Five-Year Outcome After Continuous Flow LVAD With Full-Magnetic (HeartMate 3) Versus Hybrid Levitation System (HeartWare): A Propensity-Score Matched Study From an All-Comers Multicentre Registry.

Despite the withdrawal of the HeartWare Ventricular Assist Device (HVAD), hundreds of patients are still supported with this continuous-flow pump, and the long-term management of these patients is still under debate. This study aims to analyse 5 years survival and freedom from major adverse events in patients supported by HVAD and HeartMate3 (HM3). From 2010 to 2022, the MIRAMACS Italian Registry enrolled all-comer patients receiving a LVAD support at seven Cardiac Surgery Centres. Out of 447 LVAD implantation, 214 (47.9%) received HM3 and 233 (52.1%) received HVAD. Cox-regression analysis adjusted for major confounders showed an increased risk for mortality (HR 1.5 [1.2-1.9]; p = 0.031), for both ischemic stroke (HR 2.08 [1.06-4.08]; p = 0.033) and haemorrhagic stroke (HR 2.6 [1.3-4.9]; p = 0.005), and for pump thrombosis (HR 25.7 [3.5-188.9]; p < 0.001) in HVAD patients. The propensity-score matching analysis (130 pairs of HVAD vs. HM3) confirmed a significantly lower 5 years survival (81.25% vs. 64.1%; p 0.02), freedom from haemorrhagic stroke (90.5% vs. 70.1%; p < 0.001) and from pump thrombosis (98.5% vs. 74.7%; p < 0.001) in HVAD cohort. Although similar perioperative outcome, patients implanted with HVAD developed a higher risk for mortality, haemorrhagic stroke and thrombosis during 5 years of follow-up compared to HM3 patients.

Corrigendum: Five-Year Outcome After Continuous Flow LVAD With Full-Magnetic (HeartMate 3) Versus Hybrid Levitation System (HeartWare): A Propensity-Score Matched Study From an All-Comers Multicentre Registry.

[This corrects the article DOI: 10.3389/ti.2023.11675.].

In-silico hemodynamic ramp testing of ventricular assist device implanted patients using acausal cardiovascular-VAD modeling.

BACKGROUND: While cardiovascular system and mechanical circulatory support devices are efficiently model the effect of disease and assistance, they can also lend valuable insights into clinical procedures. This study demonstrates the use of a CVS-VAD model for an invasive procedure; hemodynamic ramp testing, in-silico. METHODS: The CVS model is developed using validated models in literature, using Simscape™. An analytically derived pump model is calibrated for the HeartWare VAD. Dilated cardiomyopathy is used as an illustrative example of heart failure, and heart failure patients are created virtually by calibrating the model with requisite disease parameters obtained from published patient data. A clinically applied ramp study protocol is adopted whereby speed optimization is performed following clinically accepted hemodynamic normalization criteria. Trends in hemodynamic variables in response to pump speed increments are obtained. Optimal speed ranges are obtained for the three virtual patients based on target values of central venous pressure (CVP), pulmonary capillary wedge pressure (PCWP), cardiac output (CO), and mean arterial pressure (MAP) for hemodynamic stabilization. RESULTS: Appreciable speed changes in the mild case (300 rpm), slight changes in the moderate case (100 rpm), and no changes in the simulated severe case are possible. CONCLUSION: The study demonstrates a novel application of cardiovascular modeling using an open-source acausal model, which can be potentially beneficial for medical education and research.

Microembolic signal monitoring in patients with HeartMate 3 and HeartWare left ventricular assist devices: Association with antithrombotic treatment and cerebrovascular events.

