Ann Thorac Surg 2002;73:1933-1938
© 2002 The Society of Thoracic Surgeons
Original article: cardiovascular
Platelet activation, aggregation, and life span in calves implanted with axial flow ventricular assist devices
Trevor A. Snyder, BSEa,c,
Mary J. Watach, BSb,c,
Kenneth N. Litwak, DVM, PhDb,c,
William R. Wagner, PhD*a,b,c
a Department of Bioengineering, Pittsburgh, Pennsylvania, USA
b Department of Surgery, Pittsburgh, Pennsylvania, USA
c McGowan Institute for Regenerative Medicine, University of Pittsburgh, Pittsburgh, Pennsylvania, USA
Accepted for publication February 7, 2002.
* Address reprint requests to Dr Wagner, 328 Scaife Hall, University of Pittsburgh, 3550 Terrace St, Pittsburgh, PA 15213 USA
e-mail: wagnerwr{at}msx.upmc.edu
 |
Abstract
|
|---|
Background. A variety of rotary blood pumps are under development worldwide to serve as chronic ventricular assist devices (VADs). Historically VADs have been associated with thrombotic and thromboembolic complications, yet the ability to evaluate the thrombotic process in preclinical device testing has been limited.
Methods. We have developed and applied flow cytometric assays for activated platelets, platelet microaggregates, and platelet life span and consumption to calves implanted with an axial flow VAD and calves undergoing a sham surgical procedure.
Results. Surgical sham calves had significant increases in circulating activated platelets (p < 0.05) that resolved within 17 days, and no increases in circulating platelet microaggregates. Calves with uneventful VAD implant periods had early transient elevations in platelet microaggregates and prolonged elevations in activated platelets that did not recover to preoperative values during the study. Daily platelet consumption in VAD implanted calves was increased by 20% ± 3%. Calves with thrombotic deposition within the VAD and elevated thromboembolism observed at autopsy experienced increases in circulating activated platelets and microaggregates at the end of the implant period when VAD flow decreased.
Conclusions. This study demonstrates the ability of flow cytometry-based platelet assays to differentiate VAD implant operations from VAD support, and suggests differences that exist between uneventful VAD support and support with complications. These techniques should have value in evaluating other cardiovascular devices undergoing preclinical testing and provide insight into the temporal impact of these devices on the hemostatic system.
|
Introduction
|
|---|
For patients awaiting cardiac transplantation, the insertion of ventricular assist devices (VADs) to provide temporary circulatory support has become increasingly common. While current pulsatile VAD designs have demonstrated an ability to markedly improve patient circulatory health before the transplantation procedure, complications such as thromboembolism, bleeding, and infection remain of major concern [1, 2]. With the limited supply of donor organs placing restrictions on cardiomyopathy patients eligible for VAD bridge to transplant support, the concept of using chronically implanted VADs as an alternative to cardiac transplantation has arisen. Given the size and durability issues associated with pulsatile VADs, many investigators believe that VADs, which use centrifugal or axial flow rotary pump designs, may be more appropriate for chronic patient support. A number of such rotary VAD designs are currently under development and beginning to enter clinical trials for extended patient support [3–5].
Rotary VADs differ markedly in design from current pulsatile VADs. In terms of potential sources for thrombotic deposition and thromboemboli generation, the rotary pumps have a smaller total blood-contacting surface area and lack valves. The pumps also generally use titanium alloys for interior pump surfaces and expose blood to very high, transient shear stresses from rotors that spin in the 5,000 to 12,000 rotation per minute range.
Several groups have studied hemostatic pathway activation by pulsatile VAD designs; the motivation for these studies are often found in thromboembolic rates ranging from 5% to 30% [6–8]. The application of monoclonal antibody (MAb) technology to the assessment of coagulation (with enzyme linked immunoassays) and circulating activated platelets (with flow cytometric techniques) has permitted greater understanding of the hemostatic alterations and potential thromboembolic mechanisms associated with clinical pulsatile VAD support [9, 10]. Such studies were not possible at the time these devices were in pre-clinical testing. With the next generation of VADs it should be possible to use these methods to characterize hemostatic alterations before clinical testing, provided assays with adequate reactivity in the large animal model could be developed and applied.
