European Journal of Cardio-Thoracic Surgery 48 (2015) 158 162 doi:10.1093/ejcts/ezu406 Advance Access publication 29 October 2014 ORIGINAL ARTICLE Cite this article as: Bernhardt AM, De By TMMH, Reichenspurner H, Deuse T. Isolated permanent right ventricular assist device implantation with the HeartWare continuous-flow ventricular assist device: first results from the European Registry for Patients with Mechanical Circulatory Support. Eur J Cardiothorac Surg 2015;48:158 62. a Isolated permanent right ventricular assist device implantation with the HeartWare continuous-flow ventricular assist device: first results from the European Registry for Patients with Mechanical Circulatory Support Alexander M. Bernhardt a, *, Theo M.M.H. De By b, Hermann Reichenspurner a and Tobias Deuse a Department of Cardiovascular Surgery, University Heart Center Hamburg, Germany b EUROMACS European Registry for Patients with Mechanical Circulatory Support, Berlin, Germany * Corresponding author. University Heart Center Hamburg, Department of Cardiovascular Surgery, Martinistraße 52, 20246 Hamburg, Germany. Tel: +49-40-741052440; fax: +49-40-741054931; e-mail: al.bernhardt@uke.de (A.M. Bernhardt). Received 9 June 2014; received in revised form 11 September 2014; accepted 22 September 2014 Abstract OBJECTIVES: Isolated right ventricular (RV) dysfunction with preserved left ventricular function is difficult to treat and associated with high mortality. Temporary devices for right ventricular support [right ventricular assist device (RVAD)] are available and have been used for short-term right heart assistance. In some patients, RV function does not recover and long-term devices are needed. Recently, isolated RVAD implantation with a permanent HeartWare HeartWare ventricular assist device (HVAD) device has been reported in patients with acute RV infarction and chronic graft failure. However, isolated implantation on the right side remains rare and is still an off-label use for this pump. To gather European data, we queried the European Registry for Patients with Mechanical Circulatory Support (EUROMACS) database, in which procedures and outcome data for patients receiving mechanical circulatory support are registered. METHODS: Until May 2014, data of 8 patients (mean age 55.0 ± 17.3 years, 100% males) with an isolated HVAD for RV support were submitted to the EUROMACS registry. All patients were in INTERMACS classes 1 3. Device strategy was rescue therapy in 6 patients (75.0%) and destination therapy in 2 patients (25.0%). Indications for RVAD placement were acute myocardial infarction in 4 (50.0%), failure to wean from cardiopulmonary bypass in 2 (25.0%) and post-cardiotomy RV failure in another 2 patients (25.0%). Intra- and postoperative results of the EUROMACS registry were analysed. RESULTS: Inflow cannulas were implanted into the right atrium (RA) in 6 patients (75.0%) and into the RV in 2 patients (25.0%). CPB was used in 6 patients (75.0%). Four patients (50.0%) survived the first 30 days. During follow-up, 1 patient died after 44 days due to multiorgan failure. In the surviving three patients, 2 patients were transplanted after 29 and 419 days, respectively, and, in 1 patient, the device was explanted for pump thrombosis and recovered RV function. CONCLUSION: In this very specific and sick patient cohort within the EUROMACS registry, isolated permanent RVAD implantation is a novel and promising strategy for patients with isolated right heart failure. However, more experience is needed to identify patients who benefit from this technology. Keywords: Right ventricular failure Ventricular assist device Cardiogenic shock Myocardial infarction INTRODUCTION Isolated right ventricular (RV) failure with preserved left ventricular (LV) function is caused by either pulmonary hypertension, RVspecific cardiomyopathy, RV ischaemia or pulmonary or tricuspid valve disease [1]. In-hospital mortality rate of patients with RV infarction and cardiogenic shock is comparable with that of patients with LV failure and is over 50% [2]. Treatment with extracorporeal short-term devices for RV failure is feasible though associated with an in-hospital mortality rate of 58% [3]. In a certain number of patients, RV function does not recover and long-term support of the failing RV is needed. Recently, the first cases of isolated permanent right ventricular assist device (RVAD) implantation with the HVAD (HeartWare, Framingham, MA, USA) have been reported by our group for acute RV infarction and chronic graft failure, respectively [4, 5]. As discussed, the inflow cannulas may be placed into the right atrium (RA) or the RV and the outflow graft is anastomosed either to the main pulmonary artery or the right pulmonary artery. However, isolated HVAD implantation in the right heart is rare and still an off-label use for The Author 2014. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.