BACKGROUND: In patients with left ventricular assist devices (LVADs), ischemic and hemorrhagic stroke are dreaded complications. Predictive markers for these events are lacking. This study aimed to investigate the prevalence and predictive value of microembolic signals (MES) for stroke, detected by Transcranial Doppler sonography (TCD) in patients with HeartMate 3 (HM 3) or HeartWare (HW). METHODS: A thirty-minute bilateral TCD monitoring of the middle cerebral artery (MCA) was performed in 62 outpatients with LVAD (HM 3 N = 31, HW N = 31) and 31 healthy controls. Prevalence and quantity of MES were investigated regarding clinical and laboratory parameters. Cerebrovascular events (CVE) were recorded on follow-up at 90 and 180 days. RESULTS: MES were detected in six patients with HM 3, three patients with HW, and three controls. Within the LVAD groups, patients on monotherapy with vitamin-K-antagonist (VKA) without antiplatelet therapy were at risk for a higher count of MES (negative binomial regression: VKA: 1; VKA + ASA: Exp(B) = 0.005, 95%CI 0.001-0.044; VKA + clopidogrel: Exp(B) = 0.012, 95%CI 0.002-0.056). There was no association between the presence of MES and CVE or death on follow-up (p > 0.05). CONCLUSION: For the first time, the prevalence of MES was prospectively investigated in a notable outpatient cohort of patients with HM 3 and HW. Despite optimized properties of the latest LVAD, MES remain detectable depending on antithrombotic therapy. No association between MES and CVE could be detected.

Cerebral protection during percutaneous intervention for left ventricular assist device outflow graft obstruction.

BACKGROUND: Left ventricular assist devices (LVAD) outflow graft obstruction is an uncommon complication but carries significant morbidity and mortality. Here we provide a case series of patients with LVAD intrinsic outflow graft obstruction who are deemed to be a high surgical risk for pump exchange and, therefore, underwent percutaneous intervention with the concomitant use of neuroprotective device-Sentinel cerebral protection system (CPS) (Boston Scientific) to prevent embolic stroke. METHODS: We retrospectively analyzed patients who underwent LVAD placement in our institution and developed LVAD outflow graft obstruction. The diagnosis of LVAD outflow graft obstruction was confirmed by utilizing various cardiac imaging modalities such as echocardiography and/or computed tomography angiography. All patients were treated with percutaneous intervention and a catheter-based CPS. RESULTS: From a total of 501 LVAD implants in our institute, 6 (1.2%) patients with LVAD outflow graft obstruction who underwent percutaneous treatment were included; 4 patients with HeartMate-III LVAD, 1 patient with HeartMate-II LVAD, and 1 patient with HeartWare (HVAD). The median age of patients was 56.5 years at the time of LVAD implantation. The median time from the LVAD implantation to the episode of LVAD outflow obstruction was 1343 days. Utilization of Sentinel CPS resulted in the capture and removal of thrombus/debris in all patients. CONCLUSIONS: Percutaneous intervention of LVAD outflow graft obstruction is less invasive than surgical pump exchange and an acceptable alternative in properly selected patients. In our experience, utilization of a catheter-based CPS can help in reducing the incidence of periprocedural embolic events.

HVAD versus heartmate III bridge to heart transplantation: Waitlist and posttransplant outcomes.

BACKGROUND: This study compared outcomes of patients bridged with either the Heartware HVAD or Heartmate 3 (HM3) device to orthotopic heart transplantation (OHT). METHODS: The United Network of Organ Sharing registry was queried to perform two separate analyses of adult, isolated OHT candidates bridged with HVAD or HM3. First, waitlist outcomes were compared among patients waitlisted 1/1/2015-3/20/2020. Second, posttransplant survival was compared among those transplanted 1/1/2015-3/20/2020. RESULTS: Two thousand two hundred fifty-five candidates were waitlisted within the study period, 1587 (70.4%) bridged with HVAD and 668 (29.6%) with HM3. At 1 year from waitlisting, cumulative incidence of OHT higher in the HVAD cohort (p < .001). During the same time period, 2643 patients underwent OHT, 2154 (81.5%) with prior HVAD and 489 (18.5%) with HM3. Yearly proportions of patients bridged with HM3 increased across the study period and decreased for HVAD (p < .001). HM3-bridged recipients had shorter waitlist times, longer graft cold ischemic times, and experienced a higher rate of posttransplant dialysis requirement. Unadjusted and risk-adjusted posttransplant mortality rates were similar between both groups. CONCLUSIONS: Posttransplant survival is equivalent regardless of device type used for bridging. However, HM3 patients had lower likelihood of reaching transplantation, which may be a reflection of the recent heart allocation policy changes.

