HeartWare Research Articles

Readmissions After Continuous Flow Left Ventricular Assist Device Implantation: Differences Observed Between Two Contemporary Device Types.

Readmissions after continuous flow left ventricular assist devices implantation are common. We compared the frequency and etiology of readmissions between two continuous flow left ventricular assist devices 6 months after implant. We retrospectively assessed readmissions in 81 patients who received a bridge to transplant HeartMate-II (HM-II) n = 35, 43% or HeartWare (HW) n = 46, 57%, from 2009 to 2014. Readmissions were divided into cardiac, infection, gastrointestinal bleeding, stroke, pump thrombosis, and miscellaneous profiles. Age, gender, creatinine, INTERMACS profiles were comparable between groups (p > 0.05). Sixty-one patients accounted for 141 readmissions. At 6 months, the overall readmission rate was higher among HM-II versus HW recipients (2.3 ± 1.7 vs. 1.4 ± 1.3; p = 0.024). Multiple readmissions (≥2) occurred more frequently in HM-II recipients (HM-II 23, 66% vs. HW 20, 44%; p = 0.047) which accounted for 87% of overall readmission frequency. Cardiac profile was the most common reason for readmission (HM-II = 15, HW = 17; p = 0.95). Readmission for arrhythmia (HM-II = 10, HW = 3; p = 0.021) and overall infection rate (0.49 ± 0.70 vs. 0.17 ± 0.68; p = 0.001) were more common among HM-II recipients; however, other readmission profiles were comparable between devices (p > 0.05). Readmission frequency, multiple readmissions, and clinical profile characteristics were different between HM-II and HW recipients.

Perioperative Blood Product Use: A Comparison Between HeartWare and HeartMate II Devices

The HeartWare (HW) (Framingham, MA) and the HeartMate II (HM II) (Thoratec Inc, Pleasanton, CA) continuous-flow left ventricular assist devices (CF-LVADs) are commonly used to bridge patients to transplantation. We hypothesized that there are differences in perioperative blood product (BP) use and chest tube (CT) output between CF-LVAD types. We retrospectively evaluated BP use in 71 patients who were implanted with a CF-LVAD (HM II = 38; HW = 33) by median sternotomy for bridge to transplantation (BTT) indications from 2009 to 2013. Detailed BP use data were collected during the intraoperative and postoperative periods and included packed red blood cells, platelets, fresh frozen plasma, and cryoprecipitate. Preoperative characteristics (age, left ventricular ejection fraction, previous sternotomy, ischemic cause), and risk stratification scores (Interagency Registry for Mechanically Assisted Circulatory Support [INTERMACS]) profile, Leitz-Miller score, Kormos score) were comparable between groups (all p > 0.05). Total average intraoperative and postoperative BP use was different between device types: HW = 8.3 ± 13 versus HM II = 12.6 ± 14.0 units (p = 0.002) and HW = 6.1 ± 12.0 units compared with HM II = 13.5 ± 24.1 units (p = 0.022), respectively. Average postoperative CT output for HW (3,231 ± 3,648 mL) and HM II (3,463 ± 3,050) (p < 0.008) were different between device types. Multivariate analysis revealed that a higher preoperative Leitz-Miller score, implantation of an HM II CF-LVAD, previous sternotomy, and a longer duration of cardiopulmonary bypass (CPB) time were independently associated with increased need for BP use, whereas only use of the HM II device and a longer bypass time predicted a greater CT output. Compared with HM II, implantation of the HW CF-LVAD was associated with reduced intraoperative and postoperative BP use and decreased CT output. Increased awareness of device-related differences in bleeding and BP use may improve CF-LVAD patient outcomes.

