Search

Purpose Sudden cardiac death may occur in the setting of acute rejection or later in patients with allograft vasculopathy. The role of implantable cardioverter-defibrillators (ICDs) in heart transplant recipients with allograft failure is unknown. Methods We analyzed the United Network for Organ Sharing registry between January 2008 and March 2018 to include adults ≥18 who had undergone heart transplantation and were listed for retransplantation for the indication of allograft failure. Results Between January 2008 and March 2018, 1201 heart transplant recipients were re-listed for transplant of whom 229 (19%) had an ICD. Indications for re-listing included coronary artery disease (n = 699, 58%), primary graft failure (n = 188, 15.7%), chronic rejection (n = 141, 11.7%), acute rejection (n = 60, 5.0%), non-specific reason (n = 57, 4.7%), restrictive or constrictive disease (n = 49, 4.1%), and hyperacute rejection (n = 7, 0.6%). Allograft failure patients with ICDs were younger than those without (47.3 vs 43.5 years, p < 0.001) and less likely to be female (15.0% vs 21.6%, p = 0.006). There were no differences in body mass index, rates of diabetes, cerebral vascular disease, tobacco use, and most recent creatinine between the two groups. Waitlist mortality did not differ between the two groups with 80% survival at 1 year in the no-ICD group and 82.1% in the ICD group (p = 0.86)(Figure). Conclusion Approximately 1 in 5 heart transplant recipients re-listed for primary allograft failure received an ICD. The use of ICDs does not appear to be associated with increased waitlist survival. More research is needed regarding the role of ICDs in this unique population. [ABSTRACT FROM AUTHOR]

Purpose Extracorporeal cardiopulmonary resuscitation (ECPR) is controversial given the paucity of clear evidence for improved outcomes and a lack of clarity on appropriate candidacy in time-sensitive arrest situations. We set out to investigate what factors predicted successful outcomes in ECPR patients. Methods Between March 2007 and August 2018, 112 patients were placed on venoarterial extracorporeal membrane oxygenation (VA-ECMO) for ECPR at our institution. Primary outcomes of interest were survival to decannulation and survival to hospital discharge. Patients were compared in terms of pre-cannulation comorbidities, laboratory values, and overall outcomes. Multivariable logistic regression was performed to identify pre-cannulation predictors of in-hospital mortality. Results Among 112 patients, 44 (39%) survived to decannulation with 31 (28%) surviving to discharge. The median age was 60 (IQR 45-72 years) with a median duration of ECMO support of 2.2 days (IQR 0.6-5.1). There were no differences in age, comorbidities, or year of cannulation between those who survived to decannulation and those who did not. Patients who survived to discharge had a lower prevalence of chronic kidney disease than those who died in-hospital (19% vs. 41%, p=0.046) with lower baseline creatinine levels (median 1.2 mg/dL (IQR 0.8-1.7) vs. 1.7 (0.7-2.7), p=0.008). Median time from CPR initiation to cannulation was 40 minutes (IQR 30-50) with no differences between survivors and non-survivors (p=0.453). When controlling for age and CPR duration, multivariable logistic regression identified pre-arrest serum creatinine as predictive of mortality (OR 3.25 (95% CI: 1.22-8.70), p=0.019) while higher pre-arrest serum albumin was protective (OR 0.32 (95% CI: 0.14-0.74), p=0.007). Conclusion Despite high mortality rates, ECPR affords the opportunity to rescue patients in cardiac arrest. Reduced pre-cannulation renal function and higher serum albumin levels may have value in determining the likelihood of ECPR survival. [ABSTRACT FROM AUTHOR]

Purpose With increasing organs available in light of the national opioid epidemic and the availability of direct acting-antiviral (DAA) therapy for Hepatitis C virus (HCV), HCV antibody (Ab)+ organs serve as an important potential supply to the donor pool. We sought to examine trends over time in the utilization of HCV Ab+ organs both nationally and regionally. Methods We queried the United Network for Organ Sharing registry data in the DAA era between January 2012 and March 2018 for adults ≥18 years who had undergone heart transplantation with a HCV Ab+ donor as well as donor recovery data. Results The incidence of HCV Ab+ deceased donors have increased from 4.6% in 2012 to 8.7 in 2018 (p<0.05) in parallel with an increase in drug intoxication as cause of death during the same period (5.9% to 14.0%, p<0.05). Recovery rates for HCV Ab- deceased donor hearts were unchanged from 2012 to 2018 (range: 28.5-31.6%) however there was a significant increase in recovery of HCV Ab+ donor hearts in the same period (2.6% to 20.3%, Figure 1A). There was significant regional variation in HCV Ab+ recovery from deceased donors (Figure 1B), with highest recovery achieved in Regions 11 (23.2%), 6 (20.8%), and 1(20.6%). Nucleic acid test (NAT)- HCV Ab+ donors were more likely to be recovered as opposed to NAT+ HCV Ab+ donors (12.7% vs. 9.6%, p=0.009). Transplant rates for HCV Ab+ hearts also increased over the past three years, accounting for 6.2% in 2018. (Figure 1C). There was also regional variation in utilization of HCV Ab+ positive donors (Figure 1D), with highest utilization achieved in Regions 9 (11.6%) and 11 (10.4%). Conclusion Over the last two years, with the availability of DAA therapy, there has been increased nationwide recovery and utilization of HCV Ab+ donor hearts that is accompanied by significant regional variability. Adopting standardized protocols across UNOS regions and transplant centers may increase utilization of these donors more effectively and potentially lead to better outcomes. [ABSTRACT FROM AUTHOR]

Purpose The indications, efficacy, and safety profile of temporary, percutaneous dual lumen cannula-based right ventricular assist devices (RVAD) [TandemLife ProtekDuo] after left ventricular assist device (LVAD) placement have not been well characterized. We describe our experience with the largest series of patients reported to date. Methods We conducted a retrospective review of 24 patients who received ProtekDuo RVAD support from January 2016 through July 2018. Outcomes of interest included indications for support, procedural success, complications, and clinical outcomes. Results Patients were 79% male (n=19) with a median age of 66 (IQR 58-72). ProtekDuo RVAD support was initiated a median of 1 day (IQR 0-2) after LVAD implantation. Indications for RVAD placement included pure RV failure (n=14, 58%), vasoplegia (n=2, 8%), or mixed syndrome (n=8, 33%). Procedural success was 100% without any complications. Initial median flow was 4.1 L/min (IQR 3.8-4.5) with a median indexed flow of 2.0 L/min/m2 (IQR 1.8-2.3). During a median support of 11 days (IQR 7-14), major complications included cannula thrombosis (n=1), cannula migration into the RV requiring repositioning (n=2), and hemolysis, as defined by progressively rising LDH and plasma free hemoglobin levels (n=4). A total of 18 patients (75%) were weaned from the ProtekDuo RVAD with cannulas removed at the bedside in all patients. Three patients died while on ProtekDuo support and three were converted to central RVAD support. Overall survival to discharge was 78% (n=18/23) with one patient remaining in-hospital. Survival post-discharge was 100% at the time of most recent follow-up, with a median duration of 211 days (IQR 98-523). Conclusion The ProtekDuo RVAD affords a safe and reliable bridge to right ventricular recovery as a less invasive percutaneous option after LVAD placement. [ABSTRACT FROM AUTHOR]