Does infection predispose to thrombosis during long-term ventricular assist device support?

BACKGROUND: Infections and thrombotic events remain life-threatening complications in patients with ventricular assist devices (VAD). METHODS: We describe the relationship between both events in our cohort of patients (n = 220) supported with the HeartWare VAD (HVAD). This is a retrospective analysis of patients undergoing HVAD implantation between July 2009 and March 2019 at the Freeman Hospital, Newcastle upon Tyne, United Kingdom. RESULTS: Infection was the most common adverse event in HVAD patients, with 125 patients (56.8%) experiencing ≥ one infection (n = 168, 0.33 event per person year (EPPY)), followed by pump thrombosis (PT) in 61 patients (27.7%, 0.16 EPPY). VAD-specific infections were the largest group of infections. Of the 125 patients who had an infection, 66 (53%) had a thrombotic event. Both thrombotic events and infections were related to the duration of support, though there was only limited evidence that infections predispose to thrombosis. Those with higher than median levels of C-reactive protein during the infection were more likely to have an ischaemic stroke (IS) (34.5% vs 16.7%, p = .03), though not PT or a combined thrombotic event (CTE: first PT or IS). However, in multivariate analysis, there was no significant effect of infection predisposing to CTE. CONCLUSIONS: Infection and thrombotic events are significant adverse events related to the duration of support in patients receiving HVADs. Infections do not clearly predispose to thrombotic events.

Catheter-based system for the treatment of left ventricular assist device thrombosis.

BACKGROUND: Thrombotic complications continue to pose challenges to patients on left ventricular assist device (LVAD) support. The Hoplon system was developed to administer catheter-based lytic therapy with a novel approach to embolic protection. METHODS: Two porcine non-survival surgeries were performed in which off-pump LVAD insertion was followed by injection of thrombus into the impeller, isolation of the pump using the Hoplon system, and administration of lytic therapy to the pump chamber. Successful thrombus resolution was confirmed by pathological examination of the LVAD and brain tissue after animal sacrifice. RESULTS: Limitations of the prototype design resulted in the extrusion of thrombus from around the catheter in the first animal. Subsequent device modifications resulted in the resolution of LVAD thrombus as confirmed on removal and examination of the pump. Pathological examination of the brain tissue revealed the absence of any embolic or hemorrhagic complications. CONCLUSIONS: Early animal studies suggest feasibility in restoring function to an LVAD while at the same time preventing cerebroembolic events using the Hoplon system.

Retrograde washout of prepump LVAD thrombosis in a patient on HeartWare™ support.

The success of the left ventricular assist device (LVAD) as a treatment for terminal left-side heart failure is still restrained by some severe complications associated with mechanical circulatory support. Pump thrombus still affects many patients. It is associated with high morbidity and mortality. The therapeutic options include augmentation of anticoagulation and antiplatelet medication, intravenous or catheter-guided thrombolysis, and pump exchange. Heart transplantation would be a desirable option in this population, but unfortunately, it is only theoretical given the increasing number of LVAD implants and decreasing number of organ donors. A retrograde washout maneuver may be a treatment option in prepump thrombosis in selected patients. Therefore, the decision should be made on an individual basis after balancing the risks and benefits of different treatment approaches. In this context, we report a case of retrograde washout of prepump thrombus in a patient who has been on HeartWare™ support for more than 3 years, with a successful bailout strategy.

Recovery of a dilated left ventricle after cessation of cocaine and HVAD™ explantation using a titanium plug.

The ultimate goal in the treatment of end-stage heart failure is the recovery of cardiac function following mechanical assistance of the left ventricle. The HVAD™ pump (HeartWare Inc.) left ventricular assist device (LVAD) can be explanted without resternotomy. This article demonstrates that the use of a custom-made mechanical plug (manufactured by INNOVO Solutions GmbH), which can be inserted into the LVAD's sewing ring, is feasible. This mechanical plug explicitly designed for device explantation is a viable alternative to the current standard of care. This article adopts a less invasive technique to explant the pump. The following case illustrates this technique.