(538) - Assessment of the LVAD Inflow Cannula Position by Cardiac CT

Purpose: Position of the inflow cannula (IC) within the left ventricle may affect optimal function of the LVAD and contribute to the risk of complications. We evaluated the intraventricular position of the IC in Heartmate II (HM2) and Heartware (HW) LVADs. Methods: A single-center retrospective review of the most recent cardiac CT (CCT) of all continuous flow LVAD implants between 10/2006 and 9/2013 was performed. CT acquisitions were completed on a 64 or 256 slice platform using ECG gated helical technique. Measurements were performed in standard 2-chamber and short-axis views. Fisher’s exact and Wilcoxon-MannWhitney tests were used to determine statistical difference between groups. Results: A total of 195 patients underwent LVAD implants (101 HM2, 94 HW). At least one CCT was performed on 63 patients. Mean LV end diastolic diameter was 63 ± 11 mm. Apical tip of the IC was more frequently in contact with anterior wall in the HM2 patients (15/24, 62%) compared to HW pumps (5/38, 13%, p< 0.001). Basal tip did not abut the anterior wall in any patient. Septal wall contact was 25% vs. 5 % (p= 0.05) and lateral wall contact was 29% vs. 25% (p= 0.7) in HM2 and HW pumps, respectively. In the total population, the IC was angulated towards the mitral annulus plane (median 33 degrees, IQR 22, 46) and formed a 77-degree horizontal angle in the short axis (IQR 57, 95, figure 1). Angle of the IC was significantly different between HM2 and HW LVADs (table 1). Conclusion: Significant variation exists among patients and between two different types of continuous-flow LVAD with regards to intraventricular position of the inflow cannula. Clinical significance of these variations remains to be determined.

Pump Flow Estimation From Pressure Head and Power Uptake for the HeartAssist5, HeartMate II, and HeartWare VADs

The use of long-term mechanical circulatory support (MCS) for heart failure by means of implanted continuous-flow left ventricular assist devices (cf-LVADs) will increase, either to enable recovery or to provide a destination therapy. The effectiveness and user-friendliness of MCS will depend on the development of near-physiologic control strategies for which accurate estimation of pump flow is essential. To provide means for the assessment of pump flow, this study presents pump models, estimating pump flow (Q(lvad)) from pump speed (n) and pressure difference across the LVAD (Δp(lvad)) or power uptake (P). The models are evaluated for the axial-flow LVADs HeartAssist5 (HA5) and HeartMate II (HMII), and for a centrifugal pump, the HeartWare (HW). For all three pumps, models estimating Q(lvad) from Δp(lvad) only is capable of describing pump behavior under static conditions. For the axial pumps, flow estimation from power uptake alone was not accurate. When assuming an increase in pump flow with increasing power uptake, low pump flows are overestimated in these pumps. Only for the HW, pump flow increased linearly with power uptake, resulting in a power-based pump model that estimates static pump flow accurately. The addition of pressure head measurements improved accuracy in the axial cf-LVAD estimation models.

A Comparison between HeartMate II and HeartWare LVAD, a Single Center Experience

Purpose LVAD implantation is an established therapy for end-stage cardiac failure. The most implanted devices are HeartMate II (HMII) and HeartWare (HW). The goal of this study is to compare both devices in terms of morbidity and mortality. Methods and Materials Since 2007 100 HMII and 97 HW LVAD’s are implanted in our center. The mean age in the HW group was lower (51.0 vs 57.4, p=0.001), more patient were ventilated preoperatively (30.9% vs 13%, p=0.002), ECMO support was more frequent in the HW group (16.5% vs 6%, p=0.019). Both groups did not show differences in Intermacs levels, hemodynamic parameter, gender, blood parameters, intention to treat and cause of cardiac failure. Results Operation duration, blood loss and the need for blood transfusion were comparable in both groups. The mean duration of ICU stay postoperatively did not show a difference. Early and late right heart failure (p=0.804, p=0.602) as the need for RVAD implantation were comparable (p=0.296). Driveline infection occurred in 24.7% of the patients in the HW group vs 19% in the HMII group (p=0.66). 8.2% of the patients in the HW group vs 8.4% in the HMII group suffered an embolic CVA (p=0.965), hemorrhagic CVA were comparable (10.3% HW group, 9.5%HMII group, p=0.846). Thrombolysis, because of pump thrombosis, was performed in 3.1% in the HW group, 4.0% in the HMII group (n.s.), replacement was necessary in 3.1% of the HW patients vs 9.0% of the HMII patients (n.s.). One year mortality was 69.8% in the HW group vs 67.3% in The HMII group (n.s.). Conclusions Despite more riskfactors for morbidity and mortality in the HW group (Ventilation, ECMO preoperatively), the results were comparable for the HW and the HMII group. Both devices are feasible, HW seems to have advantages in ECMO patiens and patients on ventilation.

