| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Radiology, Tohoku University Graduate School of Medicine,
Seiryomachi, Aobaku, Sendai 980-8574, Japan.
2 Department of Cardiovascular Medicine, Tohoku University Graduate School of
Medicine, Aobaku, Sendai, Japan.
Received August 25, 2007;
accepted after revision March 17, 2008.
Address correspondence to A. Ichinose
(azusa{at}rad.med.tohoku.ac.jp).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Delayed contrast-enhanced MRI was performed in 40 patients with sarcoidosis (11 cardiac, 29 extracardiac cases). Using a 29-segment model of the left ventricle (LV), the extent of myocardial hyperenhancement was visually scored (0 = no hyperenhancement, 1 = 1–25%, 2 = 26–50%, 3 = 51–75%, 4 = 76–100% hyperenhancement) and was compared with plasma BNP level and several parameters of cardiac function.
RESULTS. Ten of the 11 patients with cardiac sarcoidosis showed myocardial hyperenhancement, whereas none of the 29 patients without cardiac sarcoidosis did. In patients with cardiac sarcoidosis, hyperenhancement was significantly more extensive in basal short axis slices than in apical short axis slices (p < 0.0005). Myocardial hyperenhancement was significantly more frequent in subepicardial layers than in subendocardial layers. The global extent of myocardial hyperenhancement was significantly correlated with plasma BNP levels and the LV end-diastolic volume index and was negatively correlated with the LV ejection fraction.
CONCLUSION. In patients with cardiac sarcoidosis, myocardial lesions detected on delayed contrast-enhanced MRI were predominantly localized in the basal and subepicardial myocardium. The extent of myocardial lesions may be related to LV dysfunction and plasma BNP level in patients with cardiac sarcoidosis.
Keywords: delayed contrast enhancement • left ventricular function • MRI • myocardium • myocardial infarction • noncaseating granulomas • sarcoidosis
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although cardiac involvement in sarcoidosis has been reported to be clinically evident in fewer than 5% of patients with sarcoidosis [3], postmortem studies have revealed cardiac involvement in 20–27% of such patients in the United States [2]. In Japan, cardiac involvement is frequently present in up to 58% of patients with sarcoidosis at autopsy and is responsible for as many as 85% of the deaths from sarcoidosis [4]. Current diagnostic imaging techniques such as 2D echocardiography and SPECT myocardial perfusion imaging have the disadvantages of low sensitivity or low specificity (or both) for diagnosing cardiac sarcoidosis [5, 6].
Delayed contrast-enhanced MRI with an inversion recovery sequence allows visualization of scar tissue in patients with myocardial infarction (MI) or hypertrophic cardiomyopathy [7–9]. Delayed contrast-enhanced MRI is also useful for evaluating and monitoring inflammatory heart disease [10]. Several reports have shown the usefulness of delayed contrast-enhanced MRI with an inversion recovery sequence in patients with card iac sarcoidosis [11–13], but the topographic and intramural localization of myocardial lesions, shown as regions of hyperenhancement, is poorly understood. Furthermore, the association between the extent of hyperenhancement and LV global function has not been clarified.
In the present study, we examined whether delayed contrast-enhanced MRI is useful for determining the topographic and intramural localization of myocardial lesions in patients with cardiac sarcoidosis. We also tested the hypothesis that the extent of myocardial lesions is correlated with LV enlargement, LV systolic dysfunction, and plasma concentrations of brain natriuretic peptide (BNP), all of which are useful prognostic markers in patients with heart failure [14, 15].
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We retrospectively studied 40 consecutive patients with systemic sarcoidosis who underwent MRI between June 2002 and May 2005. All patients were diagnosed with sarcoidosis on the basis of histologic evidence of noncaseating epithelioid granulomas with giant cells. Biopsies were obtained from lung (n = 16), lymph nodes (n = 11), skin (n = 10), muscle (n = 2), and bone (n = 1). The patients either had cardiac symptoms (n = 16) or were screened for cardiac involvement in the absence of cardiac symptoms (n = 24). All patients underwent standard 12-lead ECG, 24-hour Holter monitoring, and echocardiography.
