HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) View Related Video Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Chinchoy, E. Articles by Iaizzo, P. A. Search for Related Content PubMed PubMed Citation Articles by Chinchoy, E. Articles by Iaizzo, P. A.

Ann Thorac Surg 2000;70:1607-1614
© 2000 The Society of Thoracic Surgeons

---

Original articles: cardiovascular

Isolated four-chamber working swine heart model

^a Department of Anesthesiology, University of Minnesota, Minneapolis, Minnesota, USA
^b Department of Biomedical Engineering, University of Minnesota, Minneapolis, Minnesota, USA
^c Department of Physiology, University of Minnesota, Minneapolis, USA
^d Medtronic Inc, Minneapolis, Minnesota, USA

Address reprint requests to Dr Iaizzo, Department of Anesthesiology, Mayo Mail Code 294 UMHC, 420 Delaware St. SE, Minneapolis, MN 55455
e-mail: iaizz001{at}tc.umn.edu

	Abstract

Top Footnotes Abstract Introduction Material and methods Results Comment Acknowledgments References

 
Background. Isolated heart models separate cardiac characteristics from systemic characteristics with subsequent findings used in cardiac research, including responses to pharmacologic, mechanical, and electrical components. The model objective was to develop the ability to represent in situ physiologic cardiac function ex vivo.

Methods. Swine hearts were chosen over rat or guinea pig models due to their notably greater anatomical and physiologic similarities to humans. An in vitro apparatus was designed to work all four chambers under simulated in situ physiologic conditions. Using standard cardiac surgical techniques, 12 porcine hearts (mean weight 331 ± 18 g) were explanted into the apparatus. Preload and afterload resistances simulated in situ input and output physiologic conditions. Hemodynamic characterizations, including cardiac output, max ±dP/dt, and heart rate, were used to determine in situ function leading to explantation (prethoracic operation, postmedial sternotomy, and postperidectomy) and during in vitro function (t = 0, 60, 120, and 240 minutes).

Results. In vitro performance decayed with time, with statistical differences from base line (t = 0) function at t = 240 minutes (p > 0.05).

Conclusions. An isolation and in vitro explantation protocol has been improved to aid in the study of isolated cardiac responses, and to determine cardiac hemodynamic function during open chest operation, transplantation, and in vitro reanimation with a crystalloid perfusate. The resulting model offers similar working physiologic function, with real-time imaging capabilities. The resulting model is advantageous in representing human cardiac function with regard to anatomic and physiologic functions, and can account for atrial and ventricular interactions.

	Introduction

Top Footnotes Abstract Introduction Material and methods Results Comment Acknowledgments References

 
Numerous isolated, in situ, in vitro, and postmortem heart studies have been conducted since Langendorff first described his work in 1895 [1], each proceeding on various and different assumptions to approximate normal physiologic conditions, for eventual correlation to human function. A recent Medline search provided more than 30,000 articles on heart studies conducted by various methods of isolation.

Whereas small isolated animal heart models (guinea pigs, rabbits, etc) have primarily found use in pharmacologic, respiration, and crystalloid perfusate effects [2–4], larger animal heart models (porcine, calf, etc) provide further benefit by better approximating human hemodynamic characteristics both within the heart and for modeling circulation characteristics [5]. The improved similarities in size, musculature, and geometry may better approximate human function (cardiac output, coronary flow, etc) than developed biventricular rat or guinea pig models.

In vitro models possess specific benefits over in vivo models, one of which is the ability to separate cardiac responses from systemic responses with pharmacologic, electrical, and hemodynamic (blood flow and pressure) inputs. In addition, individual characterization of mechanical, electrical, and chemical variables is possible, with minor compromises in contractility associated with denervation [6]. Focus on swine xenotransplantation has also necessitated a further understanding of both human and swine in vivo and ex vivo physiologic function following transplantation [7, 8].

With these specific objectives, our goals were to understand cardiac changes during explantation into the developed in vitro apparatus, by which correlation to cardiac changes during transplantation could be investigated, and to maintain an interactive crystalloid perfused four-chamber working heart model and investigate its performance over time.