Purpose We have reported that severe primary graft dysfunction (PGD) can be treated by early use of extracorporeal membrane oxygenation (ECMO) support. However, little is known in detail about graft function and hemodynamics in heart transplant patients who survived severe PGD. Methods We retrospectively reviewed 20 adult patients who experienced severe PGD after heart transplant and treated with ECMO between January 2015 and October 2017. Right and left heart catheterization and echocardiographic data at 1-year after transplant was retrospectively reviewed. Results Median age of cohort was 61 years (Interquartile range (IQR) 55-66), and 17 (85%) were male. Nineteen (95%) were bridge to transplant and median duration of device support was 463 days (IQR 246-770). Median donor age was 34 years (IQR 26-43) and baseline donor ejection fraction was 65%. Median duration of ECMO support was 5.1 days (IQR 3.5-8.4) and all weaned from ECMO support. At 1-year, all patients survived (95%), except for one who died in the hospital. No one developed significant allograft vasculopathy on left heart catheterization. On echocardiogram, left ventricular ejection fraction was 60% (IQR 60-65) and 3 patients had mild right ventricular dysfunction. Right heart catheterization showed central venous pressure of 5.5 mmHg (IQR 4-10), mean pulmonary artery pressure of 22 mmHg (IQR 17-23), pulmonary capillary wedge pressure of 12.5 mmHg (IQR 8-14), and cardiac output of 6.4 L/min (IQR 4.7-6.8). Conclusion Allografts which overcame severe PGD can function normally at 1-year follow-up. Adequate perioperative management of severe PGD is mandatory to maximize survival. [ABSTRACT FROM AUTHOR]

Purpose The effect of mTOR-inhibitors initiated as maintenance immunosuppression early after cardiac transplant is not well-established. Current outcomes data are limited to small, single-center studies. Methods 22,173 adult cardiac transplant recipients from the United Network for Organ Sharing database between 2008 and 2018 were included in current study. Propensity-matching was carried out to match 184 patients started on proliferation signal inhibitor (PSI) maintainence therapy (sarolimus or everolimus) before discharge with 184 controls. Outcomes included mortality/graft failure, development of coronary artery disease, infection, and acute rejection. Results PSI and No PSI cohorts were well-matched with regards to age (40.4 vs. 50.6), gender (32% vs. 30% females), race (60% vs. 63% white), LVAD use (28% vs. 30%), UNOS status (54% vs 55% 1A), HF etiology (32% vs 31% ischemic), and other immunosuppressive medication use (all p=NS). Compared to controls, PSI group had significantly lower rates of coronary artery disease (17.8%, 95% CI 12.1%-24.4% versus 27.9%, 95% CI 20.9%-35.5; P= 0.04) (Figure). However, risk of hospitalization for rejection (36.1% vs. 23.7%, p=0.043) and hospitalization for infection (40.6% vs. 28.8%, p=0.011) at 5 years were significantly higher in the PSI group. Mortality and graft failure rates were comparable between PSI and No PSI groups. Conclusion mTOR-inhibitors initiated early as part of a maintenance immunosuppression regimen may reduce the long-term risk of developing coronary artery disease at the expense of higher risk of rejection and infection in adult cardiac transplant recipients. These findings suggest that switching to PSI at a later time point may be a more effective strategy in minimizing post-transplant complications compared to early initiation of PSI. [ABSTRACT FROM AUTHOR]

Purpose Heart transplant centers are increasingly utilizing organs from donors infected with Hepatitis C virus (HCV). Aside from single-center experiences, there is little existing data regarding the characteristics of recipients of HCV Antibody (Ab)+ organs as well as short-term outcomes and graft survival. Methods We analyzed the United Network for Organ Sharing registry from January 2016 and March 2018 to include adults ≥18 who had undergone heart transplantation with a HCV Ab+ donor. Information regarding nucleic acid amplification testing (NAT) was also available. Results Between January 2016 and March 2018, 159 recipients underwent heart transplantation with HCV Ab+ donors. When comparing recipients of HCV+ to HCV- hearts, there were no differences between recipient age (53.4 vs 53.6 years, p =0.86), donor age (33.2 vs 31.9 years, p =0.055), female donor (26.0 vs 27.7%, p =0.64), body mass index (27.3 vs 27.6, p =0.47), and serum creatinine at transplant (1.4 vs 1.4, p = 0.36). There were no differences in the use of ventricular assist devices at time of transplant between groups. Recipients of HCV+ organs were more likely to be Blood Type O (55.3% vs 38.9%, p <0.001) and spent significantly fewer days as Status 1A (25.8 vs 42.6, p <0.001). Recipients of HCV+ organs had longer ischemic times (3.6 vs 3.0 hours, p <0.001) with associated longer distances from the donor hospital to the transplant center (323 vs 146 miles, p<0.001). Kaplan-Meier survival curves showed a trend towards decreased survival in the HCV+ group when compared to the HCV- group (p =0.06). Among recipients of HCV+ organs, there was no difference in survival between recipients of NAT+ and NAT- donors (Figure). Conclusion While recipient baseline characteristics were similar between those receiving HCV+ and HCV- transplants, recipients of HCV+ organs had significantly longer ischemic times with longer distances from the donor hospital to the transplant center. Short-term survival appears similar between groups. [ABSTRACT FROM AUTHOR]