A.M. Bernhardt et al. / European Journal of Cardio-Thoracic Surgery 159 this device. To evaluate outcomes of patients undergoing permanent isolated RVAD implantation, European Registry for Patients with Mechanical Circulatory Support (EUROMACS) data were analysed. MATERIALS AND METHODS The EUROMACS registry In 2009, the EUROMACS registry was created to evaluate characteristics of implantation, long-term survival and complications of assist device recipients. By February 2014, data from 702 patients were submitted to the registry. Of those, 89 patients received mechanical circulatory support for RV failure. While 82 patients had a short-term device, 8 patients (mean age 55.4 years ± 17.3, 100% male) received a permanent HeartWare HVAD in the RVAD position and comprise our patient population. Patient population Three patients (37.5%) were in INTERMACS class 1, 3 patients (37.5%) were in INTERMACS class 2 and 2 patients (25.0%) in INTERMACS class 3. Seven patients were inotrope-dependent (87.5%), 2 patients were bridged with an intra-aortic balloon pump (25.0%) and 2 by an extracorporeal membrane oxygenator (25.0%) with or without inotropic support. Device strategy was rescue therapy in 6 patients (75.0%) and destination therapy in 2 patients (25.0%). Indications for RVAD placement were acute Table 1: Preoperative characteristics of patients undergoing permanent isolated RVAD implantation in EUROMACS Variable n =8 Age (years) 55.0 ± 17.3 Sex, male 8 (100%) Mean follow-up (days) 73.4 ± 140.3 Diabetes 2 (25.0%) Chronic obstructive pulmonary disease 3 (37.5%) Ultrafiltration/dialysis 2 (25.0%) Mechanical ventilation 4 (50.0%) Major myocardial infarction 5 (62.5%) Previous cardiac surgery 4 (50.0%) Primary diagnosis Coronary artery disease 3 (37.5%) Other 2 (25.0%) Idiopathic dilated cardiomyopathy 1 (12.5%) Valvular heart disease 1 (12.5%) Congenital heart disease 1 (12.5%) INTERMACS class 1 Critical cardiogenic shock 3 (37.5%) 2 Progressive decline 3 (37.5%) 3 Stable but inotrope-dependent 2 (25.0%) Bridging to VAD implantation With inotropes 7 (87.5%) With intra-aortic balloon pump 2 (25.0%) With extracorporeal membrane oxygenation 2 (25.0%) Device strategy Destination therapy 2 (25.0%) Rescue therapy 6 (75.0%) RVAD: right ventricular assist device; VAD: ventricular assist device. Table 2: Characteristics of 8 patients of the EUROMACS registry with isolated RVAD implantation RVAD configuration Device strategy CPB used IABP ECMO INTERMACS class Dialysis Intubation Previous cardiac surgery Gender VAD indication Cardiac arrest Patient Age (years) No RA-PA 1 64 Male Myocardial infarction No No No No No No 3 Destination therapy 2 71 Male Post-cardiotomy RV failure No No No No No No 2 Rescue therapy No RA-PA Yes RA-RPA No Yes Yes Yes No No 3 Destination therapy 3 66 Male Post-cardiotomy RV failure, tricuspid valve replacement 12 years after HTx 4 63 Male RCA dissection, CABG failure to wean from CPB Yes No Yes Yes Yes Yes 2 Rescue therapy Yes RV-PA 5 46 Male Failure to wean from CBP after postinfarction VSD No No Yes No Yes No 1 Rescue therapy Yes RA-PA closure 6 61 Male Myocardial infarction after Type A aortic dissection No Yes No Yes No No 1 Rescue therapy Yes RA-PA with RCA dissection 7 17 Male Myocardial infarction, congenital heart disease No No No Yes No Yes 2 Rescue therapy Yes RA-PA (ASD/VSD) 8 52 Male Myocardial infarction Yes No Yes No No No 1 Rescue therapy Yes RV-PA VAD: ventricular assist device; IABP: intra-aortic balloon pump; ECMO: extracorporeal membrane oxygenation; ASD: atrial septal defect; VSD: ventricular septal defect; RCA: right coronary artery; CABG: coronary artery bypass grafting; CPB: cardiopulmonary bypass; RVAD: right ventricular assist device. TX & MCS
160 A.M. Bernhardt et al. / European Journal of Cardio-Thoracic Surgery Table 3: Outcomes of 8 patients of the EUROMACS registry with isolated RVAD implantation Patient Time on device Event Outcome Cause of death 1 16 Subileus, laparotomy and caecal fistula Died Infection 2 20 Died Multiorgan failure 3 44 Died Multiorgan failure 4 419 Driveline infection Transplanted 5 40 Pump thrombosis Explanted 6 5 Reoperation after three days for bleeding Died Multiorgan failure 7 29 Transplanted 8 14 Pump thrombosis Died Pump thrombosis RVAD: right ventricular assist device. myocardial infarction in 4 (50.0%), failure to wean from cardiopulmonary bypass in 2 (25.0%) and post-cardiotomy RV failure in 2 patients (25.0%). Primary diagnoses included coronary artery disease in 3 patient (37.5%), congenital heart disease in 1 patient (12.5%), idiopathic dilated cardiomyopathy in 1 patient (12.5%), valvular heart disease in 1 patient (12.5%) and other heart diseases in 2 patients (25.0%). Patient characteristics were summarized in Tables 1 3. Anonymized intra- and postoperative results of the EUROMACS registry were retrospectively analysed. Statistical analysis Data were collected using the EUROMACS database. Categorical variables were summarized by frequencies and percentages. Continuous variables were