Search

Background: The impact of continuous flow resulting from contemporary left ventricular assist devices (LVAD) on renal vascular physiology is unknown. Renal resistive index (RRI) reflects arterial compliance, as well as renal vascular resistance, contributed by afferent and efferent arteriolar tone, the renal interstitium as well as renal venous pressures. Methods: Prospective, single center study with renal Doppler evaluation at baseline (pre‐implant) and at 3‐months support. Outcomes assessed include need for post‐operative renal replacement therapy (RRT), worsening renal function (WRF) defined as persistent increase from pre‐implant KDIGO chronic kidney disease stage, right ventricular (RV) failure, and survival to transplantation. Results: Pre‐implant RRI did not predict cardiorenal outcomes including right heart failure, need for renal replacement therapy or worsening renal function. Post‐implant RRI was significantly lower than pre‐implant RRI, with a distinct Doppler waveform characteristic of continuous flow. Post‐implant renal end‐diastolic velocity, but not RRI, correlated strongly with LVAD flow (Spearman rho −0.99, p < 0.001), with trend toward correlation with mean arterial pressure (Spearman's rho 0.63, p = 0.129). There was a negative correlation between post‐implant RRI and mean pulmonary artery pressure (Spearman's rho −0.81, p = 0.049), likely driven by elevated pulmonary capillary wedge pressure (Spearman's rho −0.83, p = 0.058). Conclusion: The hemodynamic contributors to RRI in LVAD supported patients are complex. Higher mean pulmonary artery and pulmonary capillary wedge pressures seen in lower RRI may reflect a smaller difference in systolic and diastolic flow. Future simultaneous Doppler assessment of the LVAD outflow graft and RRI may help understand the hemodynamic interactions contributing to this index. [ABSTRACT FROM AUTHOR]

The CorWave LVAD employs a novel wave membrane technology to generate low shear blood propulsion, synchronized to the native left ventricle. Here, we report the results of preclinical testing of hemodynamics, hemocompatibility, and durability for the CorWave LVAD. The pump motor, membrane, blood flow path, magnetics, and mechanical components were designed using extensive computational simulations, then rigorously tested. Durability was tested at the material, component, pump, and system level with both real-time and, when feasible, accelerated fatigue tests. Hemocompatibility was assessed in vitro and in acute (<6 hr) and chronic (up to 90 day) implants in sheep, including "thrombo-provocative" tests in which anticoagulation was reversed 10 mins after pump implant and pump output was intentionally kept to 1-3 LPM. An adaptive algorithm to automatically adjust to changes in hemodynamics was developed and tested with In vitro mock circulation loops (MCLs), acute implants in sheep with induced heart failure or RV-bypass, and chronic implants in healthy sheep. These studies were used to evaluate and improve the pump algorithm for synchronization with the LV, arrhythmia detection, suction detection and response, and responses to hemodynamic changes. Material testing and simulations predict multi-decade durability of the wave membrane. Multiple prototype pumps have completed 2-year durability tests. 12 life cycle test stands have been built and validated to commence multi-year, real-time durability testing of the clinical version of the LVAD, currently in production. Hemodynamic testing demonstrated successful LV synchronization to 95% of the native LV contractions, while the algorithm successfully detected arrhythmias, suction events, heart rate changes, and after-load changes. Responses to suction events and changes in pre-load have been revised and are undergoing confirmatory testing. Low hemolysis and preservation of von Willebrand factor activity were demonstrated in chronic implants. Thrombo-provocative studies demonstrated successful design modifications to eliminate thrombus on vulnerable pump surfaces. The CorWave LVAD has undergone extensive design efforts and a comprehensive pre-clinical test regime to establish the clinical design will have the required durability, hemocompatibility, and adaptability to commence clinical trials. [ABSTRACT FROM AUTHOR]

Minimally invasive left ventricular assist device (LVAD) implantation via left thoracotomy has previously been shown to be a safe operation with potential advantages compared to sternotomy. These advantages include improved inflow cannula positioning as well as minimising adhesions for subsequent sternal entry for heart transplantation. We wanted to assess the impact of LVAD implant technique on subsequent heart transplantation. All patients who received an LVAD via thoracotomy and subsequently transplanted were included. This group was matched 2:1 with patients who received an LVAD via sternotomy and were subsequently transplanted. Patients in the sternotomy group were excluded if they had a prior sternotomy to LVAD insertion. Parameters such as cardiopulmonary bypass time, blood usage, post-operative length of stay, morbidity and mortality measures were extracted. 12 Thoracotomy patients and 24 sternotomy patients were included in the study. There was no significant difference in gender, BMI or age between the two groups. When assessing operation variables, there was no significant difference in operation duration, cardiopulmonary bypass time or cross clamp time between the sternotomy and thoracotomy groups. Similarly, there was no significant difference in blood product administration (red blood cells, cryoprecipitate, fresh frozen plasma, platelets) intraoperatively. There was no difference in total length of stay, ICU length of stay, ventilation time and 30-day mortality. In our sample size of 36 patients, we could not demonstrate a significant difference in peri-operative variables between VAD patients who presented for their heart transplant subsequent to receiving an LVAD implanted via either sternotomy or thoracotomy. These findings suggest that while thoracotomy for LVAD insertion may present advantages such as improved inflow cannula positioning, we could not demonstrate a clear benefit in sternal entry for subsequent transplant. [ABSTRACT FROM AUTHOR]

Purpose Time in Therapeutic range (TTR) is the percentage of time an anticoagulated patient has an INR between 2-3. While studies have demonstrated a link between low TTR, bleeding and thrombosis, the mean TTR in a Ventricular Assist device (VAD) population is between 45 and 65%. We implemented a new anti- coagulation protocol using point of care devices (POC, CoaguChek®), with patient education and empowerment to control their own Warfarin dosing. Methods Before discharge all VAD patients received anticoagulation, education including drug and food interaction. Depending on INR stability, patients measured INR every 2-3 days and sent a text message to the VAD team for the appropriate Warfarin dose. After a minimum of 2 months post VAD, if the patient and team felt comfortable, Warfarin dosing was delegated to the patient. Patients updated the team weekly. Results Between February and October 2020, 24 patients were included (29% HeartMate 3 (HM3), 71% HeartWare (HW), one Bi-VAD (HW)). Median follow-up was 131 days (IQR 88-163), cumulative follow up time was 3087day or 8.5 years. Median TTR was 90% (IQR 83-97), median INR 2.41 (IQR 2.2-2.4). INR was measured 3.6 times / week (IQR 3.3-3.8). In 15 cases (62%), the VKA dosing decision was delegated to the patient. Median TTR during patient management (96%, IQR 90-97) was significantly higher than the TTR during medical control (83%, IQR 78-88) (P<0.0001). There were no deaths, major bleeding events, pump thrombosis or strokes under patient management. Ten patients (41%) received one or more doses of Enoxaparin (1mg/kg once a day) as bridge for INR < 1.8 for HW and 1.6 for HM3 pump. Conclusion A TTR higher than described in the literature is achievable in VAD patients using a POC device. If prerequisites are met, VAD patients can be empowered to control their Warfarin dosing with excellent TTR. [ABSTRACT FROM AUTHOR]

