similarly distributed between the two groups. At 12 months, the rate of moderate or severe RHF was 50.0% (No TVS) versus 51.4% (TVS). The need for RVAD was 10.8% (TVS) versus 20.6% (No TVS) but was not significant (p=0.75)(Figure 1A). Death from RHF was 5.4% (TVS) versus 8.8% (No TVS)(Figure 1B). The rates of moderate/severe TR decreased in the No TVS group (32.4%/20.6% intra-op, 42.9%/14.3% 3-6 months, 17.4%/8.7% 12 months) but modestly increased over time in the TVS group (8.1%/0% intra-op, 4.5%/0% 3-6 months, 7.1%/7.1% 12 months)(Figure 1C). By 12 months there was no significant difference in TR severity between the two groups (p=0.70). Mortality was 20.6% (No TVS) versus 18.9% (TVS). On cardiopulmonary exercise testing at 12+ months, there was no significant difference in peak VO2/kg (mean[SD]: 9.2[1.5] (No TVS) vs 8.9[3.4] (TVS), p=0.755).
Conclusions: Need for RVAD support was 10.8% in the TVS group compared to 20.6% in the No TVS group. In patients with significant pre-implant TR, TR appears to improve over time with LVAD placement alone. This improvement in TR for the control group may account for the lack of difference in late clinical or functional parameters relative to the group that received concurrent TVS.

Authors
Michelle Mendiola Pla (1), Stuart Russell (1), Carmelo Milano (1), Yuting Chiang (2), Lillian Kang (1), Emily Poehlein (1), Cynthia Green (1), Frank Benedetti (1), Han Billard (1), Benjamin Bryner (3), Jacob Schroder (1), Mani Daneshmand (4), Alina Nicoara (1), Adam DeVore (1), Chetan Patel (1), Jason Katz (1), Muath Bishawi (1)
Institutions
(1) Duke University Medical Center, Durham, NC, (2) Aurora St. Luke's Medical Center, Milwaukee, WI, (3) Northwestern University, Chicago, IL, (4) Emory University Hospital, Atlanta, GA 

Objective: Our objective was to analyze the development of aortic insufficiency (AI) in patients who received central aortic valve repair (AVr) when undergoing continuous-flow left ventricular assist device (CF-LVAD) implantation.

Methods: We conducted a retrospective review of patients who underwent HeartMate (HM) 2 or 3 implantation between 2004 and 2022. Eighty-nine patients were excluded from analysis for history of aortic valve (AV) procedures, a bicuspid AV, other concomitant AV procedure or limited follow-up echocardiographic data. Patients who had ≥ mild AI had concomitant AVr. Clinical characteristics, serial echocardiograms, and outcomes were determined.

Results: Of the 656 patients who underwent HM3/HM2 LVAD implantation, 110 patients (62 HM2 and 48 HM3) underwent AVr. Median age for AVr patients was 68 years [60-73 years], 91.8% [n=101] were male, 55.6% [n=60] were white, and 70% [n=77] received HM2/3 as DT therapy. In hospital mortality was 6.4% [n=7]. Figures showed severity of AI from preop to follow up (A) and cumulative incidence of moderate or greater AI (B). Freedom from ≥ moderate AI was 98% (± 0.01), 94.9% (± 0.03), and 92.8% (± 0.03) at 1-year, 2-year, and 3-year post-implantation, respectively. One HM2 patient experienced severe AI and left ventricular dilation at 1 month and was treated with a heart transplant 44 days after the echocardiographic measurement. Three-year survival was 64.5% [95% CI, 52.2-74.3%].

Conclusions: AVr may be an effective technique to mitigate AI in HM2 and HM3. A larger cohort study with longer duration of follow up is warranted to further investigate the clinical impact.

