The first-in-human experience with a minimally invasive, ambulatory, counterpulsation heart assist system for advanced congestive heart failure

http://www.jhltonline.org

ORIGINAL CLINICAL SCIENCE

The first-in-human experience with a minimally invasive, ambulatory, counterpulsation heart assist system for advanced congestive heart failure Valluvan Jeevanandam, MD,a Tae Song, MD,a David Onsager, MD,a Takeyoshi Ota, MD,a Colleen Juricek LaBuhn, RN, MSN,a Thomas Lammy, BS,a Gabriel Sayer, MD,a,c Gene Kim, MD,a,c Sonna Patel-Raman, PhD,a,b and Nir Uriel, MDa,c From the aDepartment of Surgery, University of Chicago Medical Center, Chicago, Illinois, USA; bNuPulseCV, Raleigh, NC, USA; and the cDepartment of Medicine, University of Chicago Medical Center, Chicago, Illinois, USA.

KEYWORDS: congestive heart failure; counterpulsation; mechanical assist device; cardiogenic shock; transplantation

BACKGROUND: The intravascular ventricular assist system (iVAS) is a new, minimally invasive, ambulatory counterpulsation heart assist system delivered via the subclavian artery and powered by a portable driver. It is designed for recovery, bridge to transplantation (BTT) or for prolonging medical therapy. We report the first-in-human (FIH) experience with iVAS. METHODS: This is a prospective, non-randomized single arm, U.S. Food and Drug Administration (FDA)-approved early feasibility trial in patients listed for cardiac transplantation. The primary endpoint was survival to transplant or stroke-free survival at 30 days. RESULTS: Fourteen patients were enrolled and 13 (92.8%) were treated with iVAS. At time of implant, the average age was 58 ± 6.7 years; 85% were male; 28% had ischemic cardiomyopathy; and 3 were Interagency Registry for Mechanically Assisted Devices (INTERMACS) Level 2, 9 were Level 3, and 1 was Level 4. The mean left ventricular ejection fraction was 22%, left ventricular internal diameter diastole was 7.13 mm, and 69% had moderate or severe mitral regurgitation. There were no intraoperative complications. Intensive care unit stay after implant was 6 ± 6 days. All patients were transplanted after 32 ± 21 days. There were no deaths or thromboembolic events: 1 patient required escalation of mechanical support, and post-implant complications included pleuritis/pericarditis (n ¼ 1) and neuropathy (n ¼ 2). No intra-operative blood transfusions were required. CONCLUSIONS: This study demonstrates a high rate of successful outcomes with an excellent risk-tobenefit profile. This FIH experience reveals that the iVAS can be successfully inserted in a standardized approach, provide hemodynamic support, can be interrupted for short periods, and allows for ambulation. A multicenter trial to investigate effectiveness and safety is warranted. J Heart Lung Transplant ]]]];]:]]]–]]] r 2017 International Society for Heart and Lung Transplantation. All rights reserved.

Each year, there are over 100,000 new patients with advanced congestive heart failure (aCHF) who could receive Reprint requests: Valluvan Jeevanandam, MD, Cardiovascular Surgery Section, Department of Surgery, University of Chicago Medical Center, 5841 South Maryland Avenue, Chicago, IL 60637. Telephone: þ773 702 2500. Fax: þ773 702 4187. E-mail address: [email protected]

a heart transplant.1 However, heart transplantation is limited by the shortage of donor organs. Worldwide, there were only 4,300 heart transplants reported in 2015, with 2,800 in the United States.2,3 As a result, over 4,000 continuous-flow LVADs (cfLVADs) are implanted per year and have become the mainstay therapy. These devices have improved survival and quality of life in aCHF patients4; however,

1053-2498/$ - see front matter r 2017 International Society for Heart and Lung Transplantation. All rights reserved. http://dx.doi.org/10.1016/j.healun.2017.10.011

