Our aim was to develop an easy-to-induce, reproducible, and low mortality clinically relevant closed-chest model of chronic myocardial infarction in swine using intracoronary ethanol
and characterize its evolution using MRI and pathology. We injected 3-4 mL of 100% ethanol into the mid-LAD of anesthetized swine. Heart function and infarct size were assessed serially using MRI.
Pigs were euthanized on days 7, 30, and 90 (
Cardiovascular diseases are a major cause of death and disability in developed countries. Myocardial infarction has an estimated annual incidence in the US of 525.000 new and 190.000 recurrent attacks. Approximately 15% of patients who suffer a coronary attack will die as a result of it [
Research in animal models is mandatory to assess safety and efficacy of any new therapy prior to clinical translation [
Clinically, intracoronary ethanol has been used for the management of symptomatic hypertrophic obstructive cardiomyopathy (HOCM) [
The aim of this study was to develop a simple, reproducible, clinically relevant, and low mortality endovascular model of chronic myocardial infarction in the swine. If successful, this model should be useful for preclinical evaluation of therapeutic approaches to HF after chronic myocardial infarction, such as regenerative therapies or electrophysiological studies, as well as for training in diagnostic and therapeutic interventions.
The study protocol was approved by the Institutional Animal Care and Use Committee, and it complied fully with the Guide for the Care and Use of Laboratory Animals: Eighth Edition (National Research Council. Washington, DC: the National Academies Press, 2010). A total of 18 female domestic swine weighing 40.9 ± 10.48 kg were used for this study. Animals received 400 mg oral amiodarone and 500 mg aspirin from two days before to 7 days after infarct induction. Surviving animals were euthanized after infarct induction at day 7 (Group I,
All the procedures were performed under general inhalant anesthesia. After being fasted for 24 hours, animals were premedicated with a combination of diazepam (0.1 mg/kg) and ketamine (10 mg/kg) injected intramuscularly (IM). Anesthesia was induced with intravenous (IV) propofol (4 mg/kg). Endotracheal intubation was performed and anesthesia was maintained with sevoflurane administration (adjusted to a 3.3% End tidal sevoflurane). A bolus (5 g/kg) of fentanyl was administered every 30 minutes. Continuous infusion of lidocaine at a rate of 1 mg/kg/h was used until the end of anesthesia. Systemic heparin was injected IV (150 IU/kg) 5 minutes before percutaneous sheath placement. A fentanyl transdermic release patch was used to assure correct analgesia in the immediate postoperative period. Prophylactic antibiotics was administered in all cases for 5 days (ceftiofur hydrochloride). Electrocardiography (ECG) (Hewlett Packard model 86S, Hewlett Packard, Geneva, Switzerland) (lead II) and pulse oximetry with a probe (Clip Tip sensor, Oximeter Sensor, Datex-Ohmeda, Louisville, CO) placed on the tongue, rectal temperature, tidal volume, end-tidal concentration of sevoflurane, end-tidal CO2 concentration, and respiratory rate (Ohmeda RGM 5250, Ohmeda, Madrid, Spain) were monitored through the procedure.
Blood samples were taken for troponin I assay immediately before infarct creation (baseline) and 2 hours, 24 hours, and 7 days after ethanol infusion.
Cardiac MR studies were performed for morphological and functional assessment, before infarct induction, immediately after and at 7, 30, and 90 days using a 1.5 T system (Intera 1.5 T, Philips Medical Systems. Best, The Netherlands). Baseline (before induction) examination was performed without gadolinium administration. A group of healthy swine (Control group), weighing the same as the study animals on days 30
Images were acquired in the intrinsic cardiac planes: short axis, vertical long axis, and horizontal long axis views. For measurement of left ventricular function and mass breath hold gradient echo cine images were obtained over the entire left ventricle (LV) with the following parameters: slice thickness: 8 mm, no gap, field of view (FOV): 320 × 320 × 80, matrix: 192 × 192, flip angle: 60°, and repetition time/echo time (TR/TE): 4.4/2.2. The number of cardiac phases acquired per slice was 25. Infarct size was measured in images acquired 5 to 15 minutes after the injection of 0.2 mmol/kg of a gadolinium-based contrast agent (Gadobutrol) using a breath-hold 3D gradient-echo inversion-recovery sequence. Inversion time was chosen selecting the time that yielded the best nulling of the myocardial signal. Imaging parameters used for the delayed enhancement (DE) images were slice thickness: 8 mm, no gap, FOV: 330 × 330 × 50, matrix: 224 × 200, flip angle: 15°, and TR/TE: 4.9/1.67. MR images were analyzed to calculate the following LV parameters: end diastolic volume (EDV), end systolic volume (ESV), ejection fraction (EF), wall mass, cardiac output (CO), and percentage of infarcted LV.
