Center Variability in Patient Outcomes Following HeartMate 3 Implantation: An Analysis of the MOMENTUM 3 Trial.

BACKGROUND: As left ventricular assist device (LVAD) survival rates continue to improve, evaluating site-specific variability in outcomes can facilitate identifying targets for quality-improvement initiative opportunities in the field. METHODS: Deidentified center-specific outcomes were analyzed for HeartMate 3 (HM3) patients enrolled in the MOMENTUM 3 pivotal and continued access protocol trials. Centers < 25th percentile for HM3 volumes were excluded. Variability in risk-adjusted center mortality was assessed at 90 days and 2 years (conditional upon 90-day survival). Adverse event (AE) rates were compared across centers. RESULTS: In the 48 included centers (1958 patients), study-implant volumes ranged between 17 and 106 HM3s. Despite similar trial-inclusion criteria, patient demographics varied across sites, including age quartile ((Q)1-Q3:57-62 years), sex (73%-85% male), destination therapy intent (60%-84%), and INTERMACS profile 1-2 (16%-48%). Center mortality was highly variable, nadiring at ≤ 3.6% (≤ 25th percentile) and peaking at ≥ 10.4% (≥ 75th percentile) at 90 days and ≤ 10.2% and ≥ 18.7%, respectively, at 2 years. Centers with low mortality rates tended to have lower 2-year AE rates, but no center was a top performer for all AEs studied. CONCLUSIONS: Mortality and AEs were highly variable across MOMENTUM 3 centers. Studies are needed to improve our understanding of the drivers of outcome variability and to ascertain best practices associated with high-performing centers across the continuum of intraoperative to chronic stages of LVAD support.

A comparison of air-cell and gel surgical table pads and an evaluation of the influence of pressure distribution and other factors on pressure injury prevention.

Guidelines for pressure injury prevention consider the use of pressure-redistributing pads to prevent tissue deformation. However, limited research exists to assess the pressure distribution provided by the operating tables and the effectiveness of pressure-redistributing pads in preventing pressure injuries. In this study, we compared the pressure distribution properties of two surgical table pads and identified parameters influencing pressure injury outcomes after a lengthy surgical procedure. Twenty-seven patients undergoing left ventricular assist device implantation surgery participated in the study. Participants were randomly assigned to use either an air cell-based pad or a gel pad. Interface pressure was recorded during the surgery. We analyzed the effect of surgical table pad type, interface pressure distribution and pressure injury outcomes and analyzed what characteristics of the patients and the interface pressure are most influential for the development of pressure injuries. Comparing the interface pressure parameters between the air-cell group and the gel group, only the peak pressure index x time was significantly different (p < 0.05). We used univariate logistic regression analysis to identify significant predictors for the pressure injury outcome. The support surface was not significant. And, among patient characteristics, only age and BMI were significant (p ≤ 0.05). Among the interface pressure parameters, pressure density maxima, peak pressure index x time, and coefficient of variation were significant for pressure injury outcome (p ≤ 0.05). Peak pressure index, average pressure, and the surgery length were not statistically significant for pressure injury outcomes.

Left Ventricular Assist Device Malfunctions: It Is More Than Just the Pump.