2

The Journal of Heart and Lung Transplantation, Vol ], No ], Month ]]]]

according to the Interagency Registry for Mechanically Assisted Devices (INTERMACS), there is a 70% incidence of a major complication (death, thrombosis, hemolysis, bleeding, stroke) within the first year.4 Newer devices, like the HeartMate 3 (Thoratec Abbott, Pleasanton, CA), have lower thrombosis rates, but other complications remain unabated.5 For non–inotrope-dependent patients, considered “less critically ill,” the medical community is reluctant to implant cfLVADs due to the need for an invasive procedure (sternotomy or thoracotomy) and the complication profile. An alternative to continuous-flow pumps is counterpulsation, which consists of supplying energy to the circulation at the beginning of diastole. This decreases the workload of the heart and increases myocardial perfusion. Intra-aortic balloon pumps (IABPs) are implanted more than 200,000 times/year worldwide, and are often the first-line therapy for patients presenting with circulatory insufficiency.6 Due to insertion in the femoral artery, and size/complexity of the drive mechanism, patients are forced to stay in an intensive care unit (ICU). This limits support duration and patient mobility. Recently, in small studies, the IABP has demonstrated safe and effective prolonged support by using the subclavian artery for access (subIABP), allowing for ambulation while bridging to LVAD, transplant or recovery.7,8 The NuPulseCV (NuPulseCV, Inc., Raleigh, NC) intravascular ventricular assist system (iVAS) combines the benefits of IABP and cfLVADs by providing counterpulsation, while allowing ambulation in and out of the hospital. The iVAS is minimally invasive and requires no access to the heart. It has been tested on the bench for over 2.5 years. Should an iVAS fail, it can be easily exchanged. If a patient’s condition deteriorates, support can easily be escalated, as the iVAS is “forward compatible” (i.e., surgical planes and access for transplant or cfLVADs are not disturbed). Although we review the results of a first-inhuman (FIH) early feasibility trial of the iVAS as a bridge to transplant device, longer term clinical studies are planned to evaluate extended duration of support in patients with New York Heart Association (NYHA) Class III/IV aCHF.

o20 mm; subclavian diameter o7 mm; abnormalities of the aorta including heavy calcification or aneurysms; and/or uncontrollable atrial or ventricular arrhythmias that prevent proper electrocardiogram (ECG) tracking.

Investigational device The iVAS is an external heart assist device that has several components (Figure 1). The intravascular component is a 50-cc displacement pump (similar to an intra-aortic balloon) placed in the descending aorta. The skin interface device (SID) is an electromechanical and pneumatic conduit with a chimney that allows for shuttling of air between the pump and external driver and communication of the captured ECG signals that are transmitted to the driver from 3 subcutaneous electrodes. The SID is placed onto the lower chest cage and connects a driver to an external drive-line. An external and wearable drive unit provides compressed ambient air to inflate and deflate the pump. Similar to an IABP, the pump can be operated in 1:1, 1:2 and 1:3 modes, and the amount of augmentation is adjustable.

Procedure A small incision is made below the right or left clavicle to access the subclavian artery. An anastomosis is performed between a custom-designed Dacron graft and the artery. A guide-wire is directed using fluoroscopy to the descending aorta. An Atrieve vascular snare (Argon Medical Devices, Plano, TX) is inserted via the femoral artery. The snare is used to engage the guide-wire and is exteriorized through the subclavian graft. The iVAS is placed within the loops of the snare and is guided into the descending aorta. A hemostatic plug is placed around the pump drive-line and is secured with sutures to the inside of the graft. A subcutaneous pocket is made for the SID along the anterior axillary line, above the ipsilateral costal margin. Using a trephine, a skin button is created and the chimney of the SID exteriorized. Three 52-cm bipolar electrodes (Capsure Novus, Medtronic, Minneapolis, MN) are tunneled subcutaneously and connected to the SID. The pump

Methods Between April 2016 and April 2017, 14 patients were enrolled in our study at the University of Chicago Medicine (UCMC). Informed consent was obtained from all patients according to the requirements of an FDA-approved clinical study, along with approval of the UCMC institutional review board.

Trial design This trial was designed to assess a device with no previous human experience in a patient population presumably in need of a short, finite period of support. The trial was conducted to demonstrate a measurable benefit by evaluating safety and performance, including adverse events, device malfunctions and failures, functional assessments, procedural success and hemodynamics. The primary end-point was either survival to transplant or stroke-free survival at 30 days. The main criterion for inclusion was for patients to be listed for cardiac transplant United Network for Organ Sharing (UNOS) Status 1A or 1B. Key exclusion criteria included: aortic diameter

Figure 1

Jeevanandam et al.

Use of iVAS for CHF

3

drive-line is tunneled and connected to the SID. A cap is placed on the top of the SID and the device activated via an external driveline connected to a drive unit. Patients were anti-coagulated with aspirin and coumadin with an international normalized ratio (INR) of 1.5 to 2.