described by mean ± standard deviation. Kaplan Meier analysis was used for time-related variables. All statistical analyses were performed using IBM SPSS Version 22.0 for Mac. RESULTS Haemodynamic characteristics All patients had isolated right heart failure with preserved left heart function (LV ejection fraction was 49 ± 5.8%). At implant, the mean right atrial pressure was 15.5 ± 4.4 mmhg. Systolic and diastolic pulmonary artery pressures were 43.4 ± 19.3 and 23.5 ± 12.8 mmhg, respectively. Mean tricuspid annular plane systolic excursion was 5.0 ± 4.1 mm. The mean heart rate was 103.5 ± 10.5 bpm. Systolic and diastolic systemic blood pressures were 91.8 ± 10.3 and 60 ± 16.4 mmhg, respectively. Five patients were in sinus rhythm at the time of RVAD implantation, 1 patient had atrial fibrillation and 2 patients were paced. Intraoperative and procedural characteristics Cardiopulmonary bypass was used in 6 patients (75.0%) with a mean run time of 115.3 ± 92.0 min. Inflow cannulas were implanted into the RA in 6 patients (75.0%) and into the RV in 2 patients (25.0%). The mean OR time, including concomitant procedures, was 315.7 ± 212.3 min. The HVAD pump speed was set to 1933 ± 780 rpm. Figure 1: Survival of patients after durable isolated RVAD implantation. Survival and events Four patients (50.0%) survived the first 30 days. During follow-up, 1 patient died after 44 days due to multiorgan failure. In the surviving 3 patients, 2 were transplanted after 29 and 419 days, respectively, and, in 1 patient, the pump was explanted due to pump thrombosis and recovered RV function. Survival rates are 50.0% at 30 days and 25.0% after 1 year (Fig. 1). Follow-up complications included pump thrombosis in 2 patients, driveline infection in 1 patient and major non-cerebral bleeding in another patient. Reasons for death were multiorgan failure in 3 patients, sepsis due to an ileus in 1 patient and pump thrombosis in another patient (Table 3). DISCUSSION Specific RVAD issue We report here the first results out of the EUROMACS registry and the first series so far on procedural characteristics and outcomes of patients undergoing isolated permanent RVAD implantation with the HVAD device. First techniques and experiences of HVAD support in RVAD position resulted from patients with two HVAD pumps used as biventricular assist devices [6]. Later, the first HVAD
A.M. Bernhardt et al. / European Journal of Cardio-Thoracic Surgery 161 implant was reported for permanent isolated RV support [4]. Modifications to the outflow graft to increase the resistance and mimic systemic values were described in the initial experience [6], but were later abandoned as it became clear that the pre- and afterload sensitivity of the HVAD device allows its unmodified use on the right heart side [4]. Five of 8 patients within this series were implanted in our centre and no outflow graft obstruction was performed. The blood flow at a constant impeller speed is determined by the differential pressure through the pump. Values are estimated and compared with reference values for a given impeller speed. Flow estimation is based on electrical current, impeller speed and a fixed viscosity value, and developed for the left heart side. Owing to lower afterload on the right heart side and low vascular resistance of the pulmonary vasculature, it should be kept in mind that calculated flows displayed by the device might overestimate the real blood flow. Because it is still an off-label use, the HeartWare HVAD as a permanent isolated RVAD will not usually be reimbursed by health insurance companies, at least in Germany, and every case needs to be negotiated on an individual basis. Patient selection and outcome All patients in this series were either inotrope-dependent or on a short-term mechanical circulatory support device for cardiogenic shock. The survival of this special high-risk cohort reported herein must thus be valued against their otherwise dismal prognosis. While RVAD support should be instituted emergently in cardiogenic shock patients before irreversible neurological or second organ failure occurs, the optimal implant timing for patients with chronic right heart failure remains totally unknown. Specifically, it remains to be seen if patients with chronic RV failure would benefit from earlier device implantation as was shown for left ventricular assist device (LVAD) implantation in patients with LV failure [7]. One of our patients was in need of RVAD support for long-standing RV failure with severe tricuspid regurgitation 12 years after orthotopic heart transplantation. He already had severe liver cirrhosis and chronic renal dysfunction at the time of presentation. The tricuspid valve was replaced with subsequent RVAD implantation. However, the patient did not recover and died after 44 days from multiorgan failure. This patient may have benefited from RVAD implantation at an earlier stage. Three of 5 patients died of multiorgan failure. All patients were in INTERMACS class 1 3 and suffered from end-organ failure secondary to cardiogenic shock. None of those patients were bridged by a short-term MCS device. The implantation of a short-term extracorporeal device for improvement of end-organ function and neurological evaluation is a reasonable concept as a bridge to LVAD implantation. Application of this concept in patients with isolated RV failure might have improved these results and prevented some of the early deaths from multiorgan failure. Therefore, patients who do not recover their end-organ function and do not improve their overall condition on a short-term device should probably not undergo implantation of a long-term device. 