Evaluation of Left Ventricular Assist Device Inflow Cannula Contact with Ventricular Wall on Cardiac CT

Purpose Optimal function of a left ventricular assist device (LVAD) requires proper positioning of the inflow cannula (IC) within the left ventricle. Cardiac computed tomography (CT) is emerging as the preferred method for evaluation of the IC position. We sought to evaluate the frequency of contact between the IC and the left ventricular wall (LV) in patients with axial flow and centrifugal flow devices. Methods and Materials This single center study included all patients who received a continuous flow implantable LVAD between 5/2008 & 8/2011 and who underwent a contrast cardiac CT. Patients were stratified into 2 groups, based on whether the indication for the CT scan was a clinical event (arrhythmia, etc) or a routine study (redo sternotomy evaluation prior to transplantation). Each CT was evaluated for the presence of contact between the IC and the LV. Results The study included 28 patients:14 patients (16 scans) with HeartMate II (HMII) and 14 patients (16 scans) with HeartWare (HW). Characteristics of the cohort are shown in Table 1 . Contact between the IC and the LV was observed in all 16 scans (100%) of patients who received a HMII. In contrast, contact between the IC and the LV was observed in only 5/16 (31%) scans of patients who received a HW: 2/7 performed for a clinical indication and 3/9 performed as routine study. In both the HMII and HW groups, wall contact was observed throughout the cardiac cycle in all but one case, where it was only noted during end-systole. In all cases, contact with the LV involved the outside tip of the IC but did not appear to cause inflow obstruction. Conclusions Contact between the IC and the LV appears to be very common among patients with HM II and less common among patients with HW. In this cohort, cannula tip contact with the LV was not associated with any obstruction to inflow and is therefore unlikely to have any clinical significance. Patient Characteristics HMII HW Age (yrs) 51 ± 11 51 ± 16 Female 2/14 5/14 Time from LVAD implant to CT (days) 233 ± 204 275 ± 203 LV dimension (cm) 6.4 ± 1.3 6.3 ± 1.9

Management of Anticoagulation in Patients with Mechanical Circulatory Support Using a Physician-Based Versus a Pharmacist Directed Anticoagulation Clinic-Based Strategy

<h3>Purpose</h3> Patients with continuous flow left ventricular assist devices (CF-LVAD) require chronic anticoagulation with warfarin. The time in therapeutic range for a physician-based (MD) versus management by an anticoagulation clinic (AC) has not been well described. <h3>Methods and Materials</h3> All INRs and bleeding/thrombotic complications for patients with CF-LVAD from 1/1/12 to 11/15/12 at a single center were reviewed. Management of anticoagulation was changed from MD to AC clinic on 7/1/12. Time in therapeutic range (TTR) was determined using the fraction of INRs in range; values during hospitalization were excluded from analysis. Goal INR: HeartMate II (HMII) 1.5-3, HeartWare (HW) 2.5-3.5. <h3>Results</h3> There were a total of 48 patients who were on support during the period studied, with 1095 INR values evaluated; mean age 48, 48% ischemic, 90% male, 60% DT, 73% HeartMateII, 27% HeartWare. There were no differences in TTR between MD and AC, overall or by device. The HW group had a significantly lower TTR (52.5% v. 71%, p<0.001). When not in TTR the INR was subtherapeutic 58% overall, 49% in HMII, and 76% in HW. Adverse events for MD v. AC: bleeding (6 v. 4), stroke (2 v. 1), and pump thrombus (1 v. 1). Of those with bleeding 5/10 had supratherapeutic INR. [figure 1] <h3>Conclusions</h3> There was no difference in the TTR for patients with CF-LVAD between physician and anticoagulation clinic management, with about a third of patients overall not at goal. When not at goal about half of the patients were subtherapeutic, with HW patients more likely to be subtherapeutic than HMII. Strategies to maximize TTR for patients with CF-LVAD are needed.

