СЕРДЕЧНАЯ И СОСУДИСТАЯ ХИРУРГИЯ ■
Имплантируемый магнитно-левитационный желудочек вспомогательного кровообращения в лечении терминальной сердечной недостаточности (одноцентровое исследование)
Ивак П., Пирк Я., Туканова З., Малий Я., Шаршой O., Меленовский В., Рига Х., Хегарова M., Доразилова З., Конарник М., Покорный M., Нетука И.
Институт клинической и экспериментальной медицины, 140 21, г. Прага, Чешская Республика
Актуальность. Вспомогательное левожелудочковое устройство (LVAD) HeartMate 3 (HM3) - это современный магнитно-левитационный желудочек вспомогательного кровообращения, устройство для поддержания функции желудочка с центробежным потоком, разработанное для повышения гемосовместимости.
Цель настоящего исследования - оценка клинических характеристик, частоты нежелательных явлений, выживаемости и исходов у пациентов, перенесших имплантацию НМ3. Материал и методы. Мы ретроспективно проанализировали группу пациентов (n=150), которым с апреля 2016 г. по июнь 2020 г. в одном и том же учреждении были имплантированы НМ3. Результаты. В исследовании приняли участие 150 (125 мужчин; 83%) пациентов в возрасте 57,2+11,9 года, большинство (n=65; 43,3%) из них соответствовали 3-й категории по классификации INTERMACS. За 4 года наблюдения 32 пациентам была выполнена трансплантация сердца, 37 умерли, 78 продолжают лечение, у 3 отмечено восстановление миокарда. Статистическая выживаемость после имплантации устройства составила 77,6; 74,6; 74,6 и 69,9% через 12, 24, 36 и 48 мес соответственно. 46 (31%) пациентам понадобилась установка временного вспомогательного пра-вожелудочкового устройства (RVAD). Другие наиболее частые нежелательные явления включали инфекции (74 пациентов; 49,3%), большие кровотечения (38; 25,3%), нейропатию (16; 10,6%). Заключение. Полученные в исследовании данные по выживаемости были сопоставимы с данными международных реестров по использованию НМ3 для длительной поддержки кровообращения. Результаты подтверждают высокую надежность системы, повышенную гемосовмести-мость и пониженную частоту нежелательных явлений. Тем не менее инфекционные осложнения остаются нерешенной проблемой.
Финансирование. Это исследование было поддержано исследовательским грантом Министерства здравоохранения Чешской Республики (Институт клинической и экспериментальной медицины, грант IN 00023001). Конфликт интересов. Все авторы получили грантовую поддержку от Abbott. Ивак П. - консультант Abbott и Carmat SA. Нетука И. является консультантом и входит в состав консультативных советов компаний Abbott, LeviticusCardio Ltd. и Evaheart Inc. Пирк Я. является консультантом Abbott.
Для цитирования: Ивак П., Пирк Я., Туканова З., Малий Я., Шаршой O., Меленовский В., Рига Х., Хегарова M., Доразилова З., Конарник М., Покорный M., Нетука И. Имплантируемый магнитно-левитационный желудочек вспомогательного кровообращения в лечении терминальной сердечной недостаточности (одноцентровое исследование) // Клиническая и экспериментальная хирургия. Журнал имени академика Б.В. Петровского. 2020. Т. 8, № 3. С. 7-16. DOI: https:// doi.org/10.33029/2308-1198-2020-8-3-7-16 (англ.) Статья поступила в редакцию 30.07.2020. Принята в печать 25.08.2020.
ДЛЯ КОРРЕСПОНДЕНЦИИ
Ивак Петер - MD, врач отделения сердечно-сосудистой хирургии Института клинической и экспериментальной медицины (Прага, Чешская Республика) E-mail: peter.ivak@ikem.cz https://orcid.org/0000-0001-6845-7080_
Ключевые слова:
левожелудочковое
вспомогательное
устройство,
HeartMate 3,
осложнения,
выживаемость,
исходы
A single-center experience with magnetically levitated left ventricular assist device for treatment of end-stage heart failure
OORRESPONDENCE
Peter Ivak - MD, PhD, Doctor of the Department of Cardiovascular Surgery, Institute of Clinical and Experimental Mediane (Prague, Czech Republic) E-mail: peter.ivak@ikem.cz https://orcid.org/0000-0001-6845-7080
Keywords:
Left Ventricular Assist Device, HeartMate 3, complications, survival, outcomes
Ivak P., Pirk J., Tucanova Z., Maly J., Szarszoi O., Melenovsky V., Riha H., Hegarova M., Dorazilova Z., Konarik M., Pokorny M., Netuka I.
Institute for Clinical and Experimental Medicine, 140 21, Prague, Czech Republic
Background. The HeartMate 3 Left Ventricular Assist System is a recent compact fully magnetically levitated centrifugal-flow continuous ventricular assist device engineered to enhance hemocom-patibility. The aim of this study was to evaluate clinical characteristics, adverse event rates, survival and outcomes of patients supported with HeartMate 3 LVAS.
