Skip Navigation


European Heart Journal Advance Access originally published online on June 2, 2006
European Heart Journal 2006 27(13):1620-1626; doi:10.1093/eurheartj/ehl059
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
27/13/1620    most recent
ehl059v1
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (12)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by van den Bos, E. J.
Right arrow Articles by Wielopolski, P. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by van den Bos, E. J.
Right arrow Articles by Wielopolski, P. A.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The European Society of Cardiology 2006. All rights reserved. For Permissions, please e-mail: journals.permissions@oxfordjournals.org

Magnetic resonance imaging of haemorrhage within reperfused myocardial infarcts: possible interference with iron oxide-labelled cell tracking?

Ewout J. van den Bos1, Timo Baks1,2, Amber D. Moelker1, Wendy Kerver1, Robert-Jan van Geuns2, Willem J. van der Giessen1, Dirk J. Duncker1 and Piotr A. Wielopolski2,*

1 Cardiovascular Research School Coeur, Experimental Cardiology, Thoraxcenter, Erasmus MC University Medical Center, Dr Molewaterplein 40, 3015 GD Rotterdam, The Netherlands
2 Department of Radiology, Erasmus MC University Medical Center, Dr Molewaterplein 40, 3015 GD Rotterdam, The Netherlands

Received 17 November 2005; revised 3 May 2006; accepted 12 May 2006; online publish-ahead-of-print 2 June 2006.

* Corresponding author. Tel: +31 10 4088001; fax: +31 10 4634033. E-mail address: p.wielopolski{at}erasmusmc.nl


   Abstract
Top
 Abstract
Introduction
 Methods
 Histology
 Results
 Discussion
 Conclusion
 Acknowledgements
 References
 
Aims Magnetic resonance imaging (MRI) has been proposed as a tool to track iron oxide-labelled cells within myocardial infarction (MI). However, infarct reperfusion aggravates microvascular obstruction (MO) and causes haemorrhage. We hypothesized that haemorrhagic MI causes magnetic susceptibility-induced signal voids that may interfere with iron oxide-labelled cell detection.

Methods and results Pigs (n = 23) underwent 2 h occlusion of the left circumflex artery. Cine, T2*-weighted, perfusion, and delayed enhancement MRI scans were performed at 1 and 5 weeks, followed by ex vivo high-resolution scanning. At 1 week, MO was observed in 17 out of 21 animals. Signal voids were observed on T2*-weighted scans in five out of eight animals, comprising 24±22% of the infarct area. A linear correlation was found between area of MO and signal voids (R2 = 0.87; P = 0.002). At 5 weeks, MO was observed in two out of 13 animals. Signal voids were identified in three out of seven animals. Ex vivo scanning showed signal voids on T2*-weighted scanning in all animals because of the presence of haemorrhage, as confirmed by histology. Signal voids interfered with the detection of iron oxide-labelled cells ex vivo (n = 21 injections).

Conclusion Haemorrhage in reperfused MI produces MRI signal voids, which may hamper tracking of iron oxide-labelled cells.

Key Words: Magnetic resonance imaging • Myocardial infarction • Haemorrhage • Cells • Iron oxide particles


   Introduction
Top
 Abstract
 Introduction
Methods
 Histology
 Results
 Discussion
 Conclusion
 Acknowledgements
 References
 
Transplantation of cells after myocardial infarction (MI) has emerged as a promising therapy to restore heart function.1–4 To evaluate cell engraftment, non-invasive tracking of cells with magnetic resonance imaging (MRI) has been proposed as a valuable tool. Cells labelled with iron oxide, paramagnetic probes cause a T2* signal void5 and can thus be detected up to single-cell resolution in vitro.6–8 In vivo, clusters of labelled cells could be detected after injection into infarcted myocardium of mouse,9,10 rabbit,11 and pig hearts.12–17

However, in clinical practice, acute MI is treated by reperfusion therapy,18 aggravating microvascular obstruction (MO) and causing haemorrhage.19,20 Haemoglobin degradation products, such as methaemoglobin and haemosiderin, have strong magnetic susceptibility effects,21 which may mimic the signal voids caused by iron oxide-labelled cells.

We hypothesized that haemoglobin degradation products in reperfused MI produce signal voids that interfere with reliable iron oxide-labelled cell tracking. Therefore, we evaluated the MRI characteristics of subacute (1 week old) and chronic (5 weeks old) infarcts in a porcine model of reperfused MI and compared the MRI findings with histology of the infarct tissue.