Purpose Stroke is a major complication with potentially devastating consequences for patients receiving veno-arterial extracorporeal membrane oxygenation (VA-ECMO). Arterial cannulation sites for VA-ECMO include ascending aorta, axillary artery, and femoral artery. However, rates of stroke according to cannulation site are currently unknown. The purpose of this study was to investigate the association between occurrence and patterns of stroke with ECMO arterial cannulation sites. Methods We retrospectively reviewed 498 consecutive patients who received VA-ECMO support for cardiogenic shock between March 2007 and May 2018. Patients were categorized by cannulation strategy. The rates, subtype and location of strokes as assessed by CT scans of the head during and after VA-ECMO support were analyzed. Results Median age was 60 years (IQR 49-68); 69% were male. 87 patients were cannulated via the ascending aorta (18%), 37 via the axillary artery (7%), and 374 (75%) via the femoral artery. In total, 27 patients (5.4%) developed 32 stroke events at a median of 4.3 (2.8-6.0) days after ECMO cannulation. Ischemic stroke was the most common subtype (63%), followed by hemorrhagic conversion (22%) and hemorrhagic stroke (15%). Location by CT was right hemispheric in 34%, left hemispheric in 34%, bilateral in 19%, and vertebrobasilar in 13%. The incidence of stroke was similar across cannulation strategies: aorta (n=5, 5.8%), axillary artery (n=2, 5.4%), and femoral artery (n=20, 5.3%), (p=0.94). Stroke subtype and location based on cannulation site are summarized in Table 1. Conclusion Stroke occurred in approximately 5% of all patients with cardiogenic shock receiving VA-ECMO, with ischemic stroke being the most common subtype. Incidence of stroke does not appear to differ among patients cannulated via the ascending aorta, axillary artery, or femoral artery, although clinical patterns may vary based on the cannulation strategy. [ABSTRACT FROM AUTHOR]

Purpose For patients with infected left ventricular support devices (LVADs) at the time of cardiac transplant, data on the risk of early infection and the effect of induction immunosuppression on post-transplant outcomes are limited. Methods In this retrospective cohort study from the UNOS database, we identified 9,294 bridge-to-transplant LVAD patients ≥18 years of age who received heart transplants between 2008 and 2018. Using propensity-scoring, 1,497 LVAD patients with device infection at the time of transplant were 1:1 matched to controls without device infection. Proportional hazards models were used to determine the effect of induction immunosuppression on outcomes including death/graft failure, infection, and acute rejection at 2 years post-transplant. Results Compared to patients without device infection, patients with infected LVADs had significantly higher rates of infection at 2 years post-transplant (22.2%, 95% CI 20.0%-24.7% versus 17.9%, 95% CI 15.8%-20.2%, P <0.009) (Figure 1A). There was no difference in rates of death/graft failure or acute rejection. Use of induction immunosuppression after transplant was not associated with increased risk of post-transplant infection (HR 1.22, 95% CI 0.93-1.61, P <0.16). In a subgroup analysis of the 1,497 LVAD patients with device infection at the time of transplant, 629 patients who received induction immunosuppression were matched to 629 patients discharged after transplant without induction using propensity scoring. There was no difference in rates of death/graft failure, infection (Figure 1B), and acute rejection between the two groups. Conclusion For LVAD patients bridged to cardiac transplant, device infection at the time of transplant is a risk factor for post-transplant infection. However, it should not necessarily be considered a contraindication to receiving induction immunosuppression, which was not associated with increased incidence of adverse post-transplant outcomes. [ABSTRACT FROM AUTHOR]

Purpose Neurological complications (NC) are potentially devastating for cardiogenic shock patients receiving veno-arterial extracorporeal membrane oxygenation (VA-ECMO) support. The purpose of this study was to investigate the overall incidence and pattern of NC and their impact on outcome. Methods We retrospectively reviewed 498 consecutive patients who received VA-ECMO support for cardiogenic shock between March 2007 and May 2018. Information on NC type, including stroke, altered mental status, seizure, encephalopathy, and transient ischemic attack and other neurological events was collected. Survival to discharge was compared between patients with and without neurological complications. Results Median age of the entire cohort was 60 years (IQR 49-68), and 69% were male. Etiology of cardiogenic shock included post-cardiotomy shock (n=154, 33%), acute myocardial infarction (n=124, 27%), post-transplant graft failure (n=60, 13%), acute decompensated heart failure (n=65, 14%), and others (n=57, 12%). Eighty-seven patients were cannulated via the ascending aorta (18%), 37 via the axillary artery (7%), and 374 (75%) via the femoral artery. In total, 125 patients (25%) were diagnosed with NC: altered mental status was most common n=62 (12%), stroke n=29 (6%), encephalopathy n=16 (3%), seizure n=15 (3%), transient ischemic attack/others n=16 (3%). NC diagnosis was significantly associated with higher in-hospital mortality rates (74%, n=93) vs. non-NC diagnosis (61%, n=229), (p=0.008). Among NC subtypes, encephalopathy had the highest in-hospital mortality rate (94%, n=15), followed by stroke (72%, n=21) and AMS (71%, n=44). Conclusion Neurological disorders are common in cardiogenic shock patients who receive VA-ECMO and significantly impact in-hospital mortality rates. [ABSTRACT FROM AUTHOR]

Purpose In postcardiotomy shock patients with prosthetic valves, the use of venoarterial extracorporeal membrane oxygenation (VA ECMO) may induce prosthetic valve thrombosis due to blood stasis in cardiac chambers. Incidence and outcomes of prosthetic valve thrombosis during VA ECMO has not been previously described. Methods Between January 2007 and June 2018, a total of 171 patients underwent VA ECMO therapy after major cardiovascular surgery at our center. Of these patients, 89 had a prosthetic valve in place, and we conducted a retrospective review of this cohort. Results Of the 89 patients with prosthetic valves placed on VA ECMO, 37 (41.5%) had central VA ECMO cannulation, and the remaining 52 patients (58.5%) underwent peripheral VA ECMO cannulation. Median age was 66 years (interquartile range [IQR], 55-74) and 53 (59.6%) were men. Average duration of ECMO support was 5.1 days (IQR 3.0, 8.5). The majority of patients (76.4%, n = 68) were on therapeutic anticoagulation within 24 hours post-cardiotomy, and inotropes were used in all patients. A total of 8 patients (9%) developed prosthetic valve thrombosis: 5 involved the aortic valve (including one mechanical aortic valve prosthesis), 2 involved bioprosthetic mitral valves, and 1 involved a bioprosthetic pulmonary valve (Table). Overall cohort survival to discharge was 41.2% (n = 35), but was reduced to 9% (n = 1) amongst those patients that developed valve thrombosis. Conclusion Prosthetic valve thrombosis is relatively common complication while patients with prosthetic valves are placed on VA-ECMO, and it is almost uniformly fatal when it does develop. [ABSTRACT FROM AUTHOR]

