| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | RSS | TABLE OF CONTENTS |
|
|
|
||||||||
|
||||||||||
Basic Science Investigation |
1 UHP-INSERM U684, Faculty of Medicine, Nancy, France; 2 Nuclear Medicine Department, University Hospital, Nancy, France; 3 In Vivo Cellular and Molecular Imaging Center, University of Brussels (VUB), Brussels, Belgium; 4 Laboratory of Surgery School, Nancy, France; and 5 Superior National School of Chemical Industries, CNRS and INPL, Nancy, France
Correspondence: For correspondence contact: Fatiha Maskali, MSc. Unité mixte INSERM-UHP U684, Faculté de Médecine de Nancy, 54505 Vandoeuvre-Les-Nancy, France. E-mail: fatiha.maskali{at}medecine.uhp-nancy.fr
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONReferences |
|---|
30% LV) in 6, limited (<30% LV) in 6, and undetectable in 8. End-diastolic volume of LAD-occluded rats was equivalent to that of sham-operated rats at 48 h (320 ± 84 µL vs. 293 ± 48 µL; not significant) but became higher at 12 wk (501 ± 191 µL vs. 343 ± 46 µL; P = 0.01). The follow-up increase in end-diastolic volume, which reflects the remodeling process, was closely related to the initial extent of necrosis revealed by the SPECT images (P < 0.001; R2 = 0.85). This increase was limited in sham-operated rats (50 ± 15 µL) and in the LAD-occluded rats with undetectable necrosis (55 ± 35 µL) but it was around 3- and 7-fold higher in the LAD-occluded rats with limited (165 ± 57 µL) and large (366 ± 113 µL) necrosis, respectively. Conclusion: The variable LV remodeling documented after coronary occlusion in rats closely relates to the variable extent of necrosis provided by this model. Pinhole gated SPECT allows this remodeling to be predicted and quantified and, hence, constitutes an original tool for the experiments scheduled on the rat infarct model.
Key Words: remodeling • myocardial infarction • rat • gated SPECT • sestamibi
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONReferences |
|---|
In humans, 99mTc-sestamibi SPECT is currently used in the setting of ischemic LV remodeling and it has been extensively validated for (i) differentiating viable from predominantly necrotic myocardium using the threshold value of 50% uptake (9–13), (ii) sizing myocardial infarction (14–16), (iii) predicting the evolution of LV function after infarction (9,10,17), and (iv) determining LV function on gated SPECT acquisitions (18). Thanks to the use of pinhole collimators and to advances in data processing, gated SPECT was recently adapted to image the rat heart (19). The present study was aimed to assess this new imaging technique for predicting and quantifying variable LV remodeling from the rat infarct model.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONReferences |
|---|
Additional experiments were conducted on cardiac rat phantoms of various sizes to assess the accuracy of LV volumes measured with pinhole gated SPECT.
Surgical Procedure
Permanent coronary occlusion of the left anterior descending coronary artery (LAD) was performed according to the usual and extensively described method (2–8). The rats were anesthetized by inhalation of an isoflurane/oxygen mixture and administration of ketamine hydrochloride (10 mg/kg). They were intubated and artificially ventilated with a rodent ventilator (Harvard Apparatus). The heart was exposed through a left lateral thoracotomy of the fifth interrib space. After pericardial incision, the proximal part of the LAD was ligated with a 7-0 Prolene suture (Ethicon, Inc.), at 2–3 mm from the left atrium tip. Finally, the chest was closed in layers with a 2-0 Vicryl suture (Ethicon, Inc.), the lungs were reinflated using positive-end expiratory pressure, and the endotracheal tube was removed once spontaneous breathing had resumed. Sham-operated rats had exactly the same surgical procedure, except that the LAD was not ligated.
Serial Pinhole Gated SPECT Acquisitions
The animals were sedated by intraperitoneal injection of 60 mg/kg of sodium pentobarbital, and 99mTc-sestamibi (400–700 MBq in a 0.3- to 0.5-mL volume) was injected intravenously 40–60 min before starting gated SPECT acquisition. During the acquisition, the animals were in prone position and were connected to a standard electrocardiogram monitor by 3 electrodes placed on the inner surfaces of limb extremities.
As previously described (19), pinhole gated SPECT acquisitions were performed using a conventional single-head 
-camera (DSX; General Electric-SMV) equipped with a 3-mm pinhole collimator (195-mm focal length; 43-mm radius of rotation). Thirty-two projections of 30 s each were acquired on a 180° anterior circular orbit centered on the heart (from 110° right-anterior to 290° left-anterior orientations). The use of 180° instead of 360° orbit allowed the rotation radius of the camera head to be reduced and, thereby, the spatial resolution of pinhole SPECT images was further enhanced. Additional acquisition parameters were as follows: 64 x 64 matrix, 2.66 zoom, 16 bins, 126- to 154-keV energy window, beat acceptance window set to ±20% of averaged R–R interval. Total acquisition time was 16 min.