	Material and methods

Top Footnotes Abstract Introduction Material and methods Results Comment Acknowledgments References

 
Animal preparation
Mongrel swine (30 to 40 kg) were used following approval from the University of Minnesota Animal Care and Use Committee. Telazol (5 mg/kg) and Xylazene (250 mg) were administered intramuscularly. Intravenous access was then obtained through an ear vein for volume administration (Ringers lactate) and delivery of medication(s) as needed. Thiopental, titrated to effect, was administered before intubation. An endotracheal tube was inserted and the animal was ventilated mechanically with 35% to 40% oxygen and 60% to 65% nitrous oxide, and general anesthesia was maintained with 0.8% halothane. Exhaled halothane and carbon dioxide levels were monitored continuously and adjusted to maintain adequate anesthetic depth (1 to 1.3 MAC) and normocapnia (40 ± 2 mm Hg) throughout the operation (SaraCap Ag, PPG BioMedical Systems, Lenexa, KS). Periodic blood gas analysis was performed (BGElectrolytes, Instrumentation Laboratory, Lexington, MA) to maintain normal electrolyte balance and pH blood gas levels (PaCO₂, PaO₂, K⁺, Na⁺, H⁺, HCO₃^-, Ca²⁺).

In vivo and in situ measurements
Hemodynamic and temperature measurements were simultaneously obtained on a PC using a multiple acquisition system (DI 410 and ATCODAS, Dataq Acquisition Systems, Akron, OH) after amplification (Gould, Valley View, OH) and signal conditioning (5900, Gould). Real-time display aided in proper placement of the catheters. A 5F microtip catheter (MPC 500, Millar, Houston, TX) was inserted through the right carotid artery and fed into the left ventricle. Another 5F Millar and a 7.5F Swan Ganz catheter (93A-931H, American Edwards Laboratories, Puerto Rico) were inserted into the right external jugular vein. The Swan Ganz catheter was floated into the pulmonary artery, while the Millar catheter was kept in the right ventricle. For the in vivo and in situ portions of the experiment, all catheters were secured following proper probe placement. A PE90 tubing catheter was inserted and secured into the femoral artery to measure arterial pressure. A five-lead electrocardiogram was used and cardiac output was estimated with a thermodilution estimator (9520A Cardiac Output Estimator, American Edwards). Core temperature, assessed by the Swan, was kept at 38.5°C ± 0.5°C throughout the in vivo/in situ process by convective surface air heating or cooling. Thermal equilibration within 0.1°C for not less than 20 minutes [20] was verified before proceeding to reduce transient physiologic temperature responses.

Acquisition of base line in vivo and in situ data
Base line in vivo control data were obtained. A medial sternotomy was performed preserving the pericardial sac intact. The ribcage was retracted and the pericardial sac freed from the surrounding tissues. In situ data were recorded to measure myocardial performance with ribcage pressure relieved. The pericardial sac was then excised, and the connective tissue and pericardial fat removed. The measurements were then repeated as outlined above.

After heparinization (10,000 U), the monitoring catheters were removed and an aortic cannula (9F double lumen, Medtronic Inc, Minneapolis, MN) was introduced. The inferior vena cava was tied off and 25 mg of adenosine was administered. Refrigerated modified St. Thomas’ Hospital cardioplegia flow was then introduced, and the aorta and the superior vena cava were cross-clamped. The pulmonary artery was vented and the heart decompressed to prevent distension. Ice and refrigerated saline were applied topically. After arrest, the heart was excised and placed in an iced saline slurry with continuous cardioplegia administration during transportation (less than 5 minutes) to the apparatus, as well as during the (re-)cannulation process (40 to 55 minutes).

After removal of excess tissue and isolation of the great vessels, the in vitro aortic exit cannula with predetermined flow resistance was inserted into the aorta (see Figs 1, 2). The edge of the cannula was 3 to 5 cm distal to the aortic valve to retain aortic compliance. Cannulas were inserted and secured into the pulmonary artery (28F), pulmonary vein (28F), inferior vena cava (28F), and directly into the superior vena cava (36F). Millar catheters were placed in the same ventricular locations as they were in vivo. A pacing/defibrillatory lead (6932, Medtronic), a reference electrode (6921S, Medtronic), an electrocardiac monitor (5358, Medtronic), and a programmer unit (9790C Vitatron, Medtronic) provided defibrillation and electrical pacing if needed.