Material and methods. We retrospectively reviewed a cohort of consecutive 150 patients implanted with HeartMate 3 LVAS at a single institution between April 2016 and June 2020. Results. The study comprised of 150 patients (125 male; 83%) aged 57.2+11.9 years, with majority in INTERMACS profile 3 (n=65; 43.3%). During 4-year follow-up 32 patients were transplanted, 37 had died, 78 are ongoing and in 3 (2%) patients had a myocardial recovery. The actuarial survival after device implantation was 77.6, 74.6, 74.6 and 69.9% at 12, 24, 36 and 48 months respectively. 46 (31%) patients were implanted temporary RVAD. Other most frequent adverse events included infections (74 patients; 49.3%), major bleeding (38; 25.3%) or neurological dysfunction (16; 10.6%). Conclusion. We have observed survival rates comparable with international registries using the HeartMate 3 LVAS for long-term circulatory support. The results of the study confirm high reliability of the system, enhanced hemocompatibility and improved adverse event rates, nevertheless, some complications remain a concern.
Funding. This study was supported by research grant from the Ministry of Health, Czech Republic (Institute for Clinical and Experimental Medicine, grant IN 00023001).
Conflict of interests. All authors received institutional grant support from Abbott. Ivak P. is a consultant for Abbott and Carmat SA. Netuka I. is a consultant and serves on advisory board for Abbott, LeviticusCardio Ltd. and Evaheart Inc. Pirk J. is a consultant for Abbott.
For citation: Ivak P., Pirk J., Tucanova Z., Maly J., Szarszoi O., Melenovsky V., Riha H., Hegarova M., Dorazilova Z., Konarik M., Pokorny M., Netuka I. A single-center experience with magnetically levitated left ventricular assist device for treatment of end-stage heart failure. Clinical and Experimental Surgery. Petrovsky Journal. 2020; 8 (3): 7-16. DOI: https:// doi.org/10.33029/2308-1198-2020-8-3-7-16 Received 30.07.2020. Accepted 25.08.2020.
Use of left ventricular assist devices (LVAD) both as bridge-to-transplant and destination therapy has become a standard of care in advanced heart failure therapy [1]. Technical developments introduced in past decades as well as improvements of patient and adverse event management and implant timing strategies have led to improved survival, quality of life and decrease of adverse event rates [1, 2]. Nevertheless, therapy with these devices is partially limited by occurrence of adverse events such as thrombosis, stroke or bleeding, related to interaction between the device and blood elements [3]. Therefore, further refinement is required to address and prevent the occurrence of such events. The novel device, HeartMate 3 (HM3) LVAS (Abbott, Chicago, IL) is a continuous centrifugal - flow pump with a fully magnetically levitated rotor. The novel key device features, including magnetically levitated rotor which eliminates the need for any bearings,
wide blood flow path gaps and fixed automatic speed modulation which allows for periodical wash-out of the device may lead to minimization or shear stress on blood elements and prevent stasis, and thus enhance hemocompatibility and decrease rates of major adverse events as reported in recent trials [4-6].
First patients implanted at our institution in 2014 were part of the HM3 CE Mark trial and the pump was used as primary choice LVAD at our center since 2016. Present retrospective study was conducted to evaluate our single-center experience with Heart-Mate 3 LVAS.
Materials and methods
Study design
We retrospectively evaluated a cohort of 150 consecutive patients implanted with HeartMate 3 at
our institution between April 2016 and June 2020. Patients who were implanted with other LVAD during this period or underwent pump exchange from a different LVAD type were excluded from the study. All primary HeartMate implantations and planned bi-ventricular support with HeartMate 3 were included in the study. Implantations of HM3 in right sided configuration in patients with preexisting LVAD support with HeartMate 3 are also reported. Institutional ethics committee waived the need for patient informed consent as this study was designed as retrospective.
Data collection
Patient baseline and follow-up data, causes of death and clinical adverse events were collected retrospectively by review of the electronic medical records database of Institute of Clinical and Experimental Medicine in Prague. Follow-up was 100% complete, maximum follow-up was 1489 days.
Surgical technique
Implantation technique. Two different surgical techniques were used. The device was implanted either via standard median sternotomy, or utilizing a less-invasive technique by accessing the apex of the heart via left anterolateral mini-thoracotomy and accessing the ascending aorta by an upper J mini-sterno-tomy (fig. 1). The implantations were performed using cardiopulmonary bypass (CPB), Extracorporeal Membrane Oxygenation (ECMO) or, in patients without need of concomitant procedures, the device was implanted off-pump. In a small series of patients we verified a feasibility to implant the HeartMate 3 directly from Impella 5-0. In general, no aortic cross-clamping was performed, unless it was necessitated by a concomitant left-sided procedure. The coring area for the apical cuff was located under transoesopha-geal Echo-CC in all cases. The apical sewing ring was sutured to the apex of the left ventricle using either "core then sew" or "sew then core" techniques, nevertheless we prefer the latter mentioned approach at our institution. In this case, the cuff is attached in the apical position by 2-T felt pledgeted polypropylene monofilament 3/0 sutures in contralateral segments and is subsequently affixed by identical non-transmural running sutures without additional felt strips. In patients with limited pericardial space we performed a V-shape pericardial incision in the projection of the left ventricle long axis-ventral to the phrenic nerve [7]. The outflow graft was positioned parallel to the diaphragm and along the right ventricle and lateral side of the right atrium upwards to the ascending aorta. In complex redo cases an alternative strategy was used to tunnel the outflow graft through left pleural space to reach the ascending aorta [7].