BACKGROUND: Reports of left ventricular assist device (LVAD) malfunction have focused on pump thrombosis. However, the device consists of the pump, driveline, and peripherals, all of which are potentially subject to failure. METHODS: Prospectively collected data were reviewed for all LVAD device malfunctions (DMs) occurring in rotary LVADs implanted at a single center between April 2004 and May 2016. Durable LVADs included 108 Heartmate II (HM II) and 105 HeartWare VAD (HVAD). DM data were categorized according to device type and into categories related to the component that failed: (1) controller, (2) peripheral components, and (3) implantable blood pump or its integral electric driveline. Pump-related events were analyzed as pump-specific (suspected or confirmed thrombosis) or nonpump-specific (driveline failure). DM rates were reported as events per 1000 patient-days, and Cox proportional hazard models were used for time-to-event analyses. Cumulative rates of malfunction were examined for the main components of each type of LVAD. RESULTS: Types of DM included controller failure (30%), battery failure (19%), or patient cable failure (14%), whereas only 13% were because of pump failure. DMs were more common in the HM II device (3.73 per 1000 patient-days versus 3.06 per 1000 patient-days for the HVAD, P<0.01). A higher rate of pump-specific malfunctions was discovered in those implanted with an HM II versus an HVAD (0.55 versus 0.39, respectively; P<0.01) and peripheral malfunctions (2.32 versus 1.78 for the HM II and HVAD, respectively; P<0.01); no difference occurred in the incidence of controller DM between the 2 LVADs. Patients with HVAD were 90% free of a pump-specific malfunction at 3 years compared with 56% for the HM II (log-rank P<0.003). Only 74% of the patients with HM II were free of pump thrombosis at 3 years compared with 90% of the patients with HVAD. Freedom from failure of the integrated driveline was 79% at 3 years for the HM II but 100% for the HVAD (log-rank P<0.02). CONCLUSIONS: Device malfunction is much broader than pump failure alone and occurs for different components at different rates based on the type of LVAD.

Substantial Reduction in Driveline Infection Rates With the Modification of Driveline Dressing Protocol.

BACKGROUND: Driveline infection (DLI) is a cause of morbidity and mortality in patients with continuous-flow left ventricular assist devices (CF-LVADs). We hypothesized that an alternate dressing protocol would decrease the rate of DLIs. METHODS AND RESULTS: A retrospective review of CF-LVAD implants at a single institution from January 2010 to October 2015 was conducted. Patients were divided into implants before (group 1) and after (group 2) the introduction of the new protocol on September 1, 2012. Patients were followed until death, transplantation, change in dressing type, or 2 years. 153 patients were included: 61 in group 1 and 92 in group 2. Group 1 had fewer HVADs than group 2 (27.9% vs 71.7%; P < .001) and more destination therapy, although the latter was not statistically significant (50.8% vs 34.8%; P = .118). At 24 months, the freedom from DLI was 53% in group 1 and 89% in group 2 (P = .01). Group 1 had a significantly greater risk of DLI than group 2 (incident rate ratio 3.18, 95% confidence interval 1.23-8.18; P = .016). CONCLUSIONS: Dramatic improvement in freedom from DLI at 2 years was achieved with a new driveline dressing protocol. This demonstrates that DLI rates can be improved with alternate percutaneous site care techniques in CF-LVAD patients.

Perioperative mortality is the Achilles heel for cardiac transplantation in adults with congenital heart disease: Evidence from analysis of the UNOS registry.

BACKGROUND: Adults with congenital heart disease may present with end-stage heart failure necessitating orthotopic heart transplant (OHT). We sought to review the United Network for Organ Sharing (UNOS) experience with this unique cohort focusing on surgical outcomes and survival. METHODS: From the UNOS registry, 737 adult congenital heart disease recipients (ACHDR) out of 26,993 OHT patients (2.7%) who underwent OHT were studied to analyze early and late outcomes and compared to non-congenital recipients (NCR) over a 15-year period (2000-2014). RESULTS: More ACHDR underwent OHT in the recent-era (3.1%; 2010-2014) as compared to the initial-era (2.5%; 2000-2004; p = 0.03). ACHDR were more likely female (40% vs. 24%; p < 0.01), younger (35 vs. 53 years; p < 0.01), less likely to have ventricular assist devices (4.2% vs. 19.3%; p < 0.01), more likely to have class-II panel-reactive antibody >10%; p < 0.01, and were listed for a longer time (249 vs. 181 days; p < 0.01). When compared to the NCR in the same period, the ACHDR cohort had longer postoperative length of stay (27 vs. 20 days; p < 0.01), higher operative mortality (11.5% vs. 4.6% p < 0.001), higher incidence of primary graft dysfunction (4.3% vs. 2.6%; p < 0.01), and higher need for dialysis (20% vs. 9%; p < 0.01). Primary graft dysfunction is the most common cause of death in (5.8%) ACHDR. Although short-term survival is poorer, long-term survival of ACHDR was found to be equivalent or better than NCR in long term. CONCLUSIONS: Perioperative morbidity and mortality adversely affects short-term survival in ACHDR. ACHDR who survive the first post-transplant year have equivalent or better long-term survival than NCR.