View larger version (83K): [in this window] [in a new window]   Fig 1. Diagram of the in vitro perfusate circuit. The superimposed image of the heart is of an actual preparation. Perfusate enters the inferior vena cava into the right atrium, flows into the right ventricle, and is ejected through the pulmonary artery. Concurrently, perfusate enters the left side of the heart though the pulmonary vein into the left atrium, into the left ventricle, and is ejected through the aortic exit.

View larger version (19K): [in this window] [in a new window]   Fig 2. The steady-state flow versus pressure curve for the in vitro aortic exit cannula. Flow was measured as a function of pressure for the aortic cannula measuring resistance loaded against the left ventricle in vitro. In a passive fluid system, flow is unique to the pressure difference across the system. Ideally, this aortic cannula fluid resistance curve is the same as the systemic resistance of the swine under physiologic conditions.

In vitro apparatus
A modified Krebs–Henseleit perfusate was used throughout the in vitro portion of the experiments. Several additions were included to maintain cardiac performance. Ethylenediaminetetraacetic acid (0.32 mmol/L) was added to chelated calcium and toxic metal ions. Insulin (10 U/L) was added to aid in glucose utilization. Sodium pyruvate (2.27 mmol/L) was added as an additional energy substrate, and mannitol (16.0 mmol/L) was added to increase perfusate osmolarity, thus reducing cardiac edema.

The perfusate entered a standard reservoir (Minimax, Medtronic) and oxygenator (3381 Hollow Fiber, Medtronic) from where it circulated through the apparatus (see Fig 1). Pharmacologic agents, if needed, were administered into this reservoir. Our protocol, however, excluded animals in which pharmacologic agents were given in vitro.

The volume of the circulating perfusate solution circuit was 3.8 L, an approximation of the swine blood volume. A water jacket maintaining circulated water at 39°C (BioCal 370 BioMedicus-Medtronic) enveloped the perfusate chambers, the oxygenator, and the chamber containing the heart, regulating perfusate temperature and surrounding the heart itself. This jacket maintained measured myocardial temperatures at 37.0°C ± 0.5°C. Langendorff’s method of perfusion was achieved by pumping the perfusate to the aortic cannula from the pump (BioConsole 550 Bio-Medicus-Medtronic) at 65 mm Hg pressure, bypassing the left filling chamber.

During working mode, left and right atrial filling was maintained through filling chambers into the two atria (preload) and against the output from the ventricles (afterload) to maintain stroke work and oxygen consumption, given the differences in ejecting versus isovolumetric and isobaric models [9–11]. In this four-chamber working heart model, it was possible to obtain data related to the role of the pericardium on hemodynamic function, as well as the interaction between both left and right ventricles. [12–14].

Methods of providing atrial filling pressures and flow, and resistance to ventricular ejection were a primary concern. The atria were supplied by passive filling using open-ended glass chambers draining through 0.952-cm diameter Nalgene tubing into the atrial cannulas. By using passive atrial fluid filling systems in both sides within tested pressure curves, filling rates were dependent on relaxation (eg -dP/dt rates and cardiac-regulated chamber volumes), and not on the delivery performance curve of the system. This type of delivery generated constant atrial preload and the circuit could be "closed" (right side output entered left side input) or "open" (left and right filling rates independent allowing intrinsic output of each side to be a function of left or right side pressure). For this protocol, the circuit was kept open. Because the limiting factor in maximal atrial filling has not yet been determined, that is, whether maximum filling flow through the atria is governed primarily by systemic or pulmonary resistance, or by atrial relaxation rate, the maximum flow rate from each fluid chamber to atria was tested. This testing was to ensure that the apparatus was not the determining factor in atrial filling.