Identification of characteristics, risk factors, and predictors of recurrent LVAD thrombosis: conditions in HeartWare devices.

BACKGROUND: Redictors of repetitive left-ventricular assist device (LVAD)-thrombosis have not been studied yet. METHODS: We identified predictors of recurrent LVAD thrombosis in HeartWare (HVAD) patients in a long-term study from 2010 until 2020. We included all patients with two or more thrombolysis treatments for repetitive HVAD thrombosis and effectiveness of thrombolytic therapy was defined as freedom from stroke, death, another HVAD thrombosis, or surgical device exchange within 30 days after the event. Study endpoints also include all-cause mortality and heart transplantation. RESULTS: A total of 534 HVAD implantations have been screened, and 73 patients (13.7%) developed first HVAD thrombosis after a median of 10 months (IQR; 6-21 months). 46 of these patients had effective thrombolysis in 71.7% (n = 33/46). After a median of 14 months (IQR 4-32 months) follow-up, 17 patients (51.5%) had developed a second HVAD thrombosis and all were treated with t-PA therapy again, resulting in effectiveness in 76.5% (n = 13/17). The four patients with ineffective t-PA therapy underwent subsequent surgical HVAD exchange. Multiple Cox regression model analysis revealed time interval between HVAD implantation and first thrombosis as an independent risk factor of recurrent thrombosis (HR, 0.93, 95% CI 0.87-0.99, p = 0.031). Kaplan-Meier analysis at 3 year follow-up showed no significant difference in overall survival for recurrent vs non-recurrent thrombosis groups (log-rank test, p = 0.959). CONCLUSION: Recurrent HVAD thrombosis mostly appears within 12 months after first thrombosis. Systemic t-PA therapy for recurrent pump thrombosis seems safe, achieving comparable effectiveness rates to initial t-PA therapy. Survival does not differ between patients with or without recurrent HVAD thrombosis.

A Novel Method to Safely De-Air a HeartWare System in a Single-Ventricle Patient by Utilizing ECMO and a Minimized CPB Circuit.

The survival of congenital heart disease (CHD) patients with single-ventricle (SV) physiology has markedly increased as a result of advances in operative techniques and postsurgical management. Nonetheless, these patients remain highly susceptible to end-stage heart failure requiring cardiac replacement therapies at early ages. Given a worldwide shortage of transplantable organs, mechanical circulatory support (MCS) represents an alternative treatment option. The significant heterogeneity of the SV population presents unique indications for MCS that have begun to be evaluated. This case study describes a 12-year-old female with heterotaxy syndrome and an SV condition, previously palliated with a Fontan operation at another institution. The patient was placed on veno-arterial (VA) extracorporeal membrane oxygenation (ECMO) during prolonged cardiopulmonary resuscitation, and later underwent HeartWare ventricular assist device (HVAD) implantation as a bridge to transplantation (BTT). A novel method was chosen to optimize careful de-airing of the heart through a minimized cardiopulmonary bypass (CPB) setup, during full ECMO support and surgical insertion of the HeartWare. The ascending aorta was vented proximal to the HVAD outflow graft anastomosis through a minimized CPB circuit at <10% of the ECMO flow rate. This circuit adaption allowed for euvolemic resuscitation via connection from the minimized CPB circuit to the venous limb of the ECMO circuit. The transition from VA-ECMO to the HeartWare was well tolerated despite a challenging sternotomy and cardiac anomaly. A minimized bypass circuit proved efficacious for the benefit of volume resuscitation and safe de-airing of the HVAD while on ECMO support. The literature is limited concerning safe practices for implantation of durable VADs in complex SV patients coupled with those transitioning from varying modalities of MCS. As SV survivability regresses to heart failure, it is essential that we share techniques that aim to improve the long-term outcomes for successful BTT or bridge to decision (BTD).

