The Impact of MOMENTUM 3 Trial Eligibility on Left Ventricular Assist Device Outcomes: A Real-World Experience.

INTRODUCTION: Although the landmark MOMENTUM 3 trial was associated with excellent short-term left ventricular assist device (LVAD) outcomes, many end-stage heart failure patients would not have met the trial eligibility criteria. Moreover, the outcomes of trial ineligible patients are poorly characterized. Therefore, we undertook this study to compare MOMENTUM 3 eligible and ineligible patients. METHODS: We conducted a retrospective review of all primary LVAD implants from 2017 to 2022. Primary stratification was according to MOMENTUM 3 inclusion and exclusion criteria. Primary outcome was survival. Secondary outcomes included complications and length of stay. Multivariable Cox proportional hazards regression models were constructed to further characterize outcomes. RESULTS: From 2017 to 2022, 96 patients underwent primary LVAD implantation. Thirty-seven (38.54%) patients were trial eligible while 59 (61.46%) were ineligible. When stratified by trial eligibility, patients who were trial eligible had higher 1-year (80.15% versus 94.52%, P = 0.04) and 2-year survival (70.17% versus 94.52%, P = 0.02). Multivariable analysis showed that trial eligibility was protective of mortality at both 1 y (HR: 0.19 [0.04-0.99], P = 0.049) and 2 y (HR: 0.17 [0.03-0.81], P = 0.03). Although the groups had similar rates of bleeding, stroke, and right ventricular failure, trial ineligibility was associated with a longer periprocedural length of stay. CONCLUSIONS: In conclusion, the majority of contemporary LVAD patients would not have been eligible for the MOMENTUM 3 trial. Ineligible patients have decreased but acceptable short-term survival. Our findings suggest that a simply reductionist approach to short-term mortality may improve outcomes but fail to capture the majority of patients who could benefit from therapy.

Novel Estimates of Renal Function are Associated with Short-Term Left Ventricular Assist Device Outcomes.

INTRODUCTION: Although preoperative kidney function has been associated with left ventricular assist device (LVAD) outcomes, most previous estimates of glomerular filtration rates (eGFRs) have utilized race in the calculation. Recently, novel eGFR equations independent of race have been suggested and validated. Therefore, we undertook this study to evaluate the predictive value of a novel, non-race-based eGFR calculation on short-term LVAD outcomes. METHODS: We conducted a retrospective review of all primary LVAD implants from 2017 to 2022 at our institution. eGFR was calculated using the novel Chronic Kidney Disease Epidemiology Collaboration 2021 formula (CKD-EPI 2021). eGFR was also calculated according to the Modification of Diet in Renal Disease equation for historical reference. Primary stratification was by eGFR: ≥60, 30-60, and <30. The primary outcome was 1-y survival. Multivariable Cox proportional hazards regression modeling was used to further evaluate the impact of kidney function on 1-y mortality. RESULTS: From 2017 to 2022, 91 patients underwent LVAD implantation with a HeartMate 3 device. The average age was 65.20 ± 11.08, 77 (84.62%) were male, and 14 (15.38%) were Black. The mean CKD-EPI 2021 eGFR was 56.07 ± 23.55 compared with 54.72 ± 26.37 as calculated by Modification of Diet in Renal Disease (P = 0.719). Overall, 30-d and 1-y survival was 96.7% and 85.0%, respectively. When stratified by eGFR, there was a significant difference in 1-y survival (≥60, 93.46%; 30-60, 87.36%; <30, 62.75%; P = 0.016). On multivariable analysis, a preoperative eGFR <30 was associated with an increased hazard of 1-y mortality (5.58 [1.06-29.17], P = 0.043). CONCLUSIONS: In conclusion, non-race-based estimates of renal function are predictive of short-term LVAD outcomes. Further investigation of this phenomenon is warranted.

Temporary Right Ventricular Assist Device Support for Acute Right Heart Failure: A Single-Center Experience.

