C

Technological Opportunities and Barriers in the Development of Mechanical Circulatory Support Systems

GERSON ROSENBERG

THE CONCEPT OF MECHANICAL CIRCULATORY SUPPORT was first postulated by LeGallois in 1812 (LeGallois, 1813). Much later, in 1934, DeBakey proposed a simple continuous flow blood transfusion instrument that was a simple roller pump (DeBakey, 1934). In 1961, Dennis and colleagues performed left heart bypass by inserting cannulae into the left atrium and returning blood through the femoral artery (Dennis, 1979). In 1961, Kolff and Moulopoulos developed the first intra-aortic balloon pump (Moulopoulos et al., 1962). In 1963, Liota performed the first clinical implantation of a pulsatile left ventricular assist device (Liota et al., 1963). In 1969, the first implant of a total artificial heart was performed by Denton Cooley (Cooley et al., 1969). Mechanical circulatory support has been used in over 1,300 patients since 1985. There have been implants of approximately 186 total artificial hearts, 600 left ventricular assist devices, and 112 right ventricular assist devices along with 409 biventricular assist devices. Of these 1,300 patients, over 600 have been weaned and approximately 365 have been discharged from the hospital (Joyce et al., 1988; Pae and Miller, 1990).

Progress in the use of ventricular assist devices and artificial hearts has been excellent. In the 1970s, animal survival with the total artificial heart

Gerson Rosenberg is Research Professor and Assistant Chief, Division of Artificial Organs, Department of Surgery, Milton S. Hershey Medical Center, Hershey, Pennsylvania, and Professor of Bioengineering, College of Engineering, The Pennsylvania State University, University Park, Pennsylvania. This appendix is based on a paper submitted to the Institute of Medicine committee in October 1990 and thus reflects developments and the literature as of that date.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 211
The Artificial Heart: Prototypes, Policies, and Patients C Technological Opportunities and Barriers in the Development of Mechanical Circulatory Support Systems GERSON ROSENBERG THE CONCEPT OF MECHANICAL CIRCULATORY SUPPORT was first postulated by LeGallois in 1812 (LeGallois, 1813). Much later, in 1934, DeBakey proposed a simple continuous flow blood transfusion instrument that was a simple roller pump (DeBakey, 1934). In 1961, Dennis and colleagues performed left heart bypass by inserting cannulae into the left atrium and returning blood through the femoral artery (Dennis, 1979). In 1961, Kolff and Moulopoulos developed the first intra-aortic balloon pump (Moulopoulos et al., 1962). In 1963, Liota performed the first clinical implantation of a pulsatile left ventricular assist device (Liota et al., 1963). In 1969, the first implant of a total artificial heart was performed by Denton Cooley (Cooley et al., 1969). Mechanical circulatory support has been used in over 1,300 patients since 1985. There have been implants of approximately 186 total artificial hearts, 600 left ventricular assist devices, and 112 right ventricular assist devices along with 409 biventricular assist devices. Of these 1,300 patients, over 600 have been weaned and approximately 365 have been discharged from the hospital (Joyce et al., 1988; Pae and Miller, 1990). Progress in the use of ventricular assist devices and artificial hearts has been excellent. In the 1970s, animal survival with the total artificial heart Gerson Rosenberg is Research Professor and Assistant Chief, Division of Artificial Organs, Department of Surgery, Milton S. Hershey Medical Center, Hershey, Pennsylvania, and Professor of Bioengineering, College of Engineering, The Pennsylvania State University, University Park, Pennsylvania. This appendix is based on a paper submitted to the Institute of Medicine committee in October 1990 and thus reflects developments and the literature as of that date.

OCR for page 211
The Artificial Heart: Prototypes, Policies, and Patients was averaging approximately two weeks; animal survivals today are approaching one year, and patients have survived for almost 600 days with artificial hearts. The use of temporary ventricular assist devices is becoming more routine, and the development of permanent left ventricular assist devices and total artificial hearts is well under way. Yet with all the progress that has been made, there are currently several complications associated with the permanent application of left ventricular assist devices and total artificial hearts. These can be broken down into durability and biocompatibility (including bleeding, thrombosis, sepsis, calcification, and hemolysis). These factors appear in various degrees in all of the devices and will be discussed, but they do not appear to be insurmountable problems; in fact, several appear to be close to solution. CURRENT STATE OF THE TECHNOLOGY IN MECHANICAL CIRCULATORY SUPPORT SYSTEMS Artificial hearts and circulatory assist devices are currently under development in the United States, Korea, Russia, Canada, Switzerland, Japan, Germany, Czechoslovakia, Italy, France, Australia, China, and other countries. A detailed description of all of these devices is beyond the scope of this document, and only those with significant design features or devices sufficiently developed to be nearing clinical trials will be discussed. The most frequent clinically used mechanical circulatory support systems (MCSSs) are those currently manufactured in the United States (Pae and Miller, 1990). The animal survival times with pneumatically powered devices are essentially the same in the United States and abroad, indicating approximately the same technology level (Total Artificial Heart, 1985; Total Artificial Heart, 1987). Electric motor-driven circulatory assist and artificial heart devices are, at the present time, more advanced in the United States than in any other country, although devices in Japan (Total Artificial Heart, 1985) and Switzerland (Odermatt, 1989) are advancing rapidly. All of these MCSSs have met with similar difficulties in development and application. These difficulties include device durability and biocompatibility. Various solutions have been implemented for these problems and have allowed devices to function for over a year in vivo and greater than two years in vitro. Short-Term-Use Devices (fewer than 180 days) Both total artificial hearts and univentricular or biventricular assist devices can be utilized temporarily for mechanical circulatory support. All of the total artificial hearts that have been utilized clinically have been pneumatically powered devices. The ventricular assist devices that have been