According to the Japanese Ministry of Health and Welfare guidelines [16] (Appendix 1), 11 patients in the study were clinically diagnosed with cardiac sarcoidosis. Coronary angiography excluded coronary artery disease in all patients. Confirmation of cardiac sarcoidosis by endomyocardial biopsy was not obtained for any of the patients. Plasma BNP level was measured in all 11 patients with cardiac sarcoidosis and in 22 of the 29 patients without cardiac sarcoidosis within 2 weeks of the MRI study. Plasma BNP levels were not measured in the seven remaining patients without cardiac sarcoidosis.
|
|
|
Using the same slice levels as described, delay ed contrast-enhanced MR images using inversion recovery turbo FLASH with a segmented inversion recovery gradient-echo pulse sequence [17] were acquired 15 minutes after the injection of gadopentetate dimeglumine (0.15 mmol/kg). The acquisition parameters were 7.5/3.4; flip angle, 15°; field of view, 300 mm; 33 lines per segment; matrix, 220 x 256; and slice thickness, 6 mm. The inversion time (200–300 milliseconds) was adjusted to null signal from normal myocardium.
MRI Analysis
Cine MR images were analyzed to determine LV function. Using a software
system (MASS, version 4.0, MEDIS Medical Imaging Systems), endo cardial and
epicardial borders in both end-diastolic and end-systolic phases were outlined
manually on all short axes and the horizontal long axis in the cine MR images.
For global LV function, the LV volume and ejection fraction were derived from
all short-axis images and the horizontal long-axis images using a modified
Simpson rule model [18]. The
LV end-diastolic volume (LVEDV) index (LVEDV / body surface area ratio) was
calculated for normalization. For assessment of regional parameters, MR images
were analyzed according to a 29-segment model of the LV (Fig.
1A,
1B,
1C).
Using MASS, version 4.0, we measured the regional end-diastolic wall thickness; the wall motion, indicating the extent of movement of the inner ventricular wall relative to the center of the cavity between the end-diastolic and end-systolic phases; and the wall thickening, indicating the increased rate (per centage) of wall thickness between the end-systolic and end-diastolic phases. We analyzed the relationship of the regional myocardial function results to the regional myocardial hyperenhancement.
Regional hyperenhancement was scored by consensus of two radiologists based on the proportion of hyperenhanced myocardial area to the total segmental area (0 = no hyperenhancement, 1 = 1–25%, 2 = 26–50%, 3 = 51–75%, and 4 = 76–100% hyperenhancement). First, the hyperenhancement scores for each short axis slice and for each myocardial wall (septal, anterior, lateral, and inferior walls in Fig. 1A, 1B, 1C) were averaged for each patient. Then, the total sum of hyperenhancement scores for all 29 segments was obtained for each patient. For example, a patient with a hyperenhancement score of 4 in all segments would have a hyperenhancement score sum of 116.
Each segment was further divided into three layers—subepicardial, midmyocardial, and subendocardial—and each layer was assessed for hyper enhancement. Partially hyperenhanced layers were deemed hyperenhanced. The patterns of hyperenhancement within each segment were defined as follows: A, no hyperenhancement; B, transmural hyperenhancement; C, subepicardial layer–dominant hyperenhancement; D, mid myocardial layer–dominant hyperenhance ment; E, subendocardial layer–dominant hyperenhancement; and F, hyperenhancement of subepicardial and subendocardial layers with an intervening zone of nonenhanced midmyocardial layer. Subepicardial layer–dominant hyperenhancement was defined as hyperenhancement predominant ly involving the subepicardial layer with possible extension into the midmyocardial layer. Similarly, subendocardial layer–dominant hyperenhancement was defined as involving the subendocardial layer with possible extension into the midmyocardial layer. Midmyocardial layer–dominant hyperenhancement was defined as being confined to the midmyocardial layer. To more clearly compare the predominant hyperenhancement among the three layers, the number of hyper enhanced segments was counted for each layer per patient.
Statistical Analyses
All values are expressed as means ± SD. We used analysis of variance
with repeated measures followed by a post-hoc Bonferroni test to analyze
statistical differences among different myocardial regions. We used analysis
of variance to analyze statistical differences among regions with different
hyperenhancement scores, whereas differences between the two groups were
analyzed using an unpaired Student's t test. We used simple linear
regression analysis for correlations between the global extent of
hyperenhanced myocardium and the LVEDV index, the LV ejection fraction, or the
plasma concentration of BNP. The proportion of subgroups in the nontransmural
hyperenhanced group was compared using the chi-square test for goodness of
fit. A p value of < 0.05 was considered statistically
significant.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Representative MR images are shown in Figures 2 and 3. Ten of the 11 cardiac sarcoidosis patients showed myocardial hyperenhancement (positive in 202 of 319 segments), whereas none of the 29 patients without cardiac sarcoidosis showed hyperenhancement.