   Methods
Top
 Abstract
 Introduction
 Methods
Histology
 Results
 Discussion
 Conclusion
 Acknowledgements
 References
 
Myocardial infarction
Experiments were performed in 23 Yorkshire–Landrace pigs (2–3 months old, ~25 kg). The study complied with the regulations of the Animal Care Committee of the Erasmus MC and the National Institutes of Health Publication 86–23, revised 1996. Animals were sedated (ketamine, 20 mg/kg IM and midazolam, 1 mg/kg IM), anaesthetized (thiopental, 12 mg/kg IV), intubated, and mechanically ventilated (mixture of oxygen and nitrogen, 1:2). Analgesia was maintained initially with fentanyl (12 µg/kg/min IV). Subsequently, animals underwent left coronary catheterization, followed by balloon occlusion of the left circumflex (LCx) coronary artery, proximal of the first margo obtusus branch. After 2 h, the balloon was deflated and the infarct reperfused, as confirmed by TIMI III flow on angiography. Anaesthesia was maintained with isoflurane (0.6–0.8%), starting after the occlusion of the LCx.

Magnetic resonance imaging
A 1.5 T MRI scanner with a dedicated four-element-phased array receiver coil was used (Signa CV/i; GE Medical Systems, Milwaukee, WI, USA). One week after MI, pigs were anaesthetized and ventilated as described earlier and underwent clinically applied,22 electrocardiogram (ECG)-gated cine, T2*-weighted, first-pass perfusion (FPP), and delayed enhancement (DE) MRI. The T2*-weighted sequence was applied in a subset of animals (n=8 and n=7 at 1 and 5 weeks, respectively) because of technical limitations of the scanner software at the start of the study. For detailed MRI protocols, see Supplementary material online. Breath-holding was achieved by interrupting the ventilation.

After scanning, eight animals were euthanized using an overdose of pentobarbital. Hearts were excised, immersed in saline, and scanned ex vivo using gradient echo (GE) and spin echo (SE) sequences with multiple echo times and flip angles to obtain T1-, proton density-, T2-, and T2*-weighted images (T1W, PDW, T2W, and T2*W, respectively). For detailed MRI protocols, see Supplementary material online. The minimum time interval between in vivo scanning and sacrifice was at least 3 h, and in most cases, >12 h (overnight), to make sure that all gadolinium-DTPA had been washed out.

One animal died following the 1 week scan. Five weeks after infarction, the remaining 14 animals underwent a follow-up MRI using a similar imaging protocol as at 1 week. In one animal, at 1 week and again at 5 weeks, no vascular access could be assured during scanning for injection of contrast; therefore, in this animal, no FPP and DE scans could be obtained. In one animal at 1 week, FPP scanning was not reliable for the assessment of MO owing to movement artefacts; therefore, it was not included in the analysis. Three infarct specimens were scanned a second time after standard paraffin embedding. Finally, of three animals, no ex vivo scans could be obtained because of technical limitations during scanning.

In order to compare signal voids induced by the presence of iron oxide-labelled cells and haemoglobin degradation products, ex vivo scanning was repeated after local injections with either 0.1, 1, or 4x106 iron oxide-labelled human umbilical vein endothelial cells. A total of 21 injections were performed in four subacute and three chronic infarct specimens. Cells were labelled as described previously,8 resulting in ~9.2 pg iron per cell, and fixed in 4% formaldehyde.

A flowchart (Figure 1) summarizes the protocol and the number of animals studied at each time point.


Figure 0591
View larger version (16K):
[in this window]
[in a new window]
  		Figure 1 Flowchart of the study protocol. Numbers between brackets indicate the number of animals or specimens used at a certain time point.

 
Image analysis
MR images were analysed using Cine Display Application Version 3.0 (GE Medical systems, Milwaukee, WI, USA). Signal voids were identified on mid-papillary, single-slice T2*W scans, and expressed as a percentage of the infarct area as defined by DE MRI. Signal void volumes were calculated by multiplying their area by slice thickness. MO was identified on mid-papillary, single-slice FPP images as an area of persistent subendocardial hypoenhancement.23 MI was identified on DE images as an area with delayed hyperenhancement and its size was expressed as a percentage of the left ventricular (LV) wall volume.

Contrast-to-noise ratio (CNR) was measured in the T2*W in vivo and ex vivo scans at 1 and 5 weeks, according to the relation [CNR=(SImyo–SIvoid)/SDnoise], where SImyo represents signal intensity of remote or infarcted myocardium, SIvoid, signal intensity of signal voids, and SDnoise, standard deviation of background noise measured in the air outside the pig.

Total haemorrhagic areas were measured at mid-papillary levels in PD-T2*W ex vivo scans using Clemex Vision PE analysis software (Clemex Technologies, Longueuil, Canada) and expressed as a percentage of the total infarct area.


   Histology
Top
 Abstract
 Introduction
 Methods
 Histology
Results
 Discussion
 Conclusion
 Acknowledgements
 References
 
Hearts were transversally cut in ~1 cm thick slices and stained with 1% triphenyltetrazolium chloride (37°C, 15 min; Sigma, St Louis, MO, USA) and subsequently with Prussian blue (PB), which stains haemosiderin deposits deep blue. After standard paraffin embedding, 5 µm serial sections were cut in the same plane as the MRI scans. Sections were stained with haematoxylin eosin (HE) or von Kossa's (VK), or restained with PB. VK was counterstained with von Gieson (VG). This approach allowed for identification of fat (HE), calcium (VK), collagen (VG), as well as haemosiderin (PB). Staining was analysed and compared with matching MR images at the same location and imaging plane.

