Kevin G. Soucy

Rotary pumps and diminished pulsatility: do we need a pulse?

Ventricular assist devices (VADs) have been successfully used as a bridge to heart transplant and destination therapy (DT) for congestive heart failure (HF) patients. Recently, continuous flow VAD (CVAD) has emerged as an attractive clinical option for long-term mechanical support of HF patients, with bridge-to-transplant outcomes comparable with pulsatile flow VAD (PVAD). Continuous flow VADs are smaller, more reliable, and less complex than the first-generation PVAD. Despite the widespread clinical use, CVAD support has been associated with gastrointestinal bleeding, hemorrhagic strokes, and aortic valve insufficiency. Speculation that diminished arterial pressure pulsatility associated with continuous flow devices may be contributing to these complications has sparked much debate over CVAD support. Studies comparing pulsatile flow and continuous flow (CF) support have presented conflicting findings, and the relevance to CVAD as DT is uncertain due to variations in device operation, support duration, and the criteria used to quantify pulsatility. Currently, there is interest in developing control algorithms for CVAD to increase the delivered pulsatility as a strategy to mitigate adverse event risks associated with CVAD therapy. There may also be the added benefit of specific control strategies for managing CVAD therapy, potentially improving the rate of myocardial recovery and successful weaning of mechanical circulatory support.

Rotary pump speed modulation for generating pulsatile flow and phasic left ventricular volume unloading in a bovine model of chronic ischemic heart failure

BACKGROUND Rotary blood pumps operate at a constant speed (rpm) that diminishes vascular pulsatility and variation in ventricular end-systolic and end-diastolic volumes, which may contribute to adverse events, including aortic insufficiency and gastrointestinal bleeding. In this study, pump speed modulation algorithms for generating pulsatility and variation in ventricular end-systolic and end-diastolic volumes were investigated in an ischemic heart failure (IHF) bovine model (n = 10) using a clinically implanted centrifugal-flow left ventricular assist device (LVAD). METHODS Hemodynamic and hematologic measurements were recorded during IHF baseline, constant pumps speeds, and asynchronous (19-60 cycles/min) and synchronous (copulse and counterpulse) pump speed modulation profiles using low relative pulse speed (±25%) of 3,200 ± 800 rpm and high relative pulse speed (±38%) of 2,900 ± 1,100 rpm. End-organ perfusion, hemodynamics, and pump parameters were measured to characterize pulsatility, myocardial workload, and LVAD performance for each speed modulation profile. RESULTS Speed modulation profiles augmented aortic pulse pressure, surplus hemodynamic energy, and end-organ perfusion (p < 0.01) compared with operation at constant speed. Left ventricular external work and myocardial oxygen consumption were significantly reduced compared with IHF baseline (p < 0.01) but at the expense of higher LVAD power consumption. CONCLUSIONS Pump speed modulation increases pulsatility and improves cardiac function and end-organ perfusion, but the asynchronous mode provides the technologic advantage of sensorless control. Investigation of asynchronous pump speed modulation during long-term support is warranted to test the hypothesis that operating an LVAD with speed modulation will minimize adverse events in patients supported by an LVAD that may be associated with long-term operation at a constant pump speed.

Defining pulsatility during continuous-flow ventricular assist device support

Continuous-flow ventricular assist devices (CVADs) have gained widespread use as an effective clinical therapy for patients with advanced-stage heart failure. Axial and centrifugal CVADs have been successfully used as bridge-to-transplant and destination therapy. CVADs are smaller, more reliable, and less complex than the first-generation pulsatile-flow ventricular assist devices. Despite their recent clinical success, arteriovenous malformations, gastrointestinal bleeding, hemorrhagic strokes, aortic valve insufficiency, and valve fusion have been reported in heart failure patients supported by CVADs. It has been hypothesized that diminished arterial pressure and flow pulsatility delivered by CVAD may be a contributing factor to these adverse events. Subsequently, the clinical significance of vascular pulsatility continues to be highly debated. Studies comparing pulsatile-flow and continuous-flow support have presented conflicting findings, largely due to variations in device operation, support duration, and the criteria used to quantify pulsatility. Traditional measurements of pulse pressure and pulsatility index are less effective at quantifying pulsatility for mechanically derived flows, particularly with the growing trend of CVAD speed modulation to achieve various pulsatile flow patterns. Kinetic measurements of energy equivalent pressure and surplus hemodynamic energy can better quantify pulsatile energies, yet technologic and conceptual challenges are impeding their clinical adaption. A review of methods for quantifying vascular pulsatility and their application as a research tool for investigating physiologic responses to CVAD support are presented.

