| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Cardiac Imaging |
1 From the Department of Diagnostic and Interventional Radiology and Nuclear Medicine, Charité, Campus Virchow Klinikum (M.G., B.S., T.T., H.B., T.D., R.F.); and Department of Cardiology, Charité, Campus Benjamin Franklin (M.N., H.P.S., U.K.), Universitätsmedizin Berlin, Berlin, Germany. From the 2006 RSNA Annual Meeting. Received December 23, 2006; revision requested February 13, 2007; revision received March 7; accepted April 13; final version accepted June 1. U.K. and H.P.S. supported in part by a grant from the Deutsche Forschungsgemeinschaft through Sonderforschungsbereich/Transregio 19 TPZ1. Address correspondence to M.G., Department of Diagnostic and Interventional Radiology, University Leipzig/Herzzentrum (Heart Institute) Leipzig, Strümpelstrasse 39, 04289 Leipzig, Germany (e-mail: matthias.gutberlet{at}herzzentrum-leipzig.de).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSION ADVANCES IN KNOWLEDGE IMPLICATION FOR PATIENT CARE...References |
|---|
Materials and Methods: Cardiac MR imaging was performed in 83 consecutive patients (55 male, 28 female; mean age, 44.8 years ± 17.7 [standard deviation]) clinically suspected of having CMC, after written informed consent was obtained according to guidelines of the local ethics committee, which approved the study. T2-weighted triple-inversion-recovery imaging to calculate the edema ratio (ER), T1-weighted imaging before and after contrast agent administration to calculate the myocardial global relative enhancement (gRE), and inversion-recovery gradient-echo imaging to evaluate areas of late gadolinium enhancement (LE) were performed. The MR results were correlated with the endomyocardial biopsy (EMB) findings to detect intramyocardial inflammation and cardiotropic viral genomes analyzed at polymerase chain reaction assay. For statistical analyses, receiver operating characteristic analysis and the Wilcoxon test for unpaired data were used because the Kolomogorov-Smirnov test revealed a distribution of data that was different from normality.
Results: Intramyocardial inflammation and cardiotropic viral persistence were confirmed at immunohistologic analysis in 48 and 49 of the 83 patients, respectively. The sensitivity, specificity, and diagnostic accuracy of the MR parameters, as compared with the immunohistologic detection of inflammation, were, respectively, 62%, 86%, and 72% for gRE; 67%, 69%, and 68% for ER; and 27%, 80%, and 49% for LE. Cardiac MR–derived gRE, ER, and LE were not associated with polymerase chain reaction proof of viral genomes.
Conclusion: In patients clinically suspected of having CMC, increased gRE and ER indicating inflammation were common findings that could be confirmed at immunohistologic analysis, whereas LE had low sensitivity and accuracy. Cardiac MR imaging may be helpful in detecting intramyocardial inflammation noninvasively, but it fails to depict viral persistence.
Supplemental material: http://radiology.rsnajnls.org/cgi/content/full/2461062179/DC1
© RSNA, 2008
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSION ADVANCES IN KNOWLEDGE IMPLICATION FOR PATIENT CARE...References |
|---|
Study investigators have concluded that cardiac magnetic resonance (MR) imaging can yield important information for the early diagnosis of AMC (9–13) and in the subsequent stages (13,14). To our knowledge, Friedrich et al (9) provided the first evidence that cardiac MR imaging enables noninvasive diagnosis of clinically suspected AMC. Their findings were not validated against EMB findings, but the diagnostic accuracy of the cardiac MR criteria used was confirmed by others (10,12,15). The visualization of regional and global myocardial edema with use of T2-weighted MR sequences has been described by using qualitative and quantitative approaches (10,15,16).
The introduction of more robust cardiac MR sequences (17,18) such as segmented inversion-recovery gradient echo has led to improved signal-to-noise and contrast-to-noise ratios for diseased and normal myocardia (9,10,12). Mahrholdt et al (11) and Rieker et al (15) used the late gadolinium enhancement (LE) technique in patients clinically suspected of having AMC and validated their findings against data derived from EMB. LE was detected in 88% of the patients. In a relatively recent study, in which a combination of all the cardiac MR imaging sequences that had been used up to that point was applied in a multisequential approach (16), it was reported that a combination of sequences had the best diagnostic accuracy in patients clinically suspected of having AMC. Again, however, these findings were not validated against EMB data on inflammation.