Statistical analysis
Data are reported as mean±SD. Data from subacute and chronic infarcts were compared using paired or unpaired t-testing as appropriate. The correlation between area of MO and signal voids was assessed using linear regression analysis, including all animals assessed by T2*W scanning. Proportions were compared using a z-test (SigmaStat software version 2.03; SPSS Inc., Chicago, IL, USA). Sample size was determined with the following assumptions: type I error of 0.05, power of 80%, an estimated proportion of animals with an in vivo MRI detectable haemorrhage-induced signal void of 35%, based on the literature, with the assumption that a proportion of 5% would not be relevant. Therefore, eight animals would yield 80% power to detect a significant proportion of animals with such a signal void. On the basis of pilot studies, the proportion of ex vivo detected signal voids was estimated to be ~90%. A value of P < 0.05 (two-tailed) was considered statistically significant.


   Results
Top
 Abstract
 Introduction
 Methods
 Histology
 Results
Discussion
 Conclusion
 Acknowledgements
 References
 
In vivo MRI
In subacute infarcts, signal voids could be identified in five out of eight animals, using the T2*W sequence (Figure 2). In the remaining three animals, signal voids were either not observed (n = 2) or could not be reliably discerned from the air–heart interface susceptibility artefact (n = 1). The size of the signal voids was 0.53±0.51 cm3. The area of signal voids comprised 24±22% of the infarct area. MO was observed in 17 out of 21 animals, comprising 35±23% of the infarct area (Table 1). A linear relationship was found between the area of signal voids on T2*W scans and the area of MO on the corresponding FPP scans at 1 week (R2=0.87; P = 0.002; n = 8). Infarct size was 25±6% of total LV wall volume (n = 22). CNR between remote myocardium and areas of signal voids was 27±16 and between infarcted, non-haemorrhagic myocardium, and areas of signal voids 34±11 (n = 5).


Figure 0592
View larger version (90K):
[in this window]
[in a new window]
  		Figure 2 In vivo scans of one animal at 1 week (subacute) and 5 weeks (chronic) after infarction. Depicted are the end-diastolic frames of cine, T2*W, FPP, and DE scans at the same level. The infarct area is indicated by white arrows. In the T2*W image, endo- and epicardial rims of signal voids are present at 1 week. The artefact caused by the heart–air interface is indicated by an asterisk. The rims of signal voids correspond with a zone of hypoperfusion in the FPP scan at 1 week, indicating MO. Furthermore, a zone of persistent hypoenhancement can be appreciated in the DE scan within the hyperenhanced infarct area. At 5 weeks, the area of signal voids in the T2*W image takes up a larger part of the infarct area. Neither MO nor an area of persistent hypoenhancement is observed in the FPP or DE scans. Bar indicates 2 cm.

 

View this table:
[in this window]
[in a new window]
  		Table 1 Results of MRI scanning

 
In chronic infarcts, an area of signal voids could be identified in three out of seven animals, comprising 15±19% of the infarct area. MO was identified in two out of 13 animals, comprising 13±31% of the infarct area (P = 0.0007 and 0.02 vs. 1 week, respectively; Table 1). Infarct size was 15±5% (P = 0.00003 vs. 1 week; n = 13). CNR between remote myocardium and areas of signal voids was 34±21 and between infarcted, non-haemorrhagic myocardium, and areas of signal voids was 23±6 (n = 3).

Ex vivo MRI and histology
In subacute infarcts, the ex vivo high-resolution scans showed rings of magnetic susceptibility-induced signal voids on T1W and PDW scans (Figure 3). The size of the signal voids increased with a longer TE (T1-T2*W). The rings surrounded large hyperintense areas on T1W scans, which appeared hypointense on T2W and PD-T2*W scans. At histology, these rings of signal voids corresponded with blue deposits on PB staining and surrounded large areas of erythrocytes.


Figure 0593
View larger version (110K):
[in this window]
[in a new window]
  		Figure 3 Ex vivo scans 1 week after infarction using different types of tissue contrast: T1W, PDW, T2W, and T2*W with a TE of 5, 20, or 60 ms (TE5, TE20, or TE60, respectively). Either a GE or an SE sequence was used. A ring of black signal voids is identified best in the T1W scan with a TE of 20 ms (T1–T2*W). It corresponds with the blue haemosiderin deposits in the PB image. The ring surrounds a hyperenhanced area on T1W scans, which is hypoenhanced on SE/T2 and PD-T2*W scans, corresponding with methaemoglobin within intact erythrocytes.