Statistical methods Data were collected using Excel 2007 (Microsoft Corp., Redmond, WA) and analyzed using SPSS categorical variables reported as frequencies, with continuous variables reported as mean ± standard deviation, or as median with interquartile range when applicable.

Table 2

Baseline Characteristics (n ¼ 13)

NYHA Class IV INTERMACS Level 4 INTERMACS Level 3 INTERMACS Level 2 Ischemic cardiomyopathy On inotropes LVEF Subclavian artery size Aorta size LVIDD Moderate and severe mitral regurgitation

100% 8% 69% 23% 28% 84% 22% 7.6 ± 0.68 mm 25 ± 2.85 mm 7.13 ± 1.3 cm 69%

INTERMACS, Interagency Registry for Mechanically Assisted Devices; NYHA, New York Heart Association; LVEF, left ventricular ejection fraction; LVIDD, left ventricular internal diameter diastole.

Results iVAS performance Twenty-six patients were screened. Fourteen patients were enrolled in the study. Reasons for exclusion included subclavian size (n ¼ 8), aortic insufficiency (n ¼ 2) and arrhythmias (n ¼ 2). One patient’s left subclavian artery was determined to be inadequate for implantation in the operating room and iVAS was not attempted (patient received standard IABP and was eventually transplanted). The focus of this analysis is on the as-treated cohort of 13 patients who were implanted with an iVAS. Patients’ demographic data and baseline characteristics are provided in Tables 1 and 2. All patients were extubated and no blood products were required in the operating room. Mean time in the ICU post-implantation was 6 (2 to 28) days; all 13 patients were ambulatory within 24 hours, and 11 patients were supported after the initial ICU stay in the stepdown unit. Two patients needed close hemodynamic monitoring and stayed in the ICU until transplantation. All patients had variable interruption of support due to routine maintenance or device issues requiring component replacement and restarts. In some cases, the patients elected to interrupt support for short periods, particularly when they started to feel better. One patient required re-initiation of milrinone for 2 hours until the device was operational. The maximum interruption of support (patient decision) was 12 hours. Two patients were discharged home for optimization until they were ready to be transplanted. There were no adverse events associated with discharge. All patients were successfully transplanted with 100% survival. We identified additional meaningful clinical outcomes, which are presented in Table 3. The table also shows the health status of patients and confirms overall successful Table 1

outcomes. One patient did not reach the primary end-point due to deteriorating heart failure requiring additional mechanical circulatory support. There were no bleeding complications. The iVAS was explanted by isolating the graft, removing the pump, and ligating the graft with silk ties. The SID and electrodes were removed and the skin closed primarily. There were no clots visualized on any of the pumps during explant. There was an improvement in 6-minute walk from 1,159 feet at baseline to 1,349 feet from baseline to last evaluation before transplant. All patients were ambulatory within 24 hours of the procedure.

Hemodynamics Hemodynamics pre- and post-iVAS are shown in Table 4. Overall, there was a significant improvement in cardiac output (CO), mean pulmonary capillary wedge pressure (PCWP) and central venous pressure (CVP). Nine patients (69%) were completely weaned off inotropes. Three patients were “super responders” with myocardial recovery (left ventricular ejection fraction [LVEF] improvement 26.2% to 38.3%) and improvement in hemodynamics from baseline to last catheterization (CVP –55%, PCWP –61%, CO þ53%). An example of a pressure volume loop from a super responder is shown in Figure 2. Only 1 patient required escalation of support after iVAS implantation. The patient had a surgical complication that caused a kink in the drive-line, resulting in inadequate support. This was discovered upon examining the device after explant. He was supported for 28 days and then required escalation to an Impella 5.0 (Abiomed, Danvers, MA) and then a CentriMag blood pump for extracorporeal membrane oxygenation

Patients’ Demographics (n ¼ 13)

Age Gender (male) Height Weight BSA Caucasian African American BSA, body surface area.

58 ± 6.7 years 85% 178.5 cm 91.2 kg 2.12 m2 70% 30%

Table 3

Primary End-point and Other Key Measures (n ¼ 13)

Met primary end-point Mean time on device Mean time in the ICU post-implant Off inotropes post-implant Improved hemodynamics from baseline Transplant delayed ICU, intensive care unit.