To our knowledge, no previous investigation had been undertaken to detect intramyocardial inflammation in patients with CMC, which is substantially lower than the inflammation in patients with AMC and can be detected only by using sensitive immunohistologic techniques rather than histologic Dallas criteria (8,19). Thus, the purpose of our study was to retrospectively compare the diagnostic accuracy of three cardiac MR imaging approaches for the detection of histologic and immunohistologic criteria (reference standard) proved myocardial inflammation in patients clinically suspected of having CMC.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSION ADVANCES IN KNOWLEDGE IMPLICATION FOR PATIENT CARE...References |
|---|
The durations of the patients' symptoms were longer than 3 months. The patients reported having dyspnea with exertion, palpitations, persisting fatigue, and/or chest discomfort at rest. There were no events suggestive of AMC, such as manifestation of acute myocardial infarction, new onset of arrhythmias, or preceding influenza-like disease. Underlying coronary artery disease was excluded by means of cardiac catheterization in all patients, and an EMB sample was taken for further evaluation to determine a specific cause of the persisting symptoms. Cardiac MR imaging was performed in all patients after biopsy. According to nested polymerase chain reaction (PCR) assay results, the patients were examined again and divided into two groups—virus positive and virus negative—and the cardiac MR findings were analyzed in relation to differences between these two groups.
Cardiac MR Imaging Protocol
For all sequences used to analyze signal intensity in our study, a body coil was used (9,16). Otherwise, the use of surface coils, owing to their inherently inhomogeneous sensitivity field, could have led to false-negative results at the inferolateral region of the left ventricle (LV) or false-positive results at the septal region if uncorrected images were used.
Transverse electrocardiographically gated T2-weighted triple-inversion-recovery images were acquired during the patient's breath hold by using the body coil with a 1.5-T MR unit (TwinSpeed-Excite; GE Healthcare, Milwaukee, Wis) to encompass the entire LV (Fig 1, A). The following parameters (16) were used: repetition time of two R-R intervals, 65-msec echo time, 150-msec inversion time, 8-mm section thickness, 350–400-mm field of view (FOV), partial FOV of 0.75, 256 x 256 matrix, echo train length of 32, receiver bandwidth of ±62.5 kHz, and acquisition time of 16 heartbeats.
|
Furthermore, cine steady-state free precession MR sequences in four- and two-chamber long-axis views, as well as the acquisition of a contiguous set of short-axis sections encompassing the entire LV, were performed by using an eight-channel cardiac coil, 3.4 (minimum)/1.5 (repetition time msec/echo time msec), a 45° flip angle, an 8-mm section thickness, a 350–400-mm FOV, a partial FOV of 1.0, a 224 x 224 matrix, 12 views per segment, retrospective gating, 50 phases per R-R interval, and an acquisition time of 16 heartbeats.
Finally, LE images were acquired, on average, 10–15 minutes after the intravenous administration of a second bolus of gadopentetate dimeglumine (0.1 mmol/kg) by using a segmented two-dimensional inversion-recovery gradient-echo technique (18,21), with constant adjustment of the inversion time to null the normal myocardium for viability assessment (21) on a complete set of short-axis sections (encompassing the entire LV) and two long-axis sections (four- and two-chamber views). LE images were acquired by using 6.5/3.1, an 8-mm section thickness, a 350–400-mm FOV, a partial FOV of 1.0, a 256 x 256 matrix, and a 20° flip angle.
Cardiac MR Analysis
First, the nonenhanced T2-weighted triple-inversion-recovery images and contrast material–enhanced T1-weighted fast spin-echo images were evaluated qualitatively for focal edema or contrast enhancement, respectively, by two independent observers (M.G., B.S., both with 5 years experience in cardiac MR imaging of myocarditis). One observer (M.G.) calculated the edema ratio (ER) (Fig 1, A) and the global relative enhancement (gRE) (Fig 1, B, C) independently by using the T2- and T1-weighted images, respectively, and an Advantage Windows (GE Healthcare) platform according to the protocol described by Friedrich et al (9). All observers were blinded to the immunohistologic EMB results, clinical course of the patients, and other evaluated clinical and MR parameters.