OCR for page 211
The Artificial Heart: Prototypes, Policies, and Patients utilized clinically include pneumatically or electrically powered pulsatile devices and nonpulsatile devices. Total Artificial Heart For the purposes of discussion, total artificial hearts will be considered those devices that are orthotopically placed with the removal of the native heart, thus distinguishing these devices from biventricular assist devices. Animal and Clinical Results Over the past decade, several short-term pneumatic artificial hearts have received significant development. These devices include the pneumatic artificial hearts of The Pennsylvania State University (Penn State) and the Free University of Berlin, the Jarvik or Symbion artificial heart originally developed at the University of Utah, the Cleveland Clinic Foundation heart, and the heart of the University of Perkinje in Bruno, Czechoslovakia. Other hearts are being developed in Tokyo, Japan (Pierce, 1986; Total Artificial Heart, 1989). The longest survival with a pneumatic artificial heart in a calf or sheep is 353 days, accomplished by the Penn State group (Aufiero et al., 1987). A sheep at the University of Utah lived for 297 days and a goat with a pneumatic artificial heart lived for 344 days at the University of Tokyo. The Tokyo heart was not orthotopically placed, but was located outside the animal's thoracic cavity. Thus, it is accurate to say that a small percentage of the artificial heart animals have been able to survive for slightly less than one year. Several factors contribute to this one-year time limit. In the growing animal such as the calf, the animals have the potential to outgrow their cardiac output. Typically, pneumatic artificial hearts are capable of pumping a maximum of 12 liters per minute. Normal healthy calves will gain as much as 1 kilogram per day. Starting with a 100-kilogram animal gaining just 1 kilogram a day, the calf will outgrow the heart in less than one year based upon a minimum cardiac output of 70 milliliters per minute per kilogram. Calcification has also been a problem in the growing animal, causing device failure through rupture of the polymeric sac. Sepsis and, to some degree, thrombosis have been present in the long-term animals. The experience at Penn State for 21 consecutive pneumatic artificial heart animals indicates that 3 died from pannus formation, a proliferation of unwanted tissue in the inlet of the artificial hearts. (This problem has been remedied and has not occurred for four years.) Of the 21 animals, only 1 died from a thromboembolic event. Technical error caused five of the animals to die. One animal died of anticoagulant-related bleeding, and two of the longest-surviving animals, living 275 and 353 days, respec-

OCR for page 211
The Artificial Heart: Prototypes, Policies, and Patients tively, died of cardiac cachexia. These animals basically outgrew their artificial hearts. Mechanical failures were responsible for 9 of the 21 animal deaths (43 percent). Blood sac or diaphragm perforations occurred in six of the nine animals that died as a result of mechanical failure, which was related to stress and/or calcification of the flexing polymer. Since Penn State instituted medical therapy with warfarin sodium and etidronate disodium to retard calcification, along with design changes to reduce the stress experienced with the flexing surfaces, there has not been any failure from these causes in pneumatically driven pumps. Although not a cause of death, sepsis was present in many of these animals (Pae et al., 1987). It should be noted that valve failure and drive system failure have not been a cause of death in these animals. Clinical results with pneumatic short-term artificial hearts are limited almost exclusively to the results with the Jarvik 100-cubic centimeter (cc) and 70-cc stroke volume devices and may not be indicative of other devices. As of 1990, there have been 186 applications of pneumatic total artificial hearts; 127 of those patients were weaned, and 62 were transplanted and discharged. As previously stated, the largest percentage of those patients received the Jarvik/Symbion-type artificial heart. Of the 186 patients to receive pneumatic total artificial hearts, the major complication was bleeding and reoperation in 28 percent of the patients. Renal failure occurred in 19 percent of the patients. Hemolysis occurred in 7 percent of the patients, respiratory failure occurred in 13 percent, thrombosis in the system occurred in 4 percent, and embolus occurred in 9 percent, for a total of 13 percent for thromboembolic complications. Infection occurred in 21 percent of the patients (Pae and Miller, 1990). Technological Development of Pneumatic Total Artificial Hearts Durability. Pneumatic artificial hearts have functioned in animals for approximately one year, and there have been devices that have functioned on the mock circulatory system, at various institutions, for periods in excess of two years. Insufficient studies have been done to accurately predict the reliability of these devices. The clinical registry data indicate that mechanical failure was present in 1 percent of the patients who received the total artificial heart. Similar sac-type blood pumps utilized for left ventricular assist devices have demonstrated a two year reliability in vitro (Jaszawalla et al., 1988). Thus, the durability for short-term application of total artificial hearts appears to be quite adequate, with approximately 1 percent mechanical failures in 186 clinical applications. It is interesting to note that the device that received the most widespread clinical use, the Symbion device, was withdrawn from the market by the Food and Drug Administration. The Food and Drug Administration discontinued the

OCR for page 211
The Artificial Heart: Prototypes, Policies, and Patients device's trials owing to the inadequacy of the methods, facilities, and controls used for the manufacturing, processing, and installation of the device and the inadequacy of the monitoring and review of the investigation. The investigational device exemption (IDE) was also withdrawn because Symbion did not provide assurance that the integrity of the device had been maintained by adhering to the original design specifications and manufacturing controls and that the clinical studies being performed were adhering to the clinical protocols approved in the original IDE. Thus, the device was withdrawn not because of poor performance but rather because of inadequate manufacturing and application of the device. Control. Throughout the development of the artificial heart, there have been various control schemes proposed for the artificial heart. The majority of artificial hearts are controlled in a Starling-like manner or fill-limited mode. Thus the device will pump blood that is returned to it within a particular range of cardiac outputs. One of the problems with this type of control system is that the gain is somewhat limited. Thus, it requires large changes in filling pressure to effect physiologic changes in cardiac output. The Penn State group has utilized an electronic automatic control system to control the devices for not only cardiac output but also actively balancing the left and right pumps. This cardiac output control system is sensitive to pump afterload, and balancing is accomplished by indirect sensing of left atrial pressure (Snyder et al., 1986). Other systems have been proposed and tested, such as systems measuring the P-wave from the remnant atrium or using various ChemFETS or other devices to measure blood chemistry values. It does not appear that animal survival has been limited by the various control schemes, since the major groups differ in control method but have essentially the same survival times. Various control schemes may require continuous monitoring and adjustment to maintain balance or cardiac output, while others perform this task automatically. It appears that as long as the control system can maintain a physiologic left atrial pressure, provide a resting cardiac output in the range of 70 milliliters per minute per kilogram, and allow for changes in cardiac output as required, the animals survive normally. In most cases, as the calf continues to grow and survive for longer periods of time, elevated central venous pressures become apparent. These central venous pressures range from 10 to 20 millimeters mercury. The etiology of the increased central venous pressure is unknown. Studies involving the measurement of atrial natriuretic peptide (ANP) have indicated that there is a disruption in the normal ANP control mechanism, and perhaps this is a contributing factor. Studies of ANP levels in these animals will prove very valuable basic knowledge about this not-well-understood physiology. It is also possible that the various control schemes that are utilized, although providing grossly adequate cardiac output, may

