| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Cardiac Imaging |
1 From the Departments of Radiology (B.G., N.R.A.M., S.D., J.B.) and Cardiology (F.E.R.), Gasthuisberg University Hospital, Herestraat 49, B-3000 Leuven, Belgium. Received April 4, 2002; revision requested June 12; final revision received November 25; accepted December 16. B.G. supported by a Marie-Curie Fellowship of the European Commission. Address correspondence to J.B. (e-mail: jan.bogaert@uz.kuleuven.ac.be).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: In 41 patients clinically suspected of having CP and 12 healthy subjects, magnetic resonance (MR) imaging yielded information about cardiac morphology and function. On short-axis cine MR images, septal motion was assessed, and the septal and left ventricular free wall (LVFW) radii of curvature were quantified and normalized to end systole. Abnormal diastolic septal motion was expressed in terms of the largest difference in normalized radius between the septum and the LVFW. Analysis of variance was used to identify significant differences in septal shape among subject groups.
RESULTS: Left-sided septal flattening was identified in 17 of the 21 patients with surgically proven CP, in none of the 20 patients without CP, and in none of the healthy subjects. CP without septal flattening was present on the left side (n = 1), on the right side (n = 1), and at the atrioventricular grooves (n = 2). Abnormal septal motion yielded a sensitivity of 81% (17 of 21 patients), specificity of 100% (20 of 20 patients), accuracy of 90% (37 of 41 patients), positive predictive value of 100% (17 of 17 patients), and negative predictive value of 83% (20 of 24 patients) in the detection of CP. The maximal difference in normalized radius of curvature between the septum and the LVFW in the patients with CP was significantly different from that in the patients without CP (P < .001) and that in the healthy subjects (P < .001).
CONCLUSION: Abnormal diastolic septal motion is a frequent phenomenon of CP. If present in patients suspected of having CP, this finding is helpful in distinguishing CP from RCM.
Supplemental material: Supplemental material: radiology.rsnajnls.org/cgi/content/full/2282020345/DC1
© RSNA, 2003
Index terms: Heart, abnormalities, 51.821, 51.824, 51.86 • Heart, cardiomyopathy, 51.86 • Heart, MR, 51.121411, 51.121412, 51.121416, 51.12144 • Pericarditis, 51.824
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
One of the main manifestations of CP is increased ventricular interdependence, a phenomenon whereby the function of one ventricle is altered by changes in the filling of the other ventricle (3,6). Because with CP the pericardial sac has a fixed volume, the position of the ventricular septum during diastole will depend on the filling characteristics of both ventricles. This phenomenon of pathologic ventricular coupling, or increased ventricular dependence, is a major direct effect of the pericardial inflexibility. The filling pressures are increased, the rate of filling is rapid in early diastole, and the rate of change in ventricular pressure at this time in the cardiac cycle is particularly rapid. This rapid change in ventricular pressure can lead to abrupt changes in septal position, with septal flattening or eventually septal inversion (ie, paradoxical motion) (3). Moreover, this ventricular septal shift is influenced by respiration (7,8). This phenomenon of abnormal septal motion, although already described at echocardiography as a finding fitting the definition of a septal "notch" (3,9), has not to our knowledge been quantified before or correlated with the distribution of pericardial abnormalities.
The purpose of the present study was to assess ventricular septal motion and quantify the septal configuration in patients clinically suspected of having CP, and to compare these patients with healthy subjects and with patients who have other diastolic heart abnormalities.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
With the results of all examinations taken into account, 21 of the 41 patients were considered to have CP and subsequently underwent pericardiectomy. Pathologic findings confirmed the presence of CP in all cases. Thirteen of the 41 patients received a diagnosis of RCM. Pericardial effusion was found in another five patients, one of whom was suspected of having an underlying restrictive heart problem. No abnormalities were found in the remaining two patients.
Twelve healthy adults (seven men, five women; mean age, 51 years ± 14; age range, 22–63 years) were also included in the study and served as the reference population. There was a small difference in mean age between the patient and healthy subject groups (P = .02). Individuals were included as healthy subjects if their clinical and echo Doppler US examinations of the heart were normal; they were physically active but not active athletes; they showed no evidence of being overweight or of having chest wall abnormalities or coronary artery, valvular, or hypertensive disease; and they were not taking medication that was known to affect cardiac contractility.
All examinations were performed according to the guidelines of the hospital committee on medical ethics and clinical investigation, and all subjects gave informed consent to be examined for the study.