Under sterile conditions, a right femoral arterial access was established using the Seldinger technique and a 7Fr introducer sheath (Terumo, Inc. Tokyo, Japan) was placed percutaneously. Under fluoroscopic guidance (Philips BV Pulsera, Philips Medical Systems, Best, The Netherlands) a 6Fr hockey stick guiding catheter (Mach 1, Boston Scientific Corporation, Natick, MA, USA) was introduced and placed at the origin of the left coronary artery. Coronary angiograms were obtained in the 40° left anterior oblique (LAO) projection to better demonstrate the length of the left anterior descending coronary artery (LAD), and a
Coronary angiogram obtained immediately after balloon inflation to assess correct occlusion of the artery prior to ethanol injection.
Immediately after the last MR examination, animals were euthanized administering a lethal dose of potassium chloride while under general anesthesia. The explanted hearts were submerged in 4% formalin for a minimum of 48 hours and, once fixed, were cut into 5 mm thick slices.
Samples were taken from the infarct and transition areas for pathological examination, embedded in paraffin, sliced into 5 µm thick sections, and stained with hematoxylin and eosin and Gallego’s trichromic.
Since the objectives of the study are descriptive in nature, no formal hypothesis testing was done. Data are presented as means and standard deviations. Comparisons between cardiac function parameters between groups were performed using the Mann-Whitney
Cardiac catheterization and ethanol injection were technically successful in all pigs. Changes in the ECG, such as premature ventricular complexes and ST segment elevation, were seen in all animals during ethanol injection. The ST wave remained elevated during the one-hour observation time in all cases. Seven pigs developed ventricular tachycardia and fibrillation after balloon deflation. Four pigs were successfully recovered using 200J biphasic defibrillation shocks and pharmacological therapy when needed, while the other 3 animals (16.6%) died during infarct creation. The remaining animals completed their allotted followup (no further deaths).
Baseline troponin I levels were 0.037 ± 0.027 µg/L. A significant
Postoperative MR studies revealed a density change in the midanteroseptal and apical septal areas of the left ventricle (Figure
Cardiac MR images obtained during the study. Cine images obtained in short axis and four chamber views at all timepoints are shown in the first two rows. The third row depicts delayed enhanced acquisitions. (a) and (b) Baseline examinations showing normal swine cardiac anatomy. (c), (d) and (e) Images acquired immediately after infarct induction. The infarcted area can be identified by a density change in the cine images, which correspond to the enhanced area on the viability study. (f), (g) and (h) Images obtained 7 days after infarction. Little changes were evidenced in the cine images at this timepoint (apart from akinesia of the septum), but the DE view shows the typical hyperenhancement appearance of the infarct. (i), (j) and (k) Study performed on the 30-day followup. Wall thickness is clearly decreased in the infarcted area (anteroseptal), corresponding with the hyperenhanced septum in the viability study. (l), (m) and (n): 90-day followup. The infarcted area is clearly identified as a very thin wall in the cine acquisitions and as a highly enhanced area in the viability study.
MR-derived cardiac function parameters are presented in Table
Cardiac parameters calculated from magnetic resonance exams performed through the study.