Severity of End Organ Damage as a Predictor of Outcomes after Implantation of Continuous Flow Left Ventricular Assist Devices (LVAD)

Purpose The optimal timing of implantation of LVAD in management of advanced heart failure remains controversial. We hypothesize that in patients with cardiogenic shock, the severity of end-organ dysfunction as determined by the sequential organ failure assessment (SOFA) score is a determinant of outcomes after LVAD implantation. Methods and Materials We determined the pre-operative SOFA score and short and long-term outcomes of 97 consecutive patients who received Heartmate II (HMII) or Heartware (HW) LVAD at our institution since January 2007. The performance of the score in predicting outcomes was evaluated using receiver operating characteristic (ROC) analysis. Kaplan-Meier analysis was used to compare long term survival across SOFA score sub-groups. Results ν 18±10 days, p=0.03). One year survival for SOFA scores 0-2, 3-5, 6-8 and ≥9 was 94%, 75%, 64% and 29% respectively. SOFA score was significantly lower in survivors at 6, 9,12, 24, and 36 months. SOFA score did not predict adverse outcomes of bleeding, cerebrovascular events, infection or pump exchange. [ figure 1 ] Conclusions These results show that pre-operative SOFA score is a powerful predictor of outcomes after LVAD implantation. Long-term outcomes of patients with advanced heart failure treated with LVAD can be significantly improved by early intervention prior to emergence of end-organ dysfunction.

Overall Outcomes of Patients with Left Ventricular Assist Devices (LVADs) in Acute Rehabilitation Facility: How Safe Are These Patients at Rehabilitation Facilities?

Purpose Patients (pts) with LVADs are challenging with issues relating to transition of care and reducing hospital length of stay. However, in severely debilitated patients this can involve transitioning to long term care facilities, subacute and acute rehabilitation facilities. Currently, there are few centers that accept patients with LVADs which indirectly affect hospital length of stay (LOS). We report the outcomes including: LOS at acute rehab facility, adverse event rate, incidence and reason for rehospitalization in a small cohort of patients discharged to acute rehabilitation facility (rehab). Methods and Materials A retrospective review of patients who were discharged to rehab from January of 2009 till October of 2012. A total of 162 LVADs were performed with 42 admissions to rehab. Results The average LOS at rehab was around 14 days without any significant difference between patients with Heart Mate (HM) II and heart ware devices. There were 12 patients transferred back to our hospital with no deaths at rehab. 30 admissions were discharged home. Patients with heart ware devices had fewer rehospitalizations compared to patients with HM II devices. Reasons for rehospitilization included GI bleeding (8 pts), Arryhthmia (1 pts) and Sepsis (3 pts). [ figure 1 ] Conclusions Patients with LVADS can be safely transitioned to an acute rehab facility with the majority of patients successfully discharged home. There were no deaths, device failures, or strokes during the rehab stay. The average LOS at rehab was 14 days. There were 12 patients readmitted to our hospital and the most common reason being anemia due to gastrointestinal bleed.

Comparison of left ventricular geometry after HeartMate II and HeartWare left ventricular assist device implantation

BackgroundHeartMate II (HM II) and HeartWare (HW) Left Ventricular Assist Devices have been successfully used in end-stage heart failure patients as a bridge to transplantation, recovery, or decision. We set out to compare their effect in off-loading the left ventricle and its geometry.MethodsThe left ventricular end diastolic (LVEDD) and end systolic (LVESD) diameters were compared between first time HM II (n = 25) and HW implantations (n = 24) before and after the operation at 1, 3, and 6 months. A p value of less than 0.05 was considered as significant.ResultsPost-operative LVEDD and LVESD at 1, 3, and 6 months were significantly reduced in comparison with pre-operative values in both HM II and HW groups. No significant difference was found comparing HM II and HW groups together before and after the operation.ConclusionsOur study shows that both HM II and HW can significantly reduce the left ventricular systolic and diastolic dimensions and off-load the left ventricle. The miniaturized nature of HW does not affect its performance and it could be as effective as HM II.