The ApixiVad study aims to test the non-inferiority of Apixaban compared to oral anticoagulation with Warfarin in patients supported by an LVAD. A secondary outcome is to test patient satisfaction according to the type of anticoagulation. Patients followed at our institution are randomised to maintain warfarin therapy or to switch to Apixaban. On the day of randomisation and follow-up, patients were asked to complete a satisfaction questionary regarding anticoagulation, the Anti-Clot-Treatment Satisfaction (ACTS). Of the first 13 patients randomised, only one was female (1/13, 3.5%), median age 45 (50-59), and all were implanted as BTT with a HeartMade 3. At the time of randomisation, the patients had been on LVAD for a median of 3.6 months (3.1-4.5); 8 were randomised to the apixaban group and 5 to the Warfarin group. The burden related to anticoagulation decreased after the switch to Apixaban, from a score of 18.5/60 to 15.5/60, whereas the patients on Marcoumar saw their burden increase from a score of 19.4/60 to 23.8/60 (Illustration, Panel A). The satisfaction with anticoagulation was not different between the two groups (illustration, Panel B). When looking at the response to the questionnaire in detail, the improvement in the score of Apixaban patients is due to the simplification of the drug administration (Figure, Panel C). The quality of life, measured by the ACTS questionnaire, increases in patients with LVAD switching from Warfarin to Apixaban. [ABSTRACT FROM AUTHOR]

The CorWave LVAD uses a unique wave membrane technology for low shear blood propulsion, operating in a pulsatile mode, synchronized with the native ventricle. Preclinical testing suggests this new LVAD could overcome several limitations of continuous flow LVADs. The device will be evaluated in a multicenter feasibility study in Australia, Europe, and Asia. Patients, Clinicians, and Device Developers share a mutual interest to confirm device safety and performance as early as possible. Up to 15 patients will be enrolled using typical LVAD inclusion/exclusion criteria: patients with worsening heart failure despite optimal medical management. As in other LVAD trials, the primary endpoint will be survival. Secondary endpoints include freedom from adverse events (AEs), (e.g. stroke, bleeding, infection, and pump replacement), quality of life and neurocognitive measures. Given that this is a feasibility study, not powered to demonstrate clinical benefit, we also plan to collect data predictive of AEs, validating preclinical data where applicable. These will include von Willebrand Factor assays, arterial pulse pressure and aortic valve opening, which are factors implicated in bleeding and aortic insufficiency. Increased aorta stiffness post-LVAD, correlated with increased AEs, will be measured. The synchronous, pulsatile flow of the CorWave LVAD is hypothesized to improve patient exercise capacity and possibly native ventricular recovery. Thus, patients will undergo exercise testing and evaluation of ventricular size and ejection fraction while on pump support. Potential outcomes include 1) Low AEs and favorable biomarker and functional test data, permitting judicious initiation of a pivotal trial, 2) Equivocal AEs and/or inconclusive biomarker and functional results, enabling Clinicians and Device Developers to collaborate to identify improvements prior to a pivotal trial, or 3) Increased AEs with negative biomarker and functional data, requiring device design and/or management revisions prior to re-commencing clinical studies. Survival rates achieved with current LVADs create a challenging standard, but the high AE rates offer opportunities for new technologies to prevent stagnation in the field. New approaches to cost-effective early LVAD trial design are required to sustain innovation and safely transition into pivotal trials. [ABSTRACT FROM AUTHOR]

Left ventricular assist device (VAD) implantation via thoracotomy is a safe operation with potential advantages compared to conventional sternotomy. One proposed advantage is improved inflow cannula (IFC) positioning due to better visualisation. We investigated the difference in IFC angles for patients with HVADs implanted via these techniques and compared post-operative as well as long term outcomes. 42 patients were retrospectively selected from our VAD database. 28 in the sternotomy group and 14 in the thoracotomy group. We included all thoracotomy implants performed at our centre and consecutive sternotomy implants from 2015-2019. Inclusion required a minimum support time of 30 days. Patients with concomitant procedures were excluded. The first post-operative CT chest scan was used to measure the angle between the IFC and mitral valve in axial and coronal planes. There was no significant difference in gender, INTERMACS score, BMI or age between the two groups. Median support time was not significantly different between the thoracotomy and sternotomy groups (328 vs 360 days, p=0.9). Median cardiopulmonary bypass (CPB) time was higher in the thoracotomy group (104 vs 78 minutes, p<0.01). Median ICU length of stay for the thoracotomy group was 5 days compared to 8 days in the sternotomy group, p=0.08. There was no significant difference in median IFC angle on the axial plane between the thoracotomy group and the sternotomy group (9 degrees v 10 degrees, p=0.3). Similarly, there was no difference in the coronal plane (9 degrees vs 12 degrees, p=0.9). Rates of pump thrombosis, stroke and GI bleeding were not significantly different between the two groups. In our sample size of 42 patients, the thoracotomy group had a slightly longer CPB time, however, there was no difference in IFC angle, pump thrombosis, GI bleeding, stroke or ICU length of stay. In saying this, there may be other benefits to the thoracotomy approach such as facilitating easier sternal entry upon transplantation. [ABSTRACT FROM AUTHOR]

Vasoplegia post heart transplantation is a well-recognised phenomenon, yet it is poorly defined within the literature, with contention regarding criteria for diagnosis. We therefore created a scoring system to retrospectively identity vasoplegic patients by rating their vasopressor requirements, cardiac function, and markers of vasodilation. A total of 45 patients (71% male) who underwent heart transplantation were chosen and grouped into 3 cohorts to isolate key risk factors for vasoplegia. 15 patients underwent ventricular assist device explant (VE), 15 patients had undergone previous sternotomies (PS), and 15 had no previous sternotomies (NPS). The scoring tool was used to allocate points based on high vasopressor requirement, preserved cardiac index, low systemic vascular resistance and low SvO2. Demographic, operative, and post-operative data was obtained from the medical records, operation reports and ICU charts. Using the Vasoplegia Scoring System, a total of 13 patients were identified as having developed vasoplegia within the first 48 hours following transplant (p=1.0). This was most apparent at 24 hours with 4 VE, 2 PS and 5 NPS patients developing vasoplegia (p=0.57). However, using the Vasoplegia Scoring System, we could not detect a difference in the rates of vasoplegia between the three cohorts which may be on account of a small sample size. With similar rates of vasoplegia, there did not appear to be any difference in 30-day mortality, ICU length of stay and overall hospital length of stay amongst the three groups. Using the Vasoplegia Scoring System, we identified that 29% of the patients developed vasoplegia in the immediate post-operative period. Using this scoring system, there was no statistically significant difference between the three groups in terms of which patients experienced vasoplegia at higher rates. As such, there did not appear to be any difference in their post operative outcomes. [ABSTRACT FROM AUTHOR]