Reliability and construct validity of PROMIS® measures for patients with heart failure who undergo heart transplant.

PURPOSE: To evaluate the reliability and construct validity of measures from the Patient-Reported Outcomes Measurement Information System(®) (PROMIS(®)) for patients with heart failure before and after heart transplantation. METHODS: We assessed reliability of the PROMIS short forms using Cronbach's alpha and the average marginal reliability. To assess the construct validity of PROMIS computerized adaptive tests and short-form measures, we calculated Pearson product moment correlations between PROMIS measures of physical function, fatigue, depression, and social function and existing PRO measures of similar domains (i.e., convergent validity) as well as different domains (i.e., discriminate validity) in patients with heart failure awaiting heart transplant. We evaluated the responsiveness of these measures to change after heart transplant using effect sizes. RESULTS: Forty-eight patients were included in the analyses. Across the many domains examined, correlations between conceptually similar domains were larger than correlations between different domains of health, demonstrating construct validity. Health status improved substantially after heart transplant (standardized effect sizes, 0.63-1.24), demonstrating the responsiveness of the PROMIS measures. Scores from the computerized adaptive tests and the short forms were similar. CONCLUSIONS: This study provides evidence for the reliability and construct validity (including responsiveness to change) of four PROMIS domains in patients with heart failure before and after heart transplant. PROMIS measures are a reasonable choice in this context and will facilitate comparisons across studies and health conditions.

Principles of HeartMate II implantation to avoid pump malposition and migration.

Proper left ventricular assist device (LVAD) insertion will help maximize LVAD flow and may reduce adverse events such as right heart failure and pump thrombosis. Although no standardized insertion technique has been universally accepted, the goals are: unobstructed inflow cannula, unobstructed outflow graft with avoidance of right ventricular compression, and prevention of pump migration. To achieve these objectives for the HeartMate II LVAD, we delineate four principles: proper pump pocket creation, optimized positioning of inflow cannula and outflow graft, proper pump position in the body, and fixation. These basic principles are easy to implement and have been beneficial in our patients, assuring long-term unobstructed LVAD flow.

Muscle-derived stem cell sheets support pump function and prevent cardiac arrhythmias in a model of chronic myocardial infarction.

Direct intracardiac cell injection for heart repair is hindered by numerous limitations including: cell death, poor spreading of the injected cells, arrhythmia, needle injury, etc. Tissue-engineered cell sheet implantation has the potential to overcome some of these limitations. We evaluated whether the transplantation of a muscle-derived stem cell (MDSC) sheet could improve the regenerative capacity of MDSCs in a chronic model of myocardial infarction. MDSC sheet-implanted mice displayed a reduction in left ventricle (LV) dilation and sustained LV contraction compared with the other groups. The MDSC sheet formed aligned myotubes and produced a significant increase in capillary density and a reduction of myocardial fibrosis compared with the other groups. Hearts transplanted with the MDSC sheets did not display any significant arrhythmias and the donor MDSC survival rate was higher than the direct myocardial MDSC injection group. MDSC sheet implantation yielded better functional recovery of chronic infarcted myocardium without any significant arrhythmic events compared with direct MDSC injection, suggesting this cell sheet delivery system could significantly improve the myocardial regenerative potential of the MDSCs.

Temporal leukocyte numbers and granulocyte activation in pulsatile and rotary ventricular assist device patients.