HVAD Flow Waveform Estimates Left Ventricular Filling Pressure.

BACKGROUND: HVAD left ventricular assist device flow waveforms provides graphical real-time information linking device performance with invasive hemodynamics. Previous studies have demonstrated a good correlation between the slopes of the ventricular filling phase slope (VFPS) and directly measured pulmonary capillary wedge pressure (PCWP). We aimed to validate the utility of VFPS to estimate PCWP and predict clinical outcomes. METHODS: In this prospective blinded study, screenshots from the HVAD monitor and simultaneous invasive hemodynamic measurements were obtained. Each screenshot was digitized and the VFPS was calculated by 2 independent reviewers who were blinded to the hemodynamic results. The equation PCWP = 7.053 +1.365 × (VFPS) was derived from a previously published dataset and the estimated PCWP was correlated to the actually measured PCWP. RESULTS: One hundred thirty-one sets of simultaneous measurements (VFPS and PCWP) were obtained from 27 HVAD patients (mean age 55 years, 47% male). A previously proposed cutoff of VFPS ≥5.8 L/min/s predicted PCWP ≥ 18 mmHg with 91.5% sensitivity and 95.2% specificity with the area under curve of 0.987. The estimated PCWP significantly correlated with measured PCWP (R2 = 0.65, P < .001) and showed acceptable agreement with measured PCWP. Patients with VFPS ≥ 5.8 L/min/s experienced significantly higher heart failure readmission rates than those without (0.24 vs 0.05 events/y, P < .001). CONCLUSIONS: VFPS of the HVAD flow waveform is a novel noninvasive parameter that can estimate PCWP.

Characteristics and outcomes of gastrointestinal bleeding in patients with continuous-flow left ventricular assist devices: A systematic review.

Gastrointestinal bleeding (GIB) is a common adverse event after continuous-flow left ventricular assist device (CF-LVAD) implantation. We sought to evaluate patterns of GIB development and related outcomes in CF-LVAD recipients. An electronic search was performed to identify all articles related to GIB in the setting of CF-LVAD implantation. A total of 34 studies involving 1087 patients were pooled for analysis. Mean patient age was 60 years (95% CI 57-64) and 24% (95% CI 21-28%) were female. The mean time from CF-LVAD implantation to the first GIB was 54 days (95% CI 24-84) with 40% (95% CI 34-45%) of patients having multiple episodes of GIB. Anemia was present in 75% (95% CI 41-93%) and the most common etiology of bleeding was arteriovenous malformations (36% [95% CI 24-50%]). The mean duration of follow-up was 14.6 months (95% CI 6.9-22.3) during which the all-cause mortality rate was 21% (95% CI 12-36%) and the mortality rate from GIB was 4% (95% CI 2-9%). Thromboembolic events occurred in 32% (95% CI 22-44%) of patients with an ischemic stroke rate of 16% (95% CI 3-51%) and a pump thrombosis rate of 8% (95%CI 3-22%). Heart transplantation was performed in 31% (95% CI 18-47%) of patients, after which 0% (95% CI 0-10%) experienced recurrent GIB. GIB is a major source of morbidity among CF-LVAD recipients. While death due to GIB is rare, cessation of anticoagulation during treatment increases the risk of subsequent thrombotic events. Heart transplant in these patients appears to reliably resolve the risk of future GIB.

Protek Duo Rapid Deployment cannulation utilized to bridge to durable heartware left ventricular assist device.

A 43-year-old gentleman was transferred for management of acute on chronic cardiogenic shock (left ventricular ejection fraction < 10%). Upon arrival, we inserted a left axillary intra-aortic balloon pump for hemodynamic support. He underwent an emergent left and right-heart catheterization which showed patent stents and coronaries, in the setting of severely elevated pulmonary artery and pulmonary capillary wedge pressure. On hospital day 35, we escalated support to Centrimag in conjunction with a 31 French Protek Duo Rapid Deployment cannula. A centrimag cannula apical sewing cuff was sewn in continuous fashion along the left ventricular apex. Via modified seldinger technique, we tunneled the Protek Duo Rapid Deployment cannula through the silastic sewing cuff and the ventricular apex, traversing the aortic valve. On hospital day 50, he underwent left anterior thoracotomy and mini-sternotomy for implantation of durable Heartware left ventricular assist device. He was discharged home off inotropes and had resumed his normal activities. He is currently listed as status four for heart transplantation.

