Ann Thorac Surg 2003;75:1942-1947
© 2003 The Society of Thoracic Surgeons
Original article: cardiovascular
A pig model of chronic heart failure by intracoronary embolization with gelatin sponge
Genichi Sakaguchi, MDa,
Yutaka Sakakibara, MDa,
Keiichi Tambara, MDa,
Fanglin Lu, MDa,
Goditha Premaratne, MDa,
Kazunobu Nishimura, MD, PhDa,
Masashi Komeda, MD, PhDa*
a Department of Cardiovascular Surgery, Graduate School of Medicine, Kyoto University, Kyoto, Japan
Accepted for publication January 17, 2003.
* Address reprint requests to Dr Komeda, Graduate School of Medicine, Department of Cardiovascular Surgery, Kyoto University, 54 Kawaharacho, Shogoin, Sakyo-ku, Kyoto, Japan 606-8507.
e-mail: masakom{at}kuhp.kyoto-u.ac.jp
 |
Abstract
|
|---|
BACKGROUND: We produced a large-animal model of left ventricular (LV) failure induced by transcatheter embolization of the left coronary artery using a gelatin sponge.
METHODS: Fourteen male pigs underwent transcatheter embolization of the left anterior descending artery (LAD) using gelatin sponge to produce anteroapical myocardial infarction. Coronary angiography was performed 1 week after the coronary embolization. The animals were followed up with echocardiography and LV pressure-volume study for the subsequent 8 weeks, and the data were compared with those of the control group (n = 13).
RESULTS: The procedure mortality was 2 of 14 (14%). Coronary angiography revealed the occluded LAD was recanalized with poor run-off. The LV end-diastolic dimension progressively increased (control versus myocardial infarction: 39 ± 2 mm versus 49 ± 4 mm, p < 0.001 at week 4; and 40 ± 2 mm versus 57 ± 6 mm, p < 0.001 at week 8). Fractional area change decreased over 8 weeks (77% ± 10% versus 43% ± 6%, p < 0.001 at week 4; and 77% ± 10% versus 40% ± 8%, p < 0.001 at week 8). End-systolic elastance progressively decreased over 8 weeks (3.04 ± 0.73 mm Hg/mL versus 1.54 ± 0.51 mm Hg/mL, p < 0.0001 at week 4; and 2.88 ± 0.44 mm Hg/mL versus 1.05 ± 0.21 mm Hg/mL, p < 0.001 at week 8). The plasma levels of brain natriuretic peptide were significantly higher in the study group (543 ± 131 pg/mL versus 1,321 ± 364 pg/mL, p < 0.001 at week 4; and 610 ± 152 pg/mL versus 1,523 ± 232 pg/mL, p < 0.001 at week 8).
CONCLUSIONS: This pig model of chronic heart failure is reliable, reproducible, and amenable to investigate other surgical procedures.
|
Introduction
|
|---|
New surgical treatments for congestive heart failure have been developed, such as left ventricular (LV) plasty, dynamic cardiomyoplasty, LV containment operation, LV myosplint implantation, cell transplantation, and so on. However, experimental studies for these treatments have been hampered by the lack of optimal animal models, which require reproducibility, easy management, and significantly impaired cardiac function. In particular, chronic heart failure models of large animals are essential because they provide hearts of sufficient size to permit detailed evaluation of regional wall motion and the fine mechanics of LV and to enable intracardiac (valve or septum) experiments with profiles similar to the human condition. However, chronic, stable, large-animal models of heart failure are relatively difficult to establish compared with acute models.
Here we report a large-animal model of LV failure induced by transcatheter embolization of the left anterior descending coronary artery (LAD) using gelatin sponge, which can be dissolved in a few days.
|
Material and methods
|
|---|
Material
Twenty-seven male pigs (approximately 10 weeks old) weighing 19 to 21 kg were randomly assigned to two groups: coronary embolization group (myocardial infarction [MI] group: n = 14) and control group (n = 13) without sham operations. All animals received humane care in accordance with the guidelines published by the National Society for Medical Research ("Principles of Laboratory Animal Care") and by the National Institutes of Health ("Guide for the Care and Use of Laboratory Animals," NIHealth publication No. 85-23, revised 1985). The study protocol was approved by the Institutional Animal Care and Ethics Committee.