OCR for page 211
The Artificial Heart: Prototypes, Policies, and Patients affect the long-term regulation of central venous pressure (Mabuchi et al., 1988). Even with elevated central venous pressure, however, it does not appear that control systems and control strategies are a limiting factor in the utilization of temporary or long-term artificial hearts. Biocompatibility. Thrombosis. In clinical applications, thrombosis or embolus occurred in 14 percent of the 186 applications of the pneumatic artificial heart. It is important to point out here, again, that this was the Symbion-type artificial heart, which may not be indicative of all pneumatic artificial hearts (Gaykowski et al., 1988). Of the three patients who received the Penn State artificial heart, the longest-surviving patient, who lived for 390 days, had one thromboembolic event 10 weeks after implantation. This patient's anticoagulant therapy was modified, and no further thromboembolic complications occurred. Based upon existing data, it would be reasonable to predict that 14 percent of the patients to receive a short-term artificial heart would have a thrombotic event with the Jarvik-type artificial heart. It is not possible to predict the thromboembolic complications with devices of other designs. Thromboembolism is a function of the material that is used in the blood pump, the cleanliness and surface characteristics of that material, the actual geometry within the blood pump (which can affect regions of stasis and blood flow), the presence of cracks or crevices, and the careful matching of materials within the blood pump, including heart valves and associated adjoining hardware. Thrombosis is very design- and manufacturing-sensitive. The only device currently available for clinical application of a total artificial heart is the Penn State device. Very careful attention has been paid to the geometry and surface characteristics, as well as careful design and choice of materials in this device, to avoid thrombosis. Other groups also have designs that are potentially superior to the Symbion/Jarvik-7 system. As previously stated, at Penn State with a pneumatic total artificial heart, 1 animal of 21—approximately 5 percent—that received the device suffered a thromboembolic event causing death. Yet evidence of thrombotic complications and organ infarction was noted in 13 of 24 calves, indicating that thrombosis is still a major complication with total artificial heart devices in animals (Al-Mondhiry et al., 1989). Similar results have been seen by other artificial heart research groups (Nojiri et al., 1989). Sepsis. In the clinical applications of artificial hearts, sepsis was present in 21 percent of the 186 patients undergoing implantation of the artificial heart (Pae and Miller, 1990). Septic complications have also been noted in the early patients receiving the Jarvik-type artificial heart (DeVries, 1988; Gristina et al., 1988; Kunin et al., 1988). In a series of 24 calves at Penn State, septic complications were documented in 10 animals, thus indicating

OCR for page 211
The Artificial Heart: Prototypes, Policies, and Patients that sepsis is a major complication with the pneumatic total artificial heart with percutaneous leads. Hemolysis. In reports from the clinical registry for the application of total artificial hearts, hemolysis was listed as a complication in 7 percent of the patients. Experience at Penn State with pneumatic artificial hearts in calves shows that compared with baseline levels, the hemoglobin was significantly lower and the plasma hemoglobin and serum lactate dehydrogenase levels significantly higher throughout the follow-up period. The platelet count decreased during the first 10 to 30 weeks, but returned to preoperative levels by week 35. Platelet survival levels in these animals in stable condition were normal and within normal limits (Al-Mondhiry et al., 1989). Similar findings have been presented by other groups (Nojiri et al., 1989). The hemolysis levels do not seem unreasonable considering there are four prosthetic heart valves within these devices as well as a large amount of prosthetic materials. The hemolysis levels, although clinically significant, are not unmanageable. Calcification. All groups utilizing calves and sheep have reported some degree of calcification within their devices (Pierce et al., 1980). Calcification within devices employed in growing animals such as calves has been very severe and a limiting factor in many of the experiments, causing stiffening and perforation of the diaphragm. This calcification does not appear to be as severe in the sheep and goat models (Portner, 1987). The calcification also appears to be a function of the material 's surface characteristics and stress on the material. Results with the clinical application of these devices, for over 600 days, indicate that calcification is not a limiting factor for a two-year device life at the present time. Certainly, in applications of these devices for short-term use, calcification is not significant. All the current short-term total artificial heart devices appear to have three significant complications in common: bleeding, thrombosis, and sepsis. In the clinical application of these devices, the most frequent complication is bleeding and reoperation in 28 percent of the patients. The next most common complication is infection, occurring in 21 percent of the patients. Thrombosis occurs in approximately 14 percent of the patients, while hemolysis occurs in only 7 percent and is easily managed in most. Univentricular or Biventricular Assist Devices Pulsatile Devices (Sac/Diaphragm) Currently four pulsatile assist devices are available for clinical application in the United States: the ABIOMED BVS System 5000, the Novacor

OCR for page 211
The Artificial Heart: Prototypes, Policies, and Patients left ventricular assist device, the Pierce-Donachy device presently manufactured by Thoratec Laboratories and Sarns, and the Thermedics (Thermo Cardiosystems) Heartmate device (McGee et al., 1989; Macoviak et al., 1990). There are other ventricular assist devices under development such as the device at the Cleveland Clinic Foundation, the device under development by ABIOMED Corporation, and the device under development by Electrocath Corporation. Ventricular assist devices are also under development in Japan (Atsumi et al., 1989). The Symbion ventricular assist device, which is no longer available, had undergone clinical trials. The most widely used device is the Pierce-Donachy ventricular assist device manufactured by Thoratec Laboratories. Animal and clinical results. In a multicenter study of 29 patients utilizing heterotopic Thoratec prosthetic ventricles as a bridge to cardiac transplantation, 21 received heart transplants and 20 were discharged from the hospital after a median 31 days. Of the 29 patients, 6 were reported to have had infections during their circulatory support period, but in only 2 did infection cause death; 11 of the 29 patients had severe bleeding complications. Two neurologic events were reported in two patients who were later discharged. In one of the patients who sustained a neurologic event, the drive console was off for 20 minutes the night before transplantation, resulting in insufficient blood flow. In both of these patients, thrombus was found in the explanted pump (Farrar et al., 1988). The ABIOMED BVS 5000 has been used on more than 170 patients in clinical centers throughout the world. It is currently in clinical trials in the United States in 11 centers. The mean patient age is 46 years, ranging between 7 and 74 years. The mean duration of support is 4 days, with the longest support duration being 30 days. The predominant support mode has been biventricular in 67 percent of the cases. In postcardiotomy support, which is the primary intended use of the system, 45 percent of the patients were weaned with 51 percent of these patients surviving. These numbers are comparable to the registry values (postcardiotomy) of 43 percent and 54 percent, respectively. Complications encountered are quite similar to the overall registry results. The key features of the BVS 5000 are its low cost and simplicity of use. A primary reason for its low cost is the use of trileaflet polyetherurethane valves. These valves were originally developed under the sponsorship of the National Heart, Lung, and Blood Institute. This technology grew out of the mechanical circulatory support program and is an important element in our efforts to expand the clinical utility of this temporary cardiac support system. The results from the clinical registry of mechanical ventricular assistance show that bleeding and reoperation are still the major complicating factors in all pneumatic mechanical ventricular assist devices, with an incidence of