Aortic valve (AV) opening in patients supported with cfLVADS may be a marker of improved myocardial contractility. We aimed to study if status of aortic valve opening is influenced by right ventricular function, preload and afterload both at rest and exercise with pump speed titration. Patients implanted with an isolated centrifugal cfLVAD undergoing routine right heart catheterisation were recruited. Mean arterial pressure (MAP), mean pulmonary capillary wedge pressure (PCWP) and pulmonary artery pressures were recorded. The pulmonary artery pressure index (PAPI) derived from formula was used as a surrogate for right ventricular function. Aortic valve status was determined by transthoracic echocardiography. Patients then performed graded exercise with supine bicycle ergometer until exhaustion (mean workload 40W) at both baseline and maximum pump speed with all measurements repeated. Patients were divided into two groups: closed or open AV. Thirteen patients were studied, mean age 37.4+/-7.4 yrs. Median NYHA class at time of study was II. Median pump speed at rest was 2600rpm (range 2500-2800). Echo derived baseline left ventricular ejection fraction was similar between both groups, closed AV valve 25+/-7.7, open 24.2+/-8.0%, p=0.86. At rest, there was no significant difference between closed AV and open AV for MAP 82.7+/-6.0 vs 85.2 +/-6.0mmHG, p=0.60, PCWP 14.7+/-0.2 vs 16.8+/-6.0mmHg, p=0.46 and PAPI 2.4+/-0.8 vs 2.9+/-0.9, p=0.33. Similarly, no significant difference was noted in these measures at maximum exercise and at maximum pump speed (median 2925rpm, range 2800-3210). Aortic valve opening in patients supported by cfLVADs is not significantly influenced by afterload or preload at both rest and exercise. The contribution of intrinsic myocardial contractility derived by load independent methods on aortic valve opening is warranted. [ABSTRACT FROM AUTHOR]

Purpose Driveline infections remain a major complication of ventricular assist device (VAD) implantation. This study aimed to characterize in vivo microbial biofilms causing driveline infections and the degree of tissue integration into the velour of explanted drivelines. Methods Drivelines were obtained from ten VAD patients undergoing heart transplantation at the Alfred Hospital, Melbourne and St Vincent Hospital, Sydney between June 2017 to October 2018. Four uninfected drivelines and six infected drivelines were aseptically sectioned into 10 pieces of 1.5 cm in length, from the smooth tube section at the exit site to the velour section adjacent to the VAD. Organisms on each section were isolated and identified. Microbial biofilms on the infected driveline sections were assessed using scanning electron microscopy (SEM) and viable counts. Computed tomography (micro-CT) and SEM were used to assess tissue integration into each section. All sections were analysed histologically to confirm human matrix protein deposition. Results No organisms were isolated from non-infected drivelines using microbiological culture. Staphylococcus aureus, Pseudomonas aeruginosa , and Staphylococcus epidermidis were found on the velour sections of infected drivelines and correlated with the microbiological culture results from patient swabs of the infected exit site. Although histological analysis found considerable collagen fiber and fibroblast deposition on both the smooth tube and the velour, SEM and micro-CT suggested insufficient tissue integration throughout the driveline velour. Microgaps were observed between the velour fibres, with evidence of microbial biofilms within these gaps. Such biofilms were morphologically distinct from in vitro biofilms grown on biomaterials and might be responsible for the antimicrobial treatment failure. Conclusion This study demonstrated inadequate tissue integration of clinical drivelines in the subcutaneous tissue tunnel, with associated microbial biofilms formed within the microgaps between velour fibres. These data provide important insights into a potentially novel therapeutic strategy against driveline infections that is focused on enhancing tissue integration into the velour, thereby preventing microbial adherence, biofilm formation and migration. [ABSTRACT FROM AUTHOR]

Renal function trajectory is well described in patients supported with ventricular assist devices (VAD), however there are limited data on major adverse kidney events (MAKE) and their impact on survival to transplantation. We describe the relationship between MAKE and adverse events in patients supported with VAD. We conducted a single-centre retrospective analysis of consecutive VAD implants between 2010-2019. Baseline demographics, INTERMACS profile, biochemistry and adverse events were collected for the duration of VAD support. MAKE was defined as the first event to occur of sustained drop in eGFR>50%, stage V chronic kidney disease, initiation of dialysis beyond index VAD implant admission, or death on dialysis at any time. 173 patients were included, median age 56.8 years, 18.5% female, 75.1% INTERMACS 1 or 2. 46 patients did not survive to transplant. MAKE occurred in 37 patients. Survival to transplant was significantly reduced in patients with MAKE (p<0.001), shown in figure 1. All components of MAKE were associated with mortality on univariate analysis. In a multivariate Cox regression analysis, female gender and MAKE, but not age, INTERMACS profile, or pre-implant estimated glomerular filtration rate, were significantly associated with mortality. Gastrointestinal bleeding was more common in patients with MAKE (odds ratio 2.95, 95% CI 1.28-6.87, p=0.009). MAKE was not associated with other adverse hemocompatibility events. MAKE is significantly associated with reduced survival to transplant in patients supported with VAD. Contributors to MAKE and means of preserving renal function should be identified in VAD supported patients to address this excess in mortality. [ABSTRACT FROM AUTHOR]

Apixaban has been proposed to be a suitable alternative to vitamin K antagonists (VKAs) in patients with left ventricular assist devices (LVADs). Important considerations for its use in bridge to transplant (BTT) patients with LVADs are perioperative transfusion requirements and intensive care unit length of stay (ICU LoS). While the recombinant competitive inhibitor andexanet-alpha has been shown to reverse factor Xa inhibition, the agent is not available globally. We report our initial experience in three LVAD patients anticoagulated with apixaban and transplanted without andexanet-alpha available. On notification of successful donor matching, patients withheld any further apixaban doses. Data for blood product requirements, post-operative ICU LoS and 30-day survival were recorded. Data from the apixaban group were compared to historical data of VKA treated BTT LVAD patients undergoing transplantation from our institution. All patients transplanted had Abbott HeartMate-3 (HM3) devices and were anticoagulated with apixaban 2.5mg twice daily. The median (range) age was 51 (10-51) years. Duration of LVAD support was 6.4 (1.8-12.7) months. The time between recipient call and operating theatre time was 17.4 (15.4-18.7) hours. Operation duration was 6.1 (5.6-7.6) hours. Intra-operative packed red blood cell (PRBC) product use in these three patients ranged from 0 to 3 units. In historical data, 132 patients transplanted after LVAD explantation at our institution received a median of 2 (IQR 1-3) PRBC units. Other peri-operative blood product usage is outlined in Table 1. The ICU LoS ranged from 5 to 9 days. All patients were alive at 30 days. Despite the lack of the competitive inhibitor at our centre, the pharmacodynamics of apixaban allowed for safe transplantation in these patients. Further studies confirming the safety of apixaban in this population are warranted. [ABSTRACT FROM AUTHOR]

Sodium-glucose cotransporter-2 inhibitors (SGLT2i) have evolved into a pillar of the heart failure therapy regimen, but their safety and tolerability in end-stage heart failure patients supported with left-ventricular-assist-devices (LVADs) has not been well-studied. We retrospectively analysed the electronic medical records of outpatients who were implanted with LVADs as bridge-to-transplant (BTT) at our institution between October 2020 and August 2022. Data on demographic profile, renal function (estimated glomerular filtration rate, eGFR) hemocompatibility related adverse events (stroke, pump thrombosis, gastrointestinal bleeding, other bleeding) and infection (sepsis, driveline infection) were analysed. Sixteen of 31 patients (51.6%) were prescribed SGLT2i, mean exposure duration of 132.7 ± 118.8 days. There were no significant differences in baseline demographics (age, weight, LVAD type, diabetes, renal function) between SGLT2i and non-SGLT2i groups. On pairwise analysis, neither group experienced significant change in eGFR from first to last clinic visits, SGLT2i 84.7 ± 25.3 to 70.1 ± 24.2 mL/min/1.73m2, p=0.11 vs non-SGLT2i group, 85.8 ± 25.8 to 72.9 ± 20.8 mL/min/1.73m2, p=0.14. In the SGLT2i group, there was no significant change in eGFR from time of SGLT2i initiation to last clinic visit, p= 0.06. Weight profile did not change significantly with SGLT2i use (88.8 ± 21.3 to 90.1 ± 14.5 kg, p=0.85). There were numerically lower hemocompatibility and infection-related adverse events in the SGLT2i group, 7 events vs 11 events. One-year freedom from adverse events is shown in Figure 1. SGLT2i are well-tolerated in BTT patients supported with LVADs with no significant difference in both renal function and adverse events in the medium-term. The effect of SGLT2i in attenuating the known decline in renal function over the long-term warrants further study. [ABSTRACT FROM AUTHOR]

