Search

In 1974, between 2 and 8% of the 50,000 adult patients undergoing cardiac surgery in this country succumbed in the early post-operative period from left ventricular failure, despite various methods of pharmacologic and/or mechanical support. Our laboratories have concentrated on the development, modification, evaluation, and validation of an abdominally positioned left ventricular assist device which has the potential of reducing these mortalities. Continuous testing in animals, for periods exceeding 2 mos, satisfied reliability, durability and longevity requirements. The cumulative results of these investigations were reviewed at the National Heart and Lung Institute on August 21, 1975. Authorization for clinical trials of the device according to specific criteria and protocols46 was approved on November 1, 1975. The ALVAD is now in the early stages of clinical testing. The results of the current experiments demonstrate that ventricular outflow impedance and prosthetic inflow impedance are the major determinants of left ventricular assist device hemodynamic effectiveness. By markedly reducing outflow impedance, the ALVAD profoundly lowers ventricular pressure-work and oxygen demands while simultaneously increasing ventricular performance and maintaining or augmenting systemic perfusion. Moreover, our studies indicate that improved device designs (intended for intermediate and long-term implantation) and maximal performance can be achieved by focusing on these central determinants.

The experimental production of stable, controlled, short-term left ventricular failure is valuable in the evaluation of implantable circulatory support systems. Acute or chronic left ventricular failure produced by occlusion or embolization of coronary arteries results in muscle dysfunction and degrees of failure that may be difficult to control. The effects of varying amounts of intracoronary lidocaine were studied during short- and long-term evaluations of intracorporeal left ventricular assist pumping. In 8 Hereford calves the left main coronary artery was cannulated with an intracoronary catheter in open and closed chest preparations. Dose-related negative inotropic effects were noted when lidocaine was injected at individual doses of 50, 75, and 100 mg. Following 100 mg doses, mean aortic pressure, cardiac output, and maximum rate of rise of left ventricular pressure decreased; left ventricular end-diastolic pressure increased fourfold. Similar effects were noted with short continuous infusions of lidocaine. The initial responses to injection or continuous infusions, if effective, were noted within 40 to 60 seconds. Several episodes of failure could be produced with either method following recovery periods of 10 to 15 minutes. In all instances, actuation of a left ventricular assist device immediately reversed the hemodynamic effects of the pharmacologically induced failure.

1) The implanted TAH offers a method of evaluating the effects of varying physiologic demands cardiac output and venous return. 2) Cardiac output and venous return measurements offer a method of evaluating system design of the device and driving and control-logic improvements in awake, unanesthetized animals. 3) The obtained cardiac function and venous return curves suggest that the present TAH is less responsive to increases in right atrial pressure than the natural heart, but is still completely controlled by venous return. Therefore, it is extremely important to make every effort to eliminate any factor causing excessive venous return.

We have initiated clinical trials with an intracorporeal (abdominal) partial artificial heart and ten preterminal postcardiotomy patients have been studied. During profound left ventricular failure, the device captures the entire cardiac output from the apex of the left ventricle at low pressures (20 to 40 mm Hg) and ejects (at 80 to 150 mm Hg) into the infrarenal abdominal aorta; the biological aortic valve opens only intermittently and the entire systemic circulation is pump generated. The device is six to ten times more effective than intra-aortic balloon pumping in man and has maintained systemic perfusion during clinical asystole and ventricular fibrillation. We have documented that the profoundly depressed postcardiotomy left ventricle, initially incapable of ejection, can recover during total left ventricular unloading with the abdominal left ventricular assist device support over a seven-day period.

