Julia Glueck

LONG-TERM IN VIVO LEFT VENTRICULAR ASSIST DEVICE STUDY WITH A TITANIUM CENTRIFUGAL PUMP

A totally implantable centrifugal artificial heart has been developed. The plastic prototype, Gyro PI 601, passed 2 day hemodynamic tests as a functional total artificial heart, 2 week screening tests for antithrombogenicity, and 1 month system feasibility. Based on these results, a metallic prototype, Gyro PI 702, was subjected to in vivo left ventricular assist device (LVAD) studies. The pump system employed the Gyro PI 702, which has the same inner dimensions and the same characteristics as the Gyro PI 601, including an eccentric inlet port, a double pivot bearing system, and a magnet coupling system. The PI 702 is driven with the Vienna DC brushless motor actuator. For the in vivo LVAD study, the pump actuator package was implanted in the preperitoneal space in two calves, from the left ventricular apex to the descending aorta. Case 1 achieved greater than 9 month survival without any complications, at an average flow rate of 6.6 L/min with 10.2 W input power. Case 2 was killed early due to the excessive growth of the calf, which caused functional obstruction of the inlet port. There was no blood clot inside the pump. During these periods, neither case exhibited any physiologic abnormalities. The PI 702 pump gives excellent results as a long-term implantable LVAD.

Physiological adaptation to a nonpulsatile biventricular assist system.

Physiological adaptation of the recipient to a nonpulsatile biventricular assist system (NPBVAS) is not well understood. The aim of this study is to evaluate the physiological adaptation of experimental animals after NPBVAS implantation. Since May 2001, four long-term NPBVAS implant experiments in calves were performed. The blood gas and hemodynamic data were analyzed retrospectively. An additional prospective experiment was performed to confirm retrospective findings. All calves (n = 5) lived longer than 5 weeks without complication. In retrospective analysis, there was not a correlation between the O2 content and total blood flow in the pulmonary artery during the 1st postoperative week, but they began to correlate within the 2nd postoperative week. Then, there was a strong correlation after the 3rd postoperative week (r = 0.753). In the prospective experiment, O2 content related to total pulmonary flow after 2 weeks (r = 0.732) was the same as in the retrospective study. Most of the hemodynamic parameters studied became normalized after 14 days. In addition, easier controllability of the blood pumps was demonstrated after the 2nd postoperative week in all five experiments. Experimental results suggested that the native healthy heart accepted NPBVAS by reducing its cardiac output in 2 weeks. In addition, complicated control of the BPVAS was not necessary after 2 weeks of implantation. These results demonstrate the possibility of physiological adaptation to the NPBVAS being established within 2 postoperative weeks.

Development of an implantable small right ventricular assist device.

Currently, at least two permanent implantable left ventricular assist devices (LVADs) are used clinically. Unfortunately, there is no small implantable right ventricular assist device (RVAD) available, even though at least 25-30% of this patient population has right ventricular failure. If a small implantable RVAD were available, biventricular assist could support patients with right ventricular failure. A small atraumatic and antithrombogenic RVAD is being developed to meet this clinical need. This small centrifugal blood pump, the Gyro PI pump, is 6.5 cm in diameter and 4.6 cm in height and has three unique characteristics to prevent thrombus formation: (1) the double pivot bearing and magnetic coupling system enable this pump to be completely sealless; (2) the secondary vanes at the bottom of the impeller accelerate the blood flow and prevent blood stagnation; and (3) the eccentric inlet port enables the top female bearing to be embedded into the top housing and decrease blood cell trauma. The inflow conduit consists of a wire reinforced tube and a hat-shaped tip that is biolized with gelatin to create a thrombus resistant material. This conduit is directly implanted into the right ventricle, and the outflow conduit is anastomosed to the PA. The pump can be implanted inside the abdominal wall or in the thoracic cavity. Biocompatibility of this pump was proved in two calves by thrombus free implantation as an LVAD for 284 days and 200 days. Two RVAD implantations were conducted, aiming for 1-month system feasibility studies. During the month, the RVADs operated satisfactorily without any thromboembolic incident. No blood clots or abnormal findings were seen inside the pump, nor were there abnormal findings in the explanted lungs except for small areas of atelectasis. The pump flow was 3.02 +/- 0.38 L/min in calf 1 and 3.75 +/- 1.18 L/min in calf 2. The power requirement was 7.28 +/- 0.43W for calf 1 and 14.52 +/- 3.93W for calf 2. The PaO2 was 72.0 +/- 3.60 mm Hg (calf 1) and 72.0 +/- 7.63 mm Hg (calf 2); PaCO2 was 38.3 +/- 2.17 mm Hg (calf 1) and 34.1 +/- 1.95 mm Hg (calf 2); and SaO2 was 94.1 +/- 1.37% (calf1) and 95.0 +/- 1.95% (calf 2). Gas exchange via the lungs was maintained. These studies indicate that the Gyro PI pump is suitable as a single implantable RVAD, and is a feasible RVAD as a part of a BiVAD system in terms of pump performance and thrombus resistance.

Development of an Implantable Centrifugal Ventricular Assist Device (CVAD)

The centrifugal ventricular assist device (CVAD) was developed for long-term circulatory support, and is capable of either intracorporeal implantation or paracorporeal placement. The pump was designed based on our antithrombogenic concepts: (1) sealless pump casing, (2) elimination of stationary parts, and (3) blood flow acceleration under the impeller. To meet conditions (1) and (2), a pivot bearing system was adopted to support the impeller. The inlet port was placed slightly off-center and inclined 60° towards the same direction as the outlet port. This port configuration not only yielded a space where an inlet cup bearing could be directly embedded but also allowed for a significant reduction of the pump height, hence, resulting in easier placement inside the body cavity. Two small secondary vanes were installed in the bottom of the impeller to satisfy condition (3). Five paracorporeal left ventricular (LV) AD studies, using calves, were performed to evaluate the antithrombogenic design of the pump. The first two cases were subjected to 2-week tests. With the activated clotting time (ACT) kept at 250 s with heparin, the initial two cases had trouble-free performances over the 2 weeks. Following these successful results, another three cases were subjected to 1-month validation studies, in which there was no device-induced thrombus formation inside the pump housing. These results confirm that the CVAD, the C1E3, meets the requirements for a 1-month paracorporeal LVAD.

Development of a Clipped Single-bag with Bicarbonate Replacement Fluid to Ensure Proper Mixing

A clipped single-bag for bicarbonate replacement fluid was developed to ensure proper mixing before administering to the patient. Nonmixture can cause imbalances of electrolytes and pH, which is a key problem for the current double-bag type bicarbonate replacement fluid sets. To resolve this problem, this single bag properly mixes the solutions before use. The new bag consists of a clip that is placed in the middle to keep the two solutions separated and sealed. When the caregiver is ready to administer treatment, the bag is simply unfolded and the clip automatically detaches, releasing the fluids. Thereby, the bicarbonate fluids are effectively mixed. An optimal clip size with an outer diameter of 16 mm and thickness of 2 mm was determined using compression tests and drop tests. This bag may be a safer and more effective way to provide proper replacement fluid supply for both hemofiltration and hemodiafiltration.