Individual ventricular assist device (VAD) design may affect leukocytes and impact immunity. Few studies have presented leukocyte and infection profiles in VAD patients over the course of the implant period. CD11b (MAC-1) expression on granulocytes is an indicator of activation during inflammation, mediating extravasation and the release of reactive oxygen species in tissue. No reported studies have presented MAC-1 expression on circulating granulocytes in VAD patients. Fifty-six patients implanted at a single center with a HeartMate II (HMII; n = 32), HeartWare (HW; n = 12), or Thoratec pneumatic VAD (PVAD; n = 12) between 1999 and 2011 were followed for 120 days of support. The leukocyte profiles and infectious events of all patients were evaluated; additionally, a subset had MAC-1 expression on circulating granulocytes was measured (HMII n = 9; HW n = 7; PVAD n = 4). All groups exhibited a significant peak in leukocyte numbers at postoperative day (POD) 14 while simultaneously experiencing a significant decrease in hematocrit. HMII patients exhibited a 3.2-fold increase in granulocyte MAC-1 expression at POD 14, and the temporal trend over the implant period differed from that experienced by HW patients. Further, HW patients experienced significantly fewer infection events. Alterations in leukocyte profiles and granulocyte activation experienced by VAD patients appear to be device-specific. Elevations in leukocyte activation may be related to an increased risk for infection, although the specific relationship between these phenomena in this patient group is not known.

Use of an intrapericardial, continuous-flow, centrifugal pump in patients awaiting heart transplantation.

BACKGROUND: Contemporary ventricular assist device therapy results in a high rate of successful heart transplantation but is associated with bleeding, infections, and other complications. Further reductions in pump size, centrifugal design, and intrapericardial positioning may reduce complications and improve outcomes. METHODS AND RESULTS: We studied a small, intrapericardially positioned, continuous-flow centrifugal pump in patients requiring an implanted ventricular assist device as a bridge to heart transplantation. The course of investigational pump recipients was compared with that of patients implanted contemporaneously with commercially available devices. The primary outcome, success, was defined as survival on the originally implanted device, transplantation, or explantation for ventricular recovery at 180 days and was evaluated for both noninferiority and superiority. Secondary outcomes included a comparison of survival between groups and functional and quality-of-life outcomes and adverse events in the investigational device group. A total of 140 patients received the investigational pump, and 499 patients received a commercially available pump implanted contemporaneously. Success occurred in 90.7% of investigational pump patients and 90.1% of controls, establishing the noninferiority of the investigational pump (P<0.001; 15% noninferiority margin). At 6 months, median 6-minute walk distance improved by 128.5 m, and both disease-specific and global quality-of-life scores improved significantly. CONCLUSIONS: A small, intrapericardially positioned, continuous-flow, centrifugal pump was noninferior to contemporaneously implanted, commercially available ventricular assist devices. Functional capacity and quality of life improved markedly, and the adverse event profile was favorable. CLINICAL TRIAL REGISTRATION: URL: http://www.clinicaltrials.gov. Unique identifier: NCT00751972.

Bariatric surgery for a patient with a HeartMate II ventricular assist device for destination therapy.

A patient with a HeartMate II left ventricular assist device who had a body mass index of 52 needed gastric bypass surgery in order to qualify for a heart transplant. Unlike previous experience in which the surgery was performed at the implant hospital, the gastric bypass surgery in this case was performed at a bariatric center of excellence that was a separate facility from the implant hospital. The artificial heart program of the University of Pittsburgh Medical Center worked with the bariatric center of excellence in scheduling the gastric bypass surgery using a multidisciplinary team approach at 2 hospitals to coordinate safe, high-quality patient care in a unique situation.

Survival benefit of implantable cardioverter-defibrillators in left ventricular assist device-supported heart failure patients.