Authors
Melissa Hynds (1), Hideyuki Hayashi (1), Paul Kurlansky, MD (1), Yanling Zhao (1), Alice Vinogradsky (1), Melana Yuzefpolskaya (1), Paolo Colombo (1), Gabriel Sayer (1), Nir Uriel (1), Yoshifumi Naka (2), Koji Takeda (1)
Institutions
(1) Columbia University Irving Medical Center, New York, NY, (2) Weill Cornell Medical Center, New York, NY 

Objective: Continuous flow fully magnetically levitated (CF-FML) left ventricular assist devices (LVAD) are a cornerstone of modern surgical treatment for end-stage heart failure. This study describes our high-volume single-center survival experience in a consecutive series of CF-FML LVAD implants.

Methods: From 10/2015-1/2023, 431 patients underwent CF-FML LVAD implant with 890 patient-years of follow-up available. Survival was estimated out to 5-years post-implant overall and stratified by Interagency Registry for Mechanically Assisted Circulatory Support (Intermacs) profile, ischemic (ICM) versus non-ischemic cardiomyopathy (NICM), and bridge to transplant (BTT) versus destination therapy (DT) strategies.

Results: Median age at implant was 61 years (15th,85th: 47.3, 71.7) and 109 (25%) patients were female. Preoperatively, at the time of CF-FML LVAD implant, 90 (21%) patients were supported with intra-aortic balloon pumps, 49 (11%) patients with temporary LVAD and 5 (1.2%) were on extracorporeal membrane oxygenation. Overall survival at 1- and 5-years post-implant was 88% and 58%. Intermacs profile was 1 in 19 (4.4%), 2 in 147 (34%), 3 in 125 (29%), and 4-7 in 145 (33%) patients. For Intermacs profile 1, 6-month survival was 66%. Survival at 1- and 5-years was 89% and 59% in profile 2, 94% and 68% in profile 3, and 84% and 59% in profiles 4-7, respectively. In NICM patients, survival at 1- and 5-years was 91% and 65%, and in ICM patients it was 84% and 53%. Survival did not differ in the early post-implant phase (≤3 weeks) between NICM and ICM patients (P=0.9) but was higher among NICM patients in the late post-implant phase beyond 3 weeks (P=0.006). For BTT patients, 1- and 5-year survival was 89% and 67% whereas DT survival was 87% and 55%, respectively.

Conclusion: This large single center experience demonstrates excellent long-term survival on CF-FML LVAD support. Survival was greatest among profile 3 patients potentially representing the benefit of timely intervention in advanced heart failure prior to deterioration. Survival was also increased among NICM compared to ICM patients and BTT compared to DT patients. Further studies to identify subpopulation-specific risk factors may allow for further improvement of outcomes.

Authors
Aaron Weiss (1), Jean-Luc Maigrot (1), Michael Tong (1), Randall Starling (1), Nicholas Smedira (1), Shinya Unai (1), David Moros (1), Jerry Estep (2), Eugene Blackstone (1), Edward Soltesz (1)
Institutions
(1) Cleveland Clinic, Cleveland, OH, (2) Cleveland Clinic Florida, Weston, FL 