To calculate the ER from the T2-weighted images, a region of interest encompassing the entire LV myocardium and a second region of interest encompassing the entire visible right erector spinae or latissimus dorsi muscle (ie, skeletal muscle), depending on the homogeneity of the muscle signal intensity, were drawn on the same section (Fig 1, A). The mean myocardial signal intensity (SImyo) was correlated with the mean skeletal muscle signal intensity (SIskm) (16) by using the equation ER = SImyo/SIskm.
The gRE was calculated by using a similar approach. A region of interest encompassing the entire LV myocardium and a second region of interest encompassing the skeletal muscle within the same section were drawn on the precontrast T1-weighted images (Fig 1, B) and copied to the postcontrast images (Fig 1, C). The mean signal intensities of the myocardium and skeletal muscle before and after contrast enhancement were used (9):
 |
As cutoff values for active inflammation, an ER of 2 or higher and a gRE of 4 or higher were used. The cine MR calculations of LV ejection fraction, LV muscle mass, LV end-diastolic volume, LV end-systolic volume, and LE extent were evaluated independently by four observers (M.G., B.S., T.T., H.B., each with 2 years experience in cardiac MR imaging) who were blinded to the immunohistologic EMB results, clinical course of the patients, and other evaluated clinical and MR parameters. These analyses were performed on the short-axis images with CAAS for Windows, version 2.0 (PIE Medical Imaging, Maastricht, the Netherlands), software (Fig 2, A, B) by using an image intensity level of 2 standard deviations above the mean signal intensity of the nonenhancing myocardium to define the LE, as previously described (11,17,18,21). Furthermore, the frequency of occurrence and localization of LE were specified and characterized as predominantly epicardial, intramyocardial, subendocardial, or transmural (Fig 2).
|
EMB Analysis of Myocardial Inflammation and Viral Infection
Histologic and immunohistologic analyses of myocardial inflammation and viral genomes were performed as previously described (3,8,11,19,22). Myocardial inflammation was defined as the detection of lymphocytes positive for CD3 and leukocyte function–associated antigen-1 (median cell count > 7.0 cells per square millimeter) in association with an enhanced expression of cell adhesion molecules (human leukocyte antigen class I or II and CD54 or intracellular adhesion molecule-1) on interstitial or endothelial cells (8,23). Macrophage-1 was used as the macrophage marker. EMB specimens containing low numbers of infiltrating lymphocytes (mean CD3+ count < 7.0 cells per square millimeter) and without an abundance of cell adhesion molecules were considered not to show substantial myocardial inflammation. Published thresholds for pathologically increased infiltration and cell adhesion molecule expression were taken into consideration (8,23–25).
Statistical Analyses
For all results, means and standard deviations were derived from the data obtained by using each imaging technique. The significance of differences was analyzed by using the Wilcoxon test for unpaired data because the Kolomogorov-Smirnov test and histograms (ie, Q-Q plots) revealed a distribution of data significantly different from normality. Differences in the results obtained were considered to be significant at P < .05 and highly significant at P < .01. Furthermore, the sensitivity, specificity, and accuracy of the results obtained by using the different cardiac MR sequences and of the results obtained by using combinations of the three MR parameters (ER, LE, and gRE) to detect inflammation were determined, with histologic and immunohistologic EMB results as the reference standard (26).
Receiver operating characteristic analysis, with corresponding measures of statistical uncertainty (ie, 95% confidence intervals [CIs]), was applied to the gRE and ER values used to detect myocardial inflammation or virus persistence. Furthermore, a power analysis was performed to calculate the minimal patient sample size necessary to achieve a minimal statistical power of at least 80%.