Mortality in cardiogenic shock (CS) remains prohibitively high despite recent advances in treatment. The extent of clinical benefit of the use of pulmonary artery catheters (PACs) in CS patients is still unclear. We hypothesize that the comprehensive assessment of PAC data is associated with reduced risk of in-hospital mortality in CS patients. 1,414 all-cause CS patients from the Cardiogenic Shock Working Group Registry, a national multicenter retrospective CS registry, were classified according to the Society for Cardiovascular Angiography and Intervention (SCAI) CS stages and evaluated for PAC use. PAC use was quantified by the presence of pulmonary artery saturation, pulmonary artery pressures and wedge pressure. We compared patients without any of these measurements (No PAC data), patients with at least one (Some PAC data) and patients with all these parameters and right atrial pressure (Complete PAC data). PAC usage was analyzed for association with in-hospital mortality in a multivariate logistic regression model. Of the total cohort of 1414 patients, 267 (18.88%) had no recorded PAC data, 549 (38.83%) had partially recorded PAC data, and 598 (42.29%) had complete PAC data. The majority of SCAI stage B (n=43, 93.5%) and C (n=132, 50.19%) had complete PAC data while the majority of SCAI stages D (n=361, 47.63%) and E (n=90, 42.45%) had some PAC data. After adjusting for institution in the total cohort, patients with complete PAC data had almost half the risk of mortality compared to those with no PAC data (OR 95%CI: 0.586, 0.418-0.822) and those with partial PAC data (OR 95%CI: 0.602, 0.467-0.813). The differences in mortality by PAC usage in the overall and SCAI stage cohorts are summarized in the figure below. These data suggest that using a PAC in CS patients, especially those in more critical condition, may decrease their risk of in-hospital mortality. This benefit could potentially be explained by the information provided by the PAC in developing a more appropriate treatment plan. [ABSTRACT FROM AUTHOR]

Purpose Left ventricular assist devices (LVAD) are widely used as a bridge to heart transplantation or destination therapy for advanced heart failure. However, hemocompatibility-related complications such as pump thrombosis, stroke and bleeding remain frequent. We previously reported that LVAD implantation is followed by a sharp increase in serum levels of IgG natural antibodies (Nabs) recognizing oxidation-specific epitopes (OSE). Nabs have been implicated in inflammatory reactions related to atherosclerosis, ischemic stroke and primary graft dysfunction following heart transplantation. However, the consequence of IgG Nabs production following LVAD is not known. We hypothesized that Nabs generated following implant may contribute to LVAD complications by modulating the coagulation cascade. Methods We studied a cohort of adult patients who received a continuous flow LVAD at Columbia University/NewYork Presbyterian Hospital. Nabs and anti-phospholipid antibodies (aPL) were measured in pre- and post-LVAD sera at 1, 6 and 12 months by assessing IgG reactivity to the OSE malondialdehyde (MDA) or β2-glycoprotein complexes by ELISA. In a subset of patients, Nabs were measured at day 1 to 7 post-implant. The effect of IgG Nabs on hemostasis was tested by rotational thromboelastometry (ROTEM) by adding IgG purified from pre- and post-LVAD serum to ABO blood type-matched donor blood. Wilcoxon's signed rank test for matched pairs or Mann-Whitney U test was used to compare measurements. Results Mean IgG Nabs levels were significantly increased in post-LVAD sera at 1 (7040±1680 ELISA units, p<0.0001), 6 (14003±1520 ELISA units, p<0.0001) and 12 months (15012±2452 ELISA units, p<0.0001) compared to pre-LVAD (1368±379.3 ELISA units). Nabs could be detected above baseline levels by 7 days. No differences were observed for aPL. The addition of Nabs from post-LVAD serum significantly increased the clotting time of healthy donor whole blood (296.7±56.52 vs 180.3±50.28 s, p<0.05). Conclusion Our studies revealed that serum IgG Nabs reactive to oxidized epitopes are elevated following LVAD implant. Appearance of Nabs was not accompanied by aPL increase. IgG Nabs prolonged the clotting time of donor whole blood, suggesting that Nabs modulate the coagulation pathway. These findings suggest that IgG Nabs may contribute to hemocompatibility-related complications in LVAD recipients. [ABSTRACT FROM AUTHOR]

Purpose Accurate estimation of renal function using glomerular filtration rate (eGFR) is key for prediction of adverse events among HF pts undergoing LVAD implant. Changes in muscle mass are frequent among HF pts and represent a known confounder of serum creatinine (sCr) based equations. Cystatin C (CysC) is independent of muscle mass and provides an alternative measure of eGFR. We aimed to prospectively compare the value of CysC vs. sCr based eGFR in the prediction of postoperative outcomes among pts undergoing LVAD implant. Methods sCr and CysC were concurrently measured pre-LVAD in 73 pts (age 61±13, F 16%). CKD stages were defined by sCr-eGFR and CysC-eGFR using MDRD4 and CysC-CKD-EPI equations, respectively. The primary endpoint was a composite of severe right ventricular failure (sRVF) or renal replacement therapy (RRT) on the index admission. Predictive values of sCr-eGFR and CysC-eGFR were compared. The independent effect of sCr and CysC on the primary endpoint was assessed using multivariable models. Results Compared to pre-LVAD sCr-eGFR, CysC-eGFR reclassified 46 (63%) pts: 7 (16%) to earlier stage CKD and 39 (84%) to later stage CKD (Fig A). For every increase in CKD stage, there was a 60% increase in the risk of the primary endpoint using CysC-eGFR (p=.02) while the increase in risk did not reach statistical significance using sCr-eGFR (p=.19) (Fig B). CysC-eGFR outperformed sCr-eGFR as predictor of the primary endpoint (AUC 0.68 vs. 0.59, p=0.04, Fig C). In a model including both CysC and sCr: i) higher CysC was associated with an increased risk for the primary endpoint (OR: 1.60 per 0.5 mg/L increase, p=.01); ii) a trend towards a paradoxical association between higher sCr and lower risk was found (OR: 0.42 per 0.5 mg/dL increase, p=.08). Conclusion CysC-eGFR is a more reliable predictor of sRVF or need for RRT among pts undergoing LVAD implant. Higher sCr in relation to CysC may represent a marker of preserved muscle mass and thus might paradoxically confer improved prognosis. [ABSTRACT FROM AUTHOR]