|
|
|
|
|
In patients with cardiac sarcoidosis, 71 myocardial segments showed transmural hyperenhancement (pattern B), and 131 myocardial segments showed nontransmural hyperenhancement (patterns C–F, Table 2). Among the nontransmural patterns, subepicardial layer–dominant hyperenhancement (pattern C) was the most frequent (p < 0.01).
|
Regional wall thickening and regional wall motion were significantly reduced in the segments with a hyperenhancement score of 4 compared with segments with hyperenhancement scores of 0–3 (p < 0.0001, Fig. 5A and p < 0.0001, Fig. 5B, respectively). Regional wall thickness did not differ significantly among the segments with different scores (Fig. 5C). In patients with cardiac sarcoidosis, the sum of the hyperenhancement scores was significantly correlated with the plasma concentration of BNP and the LVEDV index (r = 0.75, p < 0.01, and r = 0.61, p < 0.05, respectively) and was negatively correlated with the LV ejection fraction (r = –0.76, p < 0.01; Fig. 6A, 6B, 6C).
|
|
|
|
|
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Our results indicate that cardiac sarcoidosis predominantly affects the basal myocardium and the subepicardial layer. Moreover, the extent of myocardial hyperenhancement is significantly correlated with the plasma concentration of BNP and the LVEDV index and is negatively correlated with the LV ejection fraction.
The Mechanism of Myocardial Hyperenhancement in Cardiac Sarcoidosis
Delayed contrast-enhanced MRI is well established for visualizing lesions
in patients with MI [7,
8]. Although the mechanism of
myocardial hyperenhancement may differ between acute and chronic MI,
hyperenhancement is associated with an increased volume of gadolinium chelates
secondary to extracellular space expansion in acute MI and chronic myocardial
lesions [7,
8]. Mahrholdt et al.
[10] reported that the region
of myocardial hyperenhancement was closely related to the histopathology of
active myocarditis. They hypothesized that necrotic cells, characterized by
ruptured membranes, may be responsible for hyperenhancement in patients with
myocarditis.
Results from several studies have shown hyperenhancement of myocardial lesions in patients with cardiac sarcoidosis [11–13, 19–21]; however, histologic evidence showing hyperenhancement has not been reported in cardiac sarcoidosis patients to date. Pathologic studies have shown that the histologic features of cardiac sarcoidosis consist of nonspecific inflammatory changes such as infiltration of lymphocytes, interstitial edema, and damaged cardiac myocytes, resulting in interstitial fibrosis or scarring [21]. We suspect myocardial hyperenhancement in cardiac sarcoidosis may share common histologic features with myocardial hyperenhancement in MI and myocarditis (i.e., scars and inflammatory changes).
In the present study, we found that myocardial wall thickening and wall motion were decreased with an increase in hyperenhancement score, as has been reported in patients with MI [22, 23], whereas wall thickness was not significantly decreased in extensively enhanced myocardial segments. Scarring in cardiac sarcoidosis is known to be associated with granulomatous changes [24], being quite different from that in MI, and thus it may not necessarily result in wall thinning.
The Topographic Localization of Myocardial Hyperenhancement
In the present study, the analysis of regional myocardial hyperenhancement
revealed that myocardial lesions were frequently located in the basal
myocardium rather than in the apical myocardium. The localization of
hyperenhancement in the basal myocardium is consistent with earlier
morphologic and pathologic studies showing the basal localization of sarcoid
lesions [13,
24]. Several echocardiographic
studies have shown that the interventricular septum is frequently affected in
sarcoidosis [5,
25]; and in a necropsy study,
Roberts et al. [24] reported
that the LV-free wall was the most common location. In contrast, we did not
find any significant differences in the extent of hyperenhancement between the
septal, anterior, lateral, and inferior walls; this result may be explained by
differences in the diagnostic techniques between the two studies.
Echocardiography may be limited to showing the entire LV myocardium because it
is often better for depicting the interventricular septum than the LV-free
wall. This approach could lead to a disproportionate detection of
abnormalities in the interventricular septum.
