Animal models
All experimental procedures were approved by the experimental Animal Ethics Committee of West China Hospital, Sichuan University (Chengdu, China). Nineteen female Sprague–Dawley rats with body weights ranging from 250 to 300 g were divided into two groups: the reperfusion (n = 10) and sham control groups (n = 9). Before surgery, the rats were intraperitoneally anesthetized using sodium pentobarbital, and respiration was maintained using a rodent ventilator. A real-time electrocardiogram (ECG) was monitored throughout surgery.
Thoracotomy was performed to introduce the coronary occlusion. The chest was opened at the fourth intercostal space to expose the heart, the pericardium was opened using forceps, and a 6.0 suture was passed underneath the left anterior descending coronary artery at a location 2 mm proximal to the ostium of the coronary artery. Coronary occlusion was achieved by tightening the suture over a 3.0 suture. The success of occlusion was confirmed by the pale appearance of the myocardial apex area and immediate changes in the ECG profiles, including a significant increase in the amplitude of the QRS complex and ST-segment elevation [9].
In the early reperfusion model, all occlusions were maintained for 60 min, followed by reperfusion [10]. Reperfusion was achieved by untying the knot and releasing the suture from occlusion. The success of reperfusion was confirmed by ECG changes including further ST-segment elevation followed by gradual recovery of the ST-segment. In the sham surgery model, no occlusion after thoracotomy was performed.
MRI protocols
MRI was performed 48 h after surgery in both groups [10]. All MRI protocols were implemented using a 7.0 T MR system (BRUKER BIOSPEC 70/30). Each rat was anesthetized with isoflurane (2%–3%) in a small container, and anesthesia was maintained using a mixture of 100% oxygen and isoflurane (1%–2%), administered through a small mask during MRI. Body temperature was monitored using a rectal temperature probe and maintained at 37 °C using a heating blanket. Each rat was placed in the prone position within a surface coil. The ECG signal was obtained from three subcutaneous copper needles inserted in the left forelimb, hind limb, and right forelimb. The respiration signal was acquired from a respiratory pillow (SA Instruments Inc.) placed under the rat [9].
Scout imaging was initially performed using a gradient-echo sequence to localize the short-axis images at the middle level of the LV. Multi-slice cine images were obtained to cover the entire heart for functional analysis. After maintaining the respiratory rate at 30–50 cycles per minute by altering the isoflurane concentration, more than five (depending on the size of the heart) single-slice multi-echo T2*-mapping images were acquired on the short-axis slices during the mid-diastolic phase and end-inspiratory period using both ECG and respiratory gating systems. The imaging parameters were as follows: late gadolinium enhancement (LGE) imaging was performed by fast imaging with steady-state precession (FISP–cine on the same slice locations 10 min after the manual injection of gadolinium (Gd)–diethylenetriamine penta-acetic acid (Magnevist, Bayer Health Care Pharmaceuticals, 0.15 mmol/kg). Then, the same FISP–cine was repeated in identical planes after 2 min dobutamine (Gadovist, Bayer) injection at a rate of 10 mg/kg/min. Heart rate was measured before MR scan; before LDD injection; and 2, 20, and 30 min after LDD injection. Figure 1 shows the course of the examination.
The imaging parameters included: T2*-mapping: FA (Flip angle)=30°, TR/TE=1000ms/3.5,7,10.5,14,17.5,21,24.5,28 ms, Matrix size =192×192, FOV=50×50mm, and slice thickness=1.5mm without slice gap. LGE: TR/TE=5.2ms/1.8ms, FA=25°, matrix size =256×256, FOV=50×50mm, slice thickness=1.5mm, 25 frames for each slice.
Histology
Following MRI evaluation, the rats were euthanized using potassium chloride (3 mol/L, 0.5 mL), and the hearts were rapidly excised. Each heart was cut into five or more transverse slices from the apex to the base; these slices measured approximately 1.5 mm in thickness to match the MRI slices. These slices were then incubated in 4% paraformaldehyde for hematoxylin and eosin staining.
Data analysis
The strain data were analyzed using commercially available software (Circle Cardiovascular Imaging, Inc.). The peak global radial strains (PGRS), peak global circumferential strains (PGCS), peak radial strains (PRS), and peak circumferential strains (PCS) were derived using an FT analysis of the LV short-axis. The efficiency of PRS and PCS (EPRS and EPCS, respectively) were calculated using the following formula: PRS (PCS)/R-R interval. The analysis was based on the six segmentation created consisting of six hexahedral elements corresponding to the anteroseptal, anterior, anterolateral, posterolateral, posterior, and poster septal walls.
T2* maps were calculated using custom-made software written in Matlab 7.1 (The MathWorks, Inc.). More than five slices (slices with poor image quality due to motion artifacts were excluded from analysis) were selected to decide the regions of hemorrhage for each rat, which identified as a hypointense core within a hyper intense territory on T2* maps. In the sham group, the myocardium was segmented into six segments (Figure 2-3).
The hyperenhanced myocardium on LGE images was defined as an MI area; this was detected using a computer-aided threshold of >5 standard deviations (SDs) from the remote myocardium and adjusted manually and quantified using the multiple short-axis slice view. NMI areas were defined as segments without IMH and MI (The segments close to MI and IMH were exclusion for depressed strain compared with remote myocardium [11]. The MI and IMH distribution (segments) were evaluated by two experienced observers who were blinded to the experiment design. Discrepancies between the two observers were referred to another trained radiologist combined with the histopathologic results.
Normality was assessed using Kolmogorov–Smirnoff tests. Normally distributed data were expressed as mean ± SD, and comparisons between the groups were conducted using analysis of variance. Nonparametric data were expressed as median (25%–75% interquartile range) and compared using the Kruskal–Wallis test. P values of <0.05 were considered statistically significant.