Purpose Trimethylamine N-oxide (TMAO) is a novel, gut microbe-derived metabolite that directly contributes to platelet activation and thrombosis, and is associated with increased cardiovascular risk. Gastrointestinal bleeding (GIB) is the most common complication during LVAD support and is related to acquired coagulopathy. We hypothesized that higher levels of TMAO might have a protective effect against the risk of GIB in LVAD patients. Methods We prospectively enrolled 123 heart failure (HF) patients undergoing LVAD surgery at our institution between Jan 2011 and Aug 2016. TMAO was measured in blood at three time points pre- and post-LVAD: baseline (pre-LVAD), 3-6 mo post-LVAD, and >6 mo post-LVAD. Patients were categorized by > median vs ≤median TMAO level. We assessed time to GIB occurrence for each time point separately. Univariable Cox proportional hazard models were constructed, along with multivariable models adjusted for (i) age and sex and (ii) age, sex, and estimated glomerular filtration rate (eGFR). Results Among 123 pts (age 57.5 ± 14.8, 86% male, 59% white), TMAO levels were measured at baseline in 77 pts, 3-6 mo post-LVAD in 54 pts, and >6 mo post-LVAD in 60 pts. Median (IQR) TMAO levels were 7.67 (3.86, 13.80) for pre-LVAD, 5.38 (3.21, 9.82) for 3-6 mo, 5.97 (4.44, 10.78) >6 mo. In univariable analysis, there was no significant difference in time to GIB for above vs below median TMAO levels at all time points. These findings were similar in each of the multivariable models (Table). Conclusion Although TMAO levels have been associated with pro-thrombotic potentials, no association was found between pre- and post-LVAD levels of TMAO and the risk of GIB in our cohort of LVAD patients. [ABSTRACT FROM AUTHOR]

Purpose The use of veno-arterial extra-corporeal membrane oxygenation (VA-ECMO) following acute myocardial infarction (AMI) with cardiogenic shock (CS) is increasing, but the ability to predict favorable clinical response is limited. We investigated predictors of in hospital mortality and ventricular recovery in patients supported with VA-ECMO for AMI-CS. Methods We retrospectively reviewed charts of patients with AMI who were supported with VA-ECMO from 2007-2018. Data collected included AMI characteristics and measures of cardiac function. Logistic regression was used to identify predictors of survival to discharge and recovery which was defined as survival to discharge without heart replacement therapy (HRT, durable LVAD or transplant). Results 137 patients were treated with VA-ECMO for AMI during the study period. Mean cardiac index at implantation in the cohort was 1.82 ±0.53 and 26 patients (18.9%) had VA-ECMO placed during active CPR. 43 patients (31.4%) were supported with IABP either before or during VA-ECMO support; 75 patients (54.7%) were supported with Impella either before or during VA-ECMO support. 30-day survival was 60.6% and 52.6% survived to discharge. Of the 72 patients surviving to discharge, 26 (36.1%) were discharged with HRT and 46 (63.9%) had recovery. In a multivariable analysis including, age, gender, lactate, TIMI 3 flow in culprit artery, and LV ejection fraction, younger age was the only independent predictor of survival to discharge (OR 1.18, p=0.02). Independent predictors of ventricular recovery were presence of TIMI 3 flow in culprit artery (OR 5.45, p=0.03) and residual CAD (OR 0.22, p=0.04, Table). At a median follow-up of 375 days (IQR 112, 959) after discharge, 2/46 (4.3%) patients required HRT. Conclusion In patients supported with VA-ECMO for AMI-CS, younger age was associated with survival to discharge while markers of successful revascularization were associated with ventricular recovery. [ABSTRACT FROM AUTHOR]

Purpose ISHLT guidelines recommend serial right heart catheterization (RHC) to survey pulmonary hypertension in patients awaiting heart transplant. In LVAD as BTT patients the ideal surveillance frequency is unknown in those without prior pulmonary hypertension or whose pre-LVAD PVR normalized after surgery. Here, we review the utility of our surveillance RHC protocol. Methods Between January 2017 and July 2018, 69 patients who received an LVAD as BTT were studied. Routine RHC was performed every six months irrespective of prior PVR. Hemodynamic variables were retrospectively collected and reviewed. Results In 69 patients, 217 RHCs were analyzed. The median age was 55 years (IQR=15), and 55 (80%) were male. Median follow-up period was 18.4 months, and hemodynamic variables were largely unchanged post-LVAD (Table). Post-LVAD, overall PVR decreased from 3.5 ± 2.1 to 2.4 ± 1.0 Wood units (p = 7.2e-5) at six months and stabilized below 3 at all time points. Thirty-three patients (48%) had PVR ≤ 3 (1.8 ± 0.7), and 36 patients (52%) had PVR > 3 (4.9 ± 2.0) pre-LVAD. Both groups exhibited stabilization of PVR with mean < 3 at all post-LVAD time points (Figure). Of the 16 patients (23%) who had PVR > 3 at any point after LVAD implantation, 13 (81%) had a pre-LVAD PVR > 3. No patients with a PVR ≤ 3 six months post-LVAD implantation presented with a PVR > 3 at any subsequent RHC. Of the 40 patients (58%) transplanted, one patient developed severe primary graft dysfunction with right ventricular failure. Conclusion Routine RHC every six months may not be necessary for BTT patients, particularly for those with PVR ≤ 3 six months post LVAD. [ABSTRACT FROM AUTHOR]

Purpose Blood flow through the left ventricular assist device (LVAD) is dependent on the rotation speed of the impeller, and inversely dependent on the differential pressure across the LVAD (HQ relationship). We developed a novel approach to estimate intraventricular pressure difference (IVPD) in patients with LVAD and assess the change of IVPD according to pump speed during ramp study. Methods We developed in-house code of 1-dimensional incompressible Euler equation written in MATLAB (MathWorks, Natick, MA, USA) which allows us to measure IVPD by analyzing transesophageal echocardiographic images. Color M-mode images in mid-esophageal long axis view were obtained at 3 different pump speeds (low, optimized, high) in 10 patients who had LVAD implant (Figure). The optimal rotation speed was defined by stable hemodynamic conditions and the septal shape on transesophageal imaging. Pairwise comparison between each group was performed using Steel-Dwass analysis. Results Images were successfully obtained in all patients. Median IVPD values were 0.88 mmHg [IQR 0.73 - 1.02] at pre-LVAD implant, 1.86 mmHg [IQR 1.50 - 2.15] at low speed, 2.12 mmHg [IQR 1.72 - 2.71] at optimal speed, 2.09 mmHg [IQR1.72 - 2.31] at high speed, respectively (figure). There was significant difference between pre-implantation and the remaining 3 groups (p <0.005). IVPD was highest at optimal speed in 7 cases, at high speed in 2 cases, and at low speed in 1 case. Conclusion These data illustrate a novel approach to estimate IVPD during early diastole in patients with LVAD. Importantly, data suggests that setting of pump speed can be informed by analyzing IVPD during early diastole. [ABSTRACT FROM AUTHOR]

