Timothy R. Maher

In vivo evaluation of the Nimbus axial flow ventricular assist system. Criteria and methods.

Continuing in vivo trials are being conducted at the University of Pittsburgh using the Nimbus axial flow blood pump (AxiPump). To date, 14 sheep experiments have been performed to address several issues related to short-term support. Six acute experiments (< 6 hr) have been performed to assess hemodynamics related to speed regulation and to determine anatomic placement of the pump and cannulae. Eight short-term survival studies lasting up to 6 days have been performed to evaluate biocompatibility and system reliability, and to establish clinical management protocols. The AxiPump has been used as a left ventricular assist device (LVAD), right ventricular assist device (RVAD), and biventricular assist device (BiVAD) with left ventricular and right atrial cannulation. The AxiPump has demonstrated the ability to assume complete support of either the pulmonary or systemic circulation, or both. We have determined that sufficient surgical access may be obtained through left lateral thoracotomy for both LVAD and RVAD insertion. In the absence of post operative anticoagulation therapy, we have detected subclinical renal cortical infarctions in 6 of 8 short-term animals. Thrombus deposition has been observed at the ventricular cannula tip in 4 of 8 cases--necessitating design changes. Two short-term experiments have been terminated because of bleeding--one due to inflow cannula obstruction and one due to cannula failure. Plasma free hemoglobin levels were all below 15 mg/dl, except for one case complicated by inflow obstruction.(ABSTRACT TRUNCATED AT 250 WORDS)

Progress in Cleveland Clinic-Nimbus total artificial heart development

A totally implantable, Cleveland Clinic-Nimbus total artificial heart (TAH) uses electrohydraulic energy conversion and an automatic left master-alternate mode control scheme, with a filling sensitivity of 1.0 l/min/mmHg and a maximum output of 9.5 l/min. The TAHs were tested in 12 calves for 1-120 days with normal major organ and blood cell function. Post-operative suppression of platelet aggregation recovered by the second post-operative week. The gelatin-coated pump surface generally was clean without any anticoagulants and free from infection. Embolism, which occurred in two cases, was caused by complications attributable to fungal infection in a Dacron graft and by thrombus formed around a jugular vein catheter. A system with a hybridized microcircuit controller in the interventricular space has been tested successfully in the three most recent cases, with a peak device surface temperature elevation of 6.5 degrees C. Heat effects were confined to the tissues immediately adjacent to the hottest spots. The carbon fiber-reinforced epoxy housing and 60 ml butyl rubber compliance chamber showed good tissue compatibility with a thin, fibrous tissue capsule. The transcutaneous energy transmission system and the internal battery functioned well as designed in the most recent animal implant.

Fluorescent image tracking velocimetry of the Nimbus AxiPump.

High shear rates and extended residence times causing hemolysis and platelet activation can develop in an assist pump or cannula when inferior flow conditions exist. The high volume output of a miniature axial flow pump presents challenges in avoiding these adverse conditions. To assess the hemodynamics within the continuous flow Nimbus Axi-Pump, vector flow fields inside a translucent inflow cannula and a modified 12 mm AxiPump were mapped. Fluorescent image tracking velocimetry was used to track the motion of neutrally buoyant fluorescent particles (30 microns) using pulsed laser light, high resolution video cameras, and computer image analysis. An acrylic pump housing and cannula were integrated into a mock circulatory loop filled with a Newtonian, optically clear blood analog fluid. The flow parameters were controlled to yield known, physiologic loading conditions, including varying degrees of pulsatility. Cannula flow visualization results exhibited critical recirculation patterns at the bend. These results will be used to further optimize the design of the inflow. Particle impact was seen at the pump inlet in the inducer region of the rotor. Very good attachment of flow from the rotor to stator was observed when the pump operated at normal operating speeds. Intermittent regurgitant flow fields were evident in the presence of increased pulsatility and low pump speed. These results have lead to improvements in impeller design and speed control criteria to avoid potential deleterious flows.