Trend Towards Lower Rates of Gastrointestinal Bleeding in Patients Transplanted with HeartWare vs. HeartMate-2 for End Stage Heart Failure

Purpose: Left ventricular assist device (LVAD) augment systemic circulation and improves quality of life in patients suffering from end stage heart failure. Until recently, LVAD implantations had been performed with first generation pulsatile devices. Newer generation pumps work by a central rotor containing permanent magnets. The HeartMate-2 (HM-2), a 2nd generation LVAD is FDA approved for bridge to transplant and destination therapy. Their continuous flow mechanism allows reduction in pump size since external venting is no longer required for long-term pulsatile LVADs. Their generation pumps such as HeartWare (HW) have no bearings, longer durability, smaller size, and are easier to implant. They work by a central rotor containing permanent magnets which use a centrifugal pump where the rotors are shaped to accelerate the blood circumferentially and cause it to move toward the outer rim of the pump. For HM-2, several centers have reported rate of GI bleeds ranging from 15% - 40%. The purpose of this retrospective case control study is to compare the odds of GI bleed in patients implanted with HM-2 vs. HW pumps. Methods: Medical records of 98 patients (84 with HM-2 and 14 with HW), ages 28 - 77 at Penn State Hershey Medical Center from July 2006 through November 2011, were selected. GIB was defined as one guiac-positive stool with a subsequent transfusion of at least two units of PRBC. Results: In HM-2 vs. HW patients, GIB occurred in 20 (24%) vs. 1 (7%), respectively. There was no difference in median [IQR] time to onset of bleed for HM-2 vs. HW which were 51.5 [25.5- 347] vs. 49 [49, 49], p=1.0, respectively. In the multivariable logistic regression, age, sex, and extracorporeal membrane oxygenation (ECMO) prior to LVAD placement were not associated with increased incidence of GI bleed. Odd ratio showed a tread towards decreased rate of GI bleed with HW vs HM-2 (OR=0.21. p=0.190). Heart-Mate-2 Parameters Anticoagulation: Warfarin with goal INR of 1.5 - 2.0 with Aspirin 81 mg daily Speed: 9,000-10,000 rotation per minute (RMP) Flow: 3.5 - 6 L/min as appropriate for body surface area Power: 4 - 8 Watts. HeartWare Parameters Anticoagulation: Warfarin with goal INR of 2.0 - 3.0 combined with aspirin 81 mg daily Speed: 2,600 - 3,200 rpm Flow: 3.5 - 6 L/min as appropriate BSA Power: 3 - 6 watts Conclusion: There is an early trend to a decrease rate of GIB with the 3rd generation HeartWare LVAD pumps. This could be explained by the decreased rotations per minute leading to decreased shear stress. HW devices are offered solely to transplant candidates, which could add bias. Further prospective study as numbers of HW implants accrue will be necessary.

229 Toronto General Hospital Experience With Continuous Flow LVADS: A Comparison Between HeartMate II and HeartWare HVAD

The HeartWare (HW) HVAD device is a newer generation of centrifugal flow ventricular assist device (VAD) which has been recently introduced in Canada. The objective of our study was to evaluate outcomes in patients supported with a HW device in comparison to patients supported with a HeartMate II (HMII) axial flow device.

34 Impact of Driveline Material and Size on Exit Site Healing Time in Left Ventricular Assist Devices

Purpose: Left ventricular assist devices (LVADs) have proven beneficial for the treatment of advanced heart failure, however infection remains a common cause of morbidity and mortality. Slow initial driveline exit site (DLES) healing increases the risk of acquiring an infection. In this study, the transcutaneous driveline material and size were examined to determine the effect on DLES healing. Methods and Materials: The Utah Artificial Heart Program database was queried for patients implanted with a HeartMate II (HMII) or a HeartWare (HW) LVAD who had DLES assessments at 1, 2, and 3 months postimplant. To monitor healing, a score ranging from 1 to 4 was developed (1, fully incorporated with no drainage or erythema (D/E); 2, initial tear in skin with slight D/E; 3, increased D/E; 4, open wound with copious D/E). These data were then used to compare the effects of material and size on DLES healing using a Student’s t-test.

Continuous heart ware LVAD thrombosis: Pro-pump change

Aims: The HeartWare VAD, third-generation implantable VAD, includes an axial pump integrated into the device. Despite its advantages, the rate of thrombus formation and of thrombo-embolic events are not negligible. The therapeutic options are few; thrombolytic therapy with the risks of recurrence and bleeding, and surgical device exchange associating with the problems of an urgent redo procedure. Actually, there is no recommendation concerning the management of pump thrombosis. The aim of the study was to report the Bad Oeynhausen experience concerning the management of LVAD HeartWare thrombus. We related 7 cases of HeartWare thrombosis which were treated by the two strategy systemic thrombolysis and pump exchange.