HeartWare (HeartWare International, Framingham, MA) has in preclinical testing a next-generation device, the MVAD, a wide-blade axial CF pump (Fig. 26.4). The MVAD is approximately half the size of the HVAD, with 22 mL of displacement volume and 5 mL of priming volume. The MVAD is designed for versatility: it can be used as partial or full support for left, right, or biventricular applications, via either sternotomy or thoracotomy. The MVAD features a magnetic axial rotor and uses a combination of passive permanent magnets and hydrodynamic bearing surfaces to totally suspend the rotor in operation, with a displacement volume of only 15 mL. This is accomplished by a unique rotor with impeller blades that are wider than the blood flow channel between them, affording sufficient surface area for passive hydrodynamic radial rotor suspension. This design eliminates the mechanical pivot bearings typical of axial pumps and significantly reduces pump size by using passive rotor suspension rather than the larger and more complex active electromagnetic bearing designs. Also not typical of axial pump design is an outflow port, angled 90° from the pump housing inlet port, similar to centrifugal pump designs [26]. Various versions of the MVAD are being tested, implanted within the LV or at the LV apex, in animal studies (Fig. 26.5). The MVAD is currently not available for clinical application, but a clinical trial is planned soon for this investigational device.

The HeartMate III (Thoratec Corp., Pleasanton, CA) centrifugal CF rotary pump (Fig. 26.6) is now in preparation for clinical testing and Conformité Européenne mark clinical trials (expected to start by the last quarter of 2014). The pump incorporates many critical family design elements of the HeartMate II (e.g., large gaps and textured blood-contacting surfaces) and has been designed to reduce the potential for adverse events (aortic insufficiency, bleeding, thrombus, and stroke). Its large gaps also enable rapid speed changes used by the artificial pulse feature without rotor and/or housing contact. The HeartMate III addresses many of the limitations of axial pump design: it features a bearingless magnetic levitation of the pump rotor (rather than mechanical pivot bearings) to increase reliability out to 10 years, has an accurate sensorless flow estimator, and incorporates a pulse mode to increase the level of pulsatility and even produce physiological pulse pressures. This new pump may potentially reduce the need for anticoagulation agents. Wedging surfaces and other features required for hydrodynamic bearings are not required. The device’s rotor and volute have been designed to minimize shear and avoid stasis over the entire range of operation (2–10 L/min). Rotor levitation is independent of rotor speed; levitation is maintained at any rotor speed, even zero. Rotor speed independence permits flexibility in operating speeds, which could in the future enable use of the HeartMate III under low-speed conditions, e.g., pulmonary circulation and weaning protocols. These features make it feasible to implement physiologic control algorithms based on accurate flow estimates. In addition, centrifugal pumps have characteristics that generate lower pressures over a wider flow range. This capability translates to more favorable outcomes owing to much lower inlet suction at low flows; lower outlet pressures at high levels of systemic vascular resistance, preventing the creation of exceedingly high systemic arterial pressures; higher flow pulsatility that is more sensitive to LV pressure swings, providing both a more sensitive pre-suction detection capability and increased diagnostic feedback about the functional status of the LV; and safer chronic management of pump speed and output optimization. These features suggest that centrifugal rotary pumps may overtake axial-flow rotary pumps in the CF rotary pump revolution that has made first-generation pulsatile pumps all but obsolete [7].

The HeartMate X (Thoratec Corp., Pleasanton, CA), a highly energy-efficient axial-flow pump, leverages core HeartMate II bearing technology with a dramatic size reduction for rapid, less invasive implantation. The HeartMate X offers versatile cannulation options and, in this ultra-compact dimension, is considered suitable for partial to full ventricular support configurations for patients needing LVADs as well as the populations needing right or bi-ventricular assist devices (RVADs or BiVADs) [27]. The device can provide a stable 8 L/min of CF. The HeartMate X may thus further expand potential approaches for truly minimally invasive implantation techniques, especially in small size and younger patient populations.

In 2013, HeartWare International announced that it had acquired CircuLite^®, Inc. (Saddle Brook, NJ), developer of the SYNERGY^® Circulatory Support System, a “micro-pump” designed to treat less sick, ambulatory patients with chronic HF who are not yet inotrope dependent. The CircuLite Synergy LVAD (HeartWare) is a novel pump that combines axial, centrifugal, and orthogonal flow paths at a 7.5 mL volume and operating speeds of 20,000–28,000 rpm. The device is the size of a double-A battery, connects to the heart via an inflow cannula (8 mm diameter by 20 cm long), has a pumping capacity of up to 4.25 L/min, and is small enough to be placed in a subcutaneous pocket. The blood passes through narrow channels around the motor (which serves to cool the motor) and leaves the pump at a right angle to the inlet [28]. It uses a magnetically and hydrodynamically stabilized rotor design. The Synergy’s pump body is implanted via a small right-sided thoracotomy into a subcutaneous pacemaker-like pocket in the right subclavian groove. It is implanted without the use of extracorporeal circulation, with the inflow cannula placed into the left atrium and the outflow graft connected to the right subclavian artery. The device has been implanted in 25 patients on the waiting list for a heart transplant as a partial support device. With flows recorded between 2.3 and 3.5 L/min, the applications of the pump may be extended for support in less sick patients (those considered in New York Heart Association functional class III) [29].