Coronary embolization
The animals in the MI group were premedicated with an intramuscular injection of ketamine (15 mg/kg), azapelon (6 mg/kg), and atropin sulfate (0.5 mg), and then anesthesia was induced with thiopental sodium (150 mg) intravenously through an ear vein. The animals were intubated with a standard cuffed endotracheal tube (6.5 mm) and anesthesia was maintained with 3L/min of O2 and isoflurane (1.0% to 2.0%). Cefazolin sodium (500 mg) antibiotic prophylaxis was intravenously administered before skin incision. The animals were positioned in the supine position and a skin incision was made in the left side of the neck to expose the left common carotid artery. A 5F size short sheath was placed in the artery and blood pressure was monitored through the sheath. A 5F Judkins catheter (Fukuda-Denshi, Tokyo, Japan) was introduced through the arterial sheath and placed in the left main coronary artery under fluoroscopy guidance. A flexible 2.6F SP catheter (Radifocus SP Catheter [Terumo, Tokyo, Japan]) was introduced through the Judkins catheter with the guidance of core wire tracking and placed in the distal LAD between the first and second diagonal branches. A small piece (10 mg) of gelatin sponge (Spongel; Yamanouchi, Tokyo, Japan) was cut into small particles (approximately 1 mm) and homogenized by sharp scissors and then mixed in heparinized saline (0.3 mL) and contrast media (0.3 mL). After intravenous administration of prophylaxis of lidocaine (50 mg), 0.5 mL of the mixture was slowly injected through the SP catheter into the distal LAD under monitoring electrocardiogram (ECG) over a 3-minute period. After the completion of the embolization, coronary angiography confirmed the occlusion of the middle portion of the LAD with elevation of the ST segment n the ECG. Those procedures could be completed in approximately 30 minutes and then the animals were carefully observed under mild anesthesia for 15 minutes.
Coronary angiography
Coronary angiography was performed 1 week after the coronary embolization in the MI group. An anesthesia was induced in the same way as described above. The animals were positioned in the supine position and the previous skin incision was reopened in the left side of the neck to expose the left common carotid artery. In case the left common carotid artery could not be detected or occluded, the right common carotid artery was used. A 5F short sheath was placed in the artery and blood pressure was monitored through the sheath. A 5F Judkins catheter was introduced through the arterial sheath and placed in the left main coronary artery under fluoroscopy.
Echocardiography
Echocardiographic measurements were performed with a 7 MHz ultrasound transducer (connected with SSA-260A Ultrasound system; Toshiba Medical, Tokyo, Japan) at the embolization, and 4 and 8 weeks later in all animals. The animals were placed in the left lateral decubitus position. A two-dimensional short-axis view of the LV was obtained at the level of the papillary muscles and M-mode tracings were recorded at the same plane. The images were analyzed off line by a single observer blinded to the profile of the animals. The LV internal dimension was measured by the American Society for Echocardiology leading-edge method from at least three consecutive cardiac cycles [1]. The scar length fraction was expressed as the percentage of akinetic endocardial length to the whole LV endocardial circumference. Fractional area change (FAC) was calculated as follows: FAC = (LV area end-diastole - LV area end-systole)/LV area end-diastole.
Left ventricular Pressure-Volume study
The LV pressure-volume study was performed at 4 weeks and 8 weeks after the coronary catheterization in all animals. The animals in both groups were anesthetized in the same manner as described above and placed in the supine position. A 7F thermodilution catheter (CritiCath; Omeda, Singapore) was placed in the pulmonary artery through a percutaneous femoral approach. A median partial sternotomy was made and the heart was exposed. An umbilical tape was placed around the inferior vena cava for transient occlusion. A 10-electrode conductance catheter (Millar, Houston, TX) and a micromanometer-tipped catheter (Millar) were inserted into the LV for pressure and volume recording and connected to a signal-conditioner processor (Leycom Sigma-5; CardioDynamics BV, Zoetermeer, Netherlands). The parallel conductance was evaluated by the injection of hypertonic saline solution into the pulmonary artery. Pressure-volume relations were obtained at steady state and during transient inferior vena cava occlusion. Linear regression was done on the collected end-systolic pressure–volume points to calculate the slope (mm Hg/mL) of end-systolic elastance (Ees). Tau was calculated from the continuous pressure monitoring assuming a zero-pressure asymptote.
Plasma BNP levels
Blood samples were collected from all animals at the time of conductance catheter study (4 and 8 weeks). Plasma brain natriuretic peptide (BNP) level was measured by enzyme immunoassay (BNP-32 porcine EIA Kit; Peninsula Co, CA).
Histologic study
Small specimens were collected from the LV noninfarcted area and from the border zone in the MI group and from the LV free wall in the control group. The tissues were dehydrated and embedded in paraffin. The sections (4 µm) were stained with Masson’s trichrome.