Fig. 1. Patient after less-invasive implantation of HeartMate 3. The pump was implanted via anterolateral thoracotomy and upper mini sternotomy. Driveline tunnelization was performed using double tunnel technique (C-shape)
Driveline management. Intraoperative tunneling of the driveline was performed using previously described double-tunnel (C-shape) technique with the first part of the tunnel located subfascially behind right rectus abdominis muscle and second part of the tunnel located subcutaneously [8, 9]. The exit site of the driveline was standardly located in upper left quadrant of the abdomen. The driveline was then covered with a two 3 cm long rubber drains and fixed with two anchoring sutures (fig. 2). The application of the driveline cover was utilized to prevent the driveline crimping with fixation stitches. Further stabilization of the driveline was achieved using a soft anchoring device, the Centurion Foley Anchor (Centurion
Fig. 2. Fixation of the driveline with anchoring sutures at the time of implantation. The driveline is covered with a two 3 cm long rubber drains and fixed with two anchoring sutures. The application of the driveline cover is utilized to prevent the driveline crimping
with fixation stitches
Fig. 3. A soft anchoring device (Foley Anchor), used for the fixation of the driveline during the LVAD support
Medical, MI, USA), which in outpatient care fully replaces the anchoring sutures of the driveline (fig. 3). These were removed approximately one month after implantation according to our institutional driveline care protocol.
Concomitant procedures. Concomitant tricuspid valve repair was considered in patients with severe preoperative insufficiency and/or annulus size >40 mm. If more than mild aortic insufficiency was present, the valve was either replaced with a biological prosthesis or addressed by an aortic coaptation ("Park's") stitch [10]. In cases of severe aortic or mitral stenosis, the valve was replaced with biologi-
Fig. 4. A patient with simultaneous LVAD (HeartMate 3) and temporary RVAD (CentriMag) implanted via sternotomy. The outflow cannula of the RVAD is transcutaneously inserted into a Dacron graft anastomosed to pulmonary artery. The Dacron graft reaches to subcutaneous portion of the skin in subxiphoid area. Inflow cannula of the RVAD is inserted through femoral vein using Seldinger technique. This strategy allows for a mini invasive explantation of the RVAD.
(LVAD: left ventricular assist device; RVAD: right ventricular assist device)
cal prosthesis. In patients with present mechanical aortic valve we replaced the valve with biological prosthesis, and conversely, re-replacement of functional mechanical (or biological) valve in mitral position was not performed. Mitral repair with Alfieri stitch via the apex of the heart was considered in patients with severe mitral regurgitation indicated for implantation as destination therapy. In patients with severe carotid stenosis the endarterectomy was indicated according to current ESC (European Society of Cardiology) guidelines. Surgical myocardial revascularization was performed in patients with right ventricular ischemia. We also rigorously addressed the potential pre-pump risk factors for hemocompatibility adverse events (e.g. pump thrombosis, systemic embolization). Therefore surgical patent foramen ovale closure and left atrial appendage exclusion was performed in patients with history of atrial fibrillation or patent foramen ovale. In case of presence of intracardiac thrombi these were removed at the time of implant via the apex of the heart utilizing "core then sew" technique to inspect the ventricular cavity thoroughly.
Right ventricular (RV) dysfunction. In case of pre-operative or perioperative RV dysfunction requiring implantation of temporary mechanical circulatory support (MCS), we used an implant technique which allows for mini-invasive explanation in cases of RV recovery as described previously (fig. 4) [11]. In brief, the inflow cannula was inserted percutaneously via the right femoral vein using the Seldinger technique. A Dacron graft with diameter of 10mm was anastomosed to main pulmonary artery with a running suture using Satinsky clamp and the outflow cannula was inserted to the outflow graft transcutaneously with the exit site in right subcostal portion of the abdomen. The 10 mm graft diameter allows for eventual implantation of durable right ventricular assist device (RVAD) from right thoracotomy using the same graft for outflow anastomosis, in cases of refractory right heart failure (RHF). A CentriMag circulatory support system (Abbott, Chicago, IL) was used to support the right ventricle in most cases of severe RHF. The perioperative decision to implant a temporary RV support comprised echocardiography evaluation of RV function, inability to maintain sufficient LVAD flow and mean arterial pressure despite more-than moderate dose of inotropic and vasopressoric support. For a planned temporary RVAD decision we took into consideration multiple factors, including hemodynamic (CVP, PAPi, CVP:PCWP ratio), echocardiography (tricuspid regurgitation, RV:LV EDD ratio, TAPSE, FAC) and endorgan functions (renal, liver, mechanical ventilation) [12, 13].