On the basis of our experiences and data in the literature, we assumed an effect size of 0.7–1.0 for analyzed parameters (0.7 calculated for gRE, 1.1 calculated for ER). A power analysis for the t test, with identical sample numbers in both groups, was performed and adjusted so that the Wilcoxon Mann-Whitney U test could be used for unpaired data because the data distribution was different from normality. The patient number N was corrected by a factor of 0.955 (Ncorr = N/0.955). Consequently, we included a minimum of 34 patients per group to reach a minimal power of 80%. StatView, version 5.0 (SAS Institute, Cary, NC), and SPSS, version 12.0 (SPSS, Chicago, Ill), software was used for statistical analyses.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSION ADVANCES IN KNOWLEDGE IMPLICATION FOR PATIENT CARE...References |
|---|
|
|
|
|
|
|
Qualitative Assessment of Signal Intensity
A localized myocardial increase in signal intensity, similar to that described in patients with AMC (10,12), was detected in the anteroseptal region on the T2-weighted images (15) obtained in three (4%) of the 83 patients. A localized increase in signal intensity after contrast agent administration (9) was not detected on the T1-weighted images obtained in any patient. Significant LE was evident in 20 (24%) patients: in 12 (24%) of 49 patients with virus-positive findings and in eight (24%) of 34 with virus-negative findings (nonsignificant difference). In 50% (n = 10) of these 20 patients, LE localization was predominantly epicardial (Fig 2, C–E). LE localization was predominantly intramyocardial in five (25%), predominantly transmural in three (15%), and predominantly subendocardial in two (10%) patients. In the majority of patients, LE was located at the anterior (eight patients) and septal (eight patients) walls of the LV. LE was located less frequently at the lateral (six patients) and inferior (six patients) walls of the LV and at the free wall (one patient) of the right ventricle. Overlap occurred.
Quantitative Assessment of Signal Intensity
The mean ER for the patients with virus-positive findings was 1.9 ± 0.37 and therefore in the normal range. Twenty-five patients showed elevated values (Table E1, http://radiology.rsnajnls.org/cgi/content/full/2461062179/DC1). The mean gRE was 4.4 ± 2.4 and therefore slightly elevated compared with the cutoff for AMC described by Friedrich et al (9). The results in the virus-negative group did not differ significantly: The mean ER was 2.0 ± 0.4, and the mean gRE was 3.8 ± 2.4 (Table E2, http://radiology.rsnajnls.org/cgi/content/full/2461062179/DC1).
Histologic and Immunohistologic Analyses of Inflammation in EMB Specimens
All investigators blinded to the patients' clinical and cardiac MR imaging diagnoses failed to detect active or borderline myocarditis according to the Dallas criteria. There was no interobserver variability in any investigated case with respect to immunohistologic EMB evaluation based on the exact quantification of infiltrates and cell adhesion molecule expression with digital image analysis (24). In contrast, results of immunohistologic analysis, the reference standard, confirmed intramyocardial inflammation in 48 (58%) of the 83 patients (Fig 3).
Thirty-five (42%) of the 83 patients demonstrated an elevated gRE, and in 30 (86%) of these 35 patients, this elevation was consistent with the immunohistologic findings (Fig 3), yielding sensitivity, specificity, and diagnostic accuracy values of 62%, 86%, and 72%, respectively (Table). Receiver operating characteristic analysis revealed an area under the curve of 0.818 for gRE (95% CI: 0.718, 0.918) (Fig 5a).
|
Only 20 (24%) patients demonstrated LE, and in 13 (65%) of these 20 patients, the LE was consistent with the immunohistologic findings (Fig 3). These results yielded low sensitivity (27%) and diagnostic accuracy (49%), but high specificity (80%) for the detection of inflammatory processes in patients clinically suspected of having CMC (Table).
As described by Abdel-Aty et al (16) in patients with AMC, the best diagnostic performance was achieved when any two of the three MR parameters (gRE, ER, and LE) were positive for myocardial inflammation in the same patient; this approach yielded 62% sensitivity, 89% specificity, and 74% diagnostic accuracy. The use of any one of two specific parameters—gRE or ER—for MR detection of inflammation yielded similar diagnostic accuracy (72%) and higher sensitivity (79%) but lower specificity (63%).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSION ADVANCES IN KNOWLEDGE IMPLICATION FOR PATIENT CARE...References |
|---|
Although reports of cardiac MR imaging performed in patients with AMC (13) indicate a higher incidence of focal lesions (9,14,15,20), in our sample of patients with CMC, a localized signal intensity abnormality was detected on the T1- or T2-weighted images in only three patients. LE was focal, as described by other authors evaluating acute or subacute myocarditis (11,13,15), but it did not show any regional preferences, as have been observed in previous postmortem evaluations (29,30) and in vivo studies involving patients with AMC, in which LE occurred predominantly within the lateral wall (11). The epicardial preference of LE was consistent with the results of other authors (11,13).