12 (92.3%) 32 ± 21 days 6 ± 6 days 69% 92% 0

The Journal of Heart and Lung Transplantation, Vol ], No ], Month ]]]]

4 Table 4

Hemodynamics

Inotropes/dose RA (mm Hg) PAS (mm Hg) PAD (mm Hg) PA mean (mm Hg) PCWP (mm Hg) CI (liters/min/m2) BP mean (mm Hg) Cardiac power output (W)

Pre-iVAS baseline (n ¼ 13)

Pre-iVAS baseline (n ¼ 12)

Post-iVAS 2 weeks (n ¼ 13)

Post-iVAS 4 weeks (n ¼ 6)

10 ± 4.5 52 ± 18.8 25 ± 7.9 35 ± 11.3 23 ± 7.3 1.8 ± 0.33 87 ± 7.8 0.753

Milrinonea/0.35 10 ± 2.6 46 ± 17.5 25 ± 7.2 34 ± 10.7 20 ± 6.2 2.24 ± 0.65 80 ± 10.1 0.820

8 ± 5.2b 39 ± 13.2 20 ± 8.0 27 ± 10.2b 18 ± 10.1b 2.5 ± 0.60b 89 ± 13.6 1.014b

7 ± 3.5b 40 ± 10.6 19 ± 12.0 25 ± 11.3b 16 ± 9.8b 2.5 ± 0.28b 98 ± 1.4 1.172b

BP mean, mean blood pressure; CI, cardiac index; iVAS, intravascular ventricular assist system; PA mean, mean pulmonary artery pressure; PAD, pulmonary artery diastolic pressure; PAS, pulmonary artery systolic pressure; PCWP, pulmonary capillary wedge pressure; RA, right atrial pressure. a One patient was not on inotropes. b

Significant difference from baseline at p o 0.05.

(Thoratec Abbott, Pleasanton, CA). He was transplanted on Day 35.

Adverse events All patients survived stroke-free to transplant. There were no thromboembolic events. One patient received a unit of packed red cells on Post-operative Day 2 (POD2) for a hemoglobin of 7 g/dl. One patient required a reoperation to adjust the SID on POD1. Four patients developed pseudoaneurysms from the femoral artery access sites. Two patients resolved without therapy; 2 required thrombin injections. Platelet counts decreased after implant but rebounded by POD7 (pre-iVAS, 208; POD1, 163; POD7, 186; and POD14, 241). There was no evidence of hemolysis. Most patients had significant shoulder and arm pain that resolved by the first week. Two patients had neuropathy manifested as fingertip numbness that required 3 weeks and 4 months, respectively, to resolve. One patient had trauma to the drive-line and developed a superficial infection treated with oral antibiotics that did not delay transplant. Panel-reactive antibodies (PRA) did not increase from baseline in any patient.

Discussion Implant of cfLVADs are associated with significant complications, and thus are used primarily in critically ill

Figure 2

aCHF patients (INTERMACS 1 to 3).4 These devices require major open-heart surgery and, once implanted, their use is obligatory. For patients who are not as sick, the evaluation of existing cfLVADs in a randomized trial for non–inotrope-dependent patients has been challenged by the lack of clinical equipoise, as noted by the failure of the REVIVE-IT study.9 The ROADMAP study, comparing the HeartMate II with medical therapy in non–inotropedependent patients, demonstrated equivalent survival, with HeartMate II patients demonstrating improved functional status and more frequent adverse events.10 IABPs are frequently the first-line therapy in most cases of acute cardiac insufficiency. In addition, clinical use of prolonged IABP support is becoming more prevalent for bridging patients to transplant.7,8 Other counterpulsation devices, such as the CardioVAD, C-Pulse and Symphony, although different in design from IABPs, have shown effectiveness for longer periods of support in a clinical study, with some patients discharged home.11–13 We reported our own subIABP experience with a high success rate of bridging patients to transplantation (95.1%), cfLVAD (100%) and recovery with hospital discharge (50%).7 The average duration of support was 21 (13 to 34) days. Importantly, we had an excellent safety profile with the ability for robust physical rehabilitation. Other investigators have also reported success with subIABPs as bridge to transplant with similar results.8

Jeevanandam et al.