 
In chronic infarcts, ex vivo scans showed magnetic susceptibility-induced signal voids on T1W, PDW, T2W, and T1- or PD-T2*W scans throughout the infarct area (Figure 4). Hyperintense areas were observed on T1W scans, which appeared hypointense on T2W and T1- or PD-T2*W scans (data not shown). Furthermore, hyperintense areas were observed on T2W scans, which appeared hypointense on T1W scans (data not shown). At histology, areas of signal voids matched the pattern of blue haemosiderin deposits after PB staining (Figure 5). No correlation was found with collagen, fat, or calcium deposits.


Figure 0594
View larger version (121K):
[in this window]
[in a new window]
  		Figure 4 Ex vivo scans 5 weeks after infarction using various types of tissue contrast as described in Figure 3. Magnetic susceptibility-induced signal voids are observed throughout the infarct. Their size increases with T2*-weighting (TE 20 ms). Furthermore, signal voids correspond with blue haemosiderin deposits in the PB image. Epicardial fat, indicated by F, causes a hyperintense signal on the T1W scan and a hypointense signal on T2 and T2*W scans.

 

Figure 0595
View larger version (80K):
[in this window]
[in a new window]
  		Figure 5 A tissue block from the infarct border of a 5-week-old infarction was scanned at high-resolution ex vivo. (A) shows the VK (calcium) and VG's (collagen) staining. Red staining corresponds with collagenous (scar) tissue. Yellow staining corresponds with muscle fibres. Black calcium deposits were not observed in this section. In (B), a serial section is shown after PB staining. (C) shows the corresponding slice from the MRI T2*W data set. (D) shows the same section as in (B); however, blue regions are artificially enhanced. MRI signal voids correspond with blue haemosiderin particles and not with calcium, fat, or collagenous tissue. (E) shows the same section as in (B) at higher magnification. Blue haemosiderin deposits can be appreciated. (F) shows a serial section after HE staining. (G) shows the corresponding region in the image of (A). Area of signal voids corresponds with blue haemosiderin deposits in (E). (H) shows the boxed region in (F) at higher magnification; haemosiderin deposits are visible as brownish particles (black arrows). Bar indicates 4 mm in (A–D), 400 µm in (E–G), and 40 µm in (H).

 
Total haemorrhagic areas at mid-papillary level as a percentage of the total infarct area as determined by T2*W MRI were significantly larger in subacute than in chronic infarcts (P = 0.00001; n = 8 and 11, respectively; Table 1).

Post-mortem injections with iron oxide-labelled cells could be located in remote, non-infarcted myocardium. CNR between depots of 4x106 labelled cells and infarcted, non-haemorrhagic myocardium was 18±4 in subacute infarcts (n = 4) and 15±1 in chronic infarcts (n = 3; Table 1). This was comparable with the CNR between haemorrhage-induced signal voids and infarcted, non-haemorrhagic myocardium in subacute infarcts (17±9; n = 8) and chronic infarcts (14±8; n = 11; Table 1). As a cause, in the infarct region, none of the injection sites could be reliably located and discerned from haemorrhage-induced signal voids (Figure 6).


Figure 0596
View larger version (117K):
[in this window]
[in a new window]
  		Figure 6 The left panel shows the GE/PD-T2*W/TE20 scan, depicted in Figure 4, before injection with iron oxide-labelled cells. The middle panel shows the same slice after injection with 0.1, 1, or 4x106 iron oxide-labelled cells. The right panel shows a similar series of injections in remote, non-infarcted myocardium. Although the cell injections create larger areas of signal voids in the middle panel, their precise location cannot be determined because of the signal voids induced by the presence of haemoglobin degradation products. Bar indicates 0.5 cm.

 

   Discussion
Top
 Abstract
 Introduction
 Methods
 Histology
 Results
 Discussion
Conclusion
 Acknowledgements
 References
 
Myocardial infarct reperfusion and haemorrhage
Reperfusion of MI is known to induce progressive MO, also described as the ‘no-reflow’ phenomenon.19 As shown in both animal and clinical studies, MO is followed by the development of haemorrhage, probably due to loss of microvascular integrity.20,24 The extent of the haemorrhagic area has been shown to correlate with the size of MO,20 which was also observed in the present study.

Haemorrhage and MRI appearance
Few studies have been performed to evaluate the extent of the haemorrhagic area after reperfusion of MI using MRI.24–26 One study demonstrated the presence of haemorrhage by SE imaging within 72 h after reperfused MI in dogs.25 Hypointense lesions corresponding with haemorrhage were found in 14 out of 16 animals. These lesions were not found in non-reperfused animals. Furthermore, in two clinical studies, MRI was performed shortly after infarct reperfusion. The first study of 24 patients using GE and contrast-enhanced imaging showed hypointense areas within the infarct region in 38% of patients by both techniques.24 In the second study, 39 patients were imaged using a T2*W GE sequence and haemorrhage could be identified as a hypointense zone within the infarct region in 33% of patients.26 Because the two latter studies were done in patients, no histological comparison was possible. Furthermore, MRI was performed, on average, 6 days after MI, which is much shorter than the follow-up of 5 weeks used in the present study.