Statistical analysis
All data are expressed as the mean ± standard deviation (SD). Comparisons of the data between the groups were performed by two-way repeated measures analysis of variance (ANOVA) including time, group, and group-by-time interaction terms. If significance was found for group effect, post hoc comparisons between the groups at each time point were performed to explore the tendency, when appropriate, using Fisher’s protected least significant difference method. Statistical analyses were performed with Statview for Windows version 5.0 (SAS Institute, Cary, NC). A p value less than 0.05 was considered statistically significant.
|
Results
|
|---|
Fourteen animals in the MI group underwent coronary embolization, of which 2 died due to irreversible ventricular fibrillation after embolization. The remaining 12 animals in the MI group and 13 in the control group tolerated the subsequent hemodynamic studies. Body weights were similar at base line (control versus MI: 20.9 ± 1.3 kg versus 20.3 ± 2.3 kg), but at 4 weeks and 8 weeks, the MI group was significantly lower than the control group (39.6 ± 2.1 kg versus 33.1 ± 3.5, p < 0.05 at 4 weeks; and 42.5 ± 5.0 kg versus 38.0 ± 2.3Kg, p < 0.05 at 8 weeks). Although the LAD was occluded at the mid portion soon after the embolization, it was recanalized, which was confirmed in the coronary angiogram 1 week later (Fig 1).
View larger version (47K):
[in this window]
[in a new window]
|
Fig 1. Coronary angiogram soon after the embolization (left) and 1 week later (right). The left anterior descending artery (LAD) was occluded at the site of the sponge (arrow). (Cx = circumflex artery.)
|
|
Echocardiographic data are shown in Figure 2.
There were no differences in preoperative LV end-diastolic dimension (LVDd) and FAC between the groups. Two-way repeated measures ANOVA for LVDd and FAC showed high group (p < 0.0001) and time effects (p < 0.0001) and a strong group-by-time interaction (p < 0.0001). The LVDd in the MI group progressively increased, whereas that in the control group was unchanged over the 8 weeks. The FAC in the MI group decreased over the 8 weeks, whereas that in the control group was unchanged. The scar length fractions in the MI group were 29% ± 8% at 4 weeks and 30% ± 6% at 8 weeks.
View larger version (7K):
[in this window]
[in a new window]
|
Fig 2. Echocardiographic data at 0, 4, and 8 weeks. (A) Left ventricular end-diastolic dimension. (B) Fractional area shortening. *p less than 0.0001. (Circles=myocardial infarction group; squares=control group.)
|
|
Hemodynamic data are shown in Table 1.
Two-way repeated measures ANOVA for heart rate, cardiac output, LV end-diastolic pressure (LVEDP), LV positive dP/dt, negative dP/dt, tau, and Ees showed high group effects (p = 0.03 for cardiac output and p < 0.0001 for other variables) and those for LVEDP and Ees showed high time effects (p < 0.0001). The LV positive dP/dt showed a strong group-by-time interaction by two-way repeated measures ANOVA (p = 0.042). Cardiac output in the MI group was significantly lower at 8 weeks than that in the control group; however, the difference in cardiac output/body weight did not reach the statistical significance (control versus MI, 0.124 ± 0.027 versus 0.127 ± 0.018 L · min-1 · kg-1 at 4 weeks; 0.117 ± 0.018 versus 0.097 ± 0.019 L · min-1 · kg-1). The LVEDP in the MI group was significantly higher over the 8 weeks than that in the control group. The LV positive dP/dt and negative dP/dt in the MI group was significantly lower than the control group over the 8 weeks. Tau in the MI group was significantly higher than the control group over the 8 weeks. The pressure-volume loops are shown in Figure 3.
Whereas Ees in the control group was unchanged for 8 weeks, that in the MI group progressively decreased for the period. The plasma levels of BNP in the MI group were significantly higher than those in the control group (control versus MI: 543 ± 131 versus 1,321 ± 364 pg/mL, p < 0.001 at 4 weeks; 610 ± 152 versus 1,523 ± 232 pg/mL, p < 0.001 at 8 weeks). The postmortem specimen showed a transmural MI scar in the anteroseptal and apical regions in the MI group and dilatation of the LV chamber (Fig 4).
Histologic samples in the MI group stained with Masson’s trichrome showed massive fibrosis in the border zone and patchy fibrosis in the remote region in the LV free wall, whereas the control group showed no fibrosis (Fig 5).