Pericardial closure. In all patients we used the Gore Preclude pericardial membrane (W.L. Gore and
Associates Flagstaff AZ, USA) to enwrap the pump within the pericardium. Ventral closure of the pericardium was also performed using aforementioned membrane particularly in bridge-to-transplant cases to facilitate easier sternal re-entry.
Postoperative anticoagulation regimen
Continuous intravenous administration of heparin has been implemented as a bridge in all procedures until target anticoagulation range with warfarin has been reached. After implantation, the aim of anticoagulation therapy was to reach international normalized ratio (INR) of 2.0-2.7; in patients with trombophilias it was 2.5-3. Thrombophilias were defined as follows: mutation in factor V (Leiden); mutation in factor II (prothrombin); homozygosity for methylentetrahydro-pholate reductase (MTHFR); heterozygosity for MTHFR accompanied by hyperhomocysteinaemia or presence of lupus anticoagulant antibodies. Acetylsalicylic acid (ASA) has been administered as a part of antithrombotic regimen in dose of 100 mg per day. In a subgroup of patients (n=15) included in a study aimed on evaluation of low-intensity anti-coagulation, the target INR was 1.5-1.9. The inclusion and exclusion criteria and anticoagulation protocol were in detail published previously [14]. In patients (n=11) included in a different study aimed to evaluate an antithrombotic monotherapy with the HeartMate 3, warfarin was not administered and the patients were treated with a single anti-thrombotic agent only [15].
Statistical analysis
Continuous variables are expressed as mean ± standard deviation (SD) or median (interquartile range). Discrete variables are shown as absolute frequencies (percentages). Kaplan-Meier curves have been used to visualize survival without censoring at the time of heart transplant. All statistical analyses were performed using SPSS version 22 (IBM SPSS Statistics; IBM Corporation, Chicago, Ill). All statistical tests and confidence interval were 2-sided with a significance level of 0.05.
Results
Patient characteristics
In this single-center retrospective study we reviewed 150 consecutive implantations of HeartMate 3 LVAS performed between April 2016 and June 2020 at Institute for Clinical and Experimental Medicine in Prague, Czech Republic. Other implanted LVADs (Jarvik 2000; n=3) as well as LVAD exchanges from different LVAD type to HeartMate 3 (HeartMate II; n=2) were excluded from the analysis. The study population consisted of 125 (83%) men and 25 (17%) women, aged 57.2+11.9 years. Ischemic heart dis-
ease was primary cause of heart failure in 70 (46.7%) patients and dilated cardiomyopathy in 71 (47.3%) patients. Other primary causes of heart failure comprised hypertrophic cardiomyopathy (n=3; 2%), restrictive cardiomyopathy (n=1; 0.7%) and valvular heart disease (n=5; 3.3%). Device was implanted as bridge to transplantation in 107 (71.3%) and as destination therapy in 43 (28.7%) patients. Majority of the patients were classified as INTERMACS profile 3 (n=65; 43.3%). Atrial fibrillation was present in 54 patients (36%).
Detailed overview of patient baseline characteristic and pre-operative risk factors are summarized in tab. 1.
Table 1. Baseline patient characteristics
Demographics
Age (years), mean+SD 57.2+11.9
Gender male/female 125/25
Body surface area (m2), mean+SD 2.03+0.22
Body mass index, mean+SD 26.6+6.3
Diabetes, n (%) 47 (31.3)
Atrial fibrillation, n (%) 54 (36)
History of stroke, n (%) 14 (9.3)
Clinical profiles
NYHA class, n (%)
III 24 (16)
IV 126 (84)
INTERMACS profiles, n (%)
1 2 (1.3)
2 39 (26)
3 65 (43.3)
4 37 (24.7)
5 7 (4.7)
Hemodynamics, mean±SO
Left ventricular ejection fraction (%) 22.3+5.4
Left ventricular end-diastolic diameter (mm) 69.4+11.3
Mean arterial pressure (mm Hg) 78.7+9.8
Mean pulmonary artery pressure (mm Hg) 38.1+11.3
Pulmonary vascular resistance (Woods Units) 3.6+1.9
Mean capillary wedge pressure (mm Hg) 26.4+8.2
Cardiac index (l/m2/min) 1.03+0.9
Laboratory, mean±SO
Sodium (mmol/l) 134.9+12.3
Creatinine (umol/l) 122.4+56.2
Blood urea nitrogen (mmol/l) 34.5+19.2
White blood cells (109/l) 9.6+3.6
Platelets (109/l) 199.5+78.3
Hemoglobin (g/dl) 11.5+2.2
Total bilirubin (3g/dl) 1.9+1.8
Aspartate aminotransferase (U/l) 91.1+204.4
Alanine aminotransferase (U/l) 106.5+199.2
International normalized ratio 1.3+0.3
Lactate 1.6+1.6
Table 2. Preoperative mechanical circulatory support
ECMO, n (%) 16 (10,6)
ECMO + IABP, n (%) 1 (0,7)
ECMO + Impella, n (%) 3 (2)
Impella, n (%) 4 (2,7)
IABP, n (%) 1 (0,7)
LVAD CentriMag, n (%) 5 (3,3)
BiVAD CentriMag, n (%) 3 (2)
Note. ECMO - Extracorporeal Membrane Oxygenation; IABP -Intra-Aortic Balloon Pump; LVAD - left ventricular assist device; BiVAD - bi-ventricular assist device.