Minimally Invasive Implantation of Continuous Flow Left Ventricular Assist Devices: The Evolution of Surgical Techniques in a Single-Center Experience.

In this work we aimed to evaluate the evolution of our surgical experience with the implantation of a continuous flow left ventricular assist device (LVAD), from the original full sternotomy approach to less invasive surgical strategies including mini-sternotomy and/or mini-thoracotomies. We reviewed all consecutive patients implanted with a continuous flow LVAD at our Institute. To exclude the possible bias related to the device used, out of 91 collected LVADs implants, we selected only those patients (n = 42) who received, between 2012 and 2015, the HeartWare HVAD. The analysis focused on the surgical approach used for the LVAD implant. Most of the patients (95%) were affected by dilated or ischemic cardiomyopathy, with an INTERMACS class I-II in the majority of cases (77%). The LVAD implant was performed through a full sternotomy in 10 patients (24%); the remaining 32 cases (76%) were managed with minimally invasive procedures. These were left mini-thoracotomy with upper mini-sternotomy (20 patients, 62%), right and left mini-thoracotomy (7 patients, 22%), and a recently developed left mini-thoracotomy with outflow graft anastomosis to the left axillary artery (5 patients, 16%). The most common adverse event on device was right heart failure (26%). Eighteen patients (43%) were transplanted. Overall estimated 24 months survival (on device or after transplant) was 68 ± 7%. The causal analysis, adjusted by propensity score weighting baseline data and sample size, showed that left mini-thoracotomy with outflow anastomosis to the left axillary artery resulted in a significantly reduced rate of post implant right heart failure (P < 0.01), and mechanical ventilation time (P = 0.049). To conclude, in our series, by applying mini-invasive implant techniques in the majority of cases, mid-term survival of continuous flow LVADs in severely compromised patients was satisfactory. In the adjusted analysis, the left anterior mini-thoracotomy with outflow anastomosis to the left axillary artery showed the most favorable results.

Stroke Risk Following Infection in Patients with Continuous-Flow Left Ventricular Assist Device.

BACKGROUND: Infection has been associated with stroke in patients with left ventricular assist devices (LVAD); however, little data exist on the timing, type and mortality impact of infection-related stroke. METHODS: Prospectively collected data of HeartMate II (N = 332) and HeartWare (N = 70) LVAD patients from a single center were reviewed. Only strokes (ischemic or hemorrhagic) that occurred within 6 weeks following a LVAD infection were considered in analyses. The association between LVAD infections (wound, pump pocket, driveline and/or bloodstream infection [BSI]), specific pathogens and ischemic and hemorrhagic strokes was evaluated using multivariable logistic regression analysis. The impact of infection-related stroke on cumulative survival was assessed using Kaplan-Meier analysis. RESULTS: Of 402 patients, LVAD infection occurred in 158 (39%) including BSI in 107 (27%), driveline infection in 67 (17%), wound infection in 31 (8%) and pump pocket infection in 24 (6%). LVAD infection-related stroke occurred in 20/158 (13%) patients in a median of 4 days (0-36 days) from documented infection. In multivariable analysis, ischemic stroke was associated with wound infection (aOR 9.0, 95% CI 2.4-34.0, P = 0.001) and BSI (aOR 7.7, 95% CI 0.9-66.0, P = 0.064), and hemorrhagic stroke was associated with BSI in 100% of cases (P = 0.01). There was no association with driveline or pump pocket infection. The cumulative survival rate among patients with infection-related stroke was significantly lower compared to those with LVAD infection but no stroke (log-rank P < 0.001). There was a trend toward shorter stroke-free survival among patients with LVAD infection. CONCLUSIONS: LVAD infections, particularly BSI, are significantly associated with stroke, and infection-related stroke conferred significantly lower cumulative survival.