OCR for page 211
The Artificial Heart: Prototypes, Policies, and Patients approximately 30 percent. Renal failure was second with 25 percent, infection was third with 17 percent, biventricular failure occurred in 16 percent of the patients, and respiratory failure occurred in 16. Thrombosis in the system occurred in 5 percent of the patients, and embolus occurred in 7 percent. Thus, it is apparent that the problems that occur in the short-term artificial heart also occur in short-term univentricular or biventricular assist devices when all the devices are considered as a group. Some devices do much better than others. Technological development. Durability. Durability of these devices for their intended period of application has been very good. Mechanical failure has only occurred in 1.45 percent of those cases reported to the registry (Pae and Miller, 1990). Control. Devices have been run synchronously and asynchronously both partially full and full-to-empty. There have been no definitive clinical studies showing the relative advantages of one modality over the other. There are theoretical considerations, but there are no data indicating improved survival with one technique over the other. Biocompatibility. As previously stated, thrombosis, sepsis, and to some degree hemolysis are complications associated with the application of short-term univentricular and biventricular assist devices. Of the pulsatile ventricular assist devices, several designs and materials have been utilized. The Novacor, Symbion, and Thoratec devices all use a smooth segmented polyurethane surface. The device developed by Thermedics (Heartmate) utilizes a textured surface to facilitate the formation and adhesion of a biologic lining. The Heartmate pump diaphragm is fabricated of integrally textured polyurethane, and the metallic surfaces of the pump are textured by using powdered metallurgy techniques. There has been no clinical evidence of thromboembolism in any of the 17 patients for whom the device has been used, nor was there any evidence of thromboembolism in patients who came to necropsy (Graham et al., 1989). It has also been reported that in these patients, plasma hemoglobin levels remained acceptable throughout support. Blood chemistry and hematologic values returned to normal in most cases. One of the patients with this device was supported for 132 days (Nakatani et al., 1989). Thus, with the Heartmate device there is an indication of improvement in the area of thrombosis. Steady-Flow Devices Animal and clinical results. There are currently several systems available for clinical application that are of the steady-flow type. These include

OCR for page 211
The Artificial Heart: Prototypes, Policies, and Patients the Biopump manufactured by Biomedicus, the Centrimed System manufactured by Dolphen, Incorporated, and the Hemopump manufactured by Nimbus, Incorporated. There are also several other devices under development. These include devices such as the spindle pump and the axial flow pump described by Schistek and others (Schistek, 1989). A review of the clinical experience with steady-flow or centrifugal devices shows that there is a slightly higher incidence of bleeding and reoperation with centrifugal devices versus pneumatic devices: 46 percent of the patients had bleeding and reoperation with the centrifugal devices, whereas only 30 percent had bleeding as a complication with the pneumatic devices. There were other slight differences in hemolysis and sepsis, and only a very slight difference in thrombosis and embolus with the two types of devices. If one looks at the outcome of mechanical circulatory support for postcardiotomy cardiogenic shock based upon ventricular assist pump type, 24 percent of the patients with the centrifugal pump were discharged after use of the device and 21 percent of the patients with pneumatic devices were discharged. There does not appear to be a major difference. Careful analysis of all the data within the clinical registry shows that there are some slight differences between devices, but they may be related to the indications for use, i.e., postcardiotomy cardiogenic shock versus staged cardiac transplantation. Yet a review of all of the short-term devices, total artificial heart, biventricular, and univentricular devices, pulsatile and nonpulsatile devices, indicates that bleeding, infection, thrombosis, and to a lesser degree hemolysis all are problems. Other complications such as renal failure and respiratory failure are most likely associated with the poor condition of the patient at the time of surgery. In fact, many of these patients have improved renal and respiratory function with initiation of pumping. If one examines the overall outcome of staged heart transplantation based on all types of ventricular assist devices, 72 percent of the patients with pneumatic devices, 67 percent of those with centrifugal devices, and 86 percent of those with electric devices are discharged. The electric device, in this case the Novacor device, appears to have improved results, but the device is used only for bridge to transplantation. Summary When considered as a group, short-term and long-term pneumatic and centrifugal unilateral and bilateral assist devices and total artificial hearts have complications of thrombosis, sepsis, bleeding, and hemolysis, and it is not entirely clear which will be the limiting factor in the longevity of these devices. It is apparent that for short-term use, durability is not a limiting factor and devices appear to be satisfactory for use up to 180 days. Control systems do not appear, on the surface, to have a significant impact on the

OCR for page 211
The Artificial Heart: Prototypes, Policies, and Patients outcome of the use of these devices. The hemolysis levels associated with the short-term devices, in general, are not life-threatening. Bleeding may not be directly attributable to the device, but due to anticoagulant therapy or prolonged cardiopulmonary bypass. Careful anticoagulant therapy and better patient selection may affect the bleeding and reoperation rate. Although these complications exist, there are no fundamental physical laws or reasons that limit solutions to the problems of thrombosis, sepsis, and bleeding. Improved materials, as well as improved hemodynamic designs of these devices, can lower not only thrombosis but bleeding, since improved devices will require less anticoagulation therapy. Totally implanted devices as well as new techniques for encapsulating these devices will reduce infection. Permanent or Long-Term-Use Mechanical Circulatory Support Systems (more than 180 days) Unilateral Assist Devices In Vitro and In Vivo Test Results Several groups in the United States and abroad are now working on permanent electric motor- or thermal-powered ventricular assist devices. These include Nimbus in Rancho Cordova, California, Novacor Division of Baxter Healthcare Corporation in Oakland, California, Penn State in Hershey, Pennsylvania, Thermedics in Waltham, Massachusetts, and various groups in Europe and Japan. Currently, the most advanced device is the ventricular assist device developed by Novacor. The totally implantable left ventricular assist system has demonstrated a two year life in vitro with an 80 percent reliability. Implants in sheep of up to 279 days for the ventricular assist device have been accomplished, and this device has received clinical application for short-term use in patients. In its use as a temporary assist device, bleeding and reoperation occurred in 42 percent of the patients, infection occurred in 18 percent, and thrombosis and embolism occurred in 16 percent. The outcome of staged heart transplantation with the Novacor system in 41 patients that were implanted is that 22 or 54 percent were transplanted. Of the 22, 80 percent were discharged. Since this device, with a transcutaneous energy transmission system (TETS), will be utilized for long-term support, a prediction of the associated complications can be approximated based upon the short-term use. Infection rates may be improved by the elimination of percutaneous leads.