Purpose Effective tacrolimus (TAC) dosing is hampered by complex pharmacokinetics and significant intra- and interpatient variability. One prior report suggests that TAC dosing is influenced by gut microbiota, while the effects of systemic inflammation and oxidative stress are unknown. We evaluated these relationships in adult heart transplant (HT) pts. Methods We prospectively enrolled 24 HT pts (age=56±2 years, 88% male, 25% AA). Stool samples were analyzed using 16S rRNA sequencing. Alpha diversity metrics (number and evenness of bacterial taxa within samples) were estimated using Shannon index. TAC dose and serum trough level were measured at the time of gut and blood analysis. TAC doses were reported in mg/kg/day and as level-to-dose (L/D) ratio; pts were categorized as above vs. below the median. Biomarkers of inflammation (tumor necrosis factor-α (TNF-α)) and oxidative stress (Isoprostane) were measured in blood. Gut diversity and blood biomarkers were compared by TAC dose using ANOVA. DESeq regression analyses explored differences in individual taxa by weight-based TAC dose - false discovery rate (FDR) adjusted for multiple comparisons. Results The average time from HT to sample collection was 63±6 days. The median TAC dose was 0.1 mg/kg/day, while the median L/D ratio was 1.15. Above the median daily weight-based TAC dose was associated with higher Shannon index (p=0.03), lower TNF-α (p=0.05) and lower isoprostane levels (p=0.02) (Figure). These findings were similar when analyzed using the L/D ratio. We observed n=38 taxa to be significantly enriched among pts with >median TAC dose (all FDR<0.05), several of which are potential short-chain fatty acid producers with anti-inflammatory properties, including taxa from the families Ruminococcaceae or Lachnospiracae. Conclusion Our pilot study observed gut microbial diversity to be increased while inflammation and oxidative stress were reduced among pts requiring higher TAC dosing. [ABSTRACT FROM AUTHOR]

Purpose Patients with advanced HF are at increased risk for intra-cardiac thrombus formation. The impact of pre-existing left atrial (LA) or left ventricular (LV) thrombi on post-operative outcomes in LVAD patients is unknown. Methods We retrospectively reviewed 376 patients who underwent HeartMate II LVAD implantation from February 2009 through August 2018 at a single large volume center. Transthoracic echocardiography (TTE) within 1 month of implant, intra-operative transesophageal echocardiography (TEE), and operative reports were reviewed to identify patients with pre-existing LA (including appendage) or LV thrombus at the time of LVAD implantation. We assessed the composite endpoint of stroke and death at 6 months and 2 years. Results Of the 376 patients (age 57±13 years, 81% male, 47% ICM), 26 (6.9%) had a pre-existing thrombus (13 in LA and 15 in LV). Incidence of stroke at 6 months was 23% in patients with intracardiac thrombi and 9% in patients without intracardiac thrombi (p=0.02). Incidence of stroke at 2 years was 23% in patients with intracardiac thrombi and 14% in patients without intracardiac thrombi (p=0.22). Notably all strokes occurred in the first 6 months of device support in patients with intra-cardiac thrombi. Patients were significantly more likely to die or have a stroke in the first 6 months of device support if they had pre-existing intracardiac thrombi (HR (95%CI): 2.7 (1.4, 5.2), p<0.01) (Figure). This finding held true at 2 yr (HR (95%CI): 2.0 (1.1, 3.6), p=0.01). There was no difference in stroke incidence between patients with LV vs LA thrombus. It is the practice at our institution to perform thrombectomy when feasible if significant clot is identified at the time of device implantation. Conclusion Pre-existing LA or LV thrombus is associated with decreased survival free of stroke following CF-LVAD implantation which may have important implications for patient selection and management. [ABSTRACT FROM AUTHOR]

Purpose Declining renal function portends poor prognosis in advanced HF. Hemodynamic optimization with diuretics, inotropes, and temporary mechanical circulatory support is frequently performed for optimization of end-organ function before LVAD surgery. We aimed to study the relationship between long- and short-term changes in serum creatinine (sCr) that occur prior to implant and postoperative outcome in LVAD pts. Methods We retrospectively identified 146 LVAD pts with sCr at 3 time points: 1-year (1y), day of index admission, and 1-day (1d) prior to LVAD. Pts were classified based on the changes in sCr that occurred between 1y and 1d prior to LVAD as: Improved = sCr reduction >10% (n=42); Stable = sCr change ≤10% (n=50); Worsened = sCr increase >10% (n=54). Pts were also classified as Optimized using the following exploratory cutoffs: sCr reduction of ≥10, 15, and 20% during the index admission. The primary endpoint was survival free from renal replacement therapy (RRT) after LVAD. Multivariable Cox models were developed to adjust for age, sex, race and sCr 1d pre-LVAD. Results On average, sCr increased from 1 y to the time of index admission, while subsequently declined prior to LVAD (Fig A). Among 147 pts, 47 (31%) met the primary endpoint. In univariable model, Stable 1y-1d sCr was associated with a reduced risk for the primary endpoint (Fig B). After adjustment, there was only a trend towards significance in Stable vs. Improved (p=0.06). Optimization of sCr during index admission was not associated with improved outcomes. Furthermore, in an adjusted model, a decrease in sCr≥15% was significantly associated with increased risk for the primary endpoint (Fig B). Conclusion In the year preceding LVAD implant, a stable renal function is associated with a trend towards improved post-LVAD survival free from RRT. Moreover, a decrease in sCr≥15% during index admission was associated with worse outcome. The effect of sarcopenia on sCr levels in advanced HF may explain these paradoxical findings. [ABSTRACT FROM AUTHOR]