INTRODUCTION: Severe right ventricular (RV) failure is associated with significant morbidity and mortality. Although right ventricular assist devices (RVADs) are increasingly used for refractory RV failure, there is limited data on their short- and long-term outcomes. Therefore, we undertook this study to better understand our experience with temporary RVADs. METHODS: We conducted a retrospective review of all RVADS performed from 2017 to 2021. Patients supported with surgical RVADs, the Protek Duo device, and the Impella RP device were included. Patients were stratified by the type of RVAD and by etiology of RV failure. Survival was assessed by the Kaplan-Meier method and multivariable Cox proportional hazards regression models. RESULTS: From 2017 to 2021, 42 patients underwent RVAD implantation: 32 with a Protek Duo, 6 with an Impella RP, and 4 with a surgical RVAD. Majority of patients were already supported with an alternate form of mechanical support. Most patients had impaired renal function, decreased hepatic function, and lactic acidosis at the time of cannulation. The median duration of RVAD support was 8.5 [5-19] d. Survival to decannulation was 68.4%, to discharge was 47.4%, and to 1-y was 40.2%. Multivariable analysis identified elevated total bilirubin levels to be associated with 30-d mortality while increased hemoglobin levels were protective. After RVAD cannulation, the median number of pressors and inotropes was lower (P < 0.01) and the lactic acidosis was less (P < 0.01). CONCLUSIONS: In conclusion, RVAD support is associated with lower lactate levels, and decreased number of vasoactive medications, but is associated with significant morbidity and mortality.

Development of a non-transplant left ventricular assist device program.

INTRODUCTION: Although the established long-term benefit of left ventricular assist device (LVAD) therapy has led to its proliferation as destination therapy (DT), few studies have evaluated LVAD outcomes at nontransplant centers. We undertook this study to better evaluate our experience in building a nontransplant, DT LVAD program. METHODS: We conducted a retrospective review of all LVADs implanted from 2010 to 2021. Patient, operative, and outcome data were extracted from the electronic medical record. Secular trends were evaluated by organizing the data into eras of implant. Survival was assessed using the Kaplan-Meier method. Multivariable Cox proportional hazards regression models further evaluated outcomes. RESULTS: From 2010 to 2021, 100 primary LVAD implants were performed. Annual volume grew from 1 to 30 implants per year. The average age of our cohort was 65.7 years, most patients (80%) were male, 51% had an ischemic etiology, and 65 (65%) were INTERMACS profile 1 or 2. Our 1- and 2-year survival were 82% and 79%, respectively. Multivariable analysis of 1-year mortality demonstrated that decreasing renal function and increased cardiopulmonary bypass (CPB) time were associated with increased mortality while preoperative hemoglobin was protective. When stratified by era of implant, our most recent patients were more likely to be INTERMACS profile 1 or 2; had shorter CPB and aortic cross clamp times; required fewer reoperations for bleeding; and suffered less right ventricular failure requiring mechanical support. CONCLUSIONS: A single, nontransplant LVAD center can experience significant growth in volume in a high-acuity cohort while maintaining acceptable outcomes and quality of care.

Impact of preoperative Impella support on destination left ventricular assist device outcomes.

BACKGROUND: Although left ventricular assist device (LVAD) implantation is associated with improved heart failure survival, the impact of pre-implantation Impella support on outcomes is unknown. We undertook this study to evaluate the impact of preoperative Impella support on LVAD outcomes. METHODS: We conducted a retrospective review of all Heartmate 3 LVAD implants. Primary stratification was by the need for preoperative Impella support with the 5.0/5.5 device. Longitudinal survival was assessed by the Kaplan-Meier method. Multivariable Cox proportional hazards regression models were developed to evaluate mortality. Secondary outcomes included changes in laboratory values during Impella support. RESULTS: From 2017 to 2021, 87 patients underwent LVAD implantation. Sixteen were supported with a single inotrope, 36 with dual inotropes, 27 with Impella, and 3 with extracorporeal membrane oxygenation (ECMO). When stratified by the need for Impella, there was no difference in survival at 30-days (98.3 [88.2-99.8]% vs. 96.3 [76.5-99.5]%, p = .59), 1-year (91.0 [79.8-96.2] vs. 74.9 [51.7-88.2], p = .10), or at 2 years (87.9 [74.3-94.5] vs. 74.9 [51.7-88.2], p = .15). On multivariable modeling, the need for preoperative Impella was not associated with an increased hazard of 1-year (1.24 [0.23-6.73], p = .81) or 2-year mortality (1.05 [0.21-5.19], p = .95). After 7 (5-10) days of Impella support, recipient creatinine (p < .01), creatinine clearance (p = .02), and total bilirubin (p = .053) improved and lactic acidosis resolved (p < .01). CONCLUSIONS: Preoperative Impella support is not associated with increased short or long-term mortality but is associated with improved renal and hepatic function as well as total body perfusion before LVAD implantation.

Elderly Patients With Higher Acuity Have Similar Left Ventricular Assist Device Outcomes as Younger Patients at a Nontransplant Center.