Objective: Appropriate management of preoperative tricuspid valve regurgitation (TR) during durable left ventricular assist device (LVAD) implantation represents a significant clinical challenge. This study evaluates the impact of post-LVAD mitral regurgitation (MR) reduction on uncorrected moderate to severe TR.
Methods: All patients who underwent LVAD implantation at our institution between January 2018 and December 2021 with preoperative moderate or greater TR were included. No patients had concomitant mitral valve intervention at the time of LVAD implant and those who underwent tricuspid valve repair were excluded. Postoperative echocardiogram data was collected up to 2 years post-LVAD implant. The cohort was divided based on postoperative MR reduction. The primary outcomes included the relationship between TR and MR and overall survival. Kaplan-Meier curves were used to calculate overall survival and Pearson correlation coefficient was used to assess the relationship between MR and TR.
Results: One hundred and eighteen patients underwent LVAD implantation during the study period, and uncorrected moderate to severe TR was identified preoperatively in 40 (33.9%) patients. Those patients were then divided into groups based on postoperative MR reduction, with 7 (17.5%) having no change, 13 (32.5%) having 1 degree, 14 (35.0%) having 2 degrees, and 6 (15.0%) having 3 degrees of MR reduction. There were no significant differences in baseline characteristics between groups. Postoperative improvement in MR was associated with TR reduction at 1-year (r=0.596, p =0.015) and 2-years (r=0.471, p=0.03). There was no significant difference in overall survival between those with unchanged and improved MR at 1-year (85.7% vs 81.4%, p=0.780) and 2-years postoperatively (85.7% vs 65.9%, p=0.520). At 2 years, there were no significant differences in the degree of TR between groups (p=0.365). Notably, 88% (n=22) of patients had no or mild TR at 2 years.
Conclusions: For preoperative moderate to severe TR, the presence and subsequent reduction of MR after LVAD implantation appears to correlate with TR improvement. Future analyses should assess the implication of preoperative MR on optimal patient selection for tricuspid valve repair at the time of LVAD implantation

Authors
Maxwell Kilcoyne (1), Sarah Chen (1), Khaled Shorbaji (1), John Foster (1), Jennie Kwon (1), Brett Welch (1), Arman Kilic (1)
Institutions
(1) Medical University of South Carolina, Charleston, SC 

Objective: Right ventricular failure is a morbid complication after LVAD implantation which may be treated with a right ventricular assist device (RVAD). We sought to determine the effect of preemptive RVAD insertion at the time of HeartMate 3 (HM3) implantation in patients at high risk for post-LVAD RV failure.
Methods: We retrospectively reviewed 294 consecutive patients undergoing HM3 implantation at our center from November 2014 to December 2022. A preemptive RVAD strategy was applied to select patients. Outcomes were compared to patients who did not require an RVAD and who required an RVAD after HM3 insertion (reactive).
Results: Of the cohort, 33 (11.2%) patients had preemptive CentriMag (CM) RVAD insertion, while 230 (78.2%) had no RVAD and 31 (10.5%) had a reactive RVAD [CM (n=8, 25.8%) or ProTek Duo (n=23, 74.2%)]. Reasons for a preemptive RVAD were INTERMACS 1TCS (n=22, 66.7%), intractable ventricular tachycardia (n=4, 12.1%), or severe biventricular failure (n=7, 21.2%); 24.2% (n=8) of patients had an eGFR less than 45 mL/min/1.73m2. Preemptive RVAD patients were more frequently INTERMACS 1TCS (None: 10.9% vs Preemptive: 63.6% vs Reactive: 6.5%, P<0.001), had a higher CVP (9 vs 13 vs 9 mmHg, P=0.01) and higher rate of severe RV dysfunction (12.2 vs 30.3 vs 16.1%, P=0.02). In-hospital mortality was highest in the reactive group (2.2 vs 3.0 vs 25.8%, P<0.001). With a BiVAD configuration in place, 3 patients (9.1%) in the preemptive group were bridged to a heart transplant vs. 0 in the reactive group, and there were 0 mortalities for preemptive patients vs. 5 (16.1%) in the reactive group. All others (87.5%) were successfully weaned off RVAD support. Median duration of RVAD support was not different between preemptive and reactive groups (14 vs 13 days, P=0.798). Kaplan-Meier analysis showed lowest 3-year survival in the reactive group, while preemptive and non-RVAD groups had comparable survival (84.7 vs 85.5 vs 46.6%, P<0.001; Figure 1). In multivariable Cox analysis, a reactive RVAD was an independent risk factor for 3-year mortality (HR: 3.7, P<0.001). Among survivors to discharge, 3-year freedom from RV failure readmission was not different between groups (80.8 vs 85.2 vs 71.2%, P=0.23). At latest follow-up, significantly more RVAD patients had moderate or worse RV dysfunction (46.9 vs 70.4 vs 76.0%, P=1).
Conclusions: Preemptive RVAD insertion at time of HM3 may improve outcomes in patients who are at high risk for developing RV failure.