Pre-implant circulatory support
Thirty-three patients (22%) required preoperative mechanical circulatory support. Majority of the patients were supported by ECMO (n=20; 13.3%) and among these patients 4 required additional device to facilitate left ventricular unloading. Three of these patients were supported with combination of ECMO and Impella CP or Impella 5.0 and one patient was supported with ECMO in combination with intra-aortic balloon pump. 8 patients were implanted from previous CentriMag support in LVAD (n=5) or BiVAD (n=3) configuration. Impella 5.0 alone was used as bridge to implantation in 4 patients and Intra-Aortic Balloon Pump (IABP) in one patient. Overview of pre-implant MCS is provided in tab. 2.
Surgical techniques and concomitant procedures
130 (86.7%) patients were implanted via median sternotomy and in 20 (13.3%) patients the device was implanted using less-invasive technique through left antero-lateral mini-thoracotomy and upper J mini-sternotomy. Implantation after previous heart surgery was performed in 42 (28%) patients. Majority of the procedures was performed using CPB (n=128; 85%), while 22 (15%) patients were implanted off pump or directly from ECMO or Impella.
Temporary right-ventricular assist device was implanted in 46 (31%) patients, of which 31 (67%) patients received temporary RVAD during the primary procedure and 15 (33%) in early postoperative period. Durable bi-ventricular support with Heart-Mate 3 was primarily implanted in 5 (3%) patients and three patients received HeartMate 3 implantation as RVAD in latter phase due to refractory right heart failure.
A total of 50 (33%) concomitant valvular procedures were performed of which most frequent was a tricuspid repair alone, or with in combination with other valvular procedure (n=33; 22%). Aortic valve was replaced in 8 patients, and aortic coaptation ("Park's") stitch was performed in 5 patients. Mitral repair was performed in 15 (10%) patients.
Survival and outcomes
The mean duration of support was 15.6+ 14.5 months. The follow-up reached a total of 2331.5 patient-months. At the follow-up 32 patients were transplanted, 37 had died, 78 were ongoing and in 3 (2%) patients the therapy with the device was terminated due to myocardial recovery. In recovery cases a discontinuation of the device therapy with occlusion of outflow graft with Amplatzer occluder with the pump left in situ was performed in all 3 recovery patients. The driveline was terminated and surgically removed to the accessible extent.
The actuarial survival after device implantation was 80.1, 77.6 and 74.6% at 6, 12 and 24 months respectively (fig. 5). Long-term survival at 36 and 48 months was 74.6 and 69.9% respectively.
Adverse events
Mean duration of postoperative hospital stay was 40.3+32.3 days. Adverse events observed while on support are listed in tab. 3.
46 (31%) patients required temporary RVAD. Majority of the patients were successfully weaned
Fig. 5. Actuarial survival estimates after HearMate 3 implantation (n=150 patients). Patients were not censored at the time of transplantation
100
80
60
,5 40
20
Months % survival
6 80,1%
12 77,6%
24 74,6%
36 74,6%
48 69,9%
12 18 24 30 36 42 48 54 Follow-up time, months
0
Table 3. Adverse events during LVAD support
Major bleeding, n (%) 38 (25.3)
Gastrointestinal bleeding, n (%) 10 (6.6)
Major infection, n (%) 74 (49.3)
Device related, n (%) 20 (13.3)
Neurological dysfunction, n (%) 16 (10.6)
Transient ischemic attack, n (%) 5 (3.3)
Stroke, n (%) 11 (7.3)
Ischemic stroke, n (%) 6 (4)
Hemorrhagic stroke, n (%) 5 (3.3)
Cardiac arrhythmia, n (%) 18 (12)
Renal dysfunction, n (%) 35 (23.3)
Pump thrombosis, n (%) 3 (2)
Right heart failure requiring mechanical support, n (%) 46 (31)
from right ventricular support. In two patients the weaning of RVAD was not feasible due to refractory RV failure, and these patients were therefore implanted with additional HeartMate 3 for long-term RV support.
Bleeding complications were noted in 38 (25.3%) patients and the events occurred predominantly during initial hospital stay. Gastrointestinal bleeding was noted in 10 (6.7%) patients and was related to angiodysplazia in 2 patients.
74 (49.3%) patients experienced postoperative infections. Of these infections 54 (73%) were nondevice related. Pulmonary (n=23; 43%) and urinary infections (n=15; 28%) were most frequently observed non-LVAD infections.
Device related infections were observed in 20 (13.3%) patients, 18 patients developed a driveline infection (DL) and two patients had a pump pocket infection. Patients were treated by local wound care, targeted antibiotic therapy and eventually surgical debridement, vacuum assisted closure therapy and in cases of DL infection also re-position of the driveline exit site.
Neurological dysfunction was observed in 16 (10.6%) patients. 5 patients suffered from a transient ischemic attack (TIA), 11 (7.3%) patients experienced hemorrhagic (n=5; 3.3%) or ischemic (n=6; 4%) stroke.