Use of iVAS for CHF

iVAS, a hybrid of IABP and cfLVADs, was developed with the following goals: (1) minimize adverse events; (2) minimize surgical trauma associated with implant and explant; (3) shorten recovery post-implant and allow for ambulation; (4) allow for easy removal/exchange to facilitate the next level of therapy or recovery; and (5) interrupt support for variable periods of time without causing catastrophic hemodynamic compromise or thromboembolic complications. In this early feasibility study, the iVAS met the safety objective in UNOS Status 1A/1B listed patients. Twelve of 13 patients achieved survival to transplant or stroke-free survival at 30 days. The complication rates were low. One patient required reoperation to revise the SID pocket. There were no thromboembolic events in the distal aorta or arm during support or after explant. There was no significant bleeding or hematologic abnormalities. The trial also demonstrated: (1) procedural success in all patients; (2) significant diastolic augmentation with normalization of cardiac output and PCWP in 11 of 13 patients; (3) weaning off inotropes in 9 of 13 patients; (4) successful bridge to transplant with excellent post-transplant survival in 12 patients; and (5) hemocompatibility in all patients. The hemodynamic effect of iVAS is similar to that of IABP. Counterpulsation within the descending aorta facilitates myocardial performance by increasing coronary perfusion and decreasing after-load. An investigation into why certain patients demonstrated a greater benefit from counterpulsation compared with others resulted in an evaluation of cardiac output power (COP). To benefit from counterpulsation, a certain amount of systolic reserve is required. This may be predicted using COP. Although not a pre-specified end-point, COP may be a useful parameter to predict clinical response to iVAS. Early research indicates that, if the native cardiac output power index (COP ¼ cardiac output / body surface area × mean arterial pressure / 451) is less than 0.33 W/m2, then IABP may not provide adequate support.14 In acute failure with a low COP, a continuous pump, such as the Impella 5.0 or CentriMag VAD, may be preferable. Chronically low COP may require support with a cfLVAD. However, counterpulsation with iVAS may be an alternative for patients in aCHF when COP is not critically low. Device-related issues resulted in interruption of support due to failures of the cap and external drive-line. This was mitigated by elongating the neck of the SID cap and by reinforcing the external drive-line to prevent electrical wire disconnection and implementing design changes to minimize the occurrence of an unplanned disconnect between the SID, the cap and the drive-line. There are major differences that make iVAS capable of prolonged support compared with subIABP. The iVAS pump component is thicker and is manufactured without any seams. In real-time in-vitro life testing, the iVAS pump has exhibited structural integrity for up to 2.5 years. IABP is triggered with external ECG leads, whereas iVAS uses 3 electrodes that are placed subcutaneously. In addition, there are frequent reports of pump migration or kinked drive-lines associated with subIABP implants, requiring replacement. The iVAS has a drive-line that is placed internally and secured to the

5 subclavian, and is manufactured from a material that minimizes kinking. The implantable SID is designed to prevent infection by having a broad base, which stabilizes the SID under the skin and prevents motion during pumping. The chimney comes out perpendicular to the skin and promotes circumferential healing. Compared with the IABP console, the iVAS driver is portable at 5.6 pounds. Finally, the iVAS design allows for automation and adjustments that permit patient management outside the hospital. In our observational experience, the minimally invasive procedure was associated with only minor bleeding; there was no requirement for additional blood product. This may have led to decreased activation of the immune system, specifically development of HLA antibodies. In this small study, none of the patients developed new or increased existing HLA antibodies. The iVAS also demonstrated an ability to be “forward compatible.” Access to the heart and surgical planes was not disturbed by the iVAS implant. It allowed for a less traumatic transplant compared with cfLVADs with less dissection, bleeding and cardiopulmonary bypass (CPB) time. As demonstrated, the posttransplant course of these patients has been relatively smooth without any major complications. As per the protocol, the iVAS was expected to be run continuously, but there were periods when the iVAS support was interrupted. This was due to device design, such as when there was failure of the cap or electively by the patient to change clothes or exchange the drive-line. When stopped, the iVAS collapses to a very slim and smooth profile in the aorta and a high velocity of blood is maintained in the aorta; there is no place for blood to stagnate. Having support interrupted did not have significant adverse consequences, in contrast to the cfLVAD, wherein stoppage can cause regurgitation to a cardiomyopathic ventricle with catastrophic consequences. Hemocompatibility, a recently defined term to convey the complex relationship between the circulating blood and cfLVAD pump interface, is optimized in the design of the iVAS.15 Pump thrombosis, stroke and gastrointestinal bleeding complications did not occur. By design, the iVAS does not increase shear stress on the blood cells because there is no rotational component to the device. It maintains a normal, pulsatile flow that improves oxygenation to the microvasculature. There is minimal mechanical interaction between the device and the heart due to its placement in the descending aorta and access to the heart is not required.