In contrast with intramyocardial haemorrhage, MRI appearance of intracranial haemorrhage has been given extensive attention.21 Intracranial haemorrhage causes accumulation of deoxyhaemoglobin within hours. Within days, deoxyhaemoglobin is oxidized to methaemoglobin, which causes T1 and T2 shortening owing to its paramagnetic properties and magnetic susceptibility effect.27,28 Therefore, the presence of methaemoglobin within intact erythrocytes causes hyperintense signals on T1W scans and hypointense signals on T2W scans, a pattern observed in the present study in subacute infarcts. With further degradation, haemosiderin deposits are formed,21 which remain chronically present.29 Haemosiderin has a high magnetic susceptibility effect and thereby causes T2 shortening, resulting in signal voids on T1W, T2*W, and T2W images surrounding the brain haemorrhage from 2 weeks onwards, the so-called ‘ring’ pattern. In the present study, the same ring pattern was observed in subacute infarcts in the ex vivo scans. This difference in time course might be due to the higher in-plane resolution of ex vivo scanning, allowing early detection of such a pattern.

Iron oxide-labelled cell detection by MRI
Detection of iron oxide-labelled cells within MI has been described for mouse9,10 and rabbit11 models. Recently, several studies showed the feasibility of using iron oxide-labelled cells for clinically relevant, catheter-based delivery and cell tracking within MI in pigs. Infarcts were created by permanent coil occlusion16,17 or a 60–90 min balloon occlusion, followed by reperfusion of the infarct artery.12,13,15 Cells were injected immediately after infarction12,14 or into 1-day-,17 1-week-,15 or 4-week-13 old infarcts.

None of these studies mentioned the presence of MO or haemorrhage within the infarct centre. In contrast, in the present study, MO was present in 17 out of 21 animals in subacute infarcts, and signal voids were identified in all animals on the ex vivo scans corresponding with haemorrhage on histology. The high frequency of MO and haemorrhage could possibly be explained by the longer occlusion times used in our study (i.e. 2 h vs. 60 or 90 min),12,13,15 approximating the time to reperfusion in the clinical setting more closely. The frequency of MO in the present study is indeed comparable with the frequency in patients reperfused early after MI.22 Despite the longer occlusion times, it has been shown that haemorrhage already occurs in the pig heart after coronary occlusions more than 45 min.30

In a previous study, examining iron oxide-labelled cell detection in a mouse cryoinfarction model,10 the presence of ‘spontaneous’ signal voids within the infarct region was discussed. These signal voids were attributed to necrosis and fibrosis within the infarct scar. However, a recent study showed that cryoinfarction results in large areas of haemorrhage,31 which have most likely caused those signal voids. Furthermore, the problem of those spontaneous signal voids was purportedly solved by using a combination of PD- and T2*-weighted scanning: the areas with signal voids that were exclusively due to the tissue lesions would lead to similar signal void sizes in both imaging modalities. In contrast, the size of the signal voids generated by the presence of labelled cells was proposed to increase on T2*-weighted scans, thereby making unequivocal detection possible. However, Figures 3 and 4 clearly illustrate that the usefulness of this method could not be demonstrated in the present study, as the size of the haemorrhage-induced signal voids increased with a longer TE.

In studies of spinal cord regeneration by transplanted iron oxide-labelled cells,32,33 interference of haemorrhage with cell detection has been described as an important confounding factor. In one study, the discrimination between cells and haemorrhage was made by assessing the ‘blooming’ effect induced by iron oxide-labelled cells, where the distortion of the magnetic field occurs over a greater area than the presence of the contrast agent.32 In another study, the use of imaging sequences less susceptible to haemorrhage-induced signal voids was proposed as a solution.33

For cell detection in vivo, a certain number of labelled cells are necessary to obtain sufficient contrast with background tissue. For MI, the smallest number of cells needed for detection was reported to be 105 cells/150 µL,14 generating signal void volumes of 0.36 cm3. This study however was performed in a model of permanent occlusion in pigs. In three studies of porcine, reperfused infarct models, the number of cells per injection exceeded 28x106.12,13,15 In the present study, the haemorrhage-induced signal voids in the in vivo scans had a mean total size of 0.53±0.51 cm3 in subacute infarcts, and therefore would possibly obscure detection of cell groups of ~105 cells. Furthermore, CNR of signal voids identified on the T2*W in vivo scans were similar to the values previously reported for groups of iron oxide-labelled cells using similar scanning sequences.14 Finally, the ex vivo injections of iron oxide-labelled cells in the present study generated signal voids with a CNR similar to the haemorrhage-induced signal voids, thereby interfering with detection of the injected cells.