To mimic systemic aortic resistance to left ventricular ejection in vitro, the resistance of the in vitro apparatus aortic cannula to flow was measured as a function of pressure. Figure 2 provides the flow versus pressure function of the in vitro aortic cannula providing ventricular resistance, which provided a nonlinear flow and resistance as a function of pressure. This in vitro resistance is ideally matched to the in situ aortic/systemic resistance of the swine.

In vitro cardiac output(s) were measured in line using either a Transonic flow meter (T201, Transonic Systems, Ithaca, NY) or a McMillan flow meter (111 Flometer, McMillan, Austin, TX). Coronary flow was calculated as the difference between pulmonary vein flow and aortic outflow, while stroke volume was calculated as cardiac output divided by heart rate.

Imaging
A transistor based camera (IC6C-13 and ILV-C1, Olympus Optical, Tokyo, Japan) in conjunction with an endoscope (If2D5 to 12 and 3CCD, Olympus) were used to capture videos of internal cardiac movement. The 5.5-mm diameter flexible camera was inserted into the heart through one of the following: the aorta, pulmonary artery, superior vena cava, or right atrial appendage cannula. The camera allowed visual depiction of variations of input to the heart (eg, loading conditions) and for investigation of device interactions. Video was recorded online (UVW-1800 Beta, Sony Inc, Tokyo, Japan) and later digitized at 720 x 540 resolution (Model 1000, Avid Technology Inc, Tewksbury, MA). The images shown were digitized in 0.06-second increments, though much smaller increments were possible with the analog video. Imaging was not performed during monitoring periods, so as not to interfere with data collection.

	Results

Top Footnotes Abstract Introduction Material and methods Results Comment Acknowledgments References

 
In vivo event markers for data collection were: before (pre-)thoracic operation (only necessary placement of probes), postmedial sternotomy (completely intact pericardium), and postperidectomy (pericardial and connective tissue completely removed). Forty animals were sacrificed for regulation of protocol. Twelve hearts with an average isolated weight of 331 ± 18 g were used for data collection.

Mean time from in situ introduction of cardioplegia to in vitro reperfusion was 74 minutes. Upon reperfusion, 4 of 12 hearts required electrical defibrillation. With all hearts, Langendorff perfusion for not less than 30 minutes was requisite for stabilization, followed by four-chamber working mode for more than 90 minutes until stable cardiac performance was perceived (eg, heart rates, and left and right ventricular pressure waveforms).

In vitro, average diastolic pressure conditions were fixed, that is, no individual attempts were made to match pressure profiles to in situ conditions. Cardiac performance was evaluated by deriving maximum ±dP/dt (mm Hg/s). Table 1 shows hemodynamic measurements leading to explantation in conjunction with in vitro performance over time. Table 2 provides the relative changes from the previous event for each measurement, accounting for different initial pressures. In addition to systolic and diastolic pressures, time average pressures were calculated to account for possible differences due to waveform changes. Figure 3 provides typical pressure waveforms collected during each experiment phase.

View larger version (67K): [in this window] [in a new window]   Fig 3. Sample waveforms collected at three stages (stages 1, 3, and 4) leading to explantation.

In attempting to mimic physiologic conditions, predetermined preloads were used in vitro that provided mean right and left ventricular pressures of 14.7 ± 3.4 and 53.7 ± 11.2 mm Hg, respectively, compared with in vivo pressures of 13.5 ± 4.3 and 55.0 ± 17.6 mm Hg (n = 12). Similarly, pressures in the left side of the heart were well maintained: in vitro left ventricular systolic and diastolic pressures were 119.4 ± 25.8 and 8.7 ± 5.0 mm Hg, in comparison with 105.6 ± 17.0 and 4.9 ± 4.4 mm Hg in vivo (n = 12). In pilot studies, the use of lidocaine or epinephrine, or in conjunction with electrical pacing, it was possible to alter hemodynamic characteristics to better correspond in vitro with in vivo conditions (data not shown), but this was not employed as stated in the present protocol.