Pump thrombosis occurred in 3 (2%) patients. In 2 patients the cause of the thrombosis was an embolus, which was present during inspection of the device. 4 (2.6%) patients underwent surgery due to outflow graft twist, which was de-rotated via lateral thoracotomy.
The most prevalent cause of death in presented patient population after LVAD implantation was multiorgan failure or RHF (n=16; 10.6%), sepsis (n=8; 5.3%) or cerebro-vascular accident (n=4; 2.6%). Overview of causes of death is provided in tab. 4.
Discussion
In this retrospective study we present an analysis of 4-year single-center experience with centrifugal magnetically levitated left ventricular assist device, the HeartMate 3 LVAS, for treatment of advanced heart failure. The key features of the device include with wide blood flow pathways, friction free movement of magnetically levitated rotor and intrinsic pulsatility to reduce shear stress and blood stasis [6]. As previously demonstrated, Heart-Mate 3 does not degrade high-molecular weight multimers od Von Willebrand factor to the extent observed with other devices [16], which further supports the enhanced hemocompatibility of the device, and along with low lactate dehydrogenase levels reported in recent trials, suggesting lower
Table 4. Primary causes of death
Bleeding, n (%) 4 (2.6)
Cardio-pulmonary failure, n (%) 1 (0,7)
Cerebro-vascular accident, n (%) 4 (2.6)
Device failure, n (%) 1 (0,7)
Lung failure, n (%) 1 (0,7)
Multi-organ failure or Right heart failure, n (%) 16 (10.6)
Sepsis, n (%) 8 (5.3)
Unknown cause of death, n (%) 2 (1.3)
occurrence of hemolysis [6]. In a largest comparative LVAD trial, the MOMENTUM 3 the HeartMate 3 showed lower rates of hemocompatibility-related adverse event and superior survival free of disabling stroke or reoperation to replace or remove a malfunctioning device in comparison to its predecessor, the HeartMate II [6].
In contrast to HeartMate II the small size of HeartMate 3 allows for intrathoracic placement and also for less invasive implantations. Minimally invasive procedures could carry some positive effects such as reduction of trauma, blood loss and infection rates, or decrease of in-hospital stay and accelerated recovery. Observations in ELEVATE registry found that bleeding (requiring surgery), infection and arrhythmias were lower in the group implanted via the less-invasive surgical approach than those who underwent open heart surgery. Additionally, using less invasive technique may have a positive effect on the incidence of postoperative right heart failure as the pericardium remains intact and right ventricular dilation during onset of the LVAD is prevented [17]. In our series 20 (13.3%) patients were implanted using less-invasive technique, via left anterolateral mini-thoracotomy and upper J mini-sternotomy, utilizing CPB or ECMO. Several groups reported use of bilateral mini-thoracotomy and off-pump less invasive implantations [18, 19].
In this study a 4-year experience with HM 3 are presented. Postoperative survival in our patient population was 77.6 and 74.6% at 12 and 24 months respectively. These results are comparable with results reported in annual reports of Imacs (81.1 and 72.4% respectively) and INTERMACS (82 and 72% respectively). The 2 year results of overall survival are comparable not only with the 2-year follow-up in the ELEVATE registry, but also with two-year survival in MOMENTUM 3 trial (74.5 and 79% respectively) [5, 6]. However, the one-year survival rate for HM 3 patients reported in MOMENTUM 3 trial was slightly higher (86.6%). This observation may be at least partially contributed to two main reasons. First, almost one third of our patients underwent previous cardiac surgery, and second, 22% of the patients required
preoperative mechanical circulatory support (ECMO, Impella or combination). These factors could have negatively influenced the risk of perioperative complications and patient outcomes. The long-term survival in our cohort of patients at 36 and 48 months was 74.6 and 69.9% respectively. However, despite the observation of favorable patient survival, the burden of adverse event remained significant.
Right heart failure (RHF) after LVAD implantation is associated with increased morbidity and mortality and thus remains a serious concern [13]. The incidence of early RHF varies according to definitions used. According to data from EUROMACS registry it occurs in 21.7% cases, the report of 2-year results of ELEVATE registry showed the incidence of RHF up to 15% [5]. Furthermore, recent meta-analysis estimated the incidence of RV failure to approximately 35% [20]. On top of frequent occurrence of RHF, the reliability of prediction scores is questionable and the RHF consequences are fateful. In our series the incidence of use mechanical RV support was 31%, in contrast to HM 3 CE Mark Study, ELEVATE Registry or MOMENTUM 3, where the incidence of temporary RVAD use varies between 4 and 7% [4-6]. As the addition of temporary RVAD may lead to improvement of RV recovery and improved patient outcomes, we tend to be liberal in use of temporary RVAD. The implant technique with inflow cannula inserted percu-taneously through femoral vein and outflow cannula inserted via Dacron graft anastomosed to pulmonary artery is simple and allows for mini-invasive explantation, which in our opinion does not create serious additional burden to the patient.