The presence of haemorrhage and visualization with longer TEs are important in relation to the injection route, as visualization of labelled cells was reported with local injections12–17, although intracoronary injections have been used in most human studies.3 Using iron labelling to track these cells, which will presumably spread over a larger area, requires sequences with long TE, and the artefacts described in the present study are to be expected.

Study limitations
Because it is a very subjective process to match the image locations of in vivo scans with the locations of ex vivo scans on the basis of anatomical landmarks only, and because the slice thickness of in vivo scans was 6 or 8 mm compared with 0.8 mm of ex vivo scans, we chose to describe in vivo and ex vivo findings separately. Therefore, our data do not allow for a direct comparison between in vivo and ex vivo scans.

In the present study, a model of reperfused MI was used, causing large areas of haemorrhage. Therefore, it is not clear whether there is a difference in haemorrhage-induced artefacts in reperfused vs. non-reperfused infarcts in porcine models in vivo, although in similar canine models, no haemorrhage was observed when perfusion was not reinstated.25

Only intramyocardial injections were performed in the present study using fixed, iron oxide-labelled cells, which were imaged after tissue preparation. Future studies are required to investigate whether haemorrhage-induced signal voids cause similar interference with cell detection after intracoronary injection in vivo.


   Conclusion
Top
 Abstract
 Introduction
 Methods
 Histology
 Results
 Discussion
 Conclusion
Acknowledgements
 References
 
The present study demonstrates that haemorrhage in reperfused MI produces MRI signal voids, which may interfere with reliable tracking of iron oxide-labelled cells.

Supplementary material
Supplementary material is available at European Heart Journal online.


   Acknowledgements
Top
 Abstract
 Introduction
 Methods
 Histology
 Results
 Discussion
 Conclusion
 Acknowledgements
References
 
This study was supported by ZON-MW Agiko stipend 920-03-191 (to E.J.B.) and by an ISHLT 2003 Research Fellowship Award (to E.J.B.).

Conflict of interest: none declared.


   References
Top
 Abstract
 Introduction
 Methods
 Histology
 Results
 Discussion
 Conclusion
 Acknowledgements
 References
 