BACKGROUND: Implantable cardioverter-defibrillators (ICDs) reduce mortality in heart failure (HF). In patients requiring a ventricular assist device (VAD), the benefit from ICD therapy is not well established. The aim of this study was to define the impact of ICD on outcomes in VAD-supported patients. METHODS AND RESULTS: We reviewed data for consecutive adult HF patients receiving VAD as a bridge to transplantation from 1996 to 2003. The primary outcome was survival to transplantation. A total of 144 VADs were implanted [85 left ventricular (LVAD), 59 biventricular (BIVAD), mean age 50 ± 12 years, 77% male, left ventricular ejection fraction 18 ± 9%, 54% ischemic]. Mean length of support was 119 days (range 1-670); 103 patients (72%) survived to transplantation. Forty-five patients had an ICD (33 LVAD, 12 BIVAD). More LVAD patients had an appropriate ICD shock before implantation than after (16 vs 7; P = .02). There was a trend toward higher shock frequency before LVAD implant than after (3.3 ± 5.2 vs 1.1 ± 3.8 shocks/y; P = .06). Mean time to first shock after VAD implant was 129 ± 109 days. LVAD-supported patients with an ICD were significantly more likely to survive to transplantation [1-y actuarial survival to transplantation: LVAD: 91% with ICD vs 57% without ICD (log-rank P = .01); BIVAD: 54% vs 47% (log-rank P = NS)]. An ICD was associated with significantly increased survival in a multivariate model controlling for confounding variables (odds ratio 2.54, 95% confidence interval 1.04-6.21; P = .04). CONCLUSIONS: Shock frequency decreases after VAD implantation, likely owing to ventricular unloading, but appropriate ICD shocks still occur in 21% of patients. An ICD is associated with improved survival in LVAD-supported HF patients.

Long-term effects on renal function of dose-reduced calcineurin inhibitor and sirolimus in cardiac transplant patients.

UNLABELLED: Calcineurin inhibitor (CNI)-associated renal insufficiency is common after cardiac transplantation (CTX); however, the addition of sirolimus allows for CNI dose reduction and this strategy may limit CNI renal toxicity. This study examines the long-term effects of such a strategy. METHODS: Patients from a single center who had CTX from 1990 to 2007 and who were converted to sirolimus and a dose-reduced CNI were compared to group-matched controls maintained on CNI and an antiproliferative agent. RESULTS: One hundred and fifty-five patients (79 sirolimus and 76 controls) were included and had similar baseline characteristics. Sirolimus was started a mean of 1429 d post-CTX and maintained for a mean of 823 d. Reason for conversion to sirolimus was renal insufficiency (34%), vasculopathy (29%), recurrent rejection (19%), and other (18%). The eGFR was not different between groups at baseline (44.7 mL/min/1.73 m(2) vs. 46.0, p = 0.64) or at any point during follow-up: 90 d, 180 d, 1 yr, 2 yr, and 3 yr. conclusion: Patients converted to a regimen of sirolimus and a dosed-reduced CNI have stable renal function over the following three yr, but do not have an improvement in renal outcomes compared to patients maintained on full dose CNI.

Invasive aspergillosis among heart transplant recipients is rare but causes rapid death due to septic shock and multiple organ dysfunction syndrome.

Between 2000 and 2011, proven or probable invasive aspergillosis (IA) was diagnosed in 1.7% (8/455) of heart transplant (HTx) recipients at our center, in the absence of antifungal prophylaxis. All patients had invasive pulmonary infections and 75% (6/8) were diagnosed during 2 separate 3-month periods. Cases were notable for their association with septic shock and multiple organ dysfunction syndrome (MODS) (75%, 6/8 each), non-specific clinical and radiographic findings, and rapid mortality despite mould-active antifungal therapy (88%, 7/8; occuring at a median 11 days after diagnosis). All patients had predisposing conditions known to be risk factors for IA. For patients with early IA (within 90 days of HTx), conditions included hemodialysis, thoracic re-operation, and the presence of another case in the institution within the preceding 3 months. For late-onset IA, conditions included hemodialysis and receipt of augmented immunosuppression. Clinicians should suspect IA in HTx recipients with risk factors who present with non-specific and unexplained respiratory syndromes, including those in septic shock and MODS, and institute prompt antifungal therapy without waiting for the results of cultures or other diagnostic tests.

Graft Resistance Difference after HVAD to HeartMate 3 Left Ventricular Assist Device Exchange.