Purpose Right ventricular (RV) failure remains a major concern in heart failure (HF) patients undergoing LVAD surgery. We aimed to measure flow kinetic energy (KE) of the RV outflow tract (OT) - a new marker of RV global systolic function - and assess its relationship to well-established echocardiographic and hemodynamic parameters of RV performance in heart failure (HF) and LVAD pts. Methods Twenty-one subjects were prospectively enrolled: 12 HF pts (2 NYHA Class II and 10 Class IV), 3 HeartMate II (HMII) and 6 healthy controls (C). Echocardiographic parameters of RV function (TAPSE, St', Et', IVA, MPI) and RV inflow-outflow color Doppler images were recorded on all pts. Invasive hemodynamic parameters of RV function and RV failure risk score (RVFRS) were collected in all Class IV pts. KE-RVOT was derived from color Doppler images using a vector flow mapping proprietary software, Echo VFM® (Cardio Flow Design, Tokyo, Japan) (Figure). Kruskal-Wallis test was performed for comparison of each group. Correlation between KE-RVOT and echocardiographic/hemodynamic parameters as well as RVFRS was assessed by linear regression analysis. Results KERVOT (Median ± IQR) was higher in C (56.98 [37.63 to 97.21] mW/m) than in Class II HF (26.60 [26.14 to 27.05] mW/m, p <0.05) and further declined in Class IV HF (7.45 [4.87 to 10.46] mW/m, p <0.05)). LVAD therapy partially restored KE-RVOT (23.17 [9.77 to 25.03] mW/m, p <0.05)) to levels similar to Class II HF (Figure). KE-RVOT was significantly related to TAPSE, St', CVP/PCWP, and PAPi (r = 0.5887, 0.6671, -0.6942, and 0.7899, respectively, all p <0.05) but not to RVFRS. Conclusion KE-RVOT, a novel noninvasive measure of RV function, strongly correlates with well-established markers of RV performance. The utility of KE-RVOT in prediction of RV failure post LVAD requires further study. [ABSTRACT FROM AUTHOR]

Purpose Early right ventricular (RV) failure after LVAD is associated with increased morbidity and mortality. Several hemodynamic indices have been suggested to identify patients at risk for RV failure, however their predictive value remains unclear. Methods We retrospectively identified patients at our center who required mechanical RV support following LVAD between 2007-2018, with either surgical (Centrimag RVAD) or percutaneous (Impella RP, Protek Duo) devices. Those receiving RVAD at time of LVAD insertion were excluded. Predictors of RV failure were calculated with preoperative hemodynamics, including RA/PCWP (low risk (LR) <0.64, high risk (HR) >0.64), PAPI (LR >2, medium risk (MR) 1.5-2, HR <1.5), RVSWI (LR >600 mmHg*mL/m2, MR 300-600, HR <300), RV Failure Risk Score (RVFRS; LR <3, MR 4-5, HR >5.5), and EUROMACS-RHF (LR <2, MR 2-4, HR >4). Results 37 patients required mechanical RV support. 18 received percutaneous RVAD and 19 surgical RVAD. Figure 1 summarizes the proportion of patients who would have been identified as low, medium, or high risk of RV failure. Mean RA/PCWP was 0.43 ± 0.19, with 13.5% of patients identified as high risk. 70.3% of patients had PAPI suggestive of low risk, with only 10.8% and 18.9% identified as medium or high risk, respectively. RVSWI identified 51.3% of patients as medium risk of RV failure and 21.6% as high risk, with mean 519 ± 316 mmHg*mL/m2. EUROMACS-RHF classified 51.3% as medium risk and 22% as high risk, while the RVFRS identified the majority of patients as low risk (83.8%). Conclusion Among patients who ultimately required mechanical support for RV failure following LVAD insertion, the majority would not have been identified as at risk using most currently available preoperative risk assessments. RVSWI and EUROMACS-RHF best predicted RV failure, identifying the majority of patients as at medium risk. In patients not identified as at risk preoperatively, easily deployed RV support may be an important means to rapidly address RV failure after LVAD insertion. [ABSTRACT FROM AUTHOR]

Purpose When left ventricular assist device (LVAD) therapy no longer fullfills a patient's desired quality of life, the patient or their surrogate may request withdrawal. However, the details of LVAD withdrawal have not been previously studied. The aim of this study is to describe the factors associated with LVAD withdrawal. Methods Among newly implanted LVADs from 2010 until July 2018 in a single institution, those who underwent withdrawal were included. We divided this cohort into 2 groups based on whether patients initiated the decision of withdrawal (PT+) or patients could not participate due to lack of decision-making capacity (PT-). In PT+, we calculated the days between the patients' request of withdrawal and deactivation (WD_PT+). In PT-, we calculated the days between the time when patients lost capacity and deactivation (WD_PT-). Multivariate linear regression analyses were done to predict WD_PT-. Place of withdrawal, palliative care involvement, and other clinical factors were also analyzed. Results A total of 61 patients (14 in PT+, 47 in PT-) underwent LVAD withdrawal. Mean age at death was 65.4 ±10.1years and 85.2% were male. Median survival was 170 days. Palliative care was involved in the majority of LVAD withdrawals. Most LVAD withdrawals occurred in the ICU. The median WD_PT was 5.5 days (IQR 39.1), and was shorter in the ICU (n=9, 1 day, IQR 7) than non-ICU settings (n=5, 46 days, IQR 95.5). The median WD_PT- was 5.5 days (IQR 9.3), and was shorter in the ICU (n=42, 4 days, IQR 10) than in non-ICU settings (n=4, 13 days, IQR 32.3). A multivariate analysis showed that male and ICU setting were significantly associated with WD_PT- (p=0.025 for male, p=0.005 for ICU setting) Conclusion The majority of LVAD withdrawals were done in the ICU. Patients' request of withdrawal was respected sooner and surrogates' decisions for withdrawal occurred sooner in the ICU. Time to surrogates' decision for withdrawal was shorter in male patients. Palliative care was involved in the majority of withdrawals. [ABSTRACT FROM AUTHOR]