  1. Orlic D, Kajstura J, Chimenti S, Jakoniuk I, Anderson SM, Li B, Pickel J, McKay R, Nadal-Ginard B, Bodine DM, Leri A, Anversa P. (2001) Bone marrow cells regenerate infarcted myocardium. Nature 410:701–705.[CrossRef][Medline]
  2. van den Bos EJ, Thompson RB, Wagner A, Mahrholdt H, Morimoto Y, Thomson LE, Wang LH, Duncker DJ, Judd RM, Taylor DA. (2005) Functional assessment of myoblast transplantation for cardiac repair with magnetic resonance imaging. Eur J Heart Fail 7:435–443.[Abstract/Free Full Text]
  3. Schachinger V, Assmus B, Britten MB, Honold J, Lehmann R, Teupe C, Abolmaali ND, Vogl TJ, Hofmann WK, Martin H, Dimmeler S, Zeiher AM. (2004) Transplantation of progenitor cells and regeneration enhancement in acute myocardial infarction: final 1 year results of the TOPCARE-AMI Trial. J Am Coll Cardiol 44:1690–1699.[Abstract/Free Full Text]
  4. Smits PC, van Geuns RJ, Poldermans D, Bountioukos M, Onderwater EE, Lee CH, Maat AP, Serruys PW. (2003) Catheter-based intramyocardial injection of autologous skeletal myoblasts as a primary treatment of ischaemic heart failure: clinical experience with six-month follow-up. J Am Coll Cardiol 42:2063–2069.[Abstract/Free Full Text]
  5. Yeh TC, Zhang W, Ildstad ST, Ho C. (1993) Intracellular labeling of T-cells with superparamagnetic contrast agents. Magn Reson Med 30:617–625.[Web of Science][Medline]
  6. Foster-Gareau P, Heyn C, Alejski A, Rutt BK. (2003) Imaging single mammalian cells with a 1.5 T clinical MRI scanner. Magn Reson Med 49:968–971.[CrossRef][Web of Science][Medline]
  7. Shapiro EM, Skrtic S, Sharer K, Hill JM, Dunbar CE, Koretsky AP. (2004) MRI detection of single particles for cellular imaging. Proc Natl Acad Sci USA 101:10901–10906.[Abstract/Free Full Text]
  8. Zhang Z, van den Bos EJ, Wielopolski PA, de Jong-Popijus M, Bernsen M, Duncker DJ, Krestin GP. (2005) In vitro imaging of single living human umbilical vein endothelial cells with a clinical 3.0 T MRI scanner. MAGMA 18:175–185.[Medline]
  9. Himes N, Min JY, Lee R, Brown C, Shea J, Huang X, Xiao YF, Morgan JP, Burstein D, Oettgen P. (2004) In vivo MRI of embryonic stem cells in a mouse model of myocardial infarction. Magn Reson Med 52:1214–1219.[CrossRef][Web of Science][Medline]
  10. Kustermann E, Roell W, Breitbach M, Wecker S, Wiedermann D, Buehrle C, Welz A, Hescheler J, Fleischmann BK, Hoehn M. (2005) Stem cell implantation in ischaemic mouse heart: a high-resolution magnetic resonance imaging investigation. NMR Biomed 18:362–370.[CrossRef][Web of Science][Medline]
  11. van den Bos EJ, Wagner A, Mahrholdt H, Thompson RB, Morimoto Y, Sutton BS, Judd RM, Taylor DA. (2003) Improved efficacy of stem cell labeling for magnetic resonance imaging studies by the use of cationic liposomes. Cell Transplant 12:743–256.[Web of Science][Medline]
  12. Kraitchman DL, Heldman AW, Atalar E, Amado LC, Martin BJ, Pittenger MF, Hare JM, Bulte JW. (2003) In vivo magnetic resonance imaging of mesenchymal stem cells in myocardial infarction. Circulation 107:2290–2293.[Abstract/Free Full Text]
  13. Garot J, Unterseeh T, Teiger E, Champagne S, Chazaud B, Gherardi R, Hittinger L, Gueret P, Rahmouni A. (2003) Magnetic resonance imaging of targeted catheter-based implantation of myogenic precursor cells into infarcted left ventricular myocardium. J Am Coll Cardiol 41:1841–1846.[Abstract/Free Full Text]
  14. Hill JM, Dick AJ, Raman VK, Thompson RB, Yu ZX, Hinds KA, Pessanha BS, Guttman MA, Varney TR, Martin BJ, Dunbar CE, McVeigh ER, Lederman RJ. (2003) Serial cardiac magnetic resonance imaging of injected mesenchymal stem cells. Circulation 108:1009–1014.[Abstract/Free Full Text]
  15. Baklanov DV, Demuinck ED, Thompson CA, Pearlman JD. (2004) Novel double contrast MRI technique for intramyocardial detection of percutaneously transplanted autologous cells. Magn Reson Med 52:1438–1442.[Medline]
  16. Karmarkar PV, Kraitchman DL, Izbudak I, Hofmann LV, Amado LC, Fritzges D, Young R, Pittenger M, Bulte JW, Atalar E. (2004) MR-trackable intramyocardial injection catheter. Magn Reson Med 51:1163–1172.[CrossRef][Medline]
  17. Dick AJ, Guttman MA, Raman VK, Peters DC, Pessanha BS, Hill JM, Smith S, Scott G, McVeigh ER, Lederman RJ. (2003) Magnetic resonance fluoroscopy allows targeted delivery of mesenchymal stem cells to infarct borders in swine. Circulation 108:2899–2904.[Abstract/Free Full Text]
  18. Giugliano RP, Braunwald E, The TIMI Study Group. (2003) Selecting the best reperfusion strategy in ST-elevation myocardial infarction: it's all a matter of time. Circulation 108:2828–2830.[Free Full Text]
  19. Rochitte CE, Lima JA, Bluemke DA, Reeder SB, McVeigh ER, Furuta T, Becker LC, Melin JA. (1998) Magnitude and time course of microvascular obstruction and tissue injury after acute myocardial infarction. Circulation 98:1006–1014.[Abstract/Free Full Text]
  20. Reffelmann T and Kloner RA. (2002) Microvascular reperfusion injury: rapid expansion of anatomic no reflow during reperfusion in the rabbit. Am J Physiol Heart Circ Physiol 283:H1099–H1107.[Abstract/Free Full Text]
  21. Bradley WG Jr. (1993) MR appearance of hemorrhage in the brain. Radiology 189:15–26.[Abstract/Free Full Text]