BACKGROUND: A likely consequence of the discontinued distribution and sale of the HVAD System (Medtronic, Minneapolis, MN) will be an increase in replacement with the HeartMate 3 (Abbott, Chicago, IL) left ventricular assist device when device exchange is necessary. If part or all of the HVAD 10-mm-diameter outflow graft is retained during replacement, the HeartMate 3 will have to run at a higher speed than it would with its 14-mm-diameter graft. METHODS: A steady-state, in vitro study was run with 250-mm-long samples of HVAD, HeartMate 3, and half-HVAD/half-HeartMate 3 grafts and additionally 125- and 375-mm-long samples of the HVAD graft. Flows of 3.0, 3.9, 4.3, 4.7, and 6.0 L/min were applied to encompass expected clinical conditions. RESULTS: At typical and high flow rates of 4.3 and 6.0 L/min, HeartMate 3 rotor speeds with the full HVAD graft had to be increased relative to those with the HeartMate 3 graft from 5350 to 5700 and 6350 to 6900 rpm, respectively, with power consumption increases from 3.7 to 4.3 W (16%) and 5.5 to 6.8 W (24%), respectively. CONCLUSIONS: The study did not elucidate a severe consequence of using a remnant HVAD graft during pump exchange, but the incremental risks of a higher rotor speed, disadvantage to the patient in battery runtime, and the general benefit of complete conversion to the HeartMate 3 graft should be balanced against other procedural considerations. Complete graft replacement during HVAD-to-HeartMate 3 conversion remains the preferred approach from an engineering point of view.

Adipose-derived stem cell sheet under an elastic patch improves cardiac function in rats after myocardial infarction.

OBJECTIVES: Although adipose-derived stem cells (ADSCs) have shown promise in cardiac regeneration, stable engraftment is still challenging. Acellular bioengineered cardiac patches have shown promise in positively altering ventricular remodeling in ischemic cardiomyopathy. We hypothesized that combining an ADSC sheet approach with a bioengineered patch would enhance ADSC engraftment and positively promote cardiac function compared with either therapy alone in a rat ischemic cardiomyopathy model. METHODS: Cardiac patches were generated from poly(ester carbonate urethane) urea and porcine decellularized cardiac extracellular matrix. ADSCs constitutively expressing green fluorescent protein were established from F344 rats and transplanted as a cell sheet over the left ventricle 3 days after left anterior descending artery ligation with or without an overlying cardiac patch. Cardiac function was serially evaluated using echocardiography for 8 weeks, comparing groups with combined cells and patch (group C, n = 9), ADSCs alone (group A, n = 7), patch alone (group P, n = 6) or sham groups (n = 7). RESULTS: Much greater numbers of ADSCs survived in the C versus A groups (P < .01). At 8 weeks posttransplant, the percentage fibrotic area was lower (P < .01) in groups C and P compared with the other groups and vasculature in the peri-infarct zone was greater in group C versus other groups (P < .01), and hepatocyte growth factor expression was higher in group C than in other groups (P < .05). Left ventricular ejection fraction was higher in group C versus other groups. CONCLUSIONS: A biodegradable cardiac patch enhanced ADSC engraftment, which was associated with greater cardiac function and neovascularization in the peri-infarct zone following subacute myocardial infarction.

Hospitalization Patterns and Impact of a Magnetically-Levitated Left Ventricular Assist Device in the MOMENTUM 3 Trial.