In previous reports we have described the development of flow cytometric based assays for circulating activated platelets and platelet microaggregates in calves implanted with an axial flow VAD, as well as a flow cytometric technique for assessing platelet life span in calves with VADs [11, 12]. In this report we apply these previously described assays along with an annexin V-based assay for circulating activated platelets to 16 calves implanted with an axial flow ventricular device to characterize the response of calves to this VAD. To evaluate the effect of the substantial tissue damage associated with the implant operation and to attempt to separate this effect from that of VAD implantation, 8 calves underwent a sham surgical procedure. The temporal response of activated platelets and microaggregates was also related to device performance in general terms through evidence of kidney infarcts, thrombotic deposition on the device, and whether the VAD performed optimally or experienced low volumetric output during the implant period.
|
Material and methods
|
|---|
Surgical procedure
Sixteen Jersey calves were housed in the large animal facilities of the University of Pittsburgh. All procedures involving animals were conducted in compliance with the "Guide for the Care and Use of Laboratory Animals" published by the National Institutes of Health. The calves were implanted with the Heartmate II axial flow VAD (Thoratec, Pleasanton, CA), shown in Figure 1,
and recently implanted clinically [13, 14]. The system consists of a titanium alloy (Ti-6Al-4V) inflow cannula with a flexible strain relief section made of Dacron (C. R. Bard, Haverhill, PA), a titanium alloy pump, and a Dacron outflow cannula. The Dacron portions of the cannulas were preclotted with the animal’s blood before implantation. The calves were premedicated with 45 mg of atropine sulfate (American Pharmaceutical Partners, Schaumburg, IL), 45 mg. Anesthesia was induced with methohexital sodium (Jones Pharma Inc, St. Louis, MO), 10 mg/kg. Following intubation, anesthesia was maintained with isoflurane in oxygen and room air. The chest was entered through the fifth intercostal space. The descending aorta was isolated to receive the outflow cannula. All blood-contacting surfaces of the pump were rinsed three times with saline. A plastic anti-kinking guard was placed around the outflow cannula, which was then sutured to the aorta end-to-side with 4-0 monofilament polypropylene. Heparin at 1.5 mg/kg was administered when the aorta was partially clamped and then again when the anastomosis was completed. The heart was then prepared with four size 0 braided Dacron pledgeted sutures. The pump was pre-assembled to both cannulas after being filled with saline. The apex of the left ventricle was cored and the inflow cannula was inserted. The pledgeted sutures were attached to the sewing ring, then the pump was de-aired and started. A Transonic flow probe (Transonic Systems Inc, Ithaca, NY) was attached around the distal section of the outflow graft. Warfarin (DuPont Pharma, Wilmington, DE) was administered daily to maintain an international normalized ratio of 2.5 to 3.5. Typical operating measurements were volumetric flow rates of 4 to 7 L per minute at 9,000 to 12,000 revolutions per minute of the pump rotor.
View larger version (97K):
[in this window]
[in a new window]
|
Fig 1. The Heartmate II axial flow ventricular assist device (Thoratec, Pleasanton, CA) shown with the impeller removed and placed next to the housing. Power and control are provided through a percutaneous connection. The inflow cannula is connected to the right side of the pump. The cannula consists of a textured titanium blood contacting surface and a flexible polymer section (the white segment), which allows the pump to move independently of the heart to prevent cannula impingement against the ventricular wall.
|
|
In 8 surgical sham calves the same procedure was used with the following alterations. After the preclotted Dacron cannula was sutured to the aorta it was sutured shut approximately 3 centimeters distal to the anastomosis with a running suture of 4-0 monofilament polypropylene. One pledgeted 0 braid Dacron suture was placed in the apex of the heart in a horizontal mattress pattern. A stab incision was then made in the apex of the heart. The sutures were tied, thus closing the incision in the heart. Two sham cables were tunneled into the intercostal muscles before chest closure. These animals also received the same prophylactic antibiotic treatment and anticoagulation therapy as the VAD implanted calves.
At the conclusion of the implant an autopsy was performed for all calves, which included an evaluation of infarcts in the kidneys and thrombotic deposition on pump surfaces. The number and size (greater than or less than 1 cm2) of the kidney infarcts were noted.
Blood collection
Whole blood was collected by jugular venipuncture into a syringe using an 18-gauge butterfly needle. The first 5 to 8 mL of blood were collected, but not used for this study, and 2.7 mL of the remaining blood was transferred into a 3 mL S-monovette tube containing 3.8% trisodium citrate (Sarstedt, Newton, NC) at a final concentration of 1 part anticoagulant to 9 parts whole blood.