OCR for page 211
The Artificial Heart: Prototypes, Policies, and Patients were developed that utilized rare earth materials such as samarium cobalt. Energy products approaching 30,000,000 Gauss Oersted were obtained with this material. More recently, neodymium iron magnets have been developed that produce an energy product in excess of 30,000,000 Gauss Oersted. These improved magnetic materials mean smaller and lighter motors and better magnetic coupling in solenoid devices. Research continues for improved magnetic materials. Any improved magnetic materials would have a positive impact on MCSSs. They would reduce the size and weight of the device, and might also improve overall performance. Supercomputers Computational Fluid Dynamics Supercomputers such as the Cray system have had and will, in the near future, have a significant impact on the development of MCSSs. These systems can be used to solve extremely complicated, nonlinear partial differential equations that may describe stresses in the materials or fluid mechanics. These are extremely important studies in terms of (1) understanding the fluid mechanics within the blood pump and (2) determining the stresses in the various components of the system. At the present time, it is not possible to solve a three dimensional unsteady, non-Newtonian turbulent flow within these blood pumps. Thus, it is extremely difficult to know and understand the fluid mechanics that are occurring in the device and the contribution to thrombosis, hemolysis, and the mechanical stresses within these devices. Through the use of new codes and supercomputers, solutions of this problem will yield basic information related to fluid mechanics and thrombosis. Numerical Analysis Techniques The ultimate longevity of the elastomeric diaphragms and other mechanical components in these blood pumps relies on an accurate prediction of the stresses imposed on the device during operation. The determination of these stresses involves the solution of difficult differential equations. New supercomputers and new codes can be utilized to solve these differential equations. These solutions can then be used to do optimization of the size and shape of the blood pump to provide for minimum stresses and minimum hemolysis and thrombosis related to the mechanical motion and fluid mechanics of the device. Summary Developments in new materials, electronic components, magnetic materials, and supercomputers should all have a positive impact on MCSSs. Not

OCR for page 211
The Artificial Heart: Prototypes, Policies, and Patients only will the new materials and electronic components provide for smaller and more efficient and reliable devices, but the advent of new polymeric materials or surface-improved materials should also yield reduced thrombosis. New mathematical models solved on supercomputers will give a better basic understanding of the role of mechanical stresses and fluid mechanics in thrombosis, hemolysis, and mechanical component failure. This basic knowledge can be utilized to design and optimize improved devices. SPIN-OFF TECHNOLOGIES Materials and Design Research on mechanical circulatory support has generated a pool of individuals with unique expertise in the area of artificial organs. The expertise of these individuals can be utilized in the development of other devices such as grafts, valves, new biomaterials, biosensors, and implantable battery technology. Grafts The technology related to materials and blood flow within MCSSs can be applied to the design of new and improved vascular grafts. New biomaterials, manufacturing processes, or surface modification techniques can be used for these grafts. Valves Researchers working on the development of mechanical circulatory support systems are working on manufacturing polymeric trileaflet heart valves. ABIOMED currently uses a polymeric trileaflet valve that their personnel have designed and constructed for their short-term MCSS. Researchers at the University of Utah and Penn State have also manufactured polymeric heart valves (Wisman et al., 1982). As these valves are further developed, they may find clinical application as prosthetic heart valves. They potentially could offer improved biocompatibility over mechanical and tissue valves at a reduced cost. Biomaterials Certainly, any biomaterial developed for MCSSs could be utilized in other implant applications. There have really been no new major biomaterials developed in the past 10 years. Increased emphasis should be placed on the development of new biomaterials that would be applicable to mechanical circulatory assist devices and other biomedical applications.

OCR for page 211
The Artificial Heart: Prototypes, Policies, and Patients Biosensors Researchers associated with mechanical circulatory support systems are developing biosensors that can be used to control devices such as the artificial heart. These biosensors will sense quantities such as pH, carbon dioxide, carbon monoxide, and oxygen tensions within the body. These biosensors will have uses in artificial organs such as liver, kidney, and lung. They would also have application for incorporation into catheters that can be used for monitoring of hospital patients in intensive care units. Battery Technology The Honeywell Energy Systems Division has identified the biomedical market as one they would like to develop a battery for. Honeywell has previously developed a lithium technology primary battery that is utilized in the implantable defibrillator. Honeywell Energy Systems is currently working on the development of new lithium rechargeable technology for use with MCSSs. This battery technology, if developed, would be used for other high-energy, high-reliability applications such as in aerospace and other medical areas. Drug Actions Artificial heart animals and patients provide excellent models for testing the actions of various drugs. The effects of these drugs on the vascular system can be studied while the heart is controlled by the researcher. This model can then be utilized to understand more clearly the cardiac and vascular component actions of the drugs. Development of Transcutaneous Energy Transmission Systems Development of transcutaneous energy transmission has already been spun off into use into the cochlear implant, as previously described. This technology would have application for other implanted artificial organs that require high energy levels. Much of the technology that is developed for MCSSs can be used for other artificial organs or biomedical applications. Also, developments in battery technology, biomaterials, and biosensors will have uses in other high reliability situations such as aerospace applications.