BACKGROUND: In the MOMENTUM 3 (Multicenter Study of MagLev Technology in Patients Undergoing Mechanical Circulatory Support Therapy with HeartMate 3) pivotal trial, the HeartMate 3 (HM3) fully magnetically levitated left ventricular assist device (LVAD) demonstrated superiority over the axial-flow HeartMate II (HMII) LVAD. The patterns and predictors of hospitalizations with the HM3 LVAD have not been characterized. OBJECTIVES: This study sought to determine causes, predictors, and impact of hospitalizations during LVAD support. METHODS: Patients discharged after LVAD implantation were analyzed. In the pivotal trial, 485 recipients of HM3 were compared with 471 recipients of HMII. The pivotal trial HM3 group was also compared to 949 recipients of HM3 in the postapproval phase within the trial portfolio. Predictors of cause-specific rehospitalization were analyzed. RESULTS: The rates of rehospitalization were lower with HM3 LVAD than with HMII LVAD in the pivotal trial (225.7 vs 246.4 events per 100 patient-years; P < 0.05). Overall, rehospitalization rates and duration were similar in the HM3 postapproval phase and pivotal trial but prolonged hospitalizations (>7 days) were less frequent (rate ratio: 0.90 [95% CI: 0.80-0.98]; P < 0.05). In HM3 recipients, the most frequent causes of rehospitalization included infection, heart failure (HF)-related events, and bleeding. First rehospitalization caused by HF-related event versus other causes was associated with reduced survival (HR: 2.2 [95% CI: 1.3-3.9]; P = 0.0014). Male sex, non-White race, presence of cardiac resynchronization therapy/implantable cardioverter-defibrillator, obesity, higher right atrial pressure, smaller LV size, longer duration of index hospitalization, and lower estimated glomerular filtration rate at index discharge predicted HF hospitalizations. CONCLUSIONS: Contemporary support with the HM3 fully magnetically levitated LVAD is associated with a lower hospitalization burden than with prior pumps; however, rehospitalizations for infection, HF, and bleeding remain important challenges for progress in the patient journey. (MOMENTUM 3 IDE Clinical Study, NCT02224755; MOMENTUM 3 Continued Access Protocol [MOMENTUM 3 CAP], NCT02892955).

Outcomes With Phosphodiesterase-5 Inhibitor Use After Left Ventricular Assist Device: An STS-INTERMACS Analysis.

BACKGROUND: Elevated right ventricular afterload following continuous-flow left ventricular assist device (CF-LVAD) may contribute to late right heart failure (LRHF). PDE5i (phosphodiesterase-5 inhibitors) are used to treat pulmonary hypertension and right heart dysfunction after CF-LVAD, but their impact on outcomes is uncertain. METHODS: We queried Interagency Registry for Mechanically Assisted Circulatory Support from 2012 to 2017 for adults receiving a primary CF-LVAD and surviving ≥30 days from index discharge. Patients receiving early PDE5i (ePDE5i) at 1 month were propensity-matched 1:1 with controls. The primary outcome was the cumulative incidence of LRHF, defined using prevailing Interagency Registry for Mechanically Assisted Circulatory Support criteria; secondary outcomes included all-cause mortality and major bleeding. RESULTS: Among 9627 CF-LVAD recipients analyzed, 2463 (25.6%) received ePDE5i and 1600 were propensity-matched 1:1 with controls. Before implant, ePDE5i patients had more severe RV dysfunction (13.1% versus 9.6%) and higher pulmonary vascular resistance (2.8±2.7 versus 2.2±2.4 WU), both P<0.001, but clinical factors were well-balanced after propensity-matching. In the unmatched cohort, ePDE5i patients had a higher 3-year cumulative incidence of LRHF, mortality, and major bleeding, but these differences were attenuated in the propensity-matched cohort: LRHF 40.8% versus 35.7% (hazard ratio, 1.14 [95% CI, 0.99-1.32]; P=0.07); mortality 38.6% versus 35.8% (hazard ratio, 0.99 [95% CI, 0.86-1.15]; P=0.93); major bleeding 51.2% versus 46.0% (hazard ratio, 1.12 [95% CI, 0.99-1.27]; P=0.06). CONCLUSIONS: Compared with propensity-matched controls, adult CF-LVAD patients receiving ePDE5i had similar rates of LRHF, mortality, and major bleeding. While intrinsic patient risk factors likely account for more adverse outcomes with ePDE5i in the unmatched cohort, there is no obvious benefit of ePDE5i in the LVAD population.