Although in vitro hemodynamic function decreased with time, no statistical difference was observed with coronary perfusion after 240 minutes. However, total cardiac output decreased, as did ventricular contractility and relaxation. Diastolic pressures were maintained throughout the experiment duration with systolic pressure decreasing as expected, given gradual stiffening (decreasing ±dP/dt) of the heart.

Average model performance increased for the initial hour following recovery, then gradually declined. In an extended time trial (n = 3) to determine chronological performance, left ventricular contractility (LVdPdt) decreased to 87% ± 7% at 2 hours (n = 3), 85% ± 6% at 4 hours (n = 3), and 78% ± 9% at 6 hours (n = 3). Performance was maintained reliably until this period, with performance diverging to 50% ± 35% (n = 3) at 8 hours due to contingent failure. Edema always became more evident over time. However, the period of high stability for not less than 6 hours in nearly all preparations provided reliable control for interactive cardiac studies within that duration.

Imaging
The clear perfusate provides the ability to obtain video images of internal cardiac structure. This allowed internal transient cardiac motion visualization during contraction and relaxation, with varying cardiac inputs. Shown in Figure 4 are intrinsic in vitro internal images, whereas Figure 5 illustrates transient cardiac device interaction.

View larger version (104K): [in this window] [in a new window]   Fig 4. Dynamic images of cardiac movement. Serial at 0.06 second increments (smaller increments possible), obtained by use of an IC6C-13 and ILV-C1 (Olympus Optical Co, Ltd, USA) scope. Shown in series are: (A) pulmonary valve, where the camera was located in the outflow cannula of the pulmonary artery; (B) aortic valve viewed from above through the aorta; (C) aortic valve from below. For these latter images, the camera was fed through a port placed into the left atrial appendage and passed through the left atrium and into the ventricle with the camera articulated upward toward the outflow cannula (the aortic cannula can be seen when the valve is open). (D) Anterior papillary muscle of the mitral valve and movements of the mitral valve, and (E) the mitral valve.

View larger version (96K): [in this window] [in a new window]   Fig 5. Product testing and cardiac evaluation. Examples of device–heart interaction are shown using the methods described in Figure 4. (A) Pacing lead, Medtronic model 5554, implanted in the right atrial appendage. (B) Defibrillation lead, Medtronic model 6945, implanted in the right ventricular outflow tract. (C) Temporary pacing lead, Medtronic model 6416, on the right atrial lateral wall. (D) Radio frequency tissue ablation along the right atrial septal wall.

	Comment

Top Footnotes Abstract Introduction Material and methods Results Comment Acknowledgments References

In in vitro human heart experiments [26], explanted human hearts from donor recipients were used in a left univentricular model, bypassing the need for human models. However, self-described limitations included the inability to assess individual normal cardiac function of the hearts for control, and the lack of preexplant in vivo measurements. Additionally, the transitional in vitro effects were also unknown. Because models are supposed to approximate human function, it is essential to attribute hemodynamic differences to the various stages of explantation. In our model, direct correlation to human hearts is therefore currently possible. We have recently begun using human recipient hearts for correlation with swine model function.

Additionally, because apparatuses for in vivo testing of total artificial hearts have not yet been developed [27], adjustments of atrial preloads on the total artificial heart can be compared with physiologic reactions in the in vitro model. As the differences between in vitro and in vivo cardiac characteristics can be understood, such an apparatus provides a method of evaluating xenotransplants, assisted, or total artificial hearts under physiologic conditions in vivo.

A large change in ventricular performance, comparable to explantation to the apparatus, occurred with excision of the pericardium. The pericardium was removed in vivo, isolating its sole pericardial effects on ventricular performance. Because of development of left ventricular assistance devices and their contribution to right side failure [28], large animal left univentricular working models may be unable to account for these incidences.

Additionally, the clear perfusate allowed detailed imaging during ex vivo physiologic working conditions. These visual abilities aid in interactive device studies during cardiac movement in vitro, providing benefits in assessing orientations and interactions of implanted devices, chamber and valve movements to varying inputs, and reactions to other various physiologic factors. The transient ex vivo model provides interactive cardiac motion separated from systemic effects.