Diabetes mellitus is a risk factor for morbidity and mortality in patients with heart failure. The effect of pre-implant diabetes control on left ventricular device (LVAD) outcomes and specific types of infection is unclear. Prior studies have found that diabetes is associated with an increased rate of all-cause mortality and major adverse events during LVAD support, while others have shown that diabetes is not associated with major adverse events, including infections. The purpose of our study was to understand the relationship of pre-LVAD hemoglobin A1c on post implant infection. We analyzed 2054 patients from the INTERMACs registry who were implanted with an LVAD and had a hemoglobin A1c level checked between Jan 2016-Dec 2017. The association between pre-implantation hemoglobin A1c and post-implant infection within 90 days was assessed using a cox proportional hazards model. Pre-implant hemoglobin A1c levels had no significant association with post-LVAD infections within 90 days of follow-up (HR 1.02 [95% CI 0.96-1.09], p 0.544). Furthermore, this lack of association did not vary within subgroups of infection type (driveline, bloodstream, pulmonary, or urinary tract). (Figure 1) Based on this large database, pre-LVAD hemoglobin A1c is not a predictor of early post-LVAD infection. Further studies are needed to better understand the relationship of hemoglobin A1c as it relates to clinical outcomes after LVAD implantation. [ABSTRACT FROM AUTHOR]

We sought to define clinical, hemodynamic, metabolic, and both drug and device treatment parameters associated with clinical outcomes among heart failure patients presenting with cardiogenic shock (HF-CS) patients using data from the Cardiogenic Shock Working Group (CSWG) Registry. HF-CS is increasingly common however most studies have focused exclusively on acute myocardial infarction related CS. HF-CS patients were selected from the multi-center CSWG registry were divided into 3 outcome cohorts assessed at hospital discharge: mortality, heart replacement therapy (HRT: durable VAD or heart transplant), or native heart survival (NHS: without heart replacement therapy). Clinical characteristics, hemodynamic and laboratory parameters, drug therapies, acute mechanical circulatory support device (AMCS) utilization, and Society of Cardiovascular Angiography and Intervention (SCAI) stages were compared across the 3 outcome cohorts. Of the 712 HF-CS patients identified, 180 (25.3%) died during their hospitalization, 277 (38.9%) underwent HRT, and 255 (35.8%) experienced NHS without HRT. Patients who died had the highest right atrial pressure and heart rate and the lowest mean arterial pressure of the 3 outcome groups (p <0.01 for all). BiV congestion was most common among patients who died, while isolated left-ventricular congestion was most common among patients who underwent HRT. Lactate, BUN, serum creatinine, and AST were highest in HF-CS patients experiencing in-hospital death. In-hospital mortality increased with deteriorating SCAI stages (stage B: 0%, stage C: 10.7%, stage D: 29.4%, stage E: 54.5%, p <0.001), and with the use of AMCS devices (IABP: 23%, Impella: 45.3%, or ECMO: 54.7%, p<0.001). In patients with HF-CS biventricular profile, surrogates of end-organ dysfunction, increased SCAI staging and the use of AMCS devices were all associated with a higher probability of in-hospital mortality. [ABSTRACT FROM AUTHOR]

Cardiogenic shock (CS) mortality remains unacceptably high. Risk stratification of CS patients may improve trial design and treatment strategies. We tested the application of the expert consensus-based Society for Cardiovascular Angiography and Intervention (SCAI) CS stages in a retrospective dataset. Furthermore, we hypothesized that unsupervised clustering of CS patients is feasible and identifies clinically applicable phenotypes of CS. The Cardiogenic Shock Working Group (CSWG) registry is a retrospective multicenter registry of patients with all-cause CS. Descriptive analyses were performed on 1,414 CSWG patients to quantify the association of clinical variables and in-hospital mortality. In myocardial infarction-caused CS (AMICS) patients, a random forest classifier was used to identify variables that predict in-hospital mortality in CSWG. Consensus k means clustering was performed on these variables in 1,000 randomly assigned 85% samples of 408 AMICS patients from the CSWG training cohort and 701 AMICS patients from the Danish National Patient Registry (DNPR). CSWG patients with heart failure (CSWG-HF, n=480) were then assigned to the clusters. Variables associated with in-hospital mortality in all-cause CS included SCAI Stages, but also pulmonary artery catheter use, and venous congestion. In AMICS the mortality driving continuous variables comprised glomerular filtration rate, lactate, HCO3, platelets, alanine aminotransferase, and white blood cell count. Independent, unsupervised consensus k means clustering on these variables identified three phenotypes of CS that were associated with in-hospital mortality in similar patterns in CSWG, DNPR, and CSWG-HF. The phenotypes revealed unique demographic, clinical, metabolic and hemodynamic profiles. This is the first report to employ an unbiased, machine learning approach in CS and identified 3 phenotypes of patients with CS due to MI. These findings may inform treatment strategies and future CS clinical trials. [ABSTRACT FROM AUTHOR]

Left ventricular assist devices (LVAD) are widely used as a support strategy for advanced heart failure. Complications such as thrombosis and bleeding have been linked to LVAD. We observed that LVAD implantation was followed by a sharp increase in serum levels of IgG natural antibodies (Nabs) recognizing oxidation-specific epitopes (OSE) and apoptotic cells. Nabs have been implicated in inflammatory reactions related to atherosclerosis, ischemic stroke and primary graft dysfunction following heart transplantation. However, the consequence of IgG Nabs production following LVAD is not known. We hypothesized that Nabs generated following implant may be associated with LVAD complications, promoting thrombosis, stroke or bleeding. We studied a multicenter cohort (n=71) of adult patients who received a continuous flow LVAD at Columbia University Irving Medical Center (CUIMC), Virginia Commonwealth University (VCU) and Baylor University Medical Center (BUMC). Nabs were measured in pre- and longitudinal post-LVAD sera by assessing IgG reactivity to the OSE malondialdehyde (MDA) by ELISA and to apoptotic Jurkat cells by flow cytometry. Wilcoxon's signed rank test for matched pairs was used to compare measurements. Mean IgG Nabs levels were significantly increased in post-LVAD sera at 1 (0.41±0.05 Nabs units, p<0.0001), 6 (0.63±0.074, p<0.0001) and 12 months (0.49±0.69, p=0.0004) compared to pre-LVAD (0.19±0.021). Kinetics of Nabs development was found to differ between centers with 90.9% of CUIMC patients developing elevated Nabs at 6 months post-implantation compared to 30% and 29.7% for BUMC and VCU, respectively. Preliminary analysis revealed elevated Nabs in 2/3 VCU patients with pump thrombosis and 1/2 with stroke within the first year post-LVAD. Our studies revealed that serum IgG Nabs reactive to oxidized epitopes and apoptotic cells are elevated following LVAD implant. The appearance of Nabs differed between the centers. and we are investigating the possible contribution of differing perioperative procedures on Nabs development. Preliminary findings suggest that IgG Nabs may be associated with thromboembolism. Analysis of Nabs and adverse events in the other centers are forthcoming. [ABSTRACT FROM AUTHOR]