HZE 56Fe-ion irradiation induces endothelial dysfunction in rat aorta: Role of xanthine oxidase

Ionizing radiation has been implicated in the development of significant cardiovascular complications. Since radiation exposure is associated with space exploration, astronauts are potentially at increased risk of accelerated cardiovascular disease. This study investigated the effect of high atomic number, high-energy (HZE) iron-ion radiation on vascular and endothelial function as a model of space radiation. Rats were exposed to a single whole-body dose of iron-ion radiation at doses of 0, 0.5 or 1 Gy. In vivo aortic stiffness and ex vivo aortic tension responses were measured 6 and 8 months after exposure as indicators of chronic vascular injury. Rats exposed to 1 Gy iron ions demonstrated significantly increased aortic stiffness, as measured by pulse wave velocity. Aortic rings from irradiated rats exhibited impaired endothelial-dependent relaxation consistent with endothelial dysfunction. Acute xanthine oxidase (XO) inhibition or reactive oxygen species (ROS) scavenging restored endothelial-dependent responses to normal. In addition, XO activity was significantly elevated in rat aorta 4 months after whole-body irradiation. Furthermore, XO inhibition, initiated immediately after radiation exposure and continued until euthanasia, completely inhibited radiation-dependent XO activation. ROS production was elevated after 1 Gy irradiation while production of nitric oxide (NO) was significantly impaired. XO inhibition restored NO and ROS production. Finally, dietary XO inhibition preserved normal endothelial function and vascular stiffness after radiation exposure. These results demonstrate that radiation induced XO-dependent ROS production and nitroso-redox imbalance, leading to chronic vascular dysfunction. As a result, XO is a potential target for radioprotection. Enhancing the understanding of vascular radiation injury could lead to the development of effective methods to ameliorate radiation-induced vascular damage.

Left ventricular volume unloading with axial and centrifugal rotary blood pumps.

Axial (AX) and centrifugal (CFG) rotary blood pumps have gained clinical acceptance for the treatment of advanced heart failure. Differences between AX and CFG designs and mechanism of blood flow delivery may offer clinical advantages. In this study, pump characteristics, and acute physiologic responses during support with AX (HeartMate II) and CFG (HVAD) left ventricular assist devices (LVAD) were investigated in mock loop and chronic ischemic heart failure bovine models. In the mock loop model, pump performance was characterized over a range of pump speeds (HeartMate II: 7,000–11,000 rpm, HVAD: 2,000–3,600 rpm) and fluid viscosities (2.7 cP, 3.2 cP, 3.7 cP). In the ischemic heart failure bovine model, hemodynamics, echocardiography, and end-organ perfusion were investigated. CFG LVAD had a flatter HQ curve, required less power, and had a more linear flow estimation relation than AX LVAD. The flow estimation error for the AX LVAD (±0.9 L/min at 2.7 cP, ±0.7 L/min at 3.2 cP, ±0.8 L/min at 3.7 cP) was higher than the CFG LVAD (±0.5 L/min at 2.7 cP, ±0.2 L/min at 3.2 cP, ±0.5 L/min at 3.7 cP). No differences in acute hemodynamics, echocardiography, or end-organ perfusion between AX and CFG LVAD over a wide range of support were statistically discernible. These findings suggest no pronounced acute differences in LV volume unloading between AX and CFG LVAD.

Early feasibility testing and engineering development of the transapical approach for the HeartWare MVAD ventricular assist system.

Implantation of ventricular assist devices (VADs) for the treatment of end-stage heart failure (HF) falls decidedly short of clinical demand, which exceeds 100,000 HF patients per year. Ventricular assist device implantation often requires major surgical intervention with associated risk of adverse events and long recovery periods. To address these limitations, HeartWare, Inc. has developed a platform of miniature ventricular devices with progressively reduced surgical invasiveness and innovative patient peripherals. One surgical implant concept is a transapical version of the miniaturized left ventricular assist device (MVAD). The HeartWare MVAD Pump is a small, continuous-flow, full-support device that has a displacement volume of 22 ml. A new cannula configuration has been developed for transapical implantation, where the outflow cannula is positioned across the aortic valve. The two primary objectives for this feasibility study were to evaluate anatomic fit and surgical approach and efficacy of the transapical MVAD configuration. Anatomic fit and surgical approach were demonstrated using human cadavers (n = 4). Efficacy was demonstrated in acute (n = 2) and chronic (n = 1) bovine model experiments and assessed by improvements in hemodynamics, biocompatibility, flow dynamics, and histopathology. Potential advantages of the MVAD Pump include flow support in the same direction as the native ventricle, elimination of cardiopulmonary bypass, and minimally invasive implantation.

Development of an extracellular matrix delivery system for effective intramyocardial injection in ischemic tissue.

Biomaterials with direct intramyocardial injection devices have been developed and are being investigated as a potential cardiac regenerative therapy for end-stage ischemic heart failure. Decellularized extracellular matrix (ECM) has been shown to improve cardiac function and attenuate or reverse pathologic remodeling cascades. CorMatrix Cardiovascular, Inc. has developed a porcine small intestinal submucosa-derived particulate extracellular matrix (P-ECM) and ECM Delivery System to provide uniform and controlled intramyocardial delivery of the injectable P-ECM material into infarcted regions. The CorMatrix ECM Delivery System is composed of a Multi-Needle P-ECM Syringe Assembly, Automated Injection Controller, and Tissue Depth Measurement System (portable ultrasound). Feasibility of the P-ECM delivery system was tested intraoperatively in a chronic ischemic heart failure bovine model (n = 11), and demonstrated the ability to control injection volume (0.1–1.0 ml) and depth of penetration (3–5 mm) under regulated injection pressure (150 psi CO2) into the ischemic region. Targeted intramyocardial delivery of P-ECM may improve efficacy and enable development of novel patient-specific therapy.