View larger version (115K):
[in this window]
[in a new window]
|
Fig 3. Representative pressure-volume loops at 8 weeks in (A) the control group and (B) the myocardial infarction group.
|
|
View larger version (89K):
[in this window]
[in a new window]
|
Fig 5. Sections from (A) the border zone and (B) a remote area in the myocardial infarction group and (C) the anterior region in the control group (Masson’s trichrome stain, x 20).
|
|
|
Comment
|
|---|
The results of this study showed that this experimental procedure successfully produced an anteroseptal infarction of LV with impaired LV performance and provided a low mortality rate (14%) by the catheter procedure.
The echocardiography showed sufficient size of transmural anteroseptal infarction with akinetic or dyskinetic regional movement and increased chamber size. The echocardiographic data obtained at 4 weeks and 8 weeks also indicate that LV remodeling and LV dysfunction were progressive in this model.
The hemodynamic study showed significantly impaired LV performance in terms of both systolic and diastolic function. The LV positive dP/dt and Ees in the MI group were significantly lower than those in the control group at 4 weeks and 8 weeks. The results of LV negative dp/dt and tau showed the significantly impaired diastolic function in the MI group. Furthermore, the plasma levels of BNP in the MI group were significantly higher than those in the control group. The plasma BNP has been reported to increase with the severity of congestive heart failure and to be well correlated with a poor prognosis [2]. The increase in level of plasma BNP in the MI group indicates the presence of significant congestive heart failure in this model.
In this study, whereas discrete MI scar was established in the anteroseptal area at 4 weeks, the gelatin sponge mixture was dissolved and the occluded coronary artery was recanalarized in 1 week. The recanalization of the LAD may contribute to the relatively low mortality rate. Gelatin sponge mixture is now widely used in chemoembolization for liver cancer [3–5]. The chemoembolization is usually repeated in order to obtain the maximum benefits and the slow recanalization of the embolized arteries is often observed in chemoembolization therapy.
Several types of animal models of heart failure have been developed. The LAD occlusion in a porcine model, similar to our method, has been reported [6, 7]. They used a coil for the permanent occlusion of LAD. The occlusion coil is much more expensive than Spongel. Zhang and colleagues [8] demonstrated the functional abnormalities and bioenergetic characteristics in a porcine model of postinfarction LV remodeling created by ligating the circumflex artery. Ovine models with ligation of homonymous arteries have been reported [9–11]. Moainei and colleagues [11] demonstrated the myopathic process in an ovine model of postinfarction dilated cardiomyopathy created. They are reliable models of ischemic cardiomyopathy with congestive heart failure; however, they require surgical thoracotomy, which is more invasive and can disturb the next operation owing to adhesion of the heart. Multiple sequential coronary embolization with microsphere injection has been reported in canine and ovine [12, 13]. Intracoronary injection that was repeated several times produced significant heart failure. This model is stable and reproducible without surgical procedure; however, the catheter procedures should be repeated several times. An ameroid constrictor has been used to create hibernating myocardium in a pig model [14–16]. This method requires a small thoracotomy to place the ameroid constrictor, which gradually occludes coronary arteries in a few weeks. However, it was reported that resting coronary blood flow often returned to base line through collateral vessels after approximately 3 to 4 weeks of total occlusion with the ameroid constrictor, which leads to no wall motion abnormality at rest [14]. Our method can be situated between the coil embolization and the ameroid constrictor.
A rapid pacing model has been reported [17–19]. It is a simple, predictable, stable, and controllable model to induce congestive heart failure without extensive surgery. Sustained ventricular pacing can produce severe biventricular systolic and diastolic dysfunction in animal models. However, the myopathic process associated with rapid heart rates is largely reversible, which may make long-term investigations less accurate. Doxorubicin has been widely used to induce heart failure [20, 21]. This model exhibits pathologic similarity to dilated cardiomyopathy in human; however, it has the disadvantages of systemic toxic effects. The advantages of our method are firstly that our method can be performed through a catheter with a low mortality rate and secondly that it can produce significant LV failure with reproduciblity.
In conclusion this method, which does not require thoracotomy, is free from adhesion of the heart and is less invasive than other large-animal models with a low mortality. It consistently produces significant LV failure suitable for studies for the surgical treatment of LV failure.
|
Acknowledgments
|
|---|
The authors thank Mr Hashino (Taisho Biomed Instruments) for the hemodynamic study. This work was partially supported by a Grant-in-Aid for Scientific Research in Japan.
|
References
|
|---|
- Sahn D.J., DeMaria A., Kisslo J., Weyman A. Recommendations regarding quantitation in M-mode echocardiography: results of a survey of echocardiographic measurements. Circulation 1978;58:1072-1083.[Abstract/Free Full Text]
- Grantham J.A., Borgeson D.D., Burnett J.C., Jr BNP: pathophysiological and potential therapeutic roles in acute congestive heart failure. Am J Physiol 1997;272:R1077-R1083.