60 80 s and no signs of hypercoagulopathy with concomitant antiplatelet therapy. One of these patients had the pump placed into the RV and one into the RA. The different pressure and flow patterns on the right heart side may lead to a higher thrombogenicity compared with the left heart. Also, microthrombi originating from the peripheral deep venous system, which usually get absorbed in the pulmonary vasculature without haemodynamic consequences, may cause acute pump dysfunction if they get trapped between the impeller and the pump housing. Both patients had no documented suction events before their pump thrombosis. Therefore, a more aggressive anticoagulation regimen might be necessary to prevent pump thrombosis in RVAD patients. Limitations As this study was a retrospective analysis of registry data, it carries limitations inherent to this study design. Specifically, data have to be submitted to EUROMACS on a voluntary basis independent of any German federal quality management requirement, in contrast to the more verified data reporting of the INTERMACS registry. Owing to the nature of a registry, technical and surgical details of VAD implantation (e.g. modifications on the outflow graft) cannot be analysed and compared, although they might have an impact on patient outcome. Also, pathophysiological details of the reasons of death cannot be given. As the registry develops and more centres are recruited, the statistical power of the data set will improve and the data will allow more detailed analyses on predictors of successful long-term survival in the future. However, to the best of our knowledge, all permanent isolated RVAD implantations within European countries are included here. CONCLUSIONS In this very specific and sick patient cohort within the EUROMACS registry, isolated permanent RVAD implantation is a novel and promising strategy in patients with isolated right heart failure. However, more experience is needed to identify patients who benefit from this technology. Centres that implant isolated RVADs are encouraged to submit their data to the EUROMACS registry. ACKNOWLEDGEMENTS We thank the organizing and scientific board of EUROMACS for the founding and building of the registry, especially Roland Hetzer and Jan Gummert. We also thank Allison P. Levin for editing the manuscript. TX & MCS Anticoagulation and pump thrombosis All submitting centres applied the same anticoagulation protocol to their RVAD patients as used for LVAD patients. In general, Coumadin derivatives were given to achieve an International normalized ratio (INR) between 2 and 3. Antiplatelet therapy was given additionally. Two patients (25%) had early pump thrombosis despite an INR between 2 and 3 or an activated partial thromboplastin time of Funding EUROMACS was supported by a financial grant of the Friede- Springer-Heart-Foundation to build up the registry. Conflict of interest: none declared.
162 A.M. Bernhardt et al. / European Journal of Cardio-Thoracic Surgery REFERENCES [1] Haddad F, Hunt SA, Rosenthal DN, Murphy DJ. Right ventricular function in cardiovascular disease, part I: anatomy, physiology, aging, and functional assessment of the right ventricle. Circulation 2008;117:1436 48. [2] Jacobs AK, Leopold JA, Bates E, Mendes LA, Sleeper LA, White H et al. Cardiogenic shock caused by right ventricular infarction: a report from the SHOCK registry. J Am Coll Cardiol 2003;41:1273 9. [3] John R, Long JW, Massey HT, Griffith BP, Sun BC, Tector AJ et al. Outcomes of a multicenter trial of the Levitronix CentriMag ventricular assist system for short-term circulatory support. J Thorac Cardiovasc Surg 2011;141:932 9. [4] Deuse T, Schirmer J, Kubik M, Reichenspurner H. Isolated permanent right ventricular assistance using the HVAD continuous-flow pump. Ann Thorac Surg 2013;95:1434 6. [5] Deuse T, Kubik M, Reichenspurner H. HVAD implantation in right atrium-to-right pulmonary artery configuration. Artif Organs 2014;38: 519 20. [6] Krabatsch T, Potapov E, Stepanenko A, Schweiger M, Kukucka M, Huebler M et al. Biventricular circulatory support with two miniaturized implantable assist devices. Circulation 2011;124:S179 86. [7] Kirklin JK, Naftel DC, Kormos RL, Stevenson LW, Pagani FD, Miller MA et al.fifth INTERMACS annual report: risk factor analysis from more than 6,000 mechanical circulatory support patients. J Heart Lung Transplant 2013;32:141 56.