Although clinical experience is currently limited to pumping blood from the left atrium to the subclavian artery, recent cadaveric research has indicated that this micropump can also be easily positioned as an RVAD.

The SYNERGY^® Circulatory Support System is currently undergoing a redesigning process related to such issues as device thrombosis and mechanical fractures of the inflow cannula portion. The additional device development and optimization will be critical to provide proper pump reliability and function. However, the US clinical trial that has been put on the company’s agenda will most likely be further delayed.

The application of CF LVADs for refractory end-stage HF is increasing. When LV filling remains inadequate despite pharmacotherapy to enhance RV function and lower pulmonary vascular resistance, an RVAD may be required [20, 30]. The primary reason VADs have not been designed specifically for RVAD applications is that there is much less clinical need (and therefore a limited market) compared with that for LVADs [7, 30]. Although clinically approved implantable LVADs are not designed for right-sided support, LVAD implantation for right heart support is being considered simply because an effective, easily implantable RVAD is not yet available. The absence of a dedicated, clinically evaluated, long-term RVAD significantly diminishes the treatment options for patients with isolated and post-LVAD RV failure, as well as for those with biventricular HF requiring two pumps (BiVADs) [31–33]. The inherent CF pump mechanics and translation of clinical features, pump selection, and patient management issues have previously been detailed [26].

The first use of an implantable CF pump for RV support was reported in 2004; a 21-year-old patient with worsened RV failure and elevated peripheral vascular resistance after implantation of a Jarvik 2000 FlowMaker was treated by implanting a second one [20]. By directing the inlet toward the inferior vena caval-right atrial junction and keeping right atrial pressures at 5–10 cm H₂O, surgeons achieved stable RV unloading and satisfactory blood flow to the left-sided pump. Other groups consecutively reported on limited, off-label, clinical use of two LVADs. Saito et al. [34] converted a 6-month survival patient from a Toyobo paracorporeal device (Toyobo-Nipro, Osaka, Japan) to a Jarvik 2000 as RVAD and a DuraHeart (Terumo Heart, Inc., Ann Arbor, MI) for LVAD support.

Successful application of two HVADs for treatment of severe biventricular failure was reported by Strueber et al. [15] and Hetzer and collaborators [35]. Additional case reports of intrapericardial implantation of the HVAD followed, the surgeons using a similar technique of placing the pump on the RV diaphragmatic surface just below the acute margin. Two additional silicone rings were placed under the fixation ring. Artificial restriction applied to the outlet graft of the pump may compensate for the lower outlet pressures seen at the pulmonary artery [36]. HeartWare is currently working to obtain regulatory approval for modifications to the HVAD for use as RVAD support [7]. Potapov et al. [37] from the German Heart Institute in Berlin reported on ten patients with idiopathic dilated cardiomyopathy patients who received biventricular HVADs. In three of these patients, physicians have been able to turn the RVAD off following long-term biventricular support, having confirmed via echocardiography that the patients had stable hemodynamics and that no thromboembolic events and no regurgitation through the right pump had occurred. Krabatsch et al. [38] reported implantation of an HVAD, first on the RV free wall at the maximal distance from the interventricular septum. Because severe compression of the RV caused low flow of the right pump of maximally 2 L/min, the RVAD was reimplanted to the right atrium through the layer of pericardium (opening created from the right pleura) and by using two silicone rings (to prevent the inflow cannula being introduced too deeply into the chamber). That group subsequently reported 17 patients implanted with two HVADs [16]. Deuse at al. [39] recently reported the HVAD use for permanent RV support with no spacers (rings) used.

Recently, Schmitto et al. [28] investigated the use of two Synergy Micro-Pumps (CircuLite Inc., Saddle Brook, NJ) for full biventricular support in ovine models. Such use may provide another option for partial (biventricular) support in patients and, as the authors stated, for full support in small adults and children.

Currently, temporary support via RVADs/BiVADs is available by CF extracorporeal or paracorporeal circulation support systems [30, 33] such as the CentriMag^® (Thoratec Corp., Pleasanton, CA), the Bio-Pump (Medtronic, Minneapolis, MN), and the TandemHeart (CardiacAssist, Inc., Pittsburgh, PA). Progress in the development of clinically available rotary pumps now permits physicians to choose a hybrid approach for BiVAD placement, using both an implantable LVAD and temporary extracorporeal RVAD [33, 40, 41]. Since the first reported application of two Jarvik 2000 FlowMaker pumps [20], several groups have adopted this approach for biventricular assist. The adaptation of an LVAD to function as an RVAD is a technical challenge because of both ventricular distinctions and pump characteristics [30, 42].

RVAD design challenges involve pump dimensions and fit, as well as adaptability to operating flows. RVADs operate at much lower afterloads (pulmonary arterial pressures) than LVADs do and at significantly reduced hydraulic loads and power consumption rates. Because of its low-pressure working conditions and complex geometry, the RV stands in stark contrast to the LV. Under normal baseline conditions, the unique anatomy, myocardial ultrastructure, and coronary physiology of the RV reflect a high-volume, low-pressure pump. Once surgeons introduce an LVAD, they raise the expected output of the RV to that of the LVAD [43].