- Musset D., Gorse F., Roche A. Embolization of hepatic metastases. Ann Gastroenterol Hepatol (Paris) 1985;21:79-81.[Medline]
- Courtheoux P., Berthelin J.L., Segol P., et al. Chemoembolization of hepatic metastases. Preliminary study. J Radiol 1987;68:247-253.[Medline]
- Takahashi H., Tono T., Matsushita M., et al. Malignant islet cell tumor of the pancreas with multiple liver metastases effectively treated by transcatheter arterial embolization with degradable starch microspheres—a case report. Gan To Kagaku Ryoho 2000;27:1936-1939.[Medline]
- Li R.K., Weisel R.D., Mickle D.A., et al. Autologous porcine heart cell transplantation improved heart function after a myocardial infarction. J Thorac Cardiovasc Surg 2000;119:62-68.[Abstract/Free Full Text]
- Watanabe E., Smith D.M., Sun J., et al. Effect of basic fibroblast growth factor on angiogenesis in the infarcted porcine heart. Basic Res Cardiol 1998;93:30-37.
- Zhang J, Wilke N, Wang Y, et al. Functional and bioenergetic consequences of postinfarction left ventricular remodeling in a new porcine model. Circulation. 1996;94:1089–100
- Millner R.W., Mann J.M., Pearson I., Pepper J.R. Experimental model of left ventricular failure. Ann Thorac Surg 1991;52:78-83.[Abstract]
- Markovitz L.J., Savage E.B., Ratcliffe M.B., et al. Large animal model of left ventricular aneurysm. Ann Thorac Surg 1989;48:838-845.[Abstract]
- Moainie S.L., Gorman J.H., Guy T.S., et al. An ovine model of postinfarction dilated cardiomyopathy. Ann Thorac Surg 2002;74:753-760.[Abstract/Free Full Text]
- Sabbah H.N., Stein P.D., Kono T., et al. A canine model of chronic heart failure produced by multiple sequential coronary microembolizations. Am J Physiol 1991;260:H1379-H1384.
- Huang Y., Kawaguchi O., Zeng B., et al. A stable ovine congestive heart failure model. A suitable substrate for left ventricular assist device assessment. ASAIO J 1997;43:M408-M413.[Medline]
- Harada K., Grossman W., Friedman M., et al. Basic fibroblast growth factor improves myocardial function in chronically ischemic porcine hearts. J Clin Invest 1994;94:623-630.
- Mack C.A., Patel S.R., Schwarz E.A., et al. Biologic bypass with the use of adenovirus-mediated gene transfer of the complementary deoxyribonucleic acid for vascular endothelial growth factor 121 improves myocardial perfusion and function in the ischemic porcine heart. J Thorac Cardiovasc Surg 1998;115:168-177.[Abstract/Free Full Text]
- Horvath K.A., Chiu E., Maun D.C., et al. Up-regulation of vascular endothelial growth factor mRNA and angiogenesis after transmyocardial laser revascularization. Ann Thorac Surg 1999;68:825-829.[Abstract/Free Full Text]
- Shinbane J.S., Wood M.A., Jensen D.N., et al. Tachycardia-induced cardiomyopathy: a review of animal models and clinical studies. J Am Coll Cardiol 1997;29:709-715.[Abstract]
- Armstrong P.W., Stopps T.P., Ford S.E., de Bold A.J. Rapid ventricular pacing in the dog: pathophysiologic studies of heart failure. Circulation 1986;74:1075-1084.[Abstract/Free Full Text]
- Wilson J.R., Douglas P., Hickey W.F., et al. Experimental congestive heart failure produced by rapid ventricular pacing in the dog: cardiac effects. Circulation 1987;75:857-867.[Abstract/Free Full Text]
- Monnet Eorton E.C. A canine model of heart failure by intracoronary adriamycin injection: hemodynamic and energetic results. J Card Fail 1999;5:255-264.[Medline]
- Czarnecki C.M. Animal models of drug-induced cardiomyopathy. Comp Biochem Physiol C 1984;79:9-14.
Top
Abstract
Introduction
