In 1966, the first successful implant of a VAD in a patient with post-cardiopulmonary bypass cardiogenic shock was performed at Baylor College of Medicine in Houston [11]. In 2001, the HeartMate XVE was Food and Drug Administration (FDA) approved as the first pulsatile left VAD (LVAD ) for bridge to transplantation (BTT). The REMATCH trial was a prospective, randomized multicenter study that compared the HM-XVE LVAD versus optimal medical therapy for patients not considered eligible for cardiac transplantation [6] (◘ Fig. 22.2). Patients randomized to LVAD support had a 50 % increase in 1-year survival. Thus, in 2003, LVAD therapy was approved by the FDA as either BTT or destination therapy (DT). However, early pump failure limited the use of pulsatile pumps. The concept of continuous flow pumps (CF-LVADs) was developed in 1988 from a collaborative effort between Baylor College of Medicine and NASA [23]. After 10 years of development, the first CF-LVAD (MicroMed, DeBakey) was implanted in a human [24].

As the technology matured, the HeartMate II (HM II ) LVAD was the first CF-LVAD to be approved by the FDA as BTT in 2008 and then DT in 2010. This advancement had a significant impact on survival and outcomes; the 1-year survival with the HM II VAD exceeded 80 %, making it a viable alternative to heart transplantation [8]. The centrifugal CF-LVAD HeartWare pump (HVAD) was developed in 2005 and FDA approved in 2012 as BTT and in 2014 as DT. The completely redesigned HM III was introduced in 2015 and became part of the MOMENTUM III randomized controlled trial [25] (◘ Fig. 22.3).

Patients eligible for transplantation—or ineligible but with end-stage disease and unable to wait on the transplant waiting list—are referred for LVAD implantation. For 1992–2000, the International Society for Heart and Lung Transplantation reported that 12 % of transplant recipients were mechanically supported with an LVAD pretransplantation, compared to 28 % from 2006 to 2012 [26]. End-organ damage, despite maximal medical therapy, including inotropic support or inotropic support dependency with anticipated long waiting times for heart transplantation, is the major indication for CF-LVAD implantation.

These cases correspond to the Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) profiles 1–3. The INTERMACS is a US-based registry of patients receiving mechanical circulatory support (MCS) device therapy, including durable LVADs [27]. The INTERMACS scale assigns patients with stage D heart failure into seven levels according to hemodynamics and individual functional capacity (◘ Table 22.1). Patients in acute cardiogenic shock are assigned INTERMACS profile 1 (◘ Fig. 22.4). These patients are primarily supported with short-term MCS such as extracorporeal membrane oxygenation (ECMO), percutaneous axial VADs (Impella), or intra-aortic balloon counterpulsation and are then transitioned to more permanent support after they stabilize. These patients did poorly if they immediately received a durable LVAD, but most patients transitioning to a VAD in this scenario become relatively more stable and thus fall into INTERMACS profiles 2 and 3 (◘ Fig. 22.4). Irrespective of the profile, implantation can be divided into one of four groups: BTT, DT, bridge to decision (BTD) , or bridge to recovery (BTR) . The BTT designation is intended to support the patients on the waitlist until their transplant. On the other hand, the DT designation is intended to support the patients not eligible for transplantation at the time of implantation.

Separation of LVAD patients into BTT or DT can be challenging. During their acute illness, many patients may fall into a gray zone with a clinical status that improves over time and with the potential of listing for transplantation. These patients are frequently designated as BTD. In an attempt to normalize end-organ function that precludes long-term cardiac replacement therapies such as heart transplant or durable LVAD, these patients are often supported using temporary MCS such as percutaneous temporary VADs or ECMO. The risk factors for their increased mortality include large body weight (body mass index > 30 kg/m²), older age (>75 years), cardiogenic shock, and the need for a right ventricular assist device [28]. Thus, a multidisciplinary team should carefully make the decision for these subgroups.

Durable LVADs were primarily developed as a DT in patients ineligible for heart transplantation. Concerns about long-term performance and safety, however, prompted the FDA to restrict the initial use of such devices (HM-XVE) to patients who were eligible for a transplant, leading to the concept of BTT. This led to the development of more durable, smaller, and totally implantable devices, permitting a normal quality of life. The American Heart Association (AHA) guidelines recommend CF-LVAD implantation as BTT in patients awaiting heart transplantation who have become refractory to medical therapy (Class IB). Patients with refractory heart failure and hemodynamic instability expected to improve with time or who experience restoration of an improved hemodynamic profile should be considered for urgent MCS as a BTD (level of evidence C) [3, 29]. The European Society of Cardiology (ESC) recommends a CF-LVAD as BTT in selected patients with end-stage heart failure despite optimal medical therapy and who are otherwise suitable for heart transplantation to improve morbidity and mortality while awaiting transplantation (Class I, level of evidence B) [30] (◘ Table 22.2).