In past years one the most discussed complications was the LVAD thrombosis [21]. In our patient population, a suspected or confirmed pump thrombosis occurred in 3 (2%) patients. In 2 patients an embolus was present during the pump inspection at explant. In 1 patient the episode of pump thrombosis was preceded by an ischemic stroke, and related to a new onset of atrial fibrillation. We believe, that this finding further supports our strategy to address possible pre-pump factors of thrombotic adverse events, such as closure of left atrial appendage in patients with atrial fibrillation, extraction of ventricular thrombi and closure of patient foramen ovale at the time of device implantation. Furthermore, with the HeartMate 3 a new advanced diagnostics in the log files is possible, which could be instrumental in cases of pump thrombosis as reported recently [22]. Overall, in MOMENTUM 3 trial and primary cohort of the ELEVATE registry, the rate of suspected or confirmed pump thrombosis was reported at 1.4 and 1.5% respectively, which is comparable to our observation. These very encouraging results are very probably attributed to the novel design features of Heart-
Mate 3 (magnetically levitated rotor, wide blood flow paths, and artificial pulse). The overall hemo-compatibility of HeartMate 3 had been proved to be superior to HeartMate II in a recent sub-analysis of MOMENTUM 3 trial [23].
Nevertheless, bleeding remains a frequent complication after HeartMate 3 implantation [4-6].
Compared to previously reported results, the rates of bleeding events were lower in our study (25.8%). On contrary, we have observed similar rates of gastrointestinal bleeding as reported in ELEVATE registry results (9.7% at 2 years). The rather favorable bleeding rates in our study might have been influenced by our ongoing studies focused on low intensity anticoagulation protocols in HearMate 3 patients [14, 15]. In one of this studies we have shown, that targeted low-intensity anticoagulation is achievable and safe with HeartMate 3 [14]. Recently, several groups reported cessation of warfarin in patients who have suffered from bleeding complications with rather disputable outcomes [24, 25]. Further studies will have to evaluate also the ASA use in HeartMate 3 patients, as the recent analysis of MOMENTUM 3 showed that the dose of the ASA did not affect bleeding and thrombotic events rates [26]. The prospective, randomized, double-blind, placebo-controlled, international ARIES HM3 (Anti-platelet Removal and Hemocom-patibility Events with the HM3 Pump) will provide a definitive answer to question whether ASA therapy provides any meaningful therapeutic effect in HM 3 patients.
Postoperative strokes were reported in MOMENTUM 3 and ELEVATE Registry in 9.9 and 9.7% respectively at 2-years after implantation. In our study cohort we have observed slightly favorable stroke rate of 7.3%. Occurrence of hemorrhagic stroke in almost half of the stroke events may advocate for reduced anticoagulation trials and protocols, to further decrease the stroke rates with HeartMate 3.
Infection was a frequent complication after device implantation. Almost half of the patients suffered from any infection (49.3%), similarly to other reported patient cohorts [5, 6, 27, 28]. Interestingly, device related infections were observed in 13.3% patients, which is significantly superior to rates reported in MOMENTUM 3 or ELEVATE Registry at 2 years. Driveline infection was observed in 12% of patients. At our institution great attention is dedicated to driveline exit site management and staff and patient training. Our institutional protocol comprises meticulous driveline fixation, aseptic dressing exchanges in early postoperative period or, in latter phase, regular ambulatory visits to assess the exit site status. Also frequent patient re-trainings ensure strict adherence of the patients to the dressing protocol. Nevertheless, occurrence of the driveline infections emphasizes the need for fully-implantable device.
Limitations
Retrospective design of the study is a subject to limitations associated with retrospective studies. Also, reproducibility may be limited and affected by institutional experience and institutional specific characteristics, as all the patients were implanted at a single center. Moreover, this study is focused on a single device only and is not comparing outcomes of other available devices.
Conclusion
We have observed survival rates comparable with international registries using the HeartMate 3 LVAS for long-term circulatory support. The results of the study confirm high reliability of the system, enhanced hemocompatibility and have showed adverse event rates comparable or superior to previously published studies. Still, complications such as infections, bleeding or strokes remain a concern.
References
1. Goldstein D., Meyns B., Xie R., et al. Third annual report from the ISHLT mechanically assisted circulatory support registry: a comparison of centrifugal and axial continuous-flow left ventricular assist devices. J Heart Lung Tansplant. 2019; 38: 352-63.
2. Teuteberg J.J., Cleveland J.C., Cowger J., et al. The Society of Thoracic Surgeons Intermacs 2019 annual report: the changing landscape of devices and indications. Ann Thorac Surg 2020; 109: 649-60.
3. Mehra M.R. The burden of hemocompatibility with left ventricular assist systems: a complex weave. Eur Heart J. 2019; 40: 673-77.
4. Schmitto J., Pya Y., Zimpfer D., et al. Long-term evaluation of a fully magnetically levitated circulatory support device for advanced failure? Two year results from the HeartMate 3 CE Mark Study. Eur J Heart Fail. 2019; 21: 90-7.