Baseline | 2 hours after induction | Day 7 | Day 30 | Day 90 | |||
---|---|---|---|---|---|---|---|
Study group | Control | Study group | Control | ||||
Animal weight (kg) |
|
|
|
|
|
|
|
EDV (mL) |
|
|
|
|
|
|
|
ESV (mL) |
|
|
|
|
|
|
|
EF (%) |
|
|
|
|
|
|
|
Wall mass (g) |
|
|
|
|
|
|
|
CO (L/min) |
|
|
|
|
|
|
|
% Infarct MR | n/a |
|
|
|
n/a |
|
n/a |
Data presented as mean ± standard deviation. EDV: end diastolic volume; ESV: end systolic volume; EF: ejection fraction; CO: cardiac output; MR: magnetic resonance; n/a: not applicable. *indicates the existence of statistically significant differences with baseline values (
A clearly demarcated area of hyperenhancement comprising the midanterior and anteroseptal and apical septal left ventricle was visualized in all animals at the 7, 30, and 90-day follow-up MRI examinations. The size of the infarct, measured in percentage, decreased over time (Table
Pathology on day 7 revealed a clearly defined area of coagulation necrosis comprising both small vascular structures and cardiac fibers (myocardiocytes that exhibited highly eosinophilic cytoplasms and no nuclei) with moderate hemorrhage and band composed by granulation tissue with some neovessels and a cellular infiltrate rich in macrophages, lymphocytes, and fibroblasts.
On day 30, there were still some necrotic areas within granulation tissue, but most of the necrotic tissue was completely replaced by the latter. This granulation tissue was more mature, showing a decrease in the cellular component, mostly composed by fibroblasts at this time, increased collagen content, and a significant decrease in neovessels. Occasional calcifications and bundles of apparently healthy conduction cells at the subendocardial level were seen.
On day 90, injured areas exhibited a highly organized appearance with collagen fibers, consistent with fibrosis and scar formation. In transition areas, collagen fiber bundles were seen interspersed between normal myocardiocytes. A very thin wall could be seen in some areas showing preserved conduction cells near the endocardial border, some myocardiocytes groups and organized fibrous tissue.
The ideal experimental model should meet a number of criteria so that the results can be extrapolated to the clinical setting [
The model we present herein meets most of these requirements. Swine were selected for the present study because they have many similarities to the human heart in terms of cardiovascular anatomy, ventricular performance, cardiac metabolism, electrophysiology, coronary artery distribution, and collateralization after an acute event [
Generally, chronic infarctions in large animal models are obtained by surgical ligation of a coronary artery via a thoracotomy [
The intracoronary ethanol injection model has been described before [
In order to be useful a HF model should have significantly impaired left ventricular function [
Cardiac MRI allows for quantitative analysis of LV mass and infarct area and has shown high reproducibility in detecting small changes in global and regional mass. It is considered a well validated standard for myocardial function measurement, and its ability to assess chronic infarction by using delayed enhancement has also been established [
As shown in Table
A myocardial insult is associated with left ventricular remodeling including changes in geometry, structure, and function [
In the early stages of MI, infarct size can be overestimated by the addition of edema fluid and cellular components, such as hemorrhage and inflammation, which can acutely increase infarct volume by as much as 25% [
It is known that remodeling is an ongoing process, so follow-up periods of 6 weeks or shorter have been considered too short to adequately represent the chronic state of the disease in humans [
A thin surviving rim of endocardium, as seen in our model, has been described previously in human transmural infarcts [
No model is exempt from weaknesses [
We have developed a swine model of chronic myocardial infarction by intracoronary ethanol administrations and exhaustively characterized its evolution over 3 months using MRI. This technique consistently results in a transmural infarct, with a mortality analogous to those obtained with other techniques. On the chronic state (90 days), MR and pathology show an image similar to that obtained in the clinical setting. The model may thus be useful for testing therapeutic approaches or new devices in chronic heart failure, such as regenerative or interventional therapies or electrophysiologic procedures, as well as for training specialist in these fields in a setting relevant to the clinical scenario.
The authors of this paper do not have any conflict of interests to declare regarding any direct financial relation with the commercial identities mentioned in the paper.
This study was funded in part by the Extremadura Regional Government, Spain, through Grants Ref PRI08A049 and BS10031. The authors wish to thank the anesthetic team, especially Monica Garcia-Lindo, DVM, and the OR technicians Maria Borrega, Alejandra Uson, and Helena Martin for their outstanding assistance during the experimental work.