Hemodynamic changes and retrograde flow in LVAD failure.

In the event of left ventricular assist device (LVAD) failure, we hypothesized that rotary blood pumps will experience significant retrograde flow and induce adverse physiologic responses. Catastrophic LVAD failure was investigated in computer simulation with pulsatile, axial, and centrifugal LVAD, mock flow loop with pulsatile (PVAD) and centrifugal (ROTAFLOW), and healthy and chronic ischemic heart failure bovine models with pulsatile (PVAD), axial (HeartMate II), and centrifugal (HVAD) pumps. Simulated conditions were LVAD “off” with outflow graft clamped (baseline), LVAD “off” with outflow graft unclamped (LVAD failure), and LVAD “on” (5 L/min). Hemodynamics (aortic and ventricular blood pressures, LVAD flow, and left ventricular volume), echocardiography (cardiac volumes), and end-organ perfusion (regional blood flow microspheres) were measured and analyzed. Retrograde flow was observed with axial and centrifugal rotary pumps during LVAD failure in computer simulation (axial = -3.4 L/min, centrifugal = -2.8 L/min), mock circulation (pulsatile = -0.1 L/min, centrifugal = -2.7 L/min), healthy (pulsatile = -1.2 ± 0.3 L/min, axial = -2.2 ± 0.2 L/min, centrifugal = -1.9 ± 0.3 L/min), and ischemic heart failure (centrifugal = 2.2 ± 0.7 L/min) bovine models for all test conditions (p < 0.05). Differences between axial and centrifugal LVAD were statistically indiscernible. Retrograde flow increased ventricular end-systolic and end-diastolic volumes and workload, and decreased myocardial and end-organ perfusion during LVAD failure compared with baseline, LVAD support, and pulsatile LVAD failure.

Feasibility study of particulate extracellular matrix (P-ECM) and left ventricular assist device (HVAD) therapy in chronic ischemic heart failure bovine model.

Myocardial recovery with left ventricular assist device (LVAD) support is uncommon and unpredictable. We tested the hypothesis that injectable particulate extracellular matrix (P-ECM) with LVAD support promotes cell proliferation and improves cardiac function. LVAD, P-ECM, and P-ECM + LVAD therapies were investigated in chronic ischemic heart failure (IHF) calves induced using coronary embolization. Particulate extracellular matrix emulsion (CorMatrix, Roswell, GA) was injected intramyocardially using a 7 needle pneumatic delivery tool. Left ventricular assist devices (HVAD, HeartWare) were implanted in a left ventricle (LV) apex to proximal descending aorta configuration. Cell proliferation was identified using BrdU (5 mg/kg) injections over the last 45 treatment days. Echocardiography was performed weekly. End-organ regional blood flow (RBF) was quantified at study endpoints using fluorescently labeled microspheres. Before treatment, IHF calves had an ejection fraction (EF) of 33 ± 2% and left ventricular end-diastolic volume of 214 ± 18 ml with cardiac cachexia (0.69 ± 0.06 kg/day). Healthy weight gain was restored in all groups (0.89 ± 0.03 kg/day). EF increased with P-ECM + HVAD from 36 ± 5% to 75 ± 2%, HVAD 38 ± 4% to 58 ± 5%, and P-ECM 27 ± 1% to 66 ± 6%. P-ECM + HVAD demonstrated the largest increase in cell proliferation and end-organ RBF. This study demonstrates the feasibility of combined LVAD support with P-ECM injection to stimulate new cell proliferation and improve cardiac function, which warrants further investigation.

Fault detection in rotary blood pumps using motor speed response.

Clinical acceptance of ventricular assist devices (VADs) as long-term heart failure therapy requires safe and effective circulatory support for a minimum of 5 years. Yet, VAD failure beyond 2 years of support is still a concern. Currently, device controllers cannot consistently predict VAD failure modes, and undetected VAD faults may lead to catastrophic device failure. To minimize this risk, a model-based algorithm for reliable VAD fault detection that only requires VAD revolutions per minute (rpm) was developed. The algorithm was tested using computer models of the human cardiovascular system simulating heart failure and axial flow (AF) or centrifugal flow (CF) VADs. Ventricular assist device rpm was monitored after a step down of motor current for normal and simulated fault conditions (>750 faults). The ability to detect fault conditions with 1%, 5%, and 10% rpm measurement noise was evaluated. All failure modes affected the VAD rpm responses to the motor current step down. Fault detection rates were >95% for AF and >89% for CF VADs, even with 10% rpm measurement noise. The VAD rpm responses were significantly altered by blood viscosity (3.5–6.2 cP), which should be accounted for in clinical application. The proposed VAD fault detection algorithm may deliver a convenient and nonintrusive way to minimize catastrophic device failures.