Limitations
Explicit effects due to various sympathetic and parasympathetic hormones in vivo were unknown. Although blood gas analyses were performed before explantation, blood chemical analyses were not. We were therefore unaware of in vivo blood chemical composition for evaluating sympathetic, parasympathetic, and hormonal factors. Other markers, such as edema or creatine kinase levels, have also yet to be measured postmortem.

The anesthetic agents used in situ (which were not present in vitro) have also been shown to depress cardiac function. The effects of the same levels of these anesthetic agents on the isolated heart are currently being quantified. Although afterload resistance of the in vitro apparatus was a pressure head approximately equal to that of the systemic resistance, the tubing within the in vitro apparatus is stiffer than that of an active vascular system (ie, lower compliance). This caused pressure waveform characteristics to distribute more evenly, reducing ±dP/dt peaks, and accounting for a portion of the differences observed in waveform characteristics. By decreasing the resistance and increasing compliance, the pressure waveforms were altered; both maximum ±dP/dt capabilities of the heart increased (data not shown). Additional investigations are needed to better match in vitro waveform shapes to in vivo and in situ waveform shapes. For instance, large decreases in ventricular relaxation were observed following explantation into the apparatus. Part of this difference may have resulted from lack of compliance (and therefore vascular capacitance) of the filling tubing in vitro. The passive and constant filling nature of the in vitro filling pressure did not allow for elastance observed in systemic physiologic systems.

Additionally, the resistance of the in vitro aortic exit was greater than the in vivo aortic resistance. To better match in vitro with in vivo physiologic conditions, this resistance will be decreased for future experiments. The in vitro aortic cannula flow versus pressure relationship test was conducted during a steady-state relationship. Measurements of the in vitro aortic exit compliance to quantify transient flow response to pressure also requires future investigation to match that of in vivo conditions. Because cardiac output is a function of both the aortic resistance and chamber volume(s), it is considered a secondary factor of myocardial contractile properties until the aortic resistance and chamber volumes match those of physiologic values. Variation of perfusate viscosity for greater accuracy of physiologic movements should also be pursued.

Conclusion
An improvement of in vitro reproduction of in vivo cardiac function has been accomplished through documentation of a transitional explantation process into an in vitro apparatus. Benefits include the ability to better control and investigate the effects of inputs into this specific cardiac model. From this, in vivo cardiac function can eventually be established in vitro, aiding in determining in vivo effects of pharmacologic agents and for evaluating devices that were previously assessed only after implantation. The model is currently being developed to address the limitations discussed above.

The rationale of cardiac modeling is to extrapolate to human function. Although greater efforts may be required to use this model than smaller animal models, in situations in which better hemodynamic representation of human function is needed, the greater anatomic and physiologic similarities of a four-chamber working swine heart to human function may warrant the extra required effort. [16]

	Acknowledgments

Top Footnotes Abstract Introduction Material and methods Results Comment Acknowledgments References

 
This work was supported in part by a grant from Medtronic Inc. We thank Dale Wahlstrom and Clare Padgett of Medtronic. We also thank Dr Daniel Sigg and Dr Chris Kehler for their clinical input and assistance, and Gary Williams for his computer analyses.

	Footnotes

Top Footnotes Abstract Introduction Material and methods Results Comment Acknowledgments References

 
A video clip of this procedure can be viewed on the Internet at: http://www.sts.org/section/atsvideo/

	References

Top Footnotes Abstract Introduction Material and methods Results Comment Acknowledgments References

 