Establishing the time frame for implantation is critical to balance the risks and benefits. The prognostic implications of the INTERMACS profiles provide guidance for the indications and the optimal timing of implantation. Currently, BTT is a major indication for CF-LVAD implantation, with a myriad of advantages pertaining to transplantation. In the seventh annual INTERMACS registry report, 53.5 % of the primary VAD implantations in 2014 were for BTT [27]. One of the major advantages of implantation for BTT is the possibility to decrease the pulmonary pressures, leading to potential improvements in posttransplant outcomes and even to listing patients who were previously not eligible for listing (BTD) [31]. LVAD implantation also recovers end-organ perfusion and enables patients’ nutritional and functional status improvements. However, the timing of relisting heart transplant candidates who received an LVAD as a BTT remains an issue. An interval of about 6 months after implantation is recommended in cases of pulmonary hypertension [32]; however, no general consensus exists for patients who receive BTT LVAD for other reasons. Although it has been consistently shown that LVADs reduce mortality while patients are on the transplant waiting list [5, 33], controversy remains regarding whether device implantation as a BTT affects posttransplantation outcomes [34, 35]. BTT LVAD therapy is likely more cost-effective in patients at high to medium risk, with an expected long waiting time before transplantation. The excellent results achieved by CF-LVADs , the increased mortality rate for heart transplantation caused by the more liberal donor criteria, and the considerable mortality on the waiting list of nonsupported patients have led to the hypothesis that LVAD implantation should be viewed as the primary treatment of stage D heart failure, followed by heart transplantation only in eligible patients [36, 37].

With the extremely limited pool of cardiac allografts, CF-LVADs are becoming a more attractive and readily available method of support for stage D heart failure patients. Only after the development of the HM-XVE has the practice pattern begun to evolve toward DT . Although it showed improved survival, morbidity related to LVAD implantation was not insignificant. Thus, expansion of DT only began after the approval of the HM II LVAD by the FDA, and, since 2012, the number of DT implants has surpassed the number of BTT implants. In 2014, 45.7 % of implants were designated as DT [27]—interestingly, surpassing the number of heart transplants, as well.

The community of advanced heart failure specialists gain experience evaluating and treating patients, but determining which patient and the timing of device implantation continues to evolve. Indications for DT include chronic inotrope dependency, optimal medication non-tolerance, end-organ hypoperfusion, frequent rehospitalization, frailty, and exercise cardiopulmonary stress testing with VO₂ of 14 mL/kg/min or <50 % age-predicted maximum. The decision to implant a CF-LVAD as DT can be challenging, and the long-term complications associated with VAD therapy should be contrasted with current signs and symptoms of progressive heart failure treated with medical therapy alone [29]. Again, INTERMACS profiling could guide the decision-making process. For example, the benefits of CF-LVAD therapy outweigh the risks in INTERMACS profiles 1–3. In contrast, implanting a device in INTERMACS profiles 4–6 predicts greater survival after LVAD implantation compared with INTERMACS 1–3, but carries with it the adverse event burden in a patient whose mortality is less acute (◘ Fig. 22.4).

The results of the HeartMate II post-approval study for DT patients showed 1-year survival of 82 % for patients with INTERMACS levels 4–7 versus 72 % for levels 1–3 [38]. These survival rates were significantly lower than the 88 % 1-year survival for the BTT patients, but the difference may result from the younger age of the BTT patients and their fewer number of comorbidities [39]. Currently, 80 % of the approved device implants as BTT or DT are for patients in INTERMACS levels 1–3. The benefit of device implantation in more compensated INTERMACS profiles (4–7) remains unclear. This question is being evaluated in the REVIVE-IT (Randomized Evaluation of VAD InterVEntion before Inotropic Therapy) trial. This study is a National Heart, Lung, and Blood Institute (NHLBI)-sponsored prospective, randomized trial for evaluating HM II as DT in New York Heart Association (NYHA) functional class III heart failure patients [40]. The goal of this ongoing study is to determine the clinical efficacy and cost of CF-LVAD implantation as DT in a less-sick patient population, given the challenging decision of LVAD implantation as DT.