Reconstruction and Analysis of Pinhole Gated SPECT Images
As already described (19–21), projection data were corrected for the mechanical shift of the camera-head rotation and for the nonuniform sensitivity of pinhole detection. Images were thereafter reconstructed using an ordered-subset expectation maximization (OSEM) iterative process (2 iterations and 8 subsets), which was adapted, first, to take into account pinhole geometry and, second, to reduce noise by incorporating a temporal Fourier filtering after each iteration (19–21). Voxel size of the reconstructed images was 0.6 mm and the spatial resolution was 3 mm in the central slice when determined by the full width at half maximum on a point source of 99mTc.
Cardiac reorientation along the LV long axis was performed using standard clinical software (Mirage Processing Application; Segami). LV end-diastolic volume (EDV) and end-systolic volume (ESV), as well as LV ejection fraction (EF), were determined by QGS software (22,23) on contiguous gated short-axis slices. Sestamibi uptake was determined on the set of collapsed short-axis slices with QPS software (23) and the 17-segment LV division from the American Heart Association (24). The 2 proximal septal segments, corresponding to the membranous part of the septal wall, were excluded. For both QGS and QPS, constraint limits of contour detection were applied only in case of evident errors with the fully automatic process.
Phantom Study
To assess the accuracy of LV volumes measured with pinhole SPECT, the protocols of data acquisition, reconstruction, and quantitative analyses were applied to LV rat phantoms of various volumes ranging from 50 to 800 µL. These phantoms were obtained by a stereolithography process, which allows fabricating small objects of any shape with a 0.1-mm precision and by using 3-dimensional computer-aided design (25).
As illustrated in Figure 1, these phantoms were comprised of an internal cavity filled with water and an external wall cavity with a thickness of 2.1 mm, corresponding to that of rat LV walls in vivo (26,27). The wall cavity was filled with a solution containing 0.14 MBq of 99mTc per microliter.
|
For this purpose, rats were sacrificed with an overdose of sodium pentobarbital, 50 min after the intravenous injection of 180 MBq of 99mTc-sestamibi and 2–5 d after the last 12-wk pinhole SPECT. The hearts were excised, arrested in ice-cold potassium chloride (2 mol/L), and snap frozen in liquid nitrogen. Contiguous sections (8–10 µm), orientated along the vertical or horizontal long axis of the LV (depending on infarct location), were obtained in a cryostat at 22°C. One midventricular section was selected and immediately referred to a 48-h recording with a µIMAGER (Biospace) (28,29) for quantifying the distribution of 99mTc-sestamibi activity (voxel size, 20 µm). This section was thereafter stained by collagen-specific Sirius red and analyzed using light microscopy and a dedicated computer analysis system to measure the wall thickness and percentage of volume fibrosis (30).
Histologic sections, as well as their corresponding 12-wk pinhole SPECT slices, were divided into the same 7 segments, which were those visualized on long-axis slices according to the 17-segment LV division from the American Heart Association (24). Segments with <50% uptake on pinhole SPECT were identified in each section and compared with the remote segments, which were those not directly adjacent. 99mTc-Sestamibi activity was determined on µIMAGER images using hand-drawn regions of interests encompassing each segment. In segments associated with <50% uptake on pinhole SPECT, this activity was expressed by mean voxel values and in the percentage of average values from remote segments. Volume fibrosis percentage was calculated in each segment on the Sirius red–stained sections by averaging values from 4 to 6 sampling areas of 0.1–0.3 mm2 each. The wall thickness of each segment was averaged from 3 to 6 measurements. The product of wall thickness by the fraction of nonfibrotic tissue was used as an index of viable tissue (31) and percentages of cellular loss were determined in segments with <50% uptake using the formula: (V2 – V1)/V2 x 100, where V1 is the index of viable tissue of the considered segment and V2 is the mean value of indices determined in remote segments from the same slice (31).