MR Imaging
MR imaging was performed in 27 patients by using a 1.5-T unit (Magnetom Vision; Siemens Medical Systems, Erlangen, Germany) with a high-performance gradient system (maximum gradient amplitude, 25 mT/m; maximum slew rate, 300-µsec rise time), a phased-array body coil, and commercially available electrocardiographic triggering. MR imaging was performed in 14 patients by using a 1.5-T unit (Intera CV; Philips Medical Systems, Best, the Netherlands) with Powertrak 6000 (Philips Medical Systems) gradients (maximum gradient amplitude, 30 mT/m; maximum slew rate, 220-µsec rise time), a dedicated cardiac software package (Philips Medical Systems), a standard five-element synergy cardiac coil, and vector cardiographic possibilities.
The MR imaging examinations began with the acquisition of survey images in three orthogonal planes (transverse, coronal, and sagittal) to localize the heart within the chest. Next, we studied the heart by performing a T1-weighted fast spin-echo sequence in the transverse cardiac short-axis direction during breath holding. Afterward, gradient-echo cine MR images of the heart were acquired by using a breath-hold fast-field-echo (3.2/1.5 [repetition time msec/echo time msec], 20° flip angle, 10-mm section thickness, 240 x 256 matrix, 350-mm field of view) or fast low-angle shot (60/4.8, 20° flip angle, 10-mm section thickness, 180 x 256 matrix, 350-mm field of view) technique, or, more recently, a breath-hold balanced fast-field-echo technique (2.7/1.4, 55° flip angle, 8-mm section thickness, 180 x 256 matrix, 350-mm field of view) in the transverse cardiac short-axis plane.
Thirteen of the 41 MR examinations were performed by using the newer balanced fast-field-echo sequence. The temporal resolution ranged from 30 to 50 msec. The average breath-hold duration was 10–15 seconds per section, depending on the heart rate. The spin-echo and gradient-echo MR images depicted the entire heart.
MR Imaging Analysis
Analysis of cardiac and pericardial morphology.—All MR images were transferred to an offline workstation (Philips Medical Systems) for analysis. The images were analyzed in consensus by two readers, who had 10 (J.B.) and 6 (S.D.) years of experience in cardiac MR imaging studies. The images were presented to the readers for analysis in a random order. For morphologic evaluation of the pericardium, spin-echo and gradient-echo MR images were analyzed. This approach allows one to differentiate a focally thickened or calcified pericardium from pericardial effusion, both of which appear hypointense on T1-weighted fast spin-echo MR images. On gradient-echo MR images, however, a fibrotic or calcified pericardium has low signal intensity in contrast to the high signal intensity of pericardial effusion.
To define the different types of CP, we measured the pericardial thickness by using an electronic caliper and classified the distribution of thickness along the pericardium by using a system similar to that suggested by Rienmüller (10). Type I CP has a global pattern, in which the entire pericardium is thickened or at least both ventricles and part of their corresponding atria are widely involved in the process of pericardial thickening. Type II CP has a focal pattern with a localized distribution of pericardial thickening along one or both atria; when this pattern is present with atrioventricular grooves, the CP is considered to be type IIa, and when this pattern is present and the ventricles are only focally involved, the CP is considered to be type IIb. Type III CP has a left-sided pattern, which means that the pericardium of the left ventricle, and eventually that of the left atrium, is thickened. Type IV CP has a right-sided pattern, with thickening of the pericardium over the right ventricle and eventually over the right atrium. Type V CP has an effusive-constrictive pattern, in which pericardial thickening is associated with effusion.
As in previous studies (4,5,11), in this study, a pericardial thickness of greater than 4 mm was considered abnormal. In the case of pericardial effusion, the maximal transverse size of the effusion was measured. The sizes of the atria and inferior vena cava (IVC) and the shape of the ventricles were visually assessed and classified. The atria were determined to be normal or mildly, moderately, or severely enlarged; the IVC was determined to be normal or enlarged; and the ventricles were determined to be normal, narrowed, or enlarged. A ventricle was defined as narrowed if the lateral ventricular border was flattened such that the ventricle had a tubular rather than normal cone shape. Since the suprahepatic part of the IVC is normally similar in size to the descending aorta at the same level (±2 cm in diameter), we performed a visual comparison to determine whether the IVC was enlarged.