Measurement of circulating activated platelets and microaggregates with MAbs
Previously described assays to measure circulating activated platelets were used with the concentration of antibodies to detect activated platelets modified to 1.5 µg/mL instead of the 15 µg/mL previously used [11]. Four MAbs were used in this study: ColiS69A, BAQ125, GC5, and CAPP2A (Washington State University Monoclonal Antibody Center, Pullman, WA). ColiS69A served as an IgG1 isotype control. BAQ125 and GC5 served as markers for activated bovine platelets. CAPP2A was used as a platelet marker to identify circulating microaggregates.
Measurement of circulating activated platelets annexin V
Circulating platelets expressing a procoagulant surface were quantified with annexin V binding by diluting 20 µL of whole blood 1:10 with phosphate buffered saline (PBS) (Gibco BRL, Grand Island, NY) in each of four tubes containing 250 µL annexin V binding buffer (10 mmol/L Hepes buffer at pH 7.4, 140 mmol/L NaCl, 2.5 mmol/L CaCl2) and 5 µL annexin V-FITC (Pharmingen, San Diego, CA). The samples were incubated for 30 minutes in the dark at room temperature followed by the addition of 500 µL of 1% paraformaldehyde for fixation. During flow cytometric analysis of the preoperative sample, a threshold was established such that 2% of the platelets had flourescence intensities greater than the threshold level. Platelets in subsequent samples with fluorescence intensities greater than the threshold were considered activated.
In vitro verification of annexin V binding to activated platelets
To demonstrate the ability of annexin V-FITC to quantify platelet activation in bovine blood, citrated blood was incubated for 20 minutes with one of four platelet agonists: 20 µmol/L adenosine diphosphate, 0.1 U/mL bovine thrombin, 5 µmol/L calcium ionophore A23187, or 5 µmol/L calcium ionophore ionomycin (Sigma, St. Louis, MO). A control sample was allowed to stand undisturbed for 20 minutes. Samples were then prepared as described previously.
Platelet life span determination
The biotinylation procedure previously reported by Baker and colleagues [12] was used without modification to label 360 mL of collected blood, which was subsequently reinfused. Blood samples were drawn 1 hour after this reinfusion and subsequently at least once daily until no biotin-labeled platelets were detected. Flow cytometric detection of biotinylated platelets was performed as before [12] with two minor alterations: streptavidin-FITC (Calbiochem, LA Jolla, CA) was used to bind biotin on platelet surfaces and the samples were washed with 500 µL PBS before sample fixation. The modifications were made to reduce nonspecific binding. Platelet consumption was determined using the formula given by Harker and Hanson [15].
Statistical analysis
All data in the figures are reported as mean plus standard deviation. Significant differences between unactivated and in vitro activated samples for annexin V binding were determined using a paired Student’s t test with Bonferroni correction. Analysis of variance with posthoc Neuman-Keuls testing was used to determine significant differences in activated platelets or microaggregates between preoperative and postoperative values for each factor, as well as when surgical sham data were compared with uneventful pump data. Comparisons were made between the sham and uneventful pump animals for preoperative controls and postoperative days 1, 8, 17, and 27. These time points were selected to examine immediate postoperative status as well as the time course of recovery. For all evaluations statistical significance was considered to exist at p less than 0.05.
|
Results
|
|---|
Annexin V detection of activated platelets in vitro
The data in Table 1
demonstrate a significant increase in annexin V binding to bovine platelets stimulated with calcium ionophores and 20 µmol/L adenosine diphosphate, but not 0.1 u/mL thrombin.
Surgical sham procedure calves
For each of the 8 surgical sham calves, all indices of circulating activated platelets (binding of BAQ125, GC5, and annexin V) significantly increased from relatively low preoperative values after the operation (Fig 2).
The platelet activation levels remained significantly elevated versus preoperative control values for up to 13 days. No significant increase in microaggregates was observed in surgical sham animals at any time.
Uneventful VAD implanted calves
An uneventful postoperative course was one in which there were no significant bleeding events, infections, or periods of low pump flow (< 3 L per minute). At device explantation there was no evidence of thrombotic deposition on the interior VAD surfaces. In these 7 calves, all three indicators of circulating activated platelets, as well as microaggregates, increased significantly after operation as shown in Figure 3.