Although left ventricular assist device (LVAD) implantation is associated with acceptable survival, previous reports have demonstrated that advanced age is associated with increased short-term mortality. Because age is a relative contraindication to transplantation, nontransplant centers tend to implant a disproportionate number of elderly patients. We undertook this study to evaluate the impact of advanced age on LVAD outcomes at a nontransplant center. We conducted a retrospective review of all LVAD implants at our center from 2017 to 2022. Primary stratification was by age >70 years. The primary outcome was survival as assessed by the Kaplan-Meier method. The risk of 1-year mortality was further evaluated using multivariable Cox proportional hazards regression modeling. From 2017 to 2022, 93 patients underwent LVAD implantation. The mean age was 65.03 ± 11.28 years, with a median age of 68 (60 to 73) years. Most patients were INTERMACS 1 or 2 (71 patients; 76.34%). When stratified by age, 41 patients (44.09%) were aged ≥70 years. Patients aged ≥70 years had similar 30-day (96.15% vs 100.00%, p = 0.213), 1-year (90.05% vs 84.00%, p = 0.444), and 2-year survival (82.03% vs 84.00%, p = 0.870). When only the INTERMACS 1 and 2 patients with higher acuity were included, there was still no difference in 30-day, 1-year, or 2-year survival. On multivariable analysis, age >70 years was not associated with an increased hazard of 1-year mortality (0.90 [0.22 to 3.67], p = 0.878). In conclusion, in carefully selected patients, age >70 years is not associated with increased short-term mortality. Age alone should not be a contraindication to LVAD therapy.

Outcomes of surgical Impella placement in acute cardiogenic shock.

Introduction: Although a role for percutaneous Impella devices has been established, there is a paucity of data regarding the utility and outcomes of larger surgically implanted Impella devices. Methods: We conducted a retrospective review of all surgical Impella implants at our institution. All Impella 5.0 and Impella 5.5 devices were included. The primary outcome was survival. Secondary outcomes included hemodynamic and end-organ perfusion as well as commonly encountered surgical complications. Results: From 2012 to 2022, 90 surgical Impella devices were implanted. The median age was 63 [53-70] years, the mean creatinine was 2.07 ± 1.22 mg/dL, and the average lactate level was 3.32 ± 2.90 mmol/L. Prior to implantation, 47 patients (52%) were supported with vasoactive agents, while 43 (48%) were also supported with another device. The most common etiology of shock was acute on chronic heart failure (50, 56%), followed by acute myocardial infarction (22, 24%), and postcardiotomy (17, 19%). Overall, 69 patients (77%) survived to device removal, and 57 (65%) survived to hospital discharge. One-year survival was 54%. Neither etiology of heart failure nor device strategy was associated with 30-day or 1-year survival. On multivariable modeling, the number of vasoactive medications prior to device implantation was strongly associated with 30-day mortality (hazard ratio 1.94 [1.27-2.96], P < 0.01). Surgical Impella placement was associated with a significant decreased need for vasoactive infusions (P < 0.01) and decreased acidosis (P = 0.01). Conclusions: Surgical Impella support for patients in acute cardiogenic shock is associated with lower vasoactive medication use, improved hemodynamics, increased end-organ perfusion, and acceptable morbidity and mortality.

Managing Right Ventricular Failure After Left Ventricular Assist Device Implant at a Destination Therapy Center.

Although left ventricular assist device (LVAD) implant is associated with an increased survival in patients with end-stage heart failure, severe right ventricular failure requiring a right ventricular assist device (RVAD) placement is associated with increased short-term morbidity and mortality. Patients not eligible for transplant have limited options, which may impact decision-making and outcomes at nontransplant centers. We conducted a retrospective review of all LVAD implants at our nontransplant center. Primary stratification was by the need for a postoperative RVAD implant. The primary outcome was survival. The Cox proportional hazards regression modeling was used to further evaluate mortality. From 2017 to 2022, 128 patients underwent a primary LVAD implant and 24 (18.75%) required a perioperative RVAD placement. RVAD implant was associated with increased operative mortality (1.92% vs 33.33%, p <0.01) and decreased 1-year (91.29% vs 60.60%, p <0.01) and 2-year survival (84.05% vs 36.36%, p <0.01). However, in patients who survived their index hospitalization, 1-year (93.00% vs 91.67%, p = 0.78) and 2-year (86.16% vs 55.00%, p = 0.10) mortality were similar. On multivariable analysis, the need for a RVAD was associated with an increased hazard of 1-year (5.60 [1.96 to 16.01], p <0.01) and 2-year (5.17 [2.01 to 13.28], p <0.01) mortality. In conclusion, our series from a nontransplant center suggests that patients who survive the implant have acceptable short-term survival, even if they do not have a transplant option; thus, carefully selected patients with biventricular failure may benefit from an LVAD implant, even if an RVAD is needed.