  22. Baks T, van Geuns RJ, Biagini E, Wielopolski P, Mollet NR, Cademartiri F, Boersma E, van der Giessen WJ, Krestin GP, Duncker DJ, Serruys PW, de Feyter PJ. (2005) Recovery of left ventricular function after primary angioplasty for acute myocardial infarction. Eur Heart J 26:1070–1077.[Abstract/Free Full Text]
  23. Lund GK, Stork A, Saeed M, Bansmann MP, Gerken JH, Muller V, Mester J, Higgins CB, Adam G, Meinertz T. (2004) Acute myocardial infarction: evaluation with first-pass enhancement and delayed enhancement MR imaging compared with 201Tl SPECT imaging. Radiology 232:49–57.[Abstract/Free Full Text]
  24. Asanuma T, Tanabe K, Ochiai K, Yoshitomi H, Nakamura K, Murakami Y, Sano K, Shimada T, Murakami R, Morioka S, Beppu S. (1997) Relationship between progressive microvascular damage and intramyocardial hemorrhage in patients with reperfused anterior myocardial infarction: myocardial contrast echocardiographic study. Circulation 96:448–453.[Abstract/Free Full Text]
  25. Lotan CS, Bouchard A, Cranney GB, Bishop SP, Pohost GM. (1992) Assessment of postreperfusion myocardial hemorrhage using proton NMR imaging at 1.5 T. Circulation 86:1018–1025.[Abstract/Free Full Text]
  26. Ochiai K, Shimada T, Murakami Y, Ishibashi Y, Sano K, Kitamura J, Inoue S, Murakami R, Kawamitsu H, Sugimura K. (1999) Hemorrhagic myocardial infarction after coronary reperfusion detected in vivo by magnetic resonance imaging in humans: prevalence and clinical implications. J Cardiovasc Magn Reson 1:247–256.[CrossRef][Web of Science][Medline]
  27. Vymazal J, Brooks RA, Patronas N, Hajek M, Bulte JW, Di Chiro G. (1995) Magnetic resonance imaging of brain iron in health and disease. J Neurol Sci 134:19–26.[CrossRef][Medline]
  28. Anzalone N, Scotti R, Riva R. (2004) Neuroradiologic differential diagnosis of cerebral intraparenchymal hemorrhage. Neurol Sci 25:S3–S5.[Medline]
  29. Hardy PA, Kucharczyk W, Henkelman RM. (1990) Cause of signal loss in MR images of old hemorrhagic lesions. Radiology 174:549–555.[Abstract/Free Full Text]
  30. Garcia-Dorado D, Theroux P, Alonso J, Elizaga J, Botas J, Fernandez-Aviles F, Soriano J, Munoz R, Solares J. (1990) Intracoronary infusion of superoxide dismutase and reperfusion injury in the pig heart. Basic Res Cardiol 85:619–629.[Medline]
  31. van den Bos EJ, Mees BM, de Waard MC, de Crom R, Duncker DJ. (2005) A novel model of cryoinjury-induced myocardial infarction in the mouse: a comparison with coronary artery ligation. Am J Physiol Heart Circ Physiol 289:H1291–H1300.[Abstract/Free Full Text]
  32. Dunning MD, Lakatos A, Loizou L, Kettunen M, ffrench-Constant C, Brindle KM, Franklin RJ. (2004) Superparamagnetic iron oxide-labeled Schwann cells and olfactory ensheathing cells can be traced in vivo by magnetic resonance imaging and retain functional properties after transplantation into the CNS. J Neurosci 24:9799–9810.[Abstract/Free Full Text]
  33. Dunn EA, Weaver LC, Dekaban GA, Foster PJ. (2005) Cellular imaging of inflammation after experimental spinal cord injury. Mol Imaging 4:53–62.[Medline]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Eur Heart JHome page
S. Orn, C. Manhenke, O. J. Greve, A. I. Larsen, V. V. S. Bonarjee, T. Edvardsen, and K. Dickstein
Microvascular obstruction is a major determinant of infarct healing and subsequent left ventricular remodelling following primary percutaneous coronary intervention
Eur. Heart J., August 2, 2009; 30(16): 1978 - 1985.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Heart Circ. Physiol.Home page
D. P. O'Regan, R. Ahmed, C. Neuwirth, Y. Tan, G. Durighel, J. V. Hajnal, I. Nadra, S. J. Corbett, and S. A. Cook
Cardiac MRI of myocardial salvage at the peri-infarct border zones after primary coronary intervention
Am J Physiol Heart Circ Physiol, July 1, 2009; 297(1): H340 - H346.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
D. P. O'Regan, R. Ahmed, N. Karunanithy, C. Neuwirth, Y. Tan, G. Durighel, J. V. Hajnal, I. Nadra, S. J. Corbett, and S. A. Cook
Reperfusion Hemorrhage Following Acute Myocardial Infarction: Assessment with T2* Mapping and Effect on Measuring the Area at Risk
Radiology, March 1, 2009; 250(3): 916 - 922.
[Abstract] [Full Text] [PDF]

Home page
 J. Thorac. Cardiovasc. Surg.Home page
H. Zhang, P. Song, Y. Tang, X.-l. Zhang, S.-h. Zhao, Y.-j. Wei, and S.-s. Hu
Injection of bone marrow mesenchymal stem cells in the borderline area of infarcted myocardium: heart status and cell distribution.
J. Thorac. Cardiovasc. Surg., November 1, 2007; 134(5): 1234 - 1240.
[Abstract] [Full Text] [PDF]

Home page
 CirculationHome page
Y. Amsalem, Y. Mardor, M. S. Feinberg, N. Landa, L. Miller, D. Daniels, A. Ocherashvilli, R. Holbova, O. Yosef, I. M. Barbash, et al.
Iron-Oxide Labeling and Outcome of Transplanted Mesenchymal Stem Cells in the Infarcted Myocardium
Circulation, September 11, 2007; 116(11_suppl): I-38 - I-45.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
27/13/1620    most recent
ehl059v1
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (12)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by van den Bos, E. J.
Right arrow Articles by Wielopolski, P. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by van den Bos, E. J.
Right arrow Articles by Wielopolski, P. A.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?