Microaggregate levels returned to preoperative control values within the first postoperative week, however, all circulating activated platelet indices remained significantly elevated compared with preoperative controls. Circulating platelets positive for annexin V binding stabilized at 30% to 40%, and platelets positive for MAb BAQ125 and GC5 binding remained at 50% to 70% during the study period. There was no significant difference for the preoperative values for any of the assays between the sham and uneventful course animals. In postoperative comparisons on days 1, 8, 17, and 27, BAQ125 and GC5 were significantly elevated versus the sham animals on all days, annexin V was significantly increased on days 17 and 27, and microaggregates were only significantly different on day 1.
View larger version (20K):
[in this window]
[in a new window]
|
Fig 3. Circulating activated platelets and microaggregates detected in seven calves with uneventful ventricular assist device implants. The three activated platelet indicators, GC5, BAQ125 and annexin V, followed similar trends throughout the implant period and remained significantly elevated relative to preoperative values. Transient significant increases in microaggregates resolved at postoperative day 8.
|
|
Calves with early study termination caused by low VAD flow rate
Data collected from 3 calves in which the VAD implantation was terminated early because of low VAD flow rates are shown in Figure 4.
Adherent thrombi were found in the outflow region of the pumps in all three of these studies. In all 3 calves, microaggregate levels either remained elevated or increased before conclusion of the study. In 2 of the 3 calves the three circulating activated platelet indicators all increased in the final measurement before the study was stopped. In the third calf both BAQ125 and annexin V increased in the final measurement, whereas the other circulating activated platelet index, GC5, was already highly elevated.
View larger version (27K):
[in this window]
[in a new window]
|
Fig 4. Circulating activated platelets and microaggregates in three ventricular assist devices implanted in calves, in which the study was prematurely terminated because of low flow rates. In all 3 cases at least three of the four indicators increased before the study terminated. Microaggregates increased before termination in 2 calves (1.2% to 25.8% for the middle graph; 0.6% to 5.4% for the bottom graph). (Pre-op = preoperative.)
|
|
Autopsy findings
At autopsy no kidney infarcts were observed in any of the surgical sham animals. For the 7 animals with an uneventful postoperative course after VAD implantation, less than five small (< 1 cm2) infarcts were observed. For all 3 of the animals with study termination caused by VAD flow rates, at least 1 large infarct (> 1 cm2) was observed in addition to several small infarcts.
Platelet life span and consumption
Figure 5
shows the preoperative and postoperative platelet life spans for 7 VAD implanted animals. All postoperative platelet life span determinations were performed at least 14 days after the operation. This data was collected for 1 animal in which the platelet activation indicators were also measured (1 of the uneventful VAD implanted calves), and for 6 other animals that were in stable condition and in which pump performance was normal during the platelet life span determinations. A consistent decrease in platelet life span of 42% ± 5% was accompanied by an increase in platelet daily consumption by 20% ± 3%. Platelet life span determinations were performed on 4 of the surgical sham animals beginning by postoperative day 14. These platelet life spans were 5.7, 6.8, 6.0, and 5.4 days, respectively, and were significantly greater than the postoperative life spans in Figure 5, but not different than the preoperative life spans shown in the same figure.
View larger version (18K):
[in this window]
[in a new window]
|
Fig 5. Consistent change in platelet life span in seven calves implanted with ventricular assist devices. The platelet life span decreased by an average of 42%. Daily platelet consumption in these animals increased by an average of 20%.
|
|
|
Comment
|
|---|
Annexin V binding to the surface of procoagulant platelets has been used as a method to characterize circulating activated platelets in a variety of clinical studies [16–18] and has been previously reported to have the ability to detect platelet activation in the bovine model by Harasaki and colleagues [19]. We have generally confirmed the results of Harasaki and colleagues [19] showing increased bovine platelet binding of annexin V after in vitro stimulation with calcium ionophores and adenosine diphosphate, but not thrombin at 0.1 u/mL. The percent of circulating platelets binding annexin V in all of the surgical sham procedures and VAD implanted calves tended to be lower than the percent of platelets binding MAbs GC5 and BAQ125, but the trends in platelet activation reflected by these assays were largely synchronous. The MAbs have been shown to preferentially bind to activated bovine platelets [11], but the epitopes bound have not been characterized. Speculation as to specific cellular events detected by these antibodies thus is not currently possible.