Analysis of cardiac function.—Functional cardiac assessment was performed by loading the gradient-echo MR images into a cine loop to yield dynamic views of cardiac contraction and relaxation. As in the morphologic analysis, in the analysis of cardiac function, all images were studied in consensus by two readers (J.B., S.D.) with experience in cardiac MR imaging. As mentioned earlier herein, our main interest was in assessing septal motion during diastole. First, septal motion was visually assessed on the short-axis view 1 cm beneath the mitral valve plane (ie, subbasal level) and described as normal, septal flattening, or septal inversion. With the same short-axis view, we quantitatively assessed the configuration and shape of the septum by measuring the radius of curvature by means of circle best fitting the central 2 cm of the endocardial border on the left side of the ventricular septum (Fig 1). The more flattened the septum is, the larger the radius of curvature will be, and vice versa. To assess the septal changes during the cardiac cycle, we performed these measurements at all time points of the cine MR imaging examination. The radius of curvature of the left ventricular free wall (LVFW) was determined throughout the cardiac cycle in a similar way.
|
In the present study, we used the largest difference between the normalized septal radius of curvature and the normalized LVFW radius of curvature during the first half of diastole to characterize the septal configuration. In cases of paradoxical septal motion, and, thus, of inversion of the ventricular septum, no circle could be drawn on the left side of the intraventricular septum. The septal inversion was then quantified by drawing a similar circle on the right side of the intraventricular septum. Rather than using this value, however, we added the difference in radius of curvature between the right side and the left side, which was based on the last time point before septal inversion, to the left-side value. This is an objective way to express septal inversion (13,14).
Statistical Analysis
All results are expressed as means ± SDs. Analysis of variance with the Scheffé post-hoc test was performed to identify significant differences in septal shape (expressed as the largest difference in normalized radius of curvature between the septum and the LVFW during the first half of diastole) among the patients with CP, the patients without CP, and the healthy subjects. P < .05 was considered to indicate a statistically significant difference.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Of the 13 patients with RCM, only one was found to have pericardial thickening. In this 71-year-old woman, a focally thickened pericardium was found in the apicolateral part of the LVFW. In the patients with pericardial effusion, the effusion size ranged from small (pericardial width < 1 cm) to large (pericardial width > 3 cm).
Cardiac Morphology
Mild to moderate right ventricular narrowing was found in seven of the 21 patients with CP. All seven of these patients presented with right-sided or diffuse pericardial thickening. Left ventricular narrowing was found in one patient with CP. In this 71-year-old man, a large chronic pericardial hematoma compressing the left ventricle and atrium was found at surgery. None of the patients without CP demonstrated ventricular narrowing. Left atrial enlargement was seen in seven of the 21 patients with CP; the enlargement was mild in four patients, moderate in two patients, and severe in one patient. Right atrial enlargement was seen in nine of the 21 patients; the enlargement was mild in six patients, moderate in two patients, and severe in one patient. IVC enlargement was seen in 14 patients.
Of the 13 patients with RCM, nine had left atrial enlargement; the enlargement was mild in two and moderate in seven patients. Ten of the 13 patients with RCM had right atrial enlargement; the enlargement was mild in one patient, moderate in three patients, and severe in six patients. Eleven of the 13 patients had IVC enlargement. In one of the patients with pericardial effusion, important enlargement of both atria and of the IVC was found. Moreover, this patient, a 39-year-old man, had evidence of restrictive inflow physiologic features at echo Doppler US.
Ventricular Septal Motion and Configuration
Visual analysis of the short-axis cine MR images obtained during systole revealed a normal septal configuration (ie, convex to the right) with normal septal motion toward the left ventricular center in all patients. During the first half of diastole, however, an abnormal flattening of the ventricular septum was seen in 17 of the 21 patients with CP (81%) (Fig 2a) (Movie 1, radiology.rsnajnls.org/cgi/content/full/2282020345/DC1). Left-sided septal inversion (ie, convex to the left) was seen in two of these 17 patients. As shown in Figure 2 (part b) (Movie 2, radiology.rsnajnls.org/cgi/content/full/2282020345/DC1), the flattening was most obvious in the basal one-third of the septum and caused an S-like motion of the ventricular septum.
|
|
|
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
To understand the configuration of the ventricular septum in patients with CP, it is important to be familiar with the appearance of the ventricular septum in normal conditions. The position and configuration of the septum are primarily determined by using the transseptal pressure gradient. In the unloaded or unstressed heart of a living human, the septum has a flat configuration, and the normal concave shape of the septum toward the left ventricle during normal loading conditions is due to a left-to-right positive transseptal pressure gradient (16,17). On the short-axis view, the left ventricle can be seen as a relatively round structure that maintains a regular circular shape during systole because of the symmetric inward motion of its walls (6,12). During diastole, a symmetric outward motion is seen. These findings were confirmed by the present study finding of a similar pattern of change in the radii of curvature of the septum and the LVFW during the cardiac cycle.