OCR for page 211
The Artificial Heart: Prototypes, Policies, and Patients SUMMARY Current Status of Mechanical Circulatory Support Systems Short-Term Devices Several short-term ventricular assist devices are now available to clinicians under IDEs. These include the Thoratec pump, the Thermedics pump, the ABIOMED system, the Novacor system, the Sarns Centrimed system, and the Biomedicus pump. Usage of these short-term ventricular assist devices has been outlined elsewhere and is covered extensively in the Combined Registry for the Clinical Use of Mechanical Ventricular Assist Pumps and the Total Artificial Heart. It should be noted that complications associated with this class of devices, in general, include bleeding, infection, thrombosis, and to a lesser extent hemolysis. It is also quite important to note that some of the devices have much better results in certain areas than others. For example, the Thermedics device has been utilized in 17 patients without thrombosis or thromboembolic events. Although this is a small number of patients, the results appear encouraging. Also, a careful look at the registry data shows that a particular device may do better than the general population of devices. The success rate for these devices continues to improve with time, and there appears to be great interest in getting more of these devices into use. At the present time, the only short-term total artificial heart approved by the Food and Drug Administration is the Penn State heart. This device has been utilized in three patients, the longest of whom survived for 390 days. The clinical indications for the use of a short-term total artificial heart have not been well established. It appears that in most instances, biventricular support or univentricular support is adequate for short-term bridge to transplant applications. With biventricular and left ventricular assist devices as successful as they are, it is doubtful that there will be an increased use of the short-term total artificial heart. Long-Term Mechanical Circulatory Support Devices Long-term or permanent ventricular assist devices are coming quite close to clinical application. The system developed by Novacor has demonstrated a two-year life with an 80 percent reliability in vitro. Preclinical testing in vivo will begin shortly, and clinical trials under an IDE will also be beginning in the next one to two years. The Novacor system has been utilized as a short-term device clinically with results essentially similar to the average registry results. Other research groups, such as ABIOMED, Penn State, Thermedics, and the University of Washington, are all pursuing development of long-term ventricular assist devices. Devices from Nimbus, Penn

OCR for page 211
The Artificial Heart: Prototypes, Policies, and Patients State, and Thermedics have undergone limited in vitro testing and have all been utilized in vivo. Experiments at Penn State have been conducted utilizing transcutaneous energy transmission with a completely sealed system in the calf. Long-Term Implantable Total Artificial Hearts Four groups in the United States are now working under contract on long-term electric motor-driven total artificial heart devices: ABIOMED, Nimbus, Penn State, and the University of Utah. ABIOMED, Nimbus, and University of Utah have completed initial designs and have begun in vitro testing of various components of their systems. Penn State has manufactured a complete electric motor-driven total artificial heart system that transmits energy across the intact skin by inductive coupling. The system is completely sealed and totally implantable, has undergone in vitro testing, and has been utilized in two calf experiments. This system is currently in a state of development that is equivalent to the development of long-term ventricular assist devices, with the exception of the Novacor system, which is the most advanced. All of these systems have their relative advantages and disadvantages in terms of size, efficiency, and reliability, and further testing is required to determine the best system. Prospects for the Future of Mechanical Circulatory Support Systems Prediction of future prospects for MCSSs can be done with a certain degree of confidence for the next three to five years. When predicting for the next 5 to 10 years, one needs to be more cautious; in predicting the prospects for the next 10 to 30 years, one must be extremely cautious. Looking back at medical device and device-related technology in the early 1950s, it is doubtful that many would have predicted then the great usage of heart valves, pacemakers, implantable defibrillators, and vascular prostheses available today. In the 1950s, the first pacemaker had to be pushed around on a cart by the patient. Twenty years ago, patients were still changing or charging batteries in their pacemakers. Then new higherenergy-density primary cells and lower-energy-requiring C-MOS electronics made charging and changing batteries a thing of the past. Today there are smaller, lighter, programmable adaptive pacemakers. The usage of heart valves has expanded and the results have improved. In general, if one looks at the history of MCSSs and plots survival times in animals and in patients versus time, one sees a very progressive increase in both animal and patient survival times.

OCR for page 211
The Artificial Heart: Prototypes, Policies, and Patients Three to Five Years In the next three to five years, short-term ventricular assist devices should see more widespread use. Sarns/3M Healthcare Group will be making the Pierce-Donachy pump available under an IDE within the next year, and it is reasonable to predict that other companies will attempt to expand their usage under IDEs. Thus, the use of these devices should continue to increase gradually. It is doubtful that the use of short-term total artificial heart devices will expand due to the success that is occurring with the current left heart assist devices. In the next three to five years, long-term ventricular assist devices will continue to be tested for preclinical and clinical application. Novacor should be able to complete in vivo studies on its system and begin initial clinical trials within the next two to three years. Thus, clinical application of the Novacor system could occur within the next five years. Other systems being developed for long-term application may undergo in vitro testing to qualify for in vivo testing prior to clinical application within the next five years. Permanent total artificial heart devices currently under development will begin initial in vivo experiments within the next three to five years, and some fairly extensive in vivo and in vitro studies should have been completed on all four of the systems being developed under government contact. Five to Ten Years It is reasonable to assume that there will be the same trend in usage of short-term left ventricular assist and total artificial heart devices within the next 5 to 10 years. Perhaps improvements in the short-term left heart assist devices will result in more usage for cardiogenic shock support. Within the next 5 to 10 years, the permanent left heart assist system of Novacor should be well into clinical trials. Other groups manufacturing devices, such as Nimbus, Penn State, ABIOMED, and Thermedics, should within the next 5 to 10 years complete their animal in vivo studies and begin human in vivo studies. Within the next 5 to 10 years, permanent total artificial heart devices should be well on the way to becoming finalized designs. The current contracts call for the beginning of preclinical testing within the next five years. This would mean that the groups should have completed all of their preliminary in vitro and in vivo testing and have begun extensive reliability testing and animal studies. By 10 years from now, the first of these devices should receive some clinical application. Thus, it would seem reasonable to predict that within the next 10 years a permanent total artificial heart will be implanted in a human. This device should have as a minimum an 80 to 90 percent reliability for two years.