The evaluation of 8 calves undergoing a sham surgical procedure allowed for the separation of surgical effects of VAD placement from blood contact with the operating device. The invasive nature of the implantation procedure, including coring the left ventricle and creating the aortic anastomosis, would be expected to produce a nidus for platelet activation in the early postoperative period together with the soft tissue surgical wounds. The data collected from the surgical sham procedure on calves clearly demonstrated a significant increase in all of the activated platelet indicators after the operations and a return to preoperative values within 17 days. Of note, no detectable increase in circulating platelet microaggregates was observed.
These results clarify the substantial early contribution to platelet activation from the implantation procedure, demonstrating the need to be cautious in attributing increases in activated platelets (but not microaggregates), to device operation in this early postoperative period. The data have implications for cardiovascular device testing associated with similarly invasive procedures and also suggest that when rotary VADs are evaluated in a clinical setting for hemostatic disturbances, appropriate surgical control patient groups be used (e.g., those undergoing coronary artery bypass grafting).
A significant increase in all three circulating activated platelet indicators after the operations was also found in the VAD implanted calves with uneventful postoperative courses. However the activated platelet indicators were significantly elevated versus the sham implants beyond the surgical recovery period and did not return to preoperative control values during the study durations. Also in variance with the surgical sham calves, microaggregates significantly increased after VAD implantation before returning to preoperative control values within 1 week after the operations. The decrease in platelet life span after VAD implantation, and the corresponding increase in platelet consumption, suggested removal of activated platelets or microaggregates from circulation. The kidney infarct observations also indicated that a level of thromboembolic activity occurred in VAD implanted animals that was not present in the surgical sham controls. These observations support the expected conclusion that implantation of the substantial artificial surface associated with the VAD, perhaps in addition to elevated shear stresses, was associated with an increase in platelet activation and subsequent thromboembolism. However, since platelet consumption and activation is known to occur in the presence of many successful cardiovascular devices, the presence of such activation should not be considered incompatible with successful device performance.
In the 3 calves in which thrombus was visually evident in the outflow region of the pump at explant, there was generally an increase in the activated platelets and microaggregates before study termination. During the immediate postoperative period platelet activation indices were higher than for uneventful pump animals, suggesting potential to detect problematic implants early in the implant period. These animals also consistently exhibited evidence of increased thromboembolic activity in terms of renal infarcts with respect to calves with uneventful VAD implantation periods. The temporal monitoring of circulating activated platelets and platelet microaggregates may provide insight into relative device performance before study termination and may thus provide a basis for comparison between different device designs in terms of blood biocompatibility.
A limitation in this study is set by the expense associated with performing large numbers of device implants in the calf model. Thus some of the conclusions reached regarding the subset of calves with early study termination remain qualitative in nature based on the limited, but consistent findings in this small group. However this report does indicate the potential value in pursuing more sophisticated assays for monitoring thrombotic activity during the device implantation as opposed to relying largely on explant data and platelet concentration measurements during the implant period.
In summary we have demonstrated the ability of flow cytometric platelet assays to temporally characterize the implantation period of an axial flow VAD in preclinical testing and to differentiate sham control operations in calves from calves with uneventful implantation periods. Likewise 3 animals that had complications leading to termination of the implant period not only differed from calves with uneventful implants in terms of thrombotic deposition on the pump and kidney infarcts, but circulating activated platelets and microaggregates were elevated in the early implant period. Improvements in the measurement of cardiovascular device thrombosis and thromboembolism at the preclinical trial stage will offer investigators and manufacturers the opportunity to alter device designs and materials to address potential biocompatibility concerns before entering clinical trials, where the regulatory burden and expense associated with design alteration remain onerous.
|
Acknowledgments
|
|---|
The authors would like to thank Ken Butler and the Nimbus Corporation for their assistance, cooperation, and support of these studies. The authors would also like to thank the staff of the Experimental Surgery Program at the University of Pittsburgh for assistance with blood collection and sampling, as well as Patrick McGinley, Peter Hurh, Mirrhet Birru, Nicole Price, Anika Joseph, Claudia Grossman, and Aisha Moore for help in sample preparation. This study was funded by the National Institutes of Health, contract # NO1-HV-58155.
|
References
|
|---|
-
Argenziano M., Oz M.C., Rose E.A. The continuing evolution of mechanical ventricular assistance. Curr Probl Surg 1997;34:322-386.