The Intramural Localization of Myocardial Hyperenhancement
Echocardiography is unable to detect the intramural localization of sarcoid
lesions. It is also difficult to identify the intramural localization of
sarcoid lesions using SPECT or PET because of their limited spatial
resolution. Delayed contrast-enhanced MRI with an inversion recovery technique
has good spatial resolution and a good contrast ratio between diseased and
normal myocardium, making it suitable for analyzing the intramural
localization of small myocardial lesions
[17]. We have found, using
delayed contrast-enhanced MRI, that the myocardial hyperenhancement in cardiac
sarcoidosis was predominantly located transmurally or in the subepicardial
layer of the LV myocardium. This relatively higher frequency of subepicardial
involvement is consistent with the results of an autopsy study
[4] and a recent MRI study of
patients with myocarditis
[10].
Although the cause of sarcoidosis is unknown, histology of sarcoid lesions shows a variety of inflammatory cells [4, 26]. We speculate that the relatively frequent localization of hyperenhancement in the subepicardial layer might represent a characteristic feature of inflammatory myocardial diseases distinct from ischemic heart disease, in which preferential hyperenhancement of the subendocardial layer has been repeatedly reported [7, 8].
Correlations Between Hyperenhancement and LV Enlargement and LV Systolic Dysfunction
In the present study, the extent of hyperenhancement was significantly
correlated with the LVEDV index and was negatively correlated with the LV
ejection fraction. These results may be analogous to the observation that
infarct size is one of the most important determinants of LV enlargement and
LV systolic dysfunction after MI
[27]. The extent of
hyperenhancement was positively correlated with plasma BNP level, which is
correlated with the extent of ventricular dysfunction
[28], development of cardiac
arrhythmias [29], and
long-term survival [15] in
patients with heart failure. Although the mechanism of myocardial
hyperenhancement in patients with cardiac sarcoidosis is not fully understood,
these results suggest that the extent of hyperenhancement may be related to
impaired LV function in cardiac sarcoidosis, as seen in patients after MI. In
support of this theory, the plasma concentration of BNP has recently been
reported to be a useful marker for identifying patients with cardiac
sarcoidosis [30].
Clinical Implications
Our results suggest that delayed contrast-enhanced MRI with an inversion
recovery technique is useful for diagnosing myocardial lesions in patients
with cardiac sarcoidosis and that the myocardial lesions are predominantly
localized in the basal and subepicardial myocardium. Localization of
myocardial hyperenhancement to these regions may be a characteristic feature
of cardiac sarcoidosis and may be useful for differentiating cardiac
sarcoidosis from MI, which typically shows preferential hyperenhancement of
the subendocardial layer [7,
8]. Although cardiac MRI is a
useful tool for assessing myocardial damage and cardiac function in patients
with clinically suspicious cardiac sarcoidosis, MRI is not indicated for
evaluating those with sarcoidosis and no clinical signs of cardiac
involvement. Because our results show that the sum of the hyperenhancement
scores significantly correlated with LV function, more extensive
hyperenhancement may indicate worsened LV function in patients with cardiac
sarcoidosis. The long-term prognosis of patients with and those without
cardiac sarcoidosis was not evaluated in the present study and needs further
study.
Study Limitations
We could not obtain histologic confirmation of myocardial sarcoid lesions
in individual patients; therefore, it remains unclear whether subacute or
chronic myocardial sarcoid lesions might have influenced LV function
parameters and plasma levels of BNP.
Our MRI protocol did not include T2-weighted MRI, which has been shown to be useful for detecting the acute inflammatory process of myocardial sarcoidosis in patients with cardiac sarcoidosis [20]. Thus, a combination of segmented inversion recovery gradient-echo pulse sequences and T2-weighted sequences might provide increased sensitivity for detecting myocardial sarcoid lesions compared with our imaging protocol.
The relationship between the presence of heart failure in patients with cardiac sarcoidosis and the MRI findings was not assessed in this study because MRI was performed after the patients had been clinically stabilized by treatment for heart failure.
Conclusion
In patients with cardiac sarcoidosis, lesions detected by delayed
contrast-enhanced MRI were predominantly localized in the basal side of the
myocardium and tended to show subepicardial or transmural involvement.
Hyperenhancement may be related to LV dysfunction in patients with cardiac
sarcoidosis.
|
Acknowledgments
We are grateful to Ichiro Tsuji for his support in statistical analysis and
Tatsuo Nagasaka and the other radiologic technologists for their support in
performing the MRI examinations.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