1. Langendorff O. Untersuchungen am uberlebenden Saugethierherzen [Investigations on the surviving mammalian heart]. Arch Gesante Physiol 1895;61:291-332.
2. Neubauer S., Ingwall J.S. The isolated, buffer-perfused ferret heart. Lab Anim 1991;25:348-353.[Medline]
3. Jacob A.D., Elkins N., Reiss O.K., Chan L., Shapiro J.I. Effects of acetate on energy metabolism and function in the isolated perfused rat heart. Kidney Int 1997;52:755-760.[Medline]
4. Coetzee A., Kotze J., Louw J., Lochner A. Effect of oxygenated crystalloid cardioplegia on the functional and metabolic recovery of the isolated perfused rat heart. J Thorac Cardiovasc Surg 1986;91:259-269.[Abstract]
5. McCarthy P.M., Fukamachi K., Fukumura F., Muramoto K., Golding L.A., Harasaki H. The Cleveland Clinic-Nimbus total artificial heart. In vivo hemodynamic performance in calves and preclinical studies. J Thorac Cardiovasc Surg 1994;108:420-428.[Abstract/Free Full Text]
6. Von Scheidt W., Neudert J., Erdmann E., Kemkes B.M., Gokel J.M., Autenrieth G. Contractility of the transplanted, denervated human heart. Am Heart J 1991;121:1480-1488.[Medline]
7. DiSesa V.J. Cardiac xenotransplantation [Review]. Ann Thorac Surg 1997;64:1858-1865.[Abstract/Free Full Text]
8. Schmoeckel M., Nollert G., Shahmohammadi M., et al. Transgenic human decay accelerating factor makes normal pigs function as a concordant species. J Heart Lung Transplant 1997;16:758-764.[Medline]
9. Liu C.T. Techniques for isolation and performance of the perfused guinea pig working heart. Am J Vet Res 1986;47:1032-1043.[Medline]
10. Glower D.D., Spratt J.A., Snow N.D., et al. Linearity of the Frank–Starling relationship in the intact heart. Circulation 1985;71:994-1009.[Medline]
11. Neely J.R., Libermeister H. Effect of pressure development on oxygen consumption by an isolated rat heart. Am J Physiol 1967;212:804-814.[Free Full Text]
12. Fragata J.I., Areias J.C. Acute loads applied to the right ventricle. Pediatr Cardiol 1996;17:77-81.[Medline]
13. Schafer S., Schlack W., Kelm M., Deussen A., Strauer B.E. Characterisation of left ventricular relaxation in the isolated guinea pig heart. Res Exp Med (Berl) 1996;196:261-273.[Medline]
14. Hess O.M., Bhargava V., Ross J., Jr, Shabetai R. The role of the pericardium in interactions between the cardiac chambers. Am Heart J 1983;106:1377-1383.[Medline]
15. Demmy T.L., Magovern G.J., Kao R.L. Isolated biventricular working rat heart preparation. Ann Thorac Surg 1992;54:915-920.[Abstract]
16. Schouten V.J.A., Schipperheyn J.J., van Rijk-Zwikker G.L., Swier G.P.H. Calcium metabolism and depressed contractility in isolated human and porcine heart muscle. Basic Res Cardiol 1990;85:563-574.[Medline]
17. Klima U., Guerrero J.L., Levine R.A., Vlahakes G.J. A new, biventricular working heterotopic heart transplant model. Transplantation 1997;64:215-222.[Medline]
18. Igic R. The isolated perfused "working" rat heart. J Pharmacol Toxicol Methods 1996;35:63-67.[Medline]
19. Bove A.A., Santamore W.P. Ventricular interdependence. Prog Cardiovasc Dis 1981;23:365-388.[Medline]
20. Forty J., White D.G., Wallwork J. A technique for perfusion of an isolated working heart to investigate hyperacute discordant xenograft rejection. J Thorac Cardiovasc Surg 1993;106:308-316.[Abstract]
21. Dunning J., Pierson R.N., Braidley P.C., White D.J., Wallwork J. A comparison of the performance of pig hearts perfused with pig or human blood using an ex-vivo working heart model. Eur J Cardiothorac Surg 1994;8:204-206.[Abstract]
22. Van Rijk-Zwikker G.L., Schipperheyn J.J., Huysmans H.A., Bruschke A.V.G. Influence of mitral valve prosthesis or rigid mitral ring on left ventricular pump function. Circulation 1989;80(Suppl 1):I1-I7.
23. Kimose H., Ravkilde J., Helligso P., Knudsen M., Baandrup U. Influence of pre-existing ischemia on recovery from chemical cardioplegia. A study on pig hearts in an isolated blood-perfused model. Scand J Thorac Cardiovasc Surg 1992;26:23-31.[Medline]
24. Sandhu R., Diaz R.J., Wilson G.J. Comparison of ischaemic preconditioning in blood perfused and buffer perfused isolated heart models. Cardiovasc Res 1993;27:602-607.[Medline]
25. Weng Z.C., Nicolosi A.C., Detwiler P.W., et al. Effects of crystalloid, blood, and University of Wisconsin perfusates on weight, water content, and left ventricular compliance in an edema-prone, isolated porcine heart model. J Thorac Cardiovasc Surg 1992;103:504-513.[Abstract]
26. Slater J.P., Lipsitz E.C., Chen J.M., Levin H.R., Oz M.C., Burkhoff D. Use of explanted human hearts as a model for the study of cardiac pathophysiologic conditions. J Thorac Cardiovasc Surg 1995;110:239-247.[Abstract/Free Full Text]
27. Orime Y., Takatani S., Tasai K., et al. In vitro and in vivo validation tests for total artificial heart. Artif Organs 1994;18:54-72.[Medline]
28. Santamore W.P., Gray L.A., Jr Left ventricular contributions to right ventricular systolic function during LVAD support. Ann Thorac Surg 1996;61:350-356.[Abstract/Free Full Text]