Authors
Krushang Patel (1), Alice Vinogradsky (2), Melissa Hynds (3), Yuji Kaku (4), Melana Yuzefpolskaya (4), Paolo Colombo (5), Gabriel Sayer (6), Nir Uriel (5), Yoshifumi Naka (7), Koji Takeda (3)
Institutions
(1) N/A, N/A, (2) Columbia University, United States, (3) Columbia University, New York, NY, (4) Columbia University Medical Center, New York, NY, (5) Columbia University Irving Medical Center, New York, NY, (6) Columbia University Medical Center, NYC, NY, (7) New york presbyterian, New York, NY 

(Case video)
Sternum-sparing HeartMate 3 implantation simultaneous with aortic and mitral procedures
(Case)
64 years old, male
Fulminant myocarditis, 56-days Impella 5.5 support
Aortic valve pathology: deformed leaflets due to passage of Impella
Mitral valve pathology: intrinsic degenerative mitral valve regurgitation
(operation)
MICS-HeartMate 3 implantation (pump insertion: left 5th ICS, outflow graft anastomosis: right 3rd ICS) with AVP(modified Park's stitch) and MVR
(Results)
skin incision 8cm (right 3rd ICS) and 9cm (left 5th ICS)
Doing well at home 6 months after the surgery

Authors
Akira Shiose (1), Tomoki Ushijima (1), Hiromichi Sonoda (1)
Institutions
(1) Kyushu University Hospital, Fukuoka, JAPAN 

Objective: To evaluate outcomes of patients directly bridged with Veno-Arterial Extra-Corporeal Membrane Oxygenation (VA ECMO) to heart transplantation.
Methods: A single center retrospective study was realized on 1152 adult patients undergoing isolated cardiac transplantation between January 2007 and December 2021. Among these, patients bridged with a VA ECMO to transplantation (group ECMO, n=317) were compared to the other transplanted patients (group NoECMO, n=835). A period analysis (Era I 2007-13, n=97 vs Era II 2014-21, n=220) was performed for patients of group ECMO.
Results: Median duration of VA ECMO support before transplantation in the group ECMO was 8 days [4.00; 13.0]. Recipients of group ECMO were younger, with a better renal function, and a shorter time on waiting list before donor allocation. They were allocated to younger donors, with a longer ischemic time. Group ECMO and NoECMO showed similar 1-year mortality and 5-years survival: 21.07% vs. 21.13%, p=0.98, and 66.9% (95% CI 63.8%; 70.3%) vs. 67.8% (95% CI 62.8%; 73.3%, p=0.76). Period analysis of group ECMO showed improved 1-year mortality and 5-years survival in ERA II compared to ERA I: 18% vs. 29%, p=0.025 and 70.7% (95% CI 64.7%; 77.3%) vs. 60.8% (95% CI 51.8%; 71.4%, p=0.043). Better outcomes in ERA II were explained by the higher number of patients in whom ECMO support was systematically pursued after the transplantation (92% vs. 48%, p<0.001), mainly by a peripheral cannulation (99.51% vs. 57%, p<0.001).
Conclusions: VA ECMO as a direct bridge for heart transplantation shows similar outcomes than other transplanted patients, provided that time on waiting list could be short and systematic VA ECMO temporary support, with a peripheral approach, could be pursued after transplantation.

Authors
mojgan laali (1), guillaume coutance (2), Shaida Varnous (3), guillaume hekimian (4), cosimo d'alessandro (4), pierre demondion (5), guillaume lebreton (5), pascal leprince (5)
Institutions
(1) Pitie-Salpetrier Hospital, 0, France, (2) La Pitié - Salpêtrière, paris, france, (3) La Pitié - Salpêtrière, Paris, france, (4) La Pitié - Salpêtrière, Paris, France, (5) La Pitié - Salpêtrière Hospital, paris, NA