-
Goldstein D.J., Oz M.C., Rose E.A. Implantable left ventricular assist devices. N Engl J Med 1998;339:1522-1533.[Free Full Text]
-
Butler K., Thomas D., Antaki J., et al. Development of the Nimbus/University of Pittsburgh axial flow left ventricular assist system. Artif Organs 1997;21:602-610.[Medline]
-
Westaby S., Katsumata T., Houel R., et al. Jarvik 2000 heart: potential for bridge to myocyte recovery. Circulation 1998;98:1568-1574.[Abstract/Free Full Text]
-
Wieselthaler G.M., Schima H., Hiesmayr M., et al. First clinical experience with the DeBakey VAD continuous-axial-flow pump for bridge to transplantation. Circulation 2000;101:356-359.[Abstract/Free Full Text]
-
Mehta S.M., Aufiero T.X., Pae W.E., Jr, Miller C.A., Pierce W.S. Combined registry for the clinical use of mechanical ventricular assist pumps and the total artificial heart in conjunction with heart transplantation: sixth official report—1994. J Heart Lung Transplant 1995;14:585-593.[Medline]
-
De Rose J.J., Jr, Argenziano M., Sun B.C., Reemtsma K., Oz M.C., Rose E.A. Implantable left ventricular assist devices: an evolving long-term cardiac replacement therapy. Ann Surg 1997;226:461-470.[Medline]
-
El-Banayosy A., Deng M., Loisance D.Y., et al. The European experience of Novacor left ventricular assist (LVAS) therapy as a bridge to transplant: a retrospective multi-centre study. Eur J Cardiothorac Surg 1999;15:835-841.[Abstract/Free Full Text]
-
Spanier T., Oz M., Levin H., et al. Activation of coagulation and fibrinolytic pathways in patients with left ventricular assist devices. J Thorac Cardiovasc Surg 1996;112:1090-1097.[Abstract/Free Full Text]
-
Wilhelm C.R., Ristich J., Knepper L.E., et al. Measurement of hemostatic indices in conjunction with transcranial doppler sonography in ventricular assist device patients. Stroke 1999;30:2554-2561.[Abstract/Free Full Text]
-
Baker L.C., Davis W.C., Autieri J., et al. Flow cytometric assays to detect platelet activation and aggregation in device-implanted calves. J Biomed Mater Res 1998;41:312-321.[Medline]
-
Baker L.C., Kameneva M.V., Watach M.J., Litwak P., Wagner W.R. Assessment of bovine platelet life span with biotinylation and flow cytometry. Artif Organs 1998;22:799-803.[Medline]
-
Butler K.C., Dow J.J., Litwak P., Kormos R.L., Borovetz H.S. Development of the Nimbus/University of Pittsburgh innovative ventricular assist system. Ann Thorac Surg 1999;68:790-794.[Abstract/Free Full Text]
-
Griffith B., Kormos R., Borovetz H., et al. Heartmate II left ventricular assist system: from concept to first clinical use. Ann Thorac Surg 2001;71:S116-S120.[Abstract/Free Full Text]
-
Harker L.A., Hanson S.R. Experimental arterial thromboembolism in baboons. Mechanism, quantitation, and pharmacologic prevention. J Clin Invest 1979;64:559-560.
-
Musselman D.L., Tomer A., Manatunga A.K., et al. Exaggerated platelet reactivity in major depression. Am J Psychiatry 1996;153:1313-1317.[Abstract/Free Full Text]
-
Nomura S., Shouzu A., Omoto S., et al. Effect of cilostazol on soluble adhesion molecules and platelet-derived microparticles in patients with diabetes. Thromb Haemost 1998;80:388-392.[Medline]
-
Tomer A. A sensitive and specific functional flow cytometric assay for the diagnosis of heparin-induced thrombocytopenia. Br J Haematol 1997;98:648-656.[Medline]
-
Harasaki H., Fukamachi K., Benavides M., Manos J., Wika K., Massiello A. A comprehensive hematologic study in calves with total artificial hearts. ASAIO J 1995;41:M266-M271.[Medline]
This article has been cited by other articles:
|
|
|
|
 
K. Doi, L. A. R. Golding, A. L. Massiello, M. W. Kopcak Jr, R. L. Gerhart, S. Schenk, M. Inoue, Y. Ootaki, and K. Fukamachi
Preclinical readiness testing of the arrow international CorAide left ventricular assist system
Ann. Thorac. Surg.,
June 1, 2004;
77(6):
2103 - 2110.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
, and 
, respectively. There was not a significant increase in microaggregates in any of the sham procedure calves.