Changes in the position and geometry of the ventricular septum occur when there is an acute or chronic imbalance between left and right ventricular loading conditions. During diastole, the septum acts as a compliant membrane between the two ventricles, and the position and geometry of the septum respond to even small alterations in the transseptal pressure gradient. The results of a study performed by Brinker et al (12) showed that the Mueller maneuver (ie, forced inspiration against a closed airway) leads to leftward septal displacement due to acute right ventricular overload, a phenomenon that lasts for only two to three heartbeats after the onset of the Mueller maneuver but persists not only during end diastole but also during systole (6,12). Left-sided septal displacement occurs not only in patients with atrial septal defect and other causes of chronic right ventricular overload but also in patients with mitral stenosis (17,18). The delayed left ventricular filling secondary to the mitral stenosis leads to an abnormal transseptal pressure gradient and accentuated early diastolic septal movement, gradually returning rightward when left filling continues.
The abnormal shape and motion of the septum, such as those seen in the majority of patients with CP in the present study, can be explained in a similar way. Ventricular interdependence is increased owing to the presence of a noncompliant pericardium, which impedes the outward movement of the ventricular free wall during filling (19). Consequently, the instantaneous diastolic transseptal gradient changes and leads to septal reconfiguration and paradoxical motion during filling. In the present study, the flattening was always to the left and most prominent in the basal part of the septum. Since right ventricular filling starts just before left ventricular filling, the septum just below the tricuspid valve is pushed to the left by the right ventricular inflow if expansion of the right ventricular free wall is impeded. Because of the thin aspect of the right ventricular free wall, the influence of pericardial thickening is more pronounced than it is on the left side of the heart. This phenomenon is enhanced during inspiration since the intrathoracic pressure decrease in patients with CP is not transmitted to the ventricles because of the thickened fibrotic pericardium. Thus, a pressure gradient between the left ventricle and the pulmonary veins is created.
Analysis of the normalized radii of curvature in patients with CP reveals similar behavior in the LVFW and the septum during systole. The normalized radii of curvature of the LVFW and the septum are different during early filling: The septal flattening leads to a peak in the septal radius of curvature that often exceeds end-diastolic radius of curvature values, with a gradual return later in diastole to a radius of curvature that is similar to that of the LVFW. This pattern was never seen in either the patients without CP or the healthy subjects in this study.
When the pathophysiologic features of abnormal septal motion with CP are taken into account, septal flattening can be absent with CP if the pericardial thickening does not impede the ventricular expansion but restricts the ventricular filling by constricting the atrioventricular groove(s), or if the constrictive pericardium compresses parts of the heart other than the right ventricle. It should be stressed that even with obvious CP involving the right ventricle, septal flattening may be absent if the ventricular constriction is minimal, as was seen in one patient in the current study.
Compared with echo Doppler US real-time measurements of the inflow patterns and ventricular contraction patterns, MR imaging data are acquired during inspiratory breath holds that last several seconds (7). The effects of respiration on cardiac inflow dynamics, which are essential features for differentiating between CP and RCM at echocardiography, could not be evaluated by using our technique. Nevertheless, septal flattening was observed only in the patients with CP and never in the other patients or healthy subjects. Since the breath-hold cine MR images were acquired at end inspiration, and, thus, right ventricular filling was enhanced, enhancement of the septal flattening in the patients with CP was expected. Real-time MR imaging analysis of ventricular function and flow assessment are now possible and will further extend the role of MR imaging in the examination of patients suspected of having CP. Such recently developed MR imaging techniques in combination with simultaneous registration of respiratory motion will enable one to study the respiratory variations associated with ventricular filling.
In conclusion, MR imaging, by facilitating analysis of diastolic septal motion during ventricular filling, has the potential to enable not only precise visualization of the thickened fibrotic and/or calcified pericardium in patients with CP but also assessment of the severity of pericardial constriction during ventricular filling and expansion. Diastolic septal flattening in patients clinically suspected of having CP is highly suggestive of CP. It should be noted, however, that the absence of diastolic septal flattening does not exclude CP if the ventricular expansion during cardiac filling—especially that on the right side—is not impeded.
|
FOOTNOTES |
|---|
Author contributions: Guarantor of integrity of entire study, J.B.; study concepts, J.B.; study design, J.B., F.E.R.; literature research, B.G., N.R.A.M.; clinical studies, S.D., J.B.; data acquisition, S.D., J.B.; data analysis/interpretation, B.G., N.R.A.M., J.B.; statistical analysis, J.B., S.D.; manuscript preparation, B.G., N.R.A.M.; manuscript definition of intellectual content, J.B.; manuscript editing, J.B., B.G.; manuscript revision/review, F.E.R.; manuscript final version approval, J.B.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  I. Yoshioka, Y. Saiki, A. Ichinose, K. Takase, S. Takahashi, T. Ohashi, M. Sato, and K. Tabayashi Tagged cine magnetic resonance imaging with a