Following the wave of COVID and the numerous protocols of INR -self-management for patients with LVAD, it has been demonstrated that patients can monitor and adjust their anticoagulation and achieve better time in therapeutic range (TTR) compared to standard monitoring and care by physicians. To achieve this goal, more INR tests are performed. Here we compare the TTR in self-management versus the standard treatment and the costs of these two strategies in a European context. Using a protocol established at St-Vincent Hospital in Australia, patients were offered to follow a training on anticoagulation and then manage their anti-vitamin K treatment independently (Auto-management). We then compared the TTR and the cost before and under this protocol. Costs are based on Swiss prices and converted into Dollars (1CHF=1.03US $). Test-strip cost $4.8 per unit, the lancet costs a few cents, and the INR machine (CoaguCheck) is available for $810. The combined cost of the blood test and its interpretation by a general physician (GP) is about $50. Since November 2021, 11 patients have been included in the protocol; these included no women, one patient with a HeartWare, and 10 with a HeartMate 3. The patients have been included in the protocol for a cumulative time of 80 months or a mean of 7.3 months per patient. One patient was transplanted, and none died. One patient had a haemorrhagic complication following colonoscopy under therapeutic INR (2.8), no patient suffered thrombosis, and no LVAD replacement was performed. We compare this group during the Standard Care (SC) and Auto-management (AM) period. During SC, the median TTR was 72% (IQR 63-78); during AM it increased to 90% (IQR 77-93). In SC, the INR control cost per day was $9.4 (IQR 6.6-13) on average, which decreased to $2.4 (IQR 1.9-2.6) per day in AM. INR monitoring occurred on average 5.9 (IQR 4.2-8.2) times per month in SC phase; while in AM patients measured their INR 13.1 (IQR 11.7-13.4) times per month. The cost of one day of INR within the therapeutic range was $13 under standard care, $5.0 in the first year on Auto-management. Despite being associated with more INR control, Self-management of anticoagulation by LVAD- patients are less expensive and results in a higher TTR. Self-management also promotes patient independence. It should be encouraged and expanded to other LVAD implant centres. [ABSTRACT FROM AUTHOR]

The majority of patients with continuous-flow left ventricular assist devices (cf-LVADs) experience exertional symptoms despite adequate unloading at rest. We utilized pressure-volume (PV) loops to analyze the hemodynamic effects of exercise at varying pump speeds in stable HeartWare HVAD patients. Data were obtained at baseline pump speed and +400RPM, at rest and during supine cycle ergometry to 45W. PAWP, right atrial pressure (RAP) and mean arterial pressure (MAP) were measured invasively. From derived PV loops, left ventricular end-diastolic pressure (LVEDP) and volume (LVEDV), stroke volume, stroke work (SW), and maximum derivative of left ventricular pressure (dP/dt) were calculated. We also assessed the association between exercise PAWP and admissions for hemocompatibility related adverse events (HCRE). Thirteen patients were assessed, generating 52 PV loops. Exercise significantly increased PAWP (p<0.001) and RAP (p<0.001) but not MAP (p=0.69). Increased pump speed significantly reduced PAWP at baseline (p<0.001) but not during exercise (p=0.08). The change in PAWP during exercise correlated linearly with both change in RAP (p=0.008) and change in MAP (p=0.002). PV loop analysis revealed exercise-induced increase in LVEDP (p<0.001), LVEDV (p=0.001), stroke work (p<0.001), maximum dP/dt (p=0.001), and stroke volume (p=0.007), consistent with increased preload and ventricular contraction. Patients with exercise PAWP <25mmHg had significantly higher stroke work (p=0.010) and greater change in maximum dP/dt (p=0.005) with exercise, compared to those with exercise PAWP >25mmHg. Admissions for HCRE occurred more frequently in the higher PAWP group compared to the lower PAWP group (p=0.015). Patients with optimal exercise hemodynamics under cf-LVAD support have significantly greater left ventricular contractile reserve as measured by PV loops and less admissions for HCRE. [ABSTRACT FROM AUTHOR]

Continuous-flow left ventricular assist device (cfLVAD) support affects flow pathways within the microcirculation. We aimed to profile changes within retinal microvasculature and its association with adverse events. Retinal photography was performed using Topcon TRC-NW8 non-mydriatic fundus camera at monthly intervals in centrifugal cfLVAD supported patients and end-stage heart failure (ESHF) control patients. Retinal measurements including arteriolar and venular calibre, fractal dimension, branching angle, and vessel tortuosity were evaluated using the semi-automated Singapore I Vessel Assessment (SIVA) software. Demographic data and adverse event profile were documented. Overall, 172 images were recorded from 48 patients (n=29 cfLVAD, n=19 ESHF, 77.1% male, mean age 54.2 ± 11.9 years). Retinal arteriolar calibre significantly decreased in the cfLVAD group (144.87 ± 10.83 µm) versus the ESHF controls (152.31 ± 27.42 µm), p=0.045, and in multivariate-adjusted (B: -0.46 ± 0.22 µm, p=0.038) longitudinal mixed model linear regressions relative to baseline. There were no significant changes in all remaining retinal parameters (Figure 1). Amongst cfLVAD patients, there was a significant correlation between the incidence of gastrointestinal bleeding and stepwise increases in retinal arteriolar-venular calibre ratio (hazard ratio [HR]: 3.03, 95% CI: 2.06 - 4.45, p=0.005), a measure of arteriolar narrowing. We have demonstrated for the first time, decreased retinal arteriolar-venular calibre in cfLVAD-supported patients associates with gastrointestinal bleeding. Larger studies will inform the use of retinal photography as a non-invasive tool in predicting adverse events to inform future guidelines. [ABSTRACT FROM AUTHOR]

The product of pulmonary vascular resistance (PVR) and pulmonary artery compliance (PAC) describes the time constant (RC time) of the pulmonary circulation. RC time integrates parameters which describe the steady and pulsatile afterload of the right ventricle. This study investigated the effects of acute changes in left atrial pressure (LAP) on RC-time using hi-fidelity pulmonary artery (PA) pressure waveform analysis. Patients with left ventricular assist devices (LVADs) were recruited during right heart catheterization (RHC) and clinical ramp study was performed to assess change in left atrial pressure on RC time and PAC. Pulmonary arterial waveform was recorded at highest and lowest LVAD speeds using Mikrocath pressure catheter sample at 1000Hz. Phase plane analysis was then performed on 10 consecutive diastolic decay waveforms at each speed to calculate RC-time. 5 patients underwent ramp studies with simultaneous RHC. Mean pulmonary arterial wedge pressure (mPAWP) was altered acutely by a mean of 5mmHg (range 4-6). Mean pulmonary arterial pressure (mPAP) changed by 3 mmHg (range 0-5) and pulmonary vascular resistance (PVR) by 0.02 WU (range -1.3-2.0). RC time derived from waveform analysis deviated significantly from RC time estimated by the PVR and volume (SV)/pulse pressure (PP) product. Consequently, PAC also deviated significantly from SV/PP. Acute increase in PAWP resulted in shortening of the RC-time. Measuring RC time using direct pressure waveform analysis produces significantly different results from PVR*SV/PP. RC time reliably shortens with acute elevations in PAWP, suggesting that PAWP can augment pulsatile RV afterload without intrinsic pulmonary vascular remodelling. Further studies are needed to investigate whether direct of analyses of PA waveform provides more meaningful assessment of the pulsatile load. [ABSTRACT FROM AUTHOR]