Statistical Analysis
Quantitative variables were expressed as mean ± SD and compared with nonparametric tests: the Wilcoxon rank sum test was used for paired series and the Kurskal–Wallis test was used for multigroup unpaired comparisons. A P value < 0.05 was considered to be significant. Multivariate analyses were performed stepwise using ascending linear regression analyses (SPSS 10.0 statistical software). At each step, the limits for significance to enter and remove a variable were 0.05 and 0.10, respectively.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONReferences |
|---|
Characterization of Segments with <50% Sestamibi Uptake
Among the 105 segments analyzed on the histologic sections, 27 showed <50% of sestamibi uptake on the pinhole SPECT performed before sacrifice. The decrease in 99mTc-sestamibi activity, which was calculated in each of these 27 segments with regard to remote slice segments, was equivalent when determined in vitro on µIMAGER sections (–54% ± 22%) and in vivo on the corresponding SPECT slices (–56% ± 18%). Moreover, these 27 segments had a much higher volume fibrosis (in % of myocardial volume: 27% ± 20% vs. 5% ± 7%; P < 0.001) and a much lower parietal thickness (2.0 ± 1.5 mm vs. 3.9 ± 1.0 mm; P < 0.001) than remote slice segments. Finally, segments with <50% uptake on pinhole SPECT had an estimated 61% ± 33% cellular loss compared with remote segments, with >50% cellular loss being observed in 74% of these segments (20/27). Pinhole SPECT identification of necrotic myocardium is illustrated in Figure 2 by SPECT, µIMAGER, and histologic images obtained in the same animals.
|
Predominantly necrotic segments, defined as those with <50% sestamibi uptake on pinhole SPECT (nonviable myocardium), were observed in 12 of the 20 LAD-occluded rats (60%): 6 had a large necrotic area involving at least 5 segments (30% of LV) and the 6 others had a more limited necrotic area, involving <5 LV segments (<30% of LV). No segment with <50% sestamibi uptake was documented on the SPECT images from the remaining 8 LAD-occluded rats and from the 9 sham-operated rats.
EDV of LAD-occluded rats was equivalent to that of sham-operated rats at 48 h (320 ± 84 µL vs. 293 ± 48 µL; not significant). In addition, these 2 groups did not differ in ESV and EF at 48 h (Table 1). On subgroup analyses, however, significant differences were documented between the sham-operated rats and the LAD-occluded rats with large necrosis (30% of LV) for ESV, EDV, and EF (Fig. 3).
|
|
30% of LV; P = 0.08) and, to a lesser extent, for those with small necrosis (<30% of LV; P = 0.18). LAD-occluded rats had, moreover, marked increases in EDV and ESV, whereas sham-operated rats had only slight volume variations during the same period (Table 1). Consequently, EDV of LAD-occluded rats became clearly higher than that of sham-operated rats at 12 weeks (501 ± 191 µL vs. 343 ± 46 µL; P = 0.01) (Table 1).
As illustrated in Figure 3, the increase in EDV during follow-up, which reflects the remodeling process, was limited in sham-operated rats (50 ± 15 µL) and in the LAD-occluded rats with undetectable SPECT necrosis (55 ± 35 µL), but it was around 3- and 7-fold higher in the LAD-occluded rats with limited (165 ± 57 µL) and large (366 ± 113 µL) SPECT necrosis, respectively. In addition, only the LAD-occluded rats showing a large necrosis (30% of LV) had a persistent increase in EDV during the second half of follow-up (between 6- and 12-wk values). At 12 wk, finally, LV function became (i) severely affected in the LAD-occluded rats with large SPECT necrosis (EF, 29% ± 7%; EDV, 742 ± 136 µL; all P < 0.05 vs. all other groups), (ii) much less affected in those with more limited SPECT necrosis (EF, 47% ± 6%; EDV, 472 ± 84 µL; all P < 0.05 vs. all other groups), and (iii) preserved in the LAD-occluded rats with undetectable SPECT necrosis (EF, 57% ± 8%; EDV, 342 ± 40 µL), as well as in sham-operated rats (EF, 55% ± 5%; EDV, 342 ± 46 µL). These data are illustrated by the SPECT images displayed in Figure 4.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONReferences |
|---|
Pinhole ungated SPECT has already been applied in the rat infarct model (32,33) and our pinhole gated SPECT technique was previously shown to provide precise and reproducible determinations of LV volume and EF in normal adult rats (19). By using cardiac phantoms, we documented here the accuracy of volume measurement over a much larger range of values: from 100 µL, corresponding to ESV of normal adult rats, up to >800 µL, corresponding to the largest EDVs after infarction in rats (3,4,19,26).
In CAD patients, predominantly necrotic myocardium is routinely characterized by 99mTc-sestamibi uptake being <50% of the maximal LV value on rest SPECT acquisition. This corresponds to histologic evidence of chronically infarcted myocardium (extended fibrosis, cellular loss) and to a very low potential for further functional recovery (9–13). The present study gives evidence, first, that the segments with <50% uptake on pinhole SPECT truly had a dramatic decrease in 99mTc-sestamibi activity. Indeed, on average, these segments had a >50% decrease in the tracer activity when determined in vitro, after animal sacrifice, and with regard to remote segments from the same histologic slices. Second, these segments with <50% uptake on pinhole SPECT had characteristics similar to those documented in CAD patients for the predominantly necrotic myocardial segments termed nonviable (13,31,34): nearly 30% volume fibrosis and >50% cellular loss, on average. Among segments with <50% uptake, however, 26% did not show >50% cellular loss, a reliable criterion of nonviable myocardium (31). This provides evidence that nonviable myocardium is slightly overestimated by this criterion, as illustrated by the SPECT images displayed in Figure 4, and the role of the partial-volume effect might be questioned. This necrosis overestimation was, however, already documented for conventional sestamibi SPECT in CAD patients (11–13), and recent reports have shown that it might be prevented while injecting sestamibi after nitrate administration (35). The value of this method, which enhances sestamibi uptake within viable but severely ischemic myocardium, remains to be addressed on the rat infarct model.