Clinical implications Although this FIH trial was limited to transplant listed candidates, the data, especially the favorable risk-to-benefit profile, further support evaluation of safety and effectiveness of the iVAS in a wider range of aCHF patients who are not on inotropes or those well supported on low-dose inotropes (INTERMACS 3 to 7). The study will be expanded to multiple sites and more patients will allow for better understanding of the population that will be best served with this technology.

6

The Journal of Heart and Lung Transplantation, Vol ], No ], Month ]]]]

Limitations This was a small, non-randomized study of the FIH experience with iVAS. The device was evaluated in 13 patients for a short period of time at a single center. Larger studies, with longer support times, are warranted to fully assess the safety and effectiveness of iVAS. The next phase of the trial has begun enrolling patients who are not required to be listed and do not have an irreversible contraindication for transplantation. In conclusion, this trial is the first for FIH mechanical circulatory support in the U.S. since the Abiocor total artificial heart trial in 1998. For many years, most circulatory support companies conducted their FIH studies outside the U.S., despite development of these devices within the U.S. To allow aCHF patients in the U.S. access to this innovative therapy, our study was designed in collaboration with the FDA, and brings FIH and early feasibility studies back to the U.S. The NuPulseCV iVAS is a novel, minimally invasive, ambulatory heart assist system that provides counterpulsation to the failing circulation. In this FIH study of aCHF patients who are candidates for heart transplantation, we demonstrated safety of the iVAS system with excellent hemodynamic response.

3.

4.

5.

6. 7.

8.

9.

10.

11.

Disclosure statement Sonna Patel is an employee of NuPulseCV. The authors have no other conflicts of interest to disclose.

12.

References

13. 14.

1. Mozzaffarian D, Benjamin EJ, Go AS, et al. Heart disease and stroke statistics—2015 update: a report from the American Heart Association. Circulation 2015;131:e29-322. 2. International Society for Heart and Lung Transplantation. Registries— Heart/Lung Registries 4 Slides—worldwide stats. https://www.ishlt.

15.

org/registries/slides.asp?slides=heartLungRegistry/. Accessed August 21, 2017. Organ Procurement and Transplantation Network. National data. https://optn.transplant.hrsa.gov/data/viewdata-reports/national-data/#/. Accessed August 1, 2017. Kirklin JK, Naftel DC, Pagani FD, et al. Seventh INTERMACS annual report: 15,000 patients and counting. J Heart Lung Transplant 2015;34:1495-504. Mehra MR, Naka Y, Uriel N, et al. A fully magnetically levitated circulatory pump for advanced heart failure. New Engl J Med 2017;376:440-50. Webb CA, Weyker PD, Flynn BC. Management of intra-aortic balloon pumps. Sem Cardiothorac Vasc Anesth 2015;19:106-21. Tanaka A, Tuladhar SM, Onsager D, et al. The subclavian intra-aortic balloon pump: a compelling bridge device for advanced heart failure. Ann Thorac Surg 2015;100:2151-7. Estep JD, Cordero-Reyes AM, Bhimaraj A, et al. Percutaneous placement of an intra-aortic balloon pump in the left axillary/subclavian position provides safe, ambulatory long-term support as bridge to heart transplantation. JACC Heart Fail 2013;1:382-8. Pagani FD, Aaronson KD, Kormos R, et al. The NHLBI REVIVE-IT study: understanding its discontinuation in the context of current left ventricular assist device therapy. J Heart Lung Transplant 2016; 35:1277-83. Starling RC, Estep JD, Horstmanshof DA, et al. Risk assessment and comparative effectiveness of left ventricular assist device and medical management in ambulatory heart failure patients: the ROADMAP study 2-year results. JACC Heart Fail 2017;5:518-27. Jeevanandam V, Jayakar D, Anderson AS, et al. Circulatory assistance with a permanent implantable IABP: initial human experience. Circulation 2002;106(suppl I):183-8. Hayward CS, Peters WS, Merry AF, et al. Chronic extra-aortic balloon counterpulsation: first-in-human pilot study in end-stage heart failure. J Heart Lung Transplant 2010;29:1427-32. Cecere R, Dowling RD, Giannetti N. Initial clinical experience with the symphony heart assist system. Ann Thorac Surg 2015;99:298-301. Sintek MA, Gdowski M, Lindman BR, et al. Intra-aortic balloon counterpulsation in patients with chronic heart failure and cardiogenic shock: clinical response and predictors of stabilization. J Card Fail 2015;21:868-76. Mehra M. The burden of haemocompatibility with left ventricular assist systems: a complex weave. Eur Heart J 2017;38:1-5.