Renal function following ventricular assist device (VAD) implantation is a key determinant of survival. The purpose of our study was to investigate whether the development of acute kidney injury (AKI) prior to, or following, VAD implantation influenced survival to transplant. We also sought to determine if the use of extra-corporeal membrane oxygenation (ECMO) or intra-aortic balloon pump (IABP) influenced AKI pre or post-VAD. We undertook a single-centre retrospective cohort study examining outcomes in patients who underwent VAD implantation between 2010 and 2020. AKI was defined as per the KDIGO guidelines. The primary outcome was survival to transplant. 61% (96/158) of patients experienced pre-VAD AKI which predicted an increased risk of mortality (OR 2.32, p=0.032, CI 1.07 - 5.04). The incidence of post-VAD AKI was 54% (89/165) which was similarly associated with an increased risk of mortality (OR 2.49, p=0.012, CI 1.21 - 5.13). 67% (64/96) of patients who experienced pre-VAD AKI had post-VAD AKI. Patients who had both pre and post-VAD AKI had a significantly increased mortality prior to transplant with an OR of 3.5 (P=0.01, CI 1.3 - 9.41). Patients with pre-VAD AKI were more likely to require ECMO prior to VAD implantation (OR 2.94) however no difference was observed for IABP use. The use of ECMO or IABP prior to VAD implantation did not influence post-VAD AKI. Acute kidney injury both pre and post-VAD implantation reduced likelihood of survival to transplant. Patients with AKI prior to VAD insertion were more likely to require ECMO, however the use of ECMO or IABP pre-VAD insertion did not appear to influence post-VAD AKI. [ABSTRACT FROM AUTHOR]

Infection occurring during ventricular assist devices support (VAD) is the third most common cause of mortality overall and the most common cause of morbidity after three months post implantation. Understanding the epidemiology, aetiology, and response to treatment with driveline (DL) infections is crucial to their optimal management. We retrospectively analysed VAD specific infections in patients with VADs implanted at our institution between January 2009 and January 2019. Data on associated clinic and treatment were collected. During this 10-year period, we implanted 94 patients, 22 (23%) females; 9 (10%) HeartMate 2, 18 (19%) HeartMate 3, and 63 (67%) HeartWare; 23 (25%) as destination treatment. Median age was 54. A total of 216 infectious episodes were documented. 38 patients (40%) had at least one episode of VAD-specific infection, with a median of 2 (IQR: 1-3) infections per patient. All were DL infections (no pump/pocket infection). Thirty-five different organisms were recovered from swabs, most frequently 102 (41%) S. aureus and 38 (15%) P. aeruginosa. The median time to first DL-infection was 162 days (IQR: 74-258). Data on DL infection treatment was documented for 198 infections, of which 134(67%) were treated with short-term antibiotics (< 4 weeks of treatment), 28 (14%) with suppressive antibiotic therapy(> 4 weeks of treatment) and 36 (18%) surgically. The median time from VAD-implantation to the first surgical DL-treatment was 169 days (IQR: 31-958) and 262 days (IQR: 15-673) to initiation of suppressive antibiotic therapy The median time to relapse was 86 days (IQR: 53-189) for short-term antibiotic therapy, 148 days for surgical intervention and 422 days (IQR: 209-807) for long term antibiotic therapy. A conclusion on treatment efficacy is not possible here. Nevertheless, the time to first infection or to relapse may be useful as a benchmark to assess new DL dressing techniques, treatment strategies or implantation techniques. [ABSTRACT FROM AUTHOR]

Cardiorenal syndrome is associated with excess morbidity and mortality, particularly in patients requiring durable mechanical circulatory support or cardiac transplantation. There are minimal data regarding predictors of renal outcomes in these patients and identifying those who may benefit from dual organ-transplantation. We identified 18-patients with renal histology supporting the diagnosis of cardiorenal syndrome, from 665 consecutive patients listed for cardiac transplantation between 2010-2019. These were prospectively analysed for glomerular, tubular, interstitial and arteriolar changes as shown in table 1, tallied to give a severity score. The primary outcome, major adverse kidney events (MAKE), was a composite of death, need for renal replacement therapy (RRT), drop in estimated glomerular filtration rate (eGFR) >50% or stage V chronic kidney disease occurring prior to transplantation for the baseline cohort, and following transplantation for the transplanted cohort. The secondary outcomes included the individual components of MAKE at the two different time points. Mean age 52.3, 22% female. 5 patients did not survive to transplant (2 of these died on ventricular assist device). 1 patient had successful heart-kidney transplant. MAKE occurred in 7/17 of baseline; this was associated with a higher severity score, mean 5 vs. 2.3, p=0.03. There was no difference in age, creatinine or eGFR at biopsy. MAKE following transplantation occurred in 8/13 patients. There were no significant associations with this outcome. MAKE in the baseline cohort was primarily driven by need for RRT (mean 5.2 vs. 2.5, p=0.04). Contrary to limited previous literature, interstitial fibrosis did not predict any outcome, whilst presence of glomerular ischaemic changes, tubular atrophy and arteriosclerosis were associated with pre-transplant MAKE. More severe renal histology prior cardiac transplantation are associated with adverse renal endpoints. [ABSTRACT FROM AUTHOR]

An increasing number of end-stage heart failure patients are supported by continuous-flow Left Ventricular Assist Devices (cfLVADs). In these patients, accurate assessment of arterial pulsatility remains a challenge because of diminished pulse pressure. Limited pulsatility in modern cfLVADs itself can be associated with adverse events such as stroke. Our study investigates the value of simultaneous arterial pressure and arterial blood velocity assessment in cfLVAD patients. Both velocity (cm/s) and pressure (mmHg) data were collected during routine follow up. Pressure was measured using a brachial cuff oscillometry system (Sphygmocor XCel) providing a systolic and diastolic readings, while velocity was measured using a digital doppler (Huntleigh) from the brachial artery. Velocity measurements obtained included peak, minimum, average velocity and pulsatility index. Pressure measurements obtained were systolic and diastolic which allowed us to derive a mean arterial pressure (MAP) from the brachial artery. From a total of 24 patients (88% males), there were 44 velocity and pressure measurements. Mean device support duration was 240 days (SD 116). Mean derived brachial MAP was 85 mmHg (SD 9), whereas mean peak brachial artery velocity was 14 cm/s (SD 9). Mean pulsatility index was 2 (SD 2). Correlation analysis using Spearman's rho demonstrated a significant inverse relationship between Brachial derived MAP and Peak velocity (Rho=-0.41, p=0.007). Arterial doppler measurements may provide incremental value over purely pressure-based measurements. The inverse relationship between derived MAP and peak velocity in this study may be partially explained by the degree of arterial compliance. In turn, this provides significant information on the relationship between higher MAP and lower blood velocity to subsequently explain increased thrombus formation and stroke rate due to longer transit time of blood. [ABSTRACT FROM AUTHOR]