5. Zimpfer D., Gustafsson F., Potapov E., et al. Two-year outcome after implantation of a full magnetically levitated left ventricular assist device: results from the ELEVATE Registry (December 4, 2019). DOI: https:// doi.org/10.2139/ssrn.3498422
6. Mehra M., Uriel N., Naka Y., et al. A fully magnetically levitated left ventricular assist device - final report. N Engl J Med. 2019; 380: 1618-27.
7. Netuka I. HeartMate 3 left ventricular assist system implantation technique: the devil is in the detail. Interact Cardiovasc Thorac Surg. 2018; 27: 946-9.
8. Schibilsky D., Benk C., Haller C., et al. Double tunnel technique for the LVAD driveline: improved management regarding driveline infections. J Artif Organs. 2012; 15: 44-8.
9. Yarboro L.T., Bergin J.D., Kennedy J.L., et al. Technique for minimizing and treating driveline infections. Ann Cardiothorac Surg. 2014; 3: 557-62.
10. Park S.J., Liao K.K., Segurola R., et al. Management of aortic insufficiency in patients with left-ventricular assist device: a simple coaptation stitch method (Park's stitch). J Thorac Cardiovasc Surg. 2004; 127: 264-6.
11. Saeed D., Maxhera B., Kamiya H., et al. Alternative right ventricular assist device implantation technique for patients with perioperative right ventricular failure. J Thorac Cardiovasc Surg. 2015; 149: 927-32.
12. Potapov E.V., Antonides C., Crespo-Leiro M.G., et al. 2019 EACTS Expert Consensus on long-term mechanical circulatory support. Eur J Cardiothorac Surg. 2019; 56: 230-70.
13. Lapert B.C., Teuteberg J.J. Right ventricular failure after left ventricular assist devices. J Heart Lung Tans-plant. 2015; 34: 1023-30.
14. Netuka I., Ivák P., Tucanová Z., et al. Evaluation of low-intensity anti-coagulation with a fully magnetically levitated centrifugal-flow circulatory pump-the MAGENTUM 1 study. J Heart Lung Transplant. 201S; 37: 579-S6.
15. Netuka I., Ivak P., Tucanova Z., et al. A trial of complete withdrawal of anticoagulation therapy in HeartMate 3 pump. J Heart Lung Transplant. 2019; 38: S113.
16. Netuka I., Kvasnicka T., Kvasnicka J., et al. Evaluation of von Willebrand factor with a fully magnetically levitated centrifugal continuous-flow left ventricular assist device in advanced heart failure. J Heart Lung Transplant. 2016; 35: S60-7.
17. Schmitto J., Krabatch T., Damme L., Netuka I. Less invasive HeartMate 3 left ventricular assist device implantation. J Thorac Dis. 2018; 10: S15.
18. Potapov E., Kukucka M., Falk V., Krabatsch T. Off-pump implantation of the HeartMate 3 left ventricular assist device through a bilateral thoracotomy approach. J Thorac Cardiovasc Surg. 2017; 153: 104-5.
19. Saeed D., Sixt S., Albert A., et al. Minimally invasive odd pump implantation of HeartMate 3 left ventricular assist device. J Thorac Cardiovasc Surg. 2016; 152: 1446-7.
20. Bellavia D., Iacovoni A., Scardulla C., et al. Prediction of right ventricular failure after ventricular assist device implant: systematic review and meta-analysis of observational studies. Eur J Heart Fail. 2017; 19: 926-46.
21. Maltais S., Kilic A., Nathan S., et al. PREVENtion of HeartMate II pump Thrombosis through clinical management: the PREVENT multi-center study. J Heart Lung Transplant. 2017; 36: 1-12.
22. Netuka I., Mehra M.R. Ischemic stroke and subsequent thrombosis within a HeartMate 3 left ventricular
assist system: a cautionary tale. J Heart Lung Transplant. 2018; 31: 170-2.
23. Uriel N., Colombo P.C., Cleveland J.C., et al. Hemocompatibility-related outcomes in the MOMENTUM 3 Trial at 6 months: a randomized controlled study of a fully magnetically levitated pump in advanced heart failure. Circulation. 2017; 135: 2003-12.
24. Schettle S., Stulak J., Rosenbaum A., et al. Outcomes following discontinuation of warfarin in HeartMate 3 patients: a multi-center review. J Heart Lung Transplant. 2020; 39: S397.
25. Fiedler A.G., Hermsen J.L., Smith J.W., et al. Long-term discontinuation of warfarin in a patient with
HeartMate 3 left ventricular assist device without complication. J Card Surg. 2020; 35: 447-9.
26. Saeed O., Colombo P.C., Mehra M.R., et al. Effect of aspirin dose on hemocompatibility-related outcomes with a magnetically levitated left ventricular assist device: an analysis from the MOMENTUM 3 study. J Heart Lung Transplant. 2020; 39: 518-25.
27. Nowacka A., Hullin R., Tozzi P., et al. Short-term single-centre experience with the HeartMate 3 left ventricular assist device for advanced heart failure. Eur J Car-diothorac Surg. 2020; 2020; ezaa075.
28. Hanke J.S., Dogan G., Rojas S.V., et al. First experiences with HeartMate 3 follow-up and adverse events. J Thorac Cardiovasc Surg. 2017; 154: 173-8.