OCR for page 211
The Artificial Heart: Prototypes, Policies, and Patients Ten to Thirty Years Predictions for the next 10 to 30 years become much more difficult. Within the next 10 to 30 years, new materials should become available that will reduce or eliminate the complications associated with MCSSs. Almost surely new magnetic materials, new electronic components, and better conductors all will enable these devices to be improved. New electrochemical energy sources should become available and should be able to be incorporated into existing designs. Within the next 30 years, these devices should be available for widespread use and provide a satisfactory lifestyle, with five-year survival rates in excess of 50 percent. It is always worrisome to make predictions about the future and much more so when they are made in writing. With that in mind, this author has attempted to be as conservative as possible. Looking at the history of mechanical circulatory assist devices, looking at the status today, and looking at the progress that has been made, one should feel fairly comfortable with predicting their widespread usage in the future. This is not to predict that these devices will be without problems of bleeding, infection, and thromboembolic complications, but that such problems will be much reduced and the devices will provide a satisfactory lifestyle for patients in the future. REFERENCES Al-Mondhiry, H., W. E. Pae, G. Rosenberg, and W. S. Pierce. 1989. Hematologic abnormalities and thromboembolic complications in calves implanted with pneumatic artificial hearts: Long-term studies and autopsy findings . Transactions of the American Society of Artificial Internal Organs 35:238-241. Ashley, S. 1990. Ceramic metal composites. Mechanical Engineering Magazine (July), pp. 46-51. Atsumi, K., Y. Sezai, T. Fujita, S. Nitta, N. Sato, and T. Horiuchi. 1989. Chapter 12: Current status of clinical application of ventricular assist devices in Japan. In: F. Unger, ed. Assisted Circulation 3. New York: Springer-Verlag, pp. 152-159. Aufiero, T. X., J. A. Magovern, G. Rosenberg, et al. 1987. Long-term survival with a pneumatic total artificial heart (pTAH) . Transactions of the American Society of Artificial Internal Organs 33:157-161. Baldwin, J. J., J. M. Tarbell, et al. 1988. Hot film wall shear probe measurement inside a ventricular assist device. Transactions of the ASME Journal of Biomechanical Engineering 110:326-332. Baldwin, J. T., J. M. Tarbell, et al. 1989. Mean flow velocity patterns within a ventricular assist device. Transactions of the American Society of Artificial Internal Organs 35:429-433. Butler, K., T. Brown, Y. Nosé, R. Navarro, and H. Emoto. 1989. Development of a thermal powered LVAS. In: J. C. Norman, ed. Cardiovascular Science and

OCR for page 211
The Artificial Heart: Prototypes, Policies, and Patients Technology: Basic & Applied, I (précise of proceedings). Louisville, Ky.: Oxymoron Press, pp. 222-223. Capecchi, J. T., W. S. Haworth, J. A. Tucker, and P. M. Williams. 1989. Optimizing polyurethanes for fatigue resistance. In: J. C. Norman, ed. Cardiovascular Science & Technology, Basic & Applied, I (précise of proceedings). Louisville, Ky.: Oxymoron Press, pp. 163-164. Cooley, D. A., D. Liota, G. L. Hallman, R. D. Bloodwell, R. D. Leachman, and J. D. Millan. 1969. Orthotopic cardiac prosthesis for 2-stage cardiac replacement. American Journal of Cardiology 24:723-730. DeBakey, M. E. 1934. A simple continuous flow blood transfusion instrument. New Orleans Medical and Surgical Journal 87:386-389. Dennis, C. 1979. Historical background. In: F. Unger, ed. Assisted Circulation. New York: Springer-Verlag, pp. 1-2. DeVries, W. C. 1988. The permanent artificial heart: Four case reports. Journal of the American Medical Association 259:849-859. Emoto, H., H. Harasaki, et al. 1988. Systemic and local effects of heat dissipation in thermally powered LVAS. Transactions of the American Society of Artificial Internal Organs 34:361-366. Farrar, D. J., D. Hill, L. A. Gray, Jr., G. Pennington, L. R. McBride, W. S. Pierce, et al. 1988. Heterotopic prosthetic ventricles as a bridge to cardiac transplantation: A multicenter study in 29 patients. New England Journal of Medicine 318:333-340. Fukumaga, S., Y. Hamanaka, H. Ishihara, et al. 1989. Implantable artificial hearts. In: F. Unger, ed. Assisted Circulation 3. New York: Springer-Verlag, pp. 397-405. Gaykowski, R., K. Taylor, and W. G. Yates. 1988. Cumulative clinical experience with the Symbion Jarvik 7 total artificial heart. Transactions of the American Society of Artificial Internal Organs 34:455-459. Graham, T.R., M. T. Marriman, O. H. Frazier, J. A. Macoviak, K. Dasse, V. L. Poirier, et al. 1989. Current clinical experience with Medtronic-Hancock porcine xenograft valves in the Thermo Cardiosystems implanted left ventricular assist device. Presented at International Symposium for Heart Valve Diseases. Gristina, A. G., J. J. Dobbins, B. Giammara, J. C. Lewis, and W. C. DeVries. 1988. Biomaterial-centered sepsis and the total artificial heart: Microbial adhesion vs. tissue integration. Journal of the American Medical Association 259:870-874. Harasaki, H., R. Kiraly, S. Murabayashi, et al. 1979. Cross-link gelatin as a blood contacting surface. Artificial Organs 3(Suppl.):216. Himley, S., K. Butler, A. Massiello, et al. 1990. Development of the E4T electro-hydraulic total artificial heart. Transactions of the American Society of Artificial Internal Organs (in press). Jaszawalla, T. S., M. F. Daniel, et al. 1988. In vitro and in vivo testing of a totally implantable left ventricular assist system. Transactions of the American Society of Artificial Internal Organs 34:470-475. Joyce, L. D., et al. 1988. Nine years experience with clinical use of total artificial hearts and cardiac support devices. Transactions of the American Society of Artificial Internal Organs 34:703-707. Khanwilkar, P. S., K. R. Crump, et al. 1989. Development of the physiological