There is a lack of consistent approach to pump thrombosis (PT) in ventricular assist device (VAD) supported patients. Previously, bolus thrombolysis was administered at physicians' discretion, guided by changes in pump parameters. More recently, we adopted a strict protocol with invasive haemodynamic monitoring to target mean arterial pressure <80mmHg and continuous infusion of tissue plasminogen activator (tPA) to achieve sustained normalisation of pump flow and power derivatives. We compare outcomes following bolus protocol (BP) and strict infusion protocol (SIP). Single-centre retrospective analysis of all patients treated with thrombolysis for VAD PT between 2007-2020. Patients divided into two cohorts (BP and SIP) based on PT treatment. Baseline demographics, antithrombotic therapy, pump parameters and biochemistry were collected prior to PT event, at diagnosis and following thrombolysis. Outcomes reported included need for repeated thrombolysis doses, post thrombolysis major bleeding and death, and survival to transplant. Mean age 51.7y, 23.5% female. All supported with Medtronic HVAD. 21 PT events in 17 patients; 10 patients with 11 cases in BP group, 7 patients with 10 cases in SIP group. SIP group had more pre-implant renal dysfunction (mean GFR 54 vs. 81ml/min/1.73m2, p=0.017), and lower stable post-implant MAP (mean 72 vs. 76, p=0.05). Repeated thrombolytic doses were never required following SIP vs. 70% following BP (p=0.01). No patients required pump exchange, however 1 in each group required urgent heart transplant listing. Numerically fewer post thrombolysis major bleeds (2 vs. 3) and deaths (0 vs. 3) following SIP compared with BP. All SIP patients survived to transplant, compared with 70% of BP patients(figure 1). SIP is safe for treatment of PT in VAD supported patients, with excellent associated survival to transplant, and trend towards improved post-thrombolysis outcomes. [ABSTRACT FROM AUTHOR]

LVAD implantation is often utilised as a bridge to transplantation or candidacy in those with pulmonary hypertension secondary to left heart disease (PHLHD). Pulmonary hypertension portends poorer outcomes post-transplant. We compared pre and post implantation haemodynamic response and outcomes in patients who normalise transpulmonary gradient (TPG) to those who do not. Retrospective review of CF-LVAD patients was conducted for right heart catheter derived haemodynamic measurements and post-transplant outcomes from a single instituition. 219 CF-LVADs were implanted between October 2004 and September 2019. 48 patients had TPG greater than 12 prior to implantation. 1 patient was excluded for incomplete data. 28 patients normalised TPG (Gp 1) and 17 patients had elevated TPG (Gp 2) at first follow up. Table 1 describes the haemodynamic assessment of the two groups. At baseline, Gp 1 patient had significantly lower PVR, p=0.04. At follow-up, Gp 1 patients had significantly lower RAP, p<0.0001 and PP, =0.003. Gp 1 patients were more likely to survive on pump, p=0.03, more likely to be transplanted, p=0.03 and survived longer post transplant, p=0.04. Baseline pulmonary vascular resistance may provide prognostic information in CF-LVAD patients with PHLHD. Early normalization in TPG is associated with favourable haemodynamic response. [ABSTRACT FROM AUTHOR]

Mechanical circulatory support (MCS) including left ventricular assist devices (LVADs) are increasingly important in managing patients with end-stage heart failure. Current generation LVADs, such as the HeartWare HVAD operate on a continuous flow basis. Many authors have identified the lack of physiological pulsatility as a limitation in these devices. We sought to test the feasibility of implementing physiological pulsatility using sino-atrial node (SAN) electrograms. Three different rabbits had SAN tissue extracted, placed in a tissue bath and superfused with Krebs-Ringer solution. An extracellular fine metal electrode was placed into the tissue preparation for activity capture. The rate of spontaneous SAN activity was modulated by changing the temperature of the superfusing solution. Electrical activity was captured and identified using a data acquisition system and custom-made software. This was used to trigger a stepwise speed change in an HVAD running in a mock circulation loop. Signal capture was tested in three heart rate ranges (>120, 60-80 and ∼60 beats per minute). Unipolar electrogram signals were successfully recorded for all tissue extracts. To simulate physiological settings, we implemented a 100 ms delay between capture and effect, as well as a speed change duration of 30% per cycle (obtained from the beat-to-beat variation). HVAD motor speed was successfully increased from 2500 RPM to 2800 RPM in synchronicity with the firing of the SAN. Similarly, we were also able to achieve a reduction of speed from 2500 RPM to 2000 RPM in synchronicity with the firing of the SAN. We were able to successfully capture electrical activity from the SAN for the purpose of introducing physiological pulsatility in all three of our rabbit experiments. This has the potential to change fundamental MCS flow hemodynamics. [ABSTRACT FROM AUTHOR]

Purpose Stroke and ventricular assist device (VAD) pump thrombus are serious complications that have previously been thought to be related to surgical placement - namely VAD inflow cannula angle. Our aim with this study was to better elucidate the impact of post-operative HeartWare HVAD (left VAD) cannula inflow angle on stroke and thrombus rate within a single centre in Sydney, Australia. Methods Data were extracted retrospectively for all patients who underwent Heartware HVAD implantation from 2007-2017 with accessible on-site CT imaging. BiVAD patients were excluded. Angulation data was available for 95 patients. Post-implant inflow angle was measured between the middle of the inflow cannula relative to the middle of the sternum. All measurements were carried out in the axial plane on CT imaging. Data on Free Haemoglobin level was extracted at 3 intervals (3, 6, and 12-months post VAD implantation). Statistical analysis was carried out using IBM SPSS v25. Results Mean post-operative inflow cannula angle was 89 +/- 27 degrees vs 91 +/- 22 degrees for the non-stroke and stroke groups respectively, p=NS. Similarly, there was no difference in mean post-operative inflow cannula angle between those that formed pump thrombus (89 +/- 28 degrees) and those that did not (91 +/- 19 degrees), p=NS. Average 12-month post-implant Free Haemoglobin was higher in the group with pump thrombus than the group that did not form thrombus (9.1 +/- 9 vs 5.3 +/- 9), p=0.012. Conclusion There was no association between HVAD cannula inflow position and incidence of stroke or thrombus rates in this population. The previous finding of an interaction between inflow cannula angulation and pump thrombosis may be device specific. [ABSTRACT FROM AUTHOR]

Three of the main causes of early and late morbidity and mortality in Ventricular Assist Device (VAD) recipients include infections, bleeding and thrombosis. Infections directly or indirectly associated to VAD are common and often requiring prolonged courses of intravenous antibiotics. Anti-vitamin K anticoagulants are known to interact with many drugs including antimicrobial agents. This case report highlights the risks of an important drug-drug interaction that has the potential to result in serious consequences. A 45-year-old male with LVAD (HeartWare) presented with an episode of significant hypotension (MAP 40mmHg) and was admitted to the Intensive Care Unit and initiated on vancomycin and Tazocin for empiric coverage of a suspected VAD-related infection. He was subsequently diagnosed with a methicilin sensitive Staphylococcus aureus (MSSA) bacteraemia. Vancomycin was changed to I-V flucloxacillin 8 g daily. No clear source of sepsis was demonstrated and a transoesophageal echocardiogram did not reveal vegetations. In the context of sepsis, the patient's INR initially increased to 3.7. Subsequently, his INR decreased despite dramatic increases in warfarin dosing up to 22mg. He was discharged home on IV flucloxacillin. 14 doses of low molecular weight heparin and an increase of 250% of warfarin (from 7mg to 17mg average) were necessary. It was only with close home monitoring that we were able to detect this significant interaction and prevent serious complications. Flucloxacillin can reduce the anticoagulant effect of Vitamin K Antagonists. The literature about this interaction is scarce. One possible mechanism is induction of CYP3A4-mediated metabolism of R-warfarin by Flucloxacilin via Pregnane-X hormone receptor upregulation resulting in reduced warfarin concentrations and effects. This case report raises concerns about antibiotic therapy in VAD patients. It is important to be aware of this potentially serious interaction and close INR monitoring is recommended. [ABSTRACT FROM AUTHOR]