In the patients with CMC in our study, LE did not occur with such high frequency as it reportedly has in patients with AMC (11). A similar finding has been described in patients with myocardial infarcts. One possible reason for this may be that the extent of LE in the acute phase of CMC—like the LE extent in acute myocardial infarction—represents not only irreversibly injured myocardium but also regions of edema, which may recover during the natural course of the disease in some patients. During remodeling of infarcted regions, the area of infarction (31)—and therefore most likely the area of LE also (11)—shrinks. We suggest that remodeling mechanisms similar to those seen in myocardial infarction may explain the LE in myocardial inflammation. Nevertheless, the mechanism that causes a difference in the size of the LE between acute and chronic myocardial infarction also is not completely understood. There is still ongoing discussion regarding this topic, and the overestimation of infarct size in acute myocardial infarction might be due to edema.
It is likely that a similar mechanism (myocardial edema rather than necrosis) accounts for the reversible LE in patients with AMC. Furthermore, the majority of published studies have included patients who had AMC or subacute myocarditis in association with biomarkers of myocardial injury (eg, elevated cardiac troponin levels) (11), indicating a higher number of patients with LE.
To our knowledge, Friedrich et al (9) were the first to describe the gRE calculated by using pre- and postcontrast T1-weighted images as a parameter to detect intramyocardial inflammation in patients with AMC. By using a published gRE cutoff value of 4 or higher to discriminate between acute and absent inflammation, with immunohistologic EMB analysis results as the reference standard, we achieved fair sensitivity, good specificity, and diagnostic accuracy—even in patients suspected of having CMC (Figs 3, 5a). We achieved slightly higher sensitivity—but lower specificity and accuracy—by using T2-weighted sequences to calculate the ER. These results demonstrate that gRE and ER may be useful for the noninvasive detection of inflammatory processes of the myocardium in patients clinically suspected of having CMC.
Our study had limitations. Because the examined patients did not report AMC symptoms as part of their medical history or show any signs or symptoms of rare causes of CMC, such as Lyme disease, no clinical evaluation, such as troponin level or serologic analysis, was undertaken to exclude AMC. Furthermore, we did not always use the same sequence parameters that other authors who have used the described cardiac MR techniques applied, and this might be a cause of the different results. However, the parameters that we used did not differ substantially. The T2-weighted sequence that we used differed in section thickness (8 vs 6 mm) (9,16) and matrix size (512 x 512 vs 256 x 256) only (9,16), and this might have caused slightly increased sensitivity. We performed the T2-weighted sequences by using a GE Healthcare MR unit similar to that used by Abdel-Aty et al (16) but with a different section thickness (8 vs 15 mm (16) and orientation, which might have caused decreased sensitivity owing to a lower signal-to-noise ratio. Abdel-Aty et al (16) acquired three 15-mm short-axis sections for the LE examinations compared with the contiguous 8-mm sections encompassing the entire LV that we acquired. Mahrholdt et al (11) imaged the LV with 6- and 4-mm intersection gaps from the apex to the base as we did. Abdel-Aty obtained results similar to ours with use of thicker sections in patients with AMC, and Mahrholdt et al obtained different results with thinner sections and a 4-mm gap between sections. Therefore, we believe that the differing results were influenced more by differences in the included patients than by differences in the imaging parameters.
In this study, we found gRE-based (9) and ER-based cardiac MR imaging techniques to have good sensitivity, specificity, and accuracy in the detection of inflammatory processes of the myocardium in patients clinically suspected of having CMC. In addition, a combined approach previously described by Abdel-Aty et al (16) demonstrated the best overall performance, with high sensitivity, good accuracy, and reasonable specificity when any one of two specific parameters—gRE or ER—were used. Nevertheless, the sensitivities and specificities obtained were lower than those reported in patients clinically suspected of having AMC.