Device exchange from Heartmate II to HeartWare HVAD.

BACKGROUND: Despite technological advancements, pump durability and pump-related complications continue to affect and adversely impact the lives of patients with end-stage heart failure on left ventricular assist device (LVAD) support. In an attempt to avoid recurrent LVAD-related complications, there may be circumstances where it is clinically advantageous to exchange a patient's device from HeartMate II to HeartWare HVAD. However, there is a paucity of data that describes the safety and feasibility of such an approach. OBJECTIVE: We present the largest single-center series of HeartMate II (HMII) to HeartWare (HVAD) device exchanges. METHODS: A retrospective review of 11 patients who underwent HMII to HVAD exchange from 2012 to 2017 was conducted to evaluate patient characteristics, incidence of postoperative complications, and survival. RESULTS: Eleven male patients (mean age 55 ± 14.4 years) underwent HMII to HVAD device exchange. One patient expired on postoperative day 7 secondary to sepsis. One patient was lost-to-follow-up after 23 months. An additional three patients died at 5, 7, and 24 months. Mean follow-up after device exchange was 1555 ± 311 days for the remaining six patients. None of the 11 study patients underwent LVAD explant, further device exchange, or heart transplant. CONCLUSION: Exchange of an HMII LVAD to an HVAD can be performed safely with acceptable perioperative morbidity and mortality.

Clinical Outcome and Comparison of Three Different Left Ventricular Assist Devices in a High-Risk Cohort.

We present a comparison between three left ventricular assist devices (LVADs): HeartWare (HVAD) (HeartWare International Inc., Framingham, MA, USA), HeartMate II (HMII) and HeartMate III (HMIII) (Thoratec Corp., Pleasanton, CA, USA). To our knowledge, no study to date has aimed at placing these three devices in juxtaposition. Between June 2007 and June 2017, 108 consecutive patients received HMII, n = 77 (71.3%), HVAD, n = 14 (13%), or HM III, n = 17 (15.7%), for end-stage heart failure. Mean age was 63.8 ± 11.2 years (range 24-84 years), with median INTERMACS profile of 3. Preoperatively, 26 patients (24.1%) were ventilated, 17 patients (15.7%) had an intraaortic balloon pump, and 27 patients (25%) were on extracorporeal life support. Overall survival at 30 days was 70.4%, at 1 year 51.9%, and at 5 years 38% with no significant difference in survival between HMII, HVAD, and HMIII. Median cardiopulmonary bypass time was 113 min (range 50-371 min). Two patients received a minimally-invasive procedure. Most common adverse events were revision for bleeding (42.6%), tracheotomy (33.3%), acute kidney failure with new-onset dialysis (25%), sepsis (17.6%), and gastrointestinal bleeding (10.2%). The average duration of follow-up was 1.52 ± 2.11 years (range 0-7.95 years). The median number of readmissions was 2 (range 0-23), the median length of hospital stay as readmission was 17 days (range 0-158 days). Strong predictors of overall mortality (P < 0.05) were postoperative sepsis (OR = 5.729, 95%CI = 3.001-10.937), intraoperative/postoperative need for right ventricular mechanical support (OR = 5.232, 95%CI = 3.008-9.102), preoperative extracorporeal life support (OR = 2.980, 95%CI = 1.615-5.500), readmission because of suboptimal INR value (OR = 2.748, 95%CI = 1.045-7.226), need of inotropes over 7 days postoperatively (OR = 2.556, 95%CI = 1.432-4.562), new onset of temporary hemodialysis postoperatively (OR = 1.986, 95%CI = 1.084-3.635), and female gender (OR = 1.955, 95%CI = 1.062-3.598). No significant difference in mortality between HMII, HVAD, and HMIII was observed. The following predictors of overall mortality were identified (P < 0.05): postoperative sepsis, need for perioperative mechanical support, readmission because of suboptimal INR value, new onset of temporary hemodialysis postoperatively and female gender.