Standard guideline-directed neurohumoral blocking therapy is often continued or initiated after left ventricular assist device (LVAD) implantation once haemodynamics are stabilised. Angiotensin/neprilysin inhibitors (ARNI) are well integrated into international guidelines, but the role of these agents in LVAD patients is unclear. In LVAD patients with high afterload sensitivity, small preload changes can lead to suboptimal LV filling and intermittent suction-mediated inflow cannula obstruction. We investigated the haemodynamic influence of ARNI on LVAD flow, as a surrogate marker of possible suction events, using the accessible logfile data from an in-vivo HeartWare(HW) device, which captures instantaneous power and derived flow every 15-minutes. We extracted the logfile data of a patient who, after stabilisation on ARNI therapy following LVAD implantation, ultimately ceased due to labile arterial pressures. Using the patient as their own control we calculated 1) the mean flow using a running average, 2) the pulsatility index (PI) from minimum/maximum flows, and 3) potential suction events (defined as outlier flow events two standard deviations from the mean). Mean flow was similar off (3.86±0.5l L/min) compared to on (3.78±0.7L/min) ARNI therapy. The PI was numerically higher on therapy (2.1±0.2 L/min) compared to no therapy (1.7±0.3 L/min). There were numerically many more outlier flow events with ARNI therapy (78) compared to no therapy (9), which may correlate with suction events, and suggests haemodynamic instability on therapy. Whilst a few small studies have demonstrated general tolerability of ARNIs in LVAD patients, potent venodilatory and anti-hypertensive effects can cause unfavourable haemodynamics. We show evidence of increased haemodynamic instability evident from controller logfiles possibly due to a disproportionate reduction in preload relative to afterload with increased uncoupling of ventricular to arterial elastance. [ABSTRACT FROM AUTHOR]

An increasing number of end-stage heart failure patients are supported by continuous-flow Left Ventricular Assist Devices (cfLVADs) as a bridge to heart transplantation. Accurate assessment of arterial pulsatility remains a challenge because of diminished pulse pressure, which itself can be associated with adverse events such as gastrointestinal bleeding and stroke. Our study investigates the value of simultaneous arterial pressure and arterial blood velocity assessment in cfLVAD patients. Brachial artery flow velocity (cm/s, Huntleigh) and brachial cuff oscillometry pressure (mmHg, Sphygmocor XCel) data were collected during outpatient visit. Velocity measurements obtained included peak, minimum, average velocity as well as pulsatility index. Pressure measurements obtained were systolic, diastolic as well as mean arterial pressure. Twelve velocity and pressure measurements were obtained from a total of nine VAD patients. Mean age was 57 years (Range 44-71). Mean device support duration was 187 days (Range 44-471). Mean brachial MAP was 92 mmHg (Range 74-112), whereas mean peak brachial artery velocity was 11 cm/s (Range 6-26). Mean pulsatility index was 3 (range 1-8). Correlation analysis using Spearman's rho demonstrated a significant inverse relationship between Brachial MAP and Peak velocity (Rho=-0.72, p=0.08). Arterial flow measurements may provide incremental value over purely pressure-based assessment. Discrepancies between these modalities may provide further information regarding the arterial vasculature, specifically pertaining to the impact of higher MAP on blood flow. [ABSTRACT FROM AUTHOR]

Worsening renal function (WRF) in patients supported with ventricular assist devices (VAD) is associated with poor outcomes. Postulated mechanisms include right ventricular (RV) dysfunction, low pulsatility from continuous flow, and chronic haemolysis (CH). We aim to determine whether these predict WRF. Consecutive VAD implants between 2010 - 2019 were analysed. Estimated glomerular filtration rate (eGFR - ml/min/m2), lactate dehydrogenase (LDH) and transthoracic echocardiogram (TTE) were collected until death/transplant to 12-months following implant. Demographics and medications 3-months post-implant were recorded. Adverse events included mortality, gastrointestinal (GI) bleed, ischaemic stroke, intracranial haemorrhage, pump thrombosis and driveline infection at any time until death/transplant.WRF was defined as drop in Kidney Disease Improving Global Outcomes Chronic Kidney Disease (KDIGO CKD) stage from baseline to last recorded eGFR. CH defined by elevated LDH. Pulsatility index (PI) derived from mean, peak and trough VAD flows at 1, 3 and 12 months. 173 patients included, WRF occurred in 49 patients. WRF was associated with increased GI bleed (odds ratio [OR] 4.41, p=0.02) and ischemic stroke (OR 5.05, p=0.018). WRF occurred less frequently in those on beta-blockers (OR 0.31, p = 0.009) and renin-angiotensin-system inhibitors (OR 0.31, p = 0.009). Although not statistically significant, those with WRF had median higher baseline eGFR (WRF vs. no WRF 68 vs 55, p=0.598). Less profound eGFR improvement at 1 month was seen in patients with WRF (1 vs. 25.5, p = 0.021), resulting in numerically lower eGFR at 1-month in those with WRF (77 vs. 83, p=0.119) respectively. WRF was not predicted by PI, CH or change in RV function on TTE. WRF is associated with increased morbidity in patients supported with VAD, although the mechanism for this is unclear. Further studies are required to define CH and low pulsatility to determine their impact on WRF. [ABSTRACT FROM AUTHOR]

Clinical trials of the HeartWare™ HVADTM System have demonstrated a low rate of post-implant severe right ventricular failure (RVF) requiring implantation of a right ventricular assist device (RVAD). In an analysis of the North American clinical trials of the HVAD System, including the ADVANCE BTT+CAP Trial (N=382), the ENDURANCE Trial (N=296), the ENDURANCE Supplemental Trial (N=308) and the Lateral Trial (N=144), RVADs were implanted in 2.7% of patients (31/1130) over 2 years of support. However, the overall percentage of patients in INTERMACS profile 1 at implant was only 4.2% and those with anticipated need for post-operative RVAD were excluded. We will present an analysis of severe RVF requiring an RVAD in the real-world post-approval study of the HVAD System when used in destination therapy patients (DT-PAS and Apogee International study). The objective of this analysis is to examine the rate of RVAD use in the prospective DT PAS study (n=212) and the Apogee International study (n=46). At the time of presentation, we will have 258 patients with follow-up data through at least 6 months of support. Data is currently being collected from 50 centers in the US, EMEA, and Australia. Descriptive data and severe RVF data will be presented, and a comparison, accounting for baseline differences, to the HVAD clinical trials will be performed. Baseline parameters of the 258 patients includes mean age 58.3 years, with 15.4% INTERMACS profile 1 and 44.6% with ischemic etiology. Preliminary data from 212 patients in the US DT PAS cohort (with 16.3% INTERMACS 1) reveals an incidence of RVAD use of 6.1% over a mean time on support of 5 months. Data collection and analysis is ongoing. Analyses of real-world use of the HVAD System reveals its increasing use in higher risk and more complex patients. Final data on RVAD use in 258 patients through 6 months will be available at the time of presentation. [ABSTRACT FROM AUTHOR]