In the present study, predominantly necrotic LV segments, as defined by <50% of sestamibi uptake, were detected in vivo by pinhole SPECT as early as 48 h after LAD occlusion in adult rats. Using this criterion, significant necrosis was absent in as many as 40% of cases, limited (<30% LV) in 30%, and large (30% LV) in 30%. Similar proportions were previously documented by histologic postmortem analyses (2–4,7,8). Because of variability in coronary anatomy and collateral circulation, coronary occlusion is known to provide variable areas of infarction in rats (2–8).
During subsequent follow-up, there was a significant though limited enlargement of the SPECT infarct size, especially when a large infarct area (30% LV) was documented at 48 h. This observation is likely to relate to the thinning and expansion of necrotic areas. Indeed, both have been consistently documented by previous histologic studies of transmural infarction in rat (2,3) and they might be the result of myocyte slippage (a decrease in the number of cells across the wall) (36). Moreover, LV remodeling was closely related to SPECT infarct size determined at 48 h: the increase in EDV was high and sustained during the overall 12-wk follow-up in rats with large infarction, much lower and limited to the first half of follow-up in rats with smaller infarction, and, finally, rats with undetectable SPECT necrosis evolved similarly to sham-operated ones. As much as 85% of the further increase in EDV could be predicted by initial infarct size. This provides evidence that the variable LV remodeling, documented after LAD occlusion in adult male Wistar rats, is primarily related to the variable extent of the necrotic area. This situation is, therefore, different from that of CAD patients, in whom ischemic remodeling might be more variable according to inherited characteristics and to acquired diseases (diabetes, hypertension, aging, etc.) (1).
In a previous echographic study of the rat infarct model, an early decrease in LV function (LV fractional shortening) was found to be a good predictor of further remodeling (7). Similar results were achieved in the present study with the LV EF determined at 48 h. The extent of SPECT necrosis was, however, a much better predictor, as evidenced by results displayed in Figure 5. Other studies had already shown that the amount and duration of LV remodeling were related to infarct size (2–4,7,8) but this size was determined by postmortem histologic analyses and, thus, it could not be used to predict further remodeling. The dual noninvasive determinations of infarct size and LV function, which are provided by pinhole gated SPECT, might be particularly useful when assessing therapeutic interventions on the rat infarct model. It is likely that such interventions may have very different effects according to the initial infarct size and, thus, to the expected further remodeling.
|
CONCLUSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONReferences |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONReferences |
|---|
Related articles in JNM:
This article has been cited by other articles:
|
|
 |
|
  H. M. Sherif, A. Saraste, E. Weidl, A. W. Weber, T. Higuchi, S. Reder, T. Poethko, G. Henriksen, D. Casebier, S. Robinson, et al. Evaluation of a Novel 18F-Labeled Positron-Emission Tomography Perfusion Tracer for the Assessment of Myocardial Infarct Size in Rats Circ Cardiovasc Imaging, March 1, 2009; 2(2): 77 - 84. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Tran, P. R. Franken, F. Maskali, J. Nloga, P. Maureira, S. Poussier, F. Groubatch, C. Vanhove, J.-P. Villemot, and P.-Y. Marie Intramyocardial Implantation of Bone Marrow-Derived Stem Cells Enhances Perfusion in Chronic Myocardial Infarction: Dependency on Initial Perfusion Depth and Follow-up Assessed by Gated Pinhole SPECT J. Nucl. Med., March 1, 2007; 48(3): 405 - 412. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | RSS | TABLE OF CONTENTS |
| JOURNAL OF NUCLEAR MEDICINE TECHNOLOGY | THE JOURNAL OF NUCLEAR MEDICINE |
; n = 9) and rats with LAD occlusion and large (
; n = 6), limited (<30% of LV; •; n = 6), or undetectable (
; n = 8) 48-h SPECT necrosis. (A) SPECT necrosis. (B) EF. (C) EDV. (D) ESV. *P < 0.05 for comparisons vs. rats with limited necrosis (<30% LV). 
P < 0.05 for comparisons vs. rats with undetectable necrosis. 
P < 0.05 for comparisons vs. sham-operated rats.