OCR for page 211
The Artificial Heart: Prototypes, Policies, and Patients control scheme for an electrohydraulic TAH. Proceedings of 11th International Conference of IEEE & EMBS. Arlington, Va.: Core Communications. Kriett, J. M., and M. P. Kaye. 1990. The Registry of the International Society for Heart Transplantation: 7th official report—1990. Journal of Heart Transplantation 9:323-330. Kung, R. T. V., et al. 1989. A unique left-right flow imbalance compensation scheme for an implantable artificial heart. Transactions of the American Society of Artificial Internal Organs 35:468-470. Kunin, C. M., J. J. Dobbins, J. C. Melo, M. M. Levinson, K. Love, L. D. Joyce, et al. 1988. Infectious complications in four long-term recipients of the Jarvik-7 artificial heart. Journal of the American Medical Association 259:860-864. Lamson, T., O. Ojon, G. Riggins, J. Tarbell, et al. 1988. The two-phase volume compensation chamber: A potential solution to the compliance problem in ventricle assist devices . Presented at the Winter Annual Meeting American Society of Mechanical Engineers, Chicago, Ill., November 1988. LeGallois, C. J. J. 1813. Experience on the Principles of Life. Philadelphia, Pa.: Thomas. Translation of: C. J. J. LeGallois. 1812. Experiences sur les Principlesde Vie. Paris, France . Liota, D., C. W. Hall, S. H. Walter, D. A. Cooley, E. S. Crawford, and M. E. DeBakey. 1963. Prolonged assisted circulation during and after cardiac and aortic surgery. Prolonged partial left ventricular bypass by means of an intra-corporeal circulation . American Journal of Cardiology 12:399-405. Mabuchi, K., M. Najahima, K. Imachi, et al. 1988. Atrial natriuretic polypeptide: A course of pathophysiology in total artificial heart animals? Transactions of the American Society of Artificial Internal Organs 34:463-465. Macoviak, J. A., R. A. Dasse, and V. L. Poirier. 1990. Mechanical cardiac assistance and replacement. Heart Transplantation 8:39-53. McGee, M., S. Parnis, T. Nakatani, T. Myers, K. Dasse, W. E. Hare, et al. 1989. Extended clinical support with an implanted left ventricular assist device. Transactions of the American Society of Artificial Internal Organs 35:614-616. McGoon, D. C. 1984. The risk of thromboembolism following valvular operations: How does one know? Journal of Thoracic and Cardiovascular Surgery 88:782-786. Miller, C. A., W. E. Pae, Jr., and W. S. Pierce. 1990. Combined Registry for the Clinical Use of Mechanical Ventricular Assist Pumps and the Total Artificial Heart in conjunction with heart transplantation: Fourth official report—1989 . Journal of Heart Transplantation 9:453-458. Millner, A., et al. 1990. Development of an efficient electrohydraulic total artificial heart . Transactions of the American Society of Artificial Internal Organs (in press). Mitchell, R.S., D. C. Miller, E. B. Stinson, P. Oyer, S. Jamison, J. Baldwin, et al. 1988. Significant patient related determinants of prosthetic valve performance . Journal of Thoracic and Cardiovascular Surgery 91:807-817. Moulopoulos, D., S. R. Topaz, and W. J. Kolff. 1962. Extracorporeal assistance to the circulation at intra-aortic balloon pumping. Transactions of the American Society of Artificial Internal Organs 8:86-88. Nakatani, T., O. H. Frazier, M. G. McGee, S. W. Parnis, K. Dasse, M. Duncan, et al.

OCR for page 211
The Artificial Heart: Prototypes, Policies, and Patients 1989. Extended support prior to cardiac transplantation using a left ventricular assist device with textured blood contacting surfaces . Presented at the 7th World Congress of International Society for Artificial Organs, Sapporo, Japan, October 1-4, 1989. Nojiri, C., S. W. Kim, H. A. Jacobs, et al. 1989. Scanning and transmission electron microscopic evaluation of the U-100 total artificial heart blood contacting surface . Transactions of the American Society of Artificial Internal Organs 35:679-683. Odermatt, R. 1989. Personal communication. Cardiovascular Research Institute, Sion, Switzerland. Pae, W. E., Jr., and C. A. Miller. 1990. Verbal communication (August 6, 1990): American Society of Artificial Organs and International Society for Heart Transplantation clinical registry of mechanical ventricular assistance project . Pae, W. E., G. Rosenberg, and W. S. Pierce. 1987. Experimental and clinical experience with the Penn State University total artificial heart. Life Support Systems 5:259-265. Pierce, W. S. 1986. The artificial heart—1986: Partial fulfillment of a promise. Transactions of the American Society of Artificial Internal Organs 32:5-10. Pierce, W.S., J. H. Donachy, G. Rosenberg, and R. E. Baier. 1980. Calcification inside artificial hearts: Inhibition by warfarin-sodium . Science 208:601-603. Portner, P. 1987. The Novacor heart assist system: Development, testing and initial clinical evaluation. In: T. Akutsu, ed. Artificial Heart 2. Proceedings ofthe 2nd International Symposium on the Artificial Heart and Assist Devices. Tokyo, Japan: Springer-Verlag, pp. 89-97. Rosenberg, G., T. J. Cleary, A. J. Snyder, et al. 1985. A totally implantable artificial heart design. Transactions of the American Society of Mechanical Engineers 85:WA/DE-11. Rosenberg, G., A. J. Snyder, D. L. Landis, et al. 1984. An electric motor-driven total artificial heart: Seven months survival in the calf. Transactions of the American Society of Artificial Internal Organs 30:69. Schistek, R. 1989. Various designs of nonpulsatile blood pumps. In: F. Unger, ed. Assisted Circulation 3. New York: Springer-Verlag, pp. 235-242. Schuder, J. C., J. H. Gold, and H. E. Stephenson. 1971. An inductively coupled RF system for the transmission of 1 KW of power through the skin. IEEE Transactions in Biomedical EngineeringBME-18. Schwartz, M. M. 1983. Composites Materials Handbook. New York: McGraw-Hill. Sherman, C., B. Daly, W. Clay, et al. 1984. In vivo evaluation of a transcutaneous energy transmission system . Transactions of the American Society of Artificial Internal Organs 30:143. Sherman, C., B. Daly, K. Dasse, W. Clay, M. Szycher, et al. 1983. Research and development: Systems for transmitting energy through intact skin. First Technical Report XC NO1-HV-0-2903-3, Devices & Technology Branch, National Heart, Lung, and Blood Institute. Snyder, A. J., G. Rosenberg, D. L. Landis, W. Weiss, and W. S. Pierce. 1986. Control. In: E. S. Bucherl, ed. Proceedings of the Second World Symposium on the Artificial Heart. Braunschweig, Wiesbaden: Friedr. Vieweg & Son, pp. 167-190.

OCR for page 211
The Artificial Heart: Prototypes, Policies, and Patients Total Artificial Heart. 1985. Part V. In: T. Akutsu, ed. Artificial Heart 1:Proceedings of the 2nd International Symposium on Artificial Heart and Assist Devices. Tokyo, Japan: Springer-Verlag, pp. 161-192. Total Artificial Heart. 1987. Part VI. In: T. Akutsu, ed. Artificial Heart 2. Proceedings of the 2nd International Symposium on Artificial Heart and Assist Devices. Tokyo, Japan: Springer-Verlag, pp. 283-362. Total Artificial Heart. 1989. Part 4. In: F. Unger, ed. Assisted Circulation 3. New York: Springer-Verlag, pp. 329-407. Weiss, W., G. Rosenberg, and A. J. Snyder. 1990. Permanent circulatory support system at The Pennsylvania State University . IEEE Transactions on Biomedical Engineering 37:138-145. Wisman, C. B., W. S. Pierce, J. H. Donachy, et al. 1982. A polyurethane trileaflet cardiac valve prosthesis: In vitro and in vivo studies. Transactions of the American Society of Artificial Internal Organs 28:164-168.