In general, a number of clinical risk factors portend poor survival and might exclude a patient from LVAD candidacy. These risk factors include advanced age, previous cardiac surgery, renal failure, and right ventricular failure. Elderly patients (>65 years of age) tolerate CF-LVAD implantation poorly during acute decompensation. Nonetheless, outcomes in this population are generally favorable over a long period. Most importantly, the decision to implant a DT CF-LVAD, as opposed to pursuing ongoing medical therapy, implies that the patient is expected to make a successful postimplant recovery, is at a point in disease progression at which end-organ damage is reversible, is expected to benefit from a reduction in mortality, and, importantly, has an improved quality of life .

As clinical experience expands and patients live longer with VADs, more complications are being encountered, and management methods continue to evolve. The most common complications of VAD therapy include gastrointestinal (GI) bleeding, pump thrombosis, stroke, aortic valve insufficiency, and right ventricular failure.

The mechanisms of GI bleeding, which occurs in up to 40 % of cases [41], include acquired von Willebrand factor deficiency, mucosal arteriovenous malformations (AVMs ), and chronic anticoagulation therapy [42]. Decreased or absent pulsatility with increased shear stress leads to subsequent angiodysplasia and the development of AVMs. The fragility of these capillaries leads to bleeding, which is exacerbated with anticoagulation [43]. Unfortunately, treatment options remain limited and predominantly include cessation of anticoagulation and antiplatelet therapy, endoscopy with control of visualized bleeding sources, and off-label medication use including subcutaneous octreotide injections [44] and oral thalidomide [45]. Effective use of these agents is being extrapolated from small retrospective studies; randomized studies are being undertaken.

Another major complication is pump thrombosis (◘ Fig. 22.5). Proper inflow and outflow cannula positioning, anticoagulation, and antiplatelet therapies are needed to prevent thrombus formation within the pump. Delayed initiation, inadequate dosing, or cessation of anticoagulation therapy may be associated with thrombus formation within the pump, leading to pump failure. The initial HM II trials considered an international normalized ratio (INR) of 2–3 and a full-dose 325 mg aspirin (ASA) as standard therapy. Low thrombotic and high bleeding rates led to modification of these standards; an INR of 1.8–2.2 and an ASA dose of 81 mg were proposed. However, the recent increase in the incidence of thrombosis [46] led to new guidelines [47]. All risk factors remain unclear, but include driveline infection, obesity, younger age, and female sex [48]. Increased inflow cannula angulation and increased depth of the pump pocket have also been contemplated as risk factors of pump thrombosis [49]. Work-up includes measuring the serum levels of biomarkers of hemolysis such as lactate dehydrogenase and plasma-free hemoglobin. Newer ramp echocardiographic studies have been used with the notion that high speeds do not decompress the left ventricle as well in the presence of thrombosis [50]. Management includes intravenous unfractionated heparin or argatroban infusion, thrombolytic therapy, intensified chronic anticoagulation, and pump exchange via the subcostal/substernal approaches [51].

Self-administered hardware maintenance remains the cornerstone of device longevity. Patients with CF-LVADs are tethered by a driveline that emerges from the anterior abdominal wall and is connected to the portable controller (◘ Fig. 22.6). Care of the driveline is critical to avoid infection that can have devastating consequences. Regimens of weekly or biweekly sterile dressing changes are required [52]. With the onset of a driveline infection, an aggressive treatment plan is implemented to avoid the need for pump exchange or urgent transplantation. Oral or parenteral antibiotics, surgical debridement, and pump exchange or explantation might be eventually needed [53].

The CF-LVAD is designed to be implanted at the apex (inflow cannula), with the outflow cannula anastomosed to the ascending aorta. Flow from this cannula is antegrade with some retrograde flow to the aortic valve, leading to increased pressure in the dependent sinuses of Valsalva. This increased pressure likely leads to leaflet fusion and prolapse, culminating in the development of late aortic insufficiency. This is primarily due to complete unloading of the ventricle when the pump speed is high enough, such that there is little or no ejection through the native aortic valve. Incidence of late de novo insufficiency has been estimated to be as high as 43.1 % in HM II vs. 65.7 % in HW recipients [54]. Non-opening of the aortic valve was strongly associated with de novo regurgitation development [55]. Treatment options include decreasing the flow through the device, afterload reduction, and valvular replacement. Surgical techniques include complete valve closure, repair, or complete replacement with tissue prosthesis [56, 57].

Early right ventricular (RV) failure and progressive decline of RV function are common complications of CF-LVAD implantation. With an incidence of up to 44 % [58], RV failure has been associated with higher mortality and morbidity [59]. It may lead to impaired LVAD flow, difficulty in weaning from cardiopulmonary bypass, decreased tissue perfusion, and, ultimately, multi-organ failure and death. Thus, identifying CF-LVAD patients at risk for RV failure postoperatively is crucial and most often unclear.