An abdominal left ventricular assist device (ALVAD) is undergoing controlled clinical trials in our institution. The ALVAD is pneumatically-actuated, synchronously or asynchronously with an external console and is interposed between the apex of the left ventricle and the infrarenal abdominal aorta. It is an order of magnitude more effective than conventional intraaortic balloon pumping. Thus far, we have implanted this pump in 21 patients (15 males and six females). The average age has been 50. The duration of cardiopulmonary bypass with intensive pharmacologic support and IABP until ALVAD implantation has been nearly 4 hours. The plasma hemoglobins prior to ALVAD implantation have averaged 168 mg%. The platelet counts at implantation have averaged 68,000 mm(3). The average duration of ALVAD support has been in excess of one day and the longest trial extended for one week. We have been able to remove the pump after ventricular recovery in two instances and effected cardiac allografting in one instance of ALVAD dependency. We have found that (1) the profoundly depressed left (and right) ventricles can recover if totally supported with this device; (2) the device can function in the presence of ventricular fibrillation and/or standstill; (3) the device can effectively replace both left and right ventricular function in the presence of normal pulmonary vascular resistance and microcirculatory hemodynamics; and (4) in the presence of impending multiple organ failure, procrastination in use is to be avoided.

From January 1984 to September 1988, 245 heart transplants were performed at our institution. During this period, 53 staged cardiac transplant procedures were performed using an intraaortic balloon pump (IABP), an extracorporeal centrifugal blood pump (ECBP), an intracorporal left ventricular assist device, and a total artificial heart. Thirty-four (64%) patients underwent subsequent cardiac transplantation and 28 (53%) were discharged. Our data indicate that patients who suffer rapid onset of heart failure and undergo early circulatory support are suitable candidates for successful staged cardiac transplantation. We have also used mechanical support to treat severe heart failure after transplantation, using IABP and ECBP support, as well as an intraarterial axial flow pump. In immunosuppressed patients, infection is the critical issue affecting outcome. There is a need for development of smaller, simpler, less invasive devices with the capacity to provide both systemic and pulmonary circulatory support.

To determine anatomic parameters for a permanent, electrically actuated left ventricular assist device (LVAD), the effects of abdominal placement of pneumatic LVADs used as temporary support for patients awaiting heart transplantation was studied. Understanding the anatomic constraints imposed by the abdominal viscera in LVAD placement is crucial, because improper placement can result in compression or obstruction of adjacent structures. Anatomic compatibility was assessed in four men (age 22-48 years) who were supported by the LVAD for over 1 month (range 35-132 days). The pump was intraperitoneally placed in the left upper quadrant. Radiographic techniques were employed, including CT scanning (with patients supine) and contrast imaging (patients in anatomical position), and the pump and conduits appeared to be properly positioned, with minimal compression of the body of the stomach, and no obstruction of adjacent organs. Three patients returned to a solid food diet and exercised daily by stationary cycling and walking. No signs of migration or erosion of the pump were present at the time of LVAD removal and cardiac transplantation. Successful clinical experience with short-term use of the LVAD suggests that the electrically actuated device can be well tolerated in patients requiring permanent left ventricular assistance.

A prosthetic vascular graft should be nonthrombogenic, compliant with anastomosed vessels and resist excessive dilatation or aneurysm formation. Mechanical performance of a negatively charged, glutaraldehyde-tanned bovine heterograft was evaluated, in vitro, during dynamic loading conditions. The grafts were 16 cm in length with internal diameters of 6 mm (n = 10), 8 mm (n = 12), and 10 mm (n = 8). The prostheses were pneumatically cycled, using intragraft balloons, at a pulse pressure of 120/70 mmHg and a rate of 80 beats/min in 37 degrees C saline for 2.6 million cycles (22.6 days). At selected intervals, the diastolic and pulse volumes of the grafts were measured and compliance calculated. These studies demonstrated that this vascular prosthesis had a high initial compliance, independent of graft diameter. During dynamic loading, diastolic volume increased whereas pulse volume and compliance decreased 12%, 43%, and 49%, respectively (P less than 0.001). Greater than 75% of the changes occurred within 5 days after the onset of testing. Despite these changes, isocompliance was demonstrated for the BioPolyMeric grafts following 22 days of cycling, yielding values similar to those reported for human saphenous vein.