This article has been cited by other articles:

				T. Mihaljevic, Y. Ootaki, J. O. Robertson, A. K. Durrani, K. Kamohara, M. Akiyama, F. Cingoz, C. Ootaki, R. Dessoffy, M. Kopcak, et al. Beating Heart Cardioscopy: A Platform for Real-Time, Intracardiac Imaging Ann. Thorac. Surg., March 1, 2008; 85(3): 1061 - 1065. [Abstract] [Full Text] [PDF]

				F. Lu, P. A. Iaizzo, D. G. Benditt, R. Mehra, E. N. Warman, and B. T. McHenry Isolated Atrial Segment Pacing: An Alternative to His Bundle Pacing After Atrioventricular Junctional Ablation J. Am. Coll. Cardiol., April 3, 2007; 49(13): 1443 - 1449. [Abstract] [Full Text] [PDF]

				S. Saito, Y. Araki, A. Usui, T. Akita, H. Oshima, J. Yokote, and Y. Ueda Mitral valve motion assessed by high-speed video camera in isolated swine heart. Eur. J. Cardiothorac. Surg., October 1, 2006; 30(4): 584 - 591. [Abstract] [Full Text] [PDF]

				K. Ikeda, Y. Okazaki, K. Furukawa, S. Ohtsubo, J. Yunoki, M. Natsuaki, and T. Itoh Direct imaging of bileaflet mechanical valve behavior in the tricuspid position. Eur. J. Cardiothorac. Surg., June 1, 2006; 29(6): 1014 - 1019. [Abstract] [Full Text] [PDF]

				Y. Araki, A. Usui, O. Kawaguchi, S. Saito, M.-H. Song, T. Akita, and Y. Ueda Pressure-volume relationship in isolated working heart with crystalloid perfusate in swine and imaging the valve motion Eur. J. Cardiothorac. Surg., September 1, 2005; 28(3): 435 - 442. [Abstract] [Full Text] [PDF]

				A. J. Hill, T. G. Laske, J. A. Coles Jr, D. C. Sigg, N. D. Skadsberg, S. A. Vincent, C. L. Soule, W. J. Gallagher, and P. A. Iaizzo In Vitro Studies of Human Hearts Ann. Thorac. Surg., January 1, 2005; 79(1): 168 - 177. [Abstract] [Full Text] [PDF]

				D. C. Sigg, J. A. Coles Jr, W. J. Gallagher, P. R. Oeltgen, and P. A. Iaizzo Opioid preconditioning: myocardial function and energy metabolism Ann. Thorac. Surg., November 1, 2001; 72(5): 1576 - 1582. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) View Related Video Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Chinchoy, E. Articles by Iaizzo, P. A. Search for Related Content PubMed PubMed Citation Articles by Chinchoy, E. Articles by Iaizzo, P. A.