Viral persistence itself cannot be detected with cardiac MR imaging. Therefore, EMB is still mandatory for virus analysis, and immunohistologic analysis is required for differentiation of the subtypes of infiltrating inflammatory cells. Even with these shortcomings, cardiac MR imaging may enable evaluation of the functional recovery of the myocardium and of the development of persistent inflammation and thus render additional EMB unnecessary. Therefore, cardiac MR imaging might also have a role in clinical trials to evaluate immunomodulatory treatment strategies (32,33).
|
ADVANCES IN KNOWLEDGE |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSION ADVANCES IN KNOWLEDGE IMPLICATION FOR PATIENT CARE...References |
|---|
|
IMPLICATION FOR PATIENT CARE |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSION ADVANCES IN KNOWLEDGE IMPLICATION FOR PATIENT CARE...References |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Abbreviations: AMC = acute human myocarditis • CI = confidence interval • CMC = chronic myocarditis • EMB = endomyocardial biopsy • ER = edema ratio • FOV = field of view • gRE = global relative enhancement • LE = late gadolinium enhancement • LV = left ventricle • PCR = polymerase chain reaction
Guarantors of integrity of entire study, M.G., R.F., H.P.S., U.K.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, M.G., B.S., T.T., H.B., T.D., M.N., U.K.; clinical studies, M.G., B.S., T.T., H.B., R.F., M.N., H.P.S., U.K.; experimental studies, M.G.; statistical analysis, M.G., T.D.; and manuscript editing, M.G., R.F., M.N., H.P.S., U.K.
Authors stated no financial relationship to disclose.
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSION ADVANCES IN KNOWLEDGE IMPLICATION FOR PATIENT CARE...References |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. G. Friedrich, U. Sechtem, J. Schulz-Menger, G. Holmvang, P. Alakija, L. T. Cooper, J. A. White, H. Abdel-Aty, M. Gutberlet, S. Prasad, et al. Cardiovascular magnetic resonance in myocarditis: A JACC White Paper. J. Am. Coll. Cardiol., April 28, 2009; 53(17): 1475 - 1487. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. T. Cooper Jr. Myocarditis N. Engl. J. Med., April 9, 2009; 360(15): 1526 - 1538. [Full Text] [PDF]
|
|
|
|
 |
|
 A. Zagrosek, H. Abdel-Aty, P. Boye, R. Wassmuth, D. Messroghli, W. Utz, A. Rudolph, S. Bohl, R. Dietz, and J. Schulz-Menger Cardiac magnetic resonance monitors reversible and irreversible myocardial injury in myocarditis. J. Am. Coll. Cardiol. Img., February 1, 2009; 2(2): 131 - 138. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Mahrholdt and U. Sechtem Noninvasive differentiation between active and healed myocarditis by cardiac magnetic resonance: are we there yet? J. Am. Coll. Cardiol. Img., February 1, 2009; 2(2): 139 - 142. [Full Text] [PDF]
|
|
|
|
|
|
M. G. Friedrich There Is More Than Shape and Function J. Am. Coll. Cardiol., November 4, 2008; 52(19): 1581 - 1583. [Full Text] [PDF]
|
|
|
|
 |
|
 U. M. Muller, S. Gielen, G. C. Schuler, and M. Gutberlet Endocardial Calcification of Left Atrium, Tracheobronchopathia Osteoplastica, and Calcified Aortic Arch in a Patient With Dyspnea Circ Heart Fail, November 1, 2008; 1(4): 290 - 292. [Full Text] [PDF]
|
|
|
|
|
|
M. G. Friedrich Tissue characterization of acute myocardial infarction and myocarditis by cardiac magnetic resonance. J. Am. Coll. Cardiol. Img., September 1, 2008; 1(5): 652 - 662. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Dennert, H. J. Crijns, and S. Heymans Acute viral myocarditis Eur. Heart J., September 1, 2008; 29(17): 2073 - 2082. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
4) and ER-based (dotted line) (cutoff value, 