BiVACOR is a novel total artificial heart utilising a single centrifugal magnetically levitated rotor with the ability to modulate pulsatile flow through speed variation. The device has been successfully tested in cattle with encouraging 90-day results. We sought to undertake a multicentre virtual and anatomical fitting study of the BiVACOR in patients undergoing heart transplantation. 10 patients on the heart transplant waiting list were recruited (5 Females, 5 Males) across two heart transplant centres. A 1:1 titanium model of the device with correct port configuration (Aorta, Pulmonary Artery, Right Atrium, Left Atrium) was provided by BiVACOR. Chest size and major vessel measurements were recorded from preoperative CT Scans. The sterilised model was then inserted into the patient's chest cavity post heart explant, prior to implantation of the donor heart. The atrial ports (right and left) on the device were lined up with their native counterparts. Measurements were recorded in-situ. The Device was then removed. 3D reconstructions were also created to simulate a virtual fitting. Mean age was 53 years (Range 38-67). Mean BMI was 28 (Range 20-37). Aetiology was varied - 5 Patients had Ischaemic Cardiomyopathy, whereas 1 was restrictive, 1 was giant cell myocarditis and 3 were dilated cardiomyopathy. Four had re-do sternotomies and 3 of these involved a ventricular assist device explant. Mean Spine-to-Sternum distance at the level of T10 was 14cm (range 11-18). Mean Ao diameter was 3.1cm (range 2.5-3.8). Mean PA Diameter was 3.4cm (range 2.0-4.2). Mean Ao to Ao Port distance was 0.2cm (range 0-0.5). Mean PA to PA Port distance was 4.2cm (range 1-7). The device fit in all patients. There was no gross distortion to the geometry between native vessels/chambers and ports. This study described the virtual and anatomical fitting of the BiVACOR total artificial heart. The device fit within the chest cavities of all 10 patients, who represented a variety of body morphologies and heart failure aetiology. [ABSTRACT FROM AUTHOR]

Ventricular Assist device (VAD) patients are at higher risk of bleeding at the time of heart transplant. Moreover, the quantity of packed red cells (RBC) received during the first 24 hours postoperatively correlates with increased mortality. Whether this is related to surgical issues or due to the anticoagulation associated with VADs is unknown. Operative transfusion requirements and demographics data was obtained from our hospital cardiothoracic database for all patients who underwent heart transplantation between 2006 and 2020. Patients were divided into three groups - No previous sternotomy (NPS), VAD explant (VE) and Previous (non-VAD) Sternotomy (PS). Biventricular VAD, total artificial heart explants and multi-organ transplants were excluded. 428 patients were included in the analysis. Median age was 54 years (IQR 43-61), 131 patients were female (31%), median eGFR was 78ml/min/1.73m2 (IQR 54-95). 236 patients (54.2%) were in the NPS group, 133 (31.1%) in the VE group and 62 patients (14.6%) in the PS group. There was a strong association between blood transfusion and re-do sternotomy (both VE and PS groups) (p<0.001), but no difference between the VE and PS groups. A lower mean eGFR (<78ml/min/1.73m2) was associated with higher transfusion requirement (P<0.001). Sternotomy patients (with or without VAD explant) demonstrated a higher RBC transfusion requirement compared to clean skin and these patients tended to have a lower renal function post-operatively. The lack of difference in RBC requirement between VE and PS patients suggest that anticoagulated VAD patients who have received Vitamin K do not necessarily have higher bleeding risk than other re-do sternotomy patients. [ABSTRACT FROM AUTHOR]

Hypertension (HTN) has been implicated as a risk factor for stroke in continuous-flow Left Ventricular Assist Device (LVAD) populations. Active treatment of HTN has been shown to decrease cerebrovascular accident rates in the recent trial. Current guidelines recommend a mean arterial pressure (MAP) of 80mmHg in LVAD patients; however, it is difficult to reliably measure Blood Pressure (BP) in this cohort. Increased BP will further affect pump output since LVAD flows are altered by both preload and afterload. We studied the impact of both HTN (adjusting systemic vascular resistance) and arterial compliance (Windkessel chamber) on LVAD flow waveforms in an in-vitro mock circulatory loop. We also analysed the impact of MAP and compliance on Aortic (Ao) pressure, Ao Pulse Pressure (PP = peak - trough), mean LVAD flow (Q) and flow Pulsatility Index (PI). Reduced compliance (HC (High Compliance) to LC (Low Compliance)) increased Ao pulsatility (Ao PP) (table) due to higher peak Ao pressure and lower minimum pressure (figure), while LVAD flow contour was altered with increased slope of the diastolic segment (figure). Increase in MAP (MAP80 to MAP110) as an indication of HTN resulted in reduced peak, trough and mean LVAD flows (table). Higher flow PI was also apparent with elevated MAP, and this can be attributed to lower pressure trough. Similar results were found across a range of preloads. Relationships between higher flow PI and Ao pulsatility with lower Ao compliance may partly explain the increased risk of stroke due to HTN. In summary, increase in MAP has a significant effect on the LVAD flow waveform. The relatively lower pump flow seen with HTN state may be important in contributing to greater risk of thrombus formation as well as cerebrovascular accident. [ABSTRACT FROM AUTHOR]

BiVACOR is a new and novel total artificial heart utilising a single centrifugal magnetically levitated motor featuring a pulsatile flow. The device has been successfully tested in cattle with encouraging 90-day results. We sought to undertake the first-in-man fitting of the device. Six patients awaiting heart transplant were recruited (2 Females, 4 Males). Chest size, atria size, pulmonary artery (PA) and aortic (Ao) measurements were recorded from preoperative CT Scans and 3D reconstructions were created. We generated a 1:1 3D printed titanium prototype of the BiVACOR device. The sterilised prototype with correct port configuration (Aortic, Pulmonary Artery, Right Atrium, Left Atrium) was then inserted into the patient's chest cavity at surgery post heart explant and before implantation of the donor heart. The atrial ports (right and left) on the device were lined up with their native counterparts. Measurements were recorded in-vivo. The device was then removed and compared in size to the explanted heart. Mean age was 55 (SD 9) years. Mean BMI was 28 (SD 6) kg/m2. Aetiology was varied - 3 Patients had Ischaemic cardiomyopathy, 1 restrictive, 1 active giant cell and 1 idiopathic dilated cardiomyopathy. Two had re-do sternotomies and one of these was a ventricular assist device (HVAD) explant. Mean thoracic cage area was 377cm2 (range 343-413) at the level of the 5th rib. Mean Spine-to-Sternum distance was 13cm (range 11-16) measured between the 6th/7th ribs. Mean Ao diameter was 3.6cm (range 2.5-3.8). Mean PA diameter was 3.6cm (range 2.9-4.2). Mean Ao to Ao Port distance was 0.25cm (range 0-0.5). Mean PA to PA Port distance was 5.3cm (range 3.5-7). The device fit easily in all patients. There was no gross distortion to the geometry between native structures and ports. This study described the first in man anatomical fitting of the BiVACOR total artificial heart. The device fit well within the chest cavities of all 6 patients, who represented a variety of body morphologies and heart failure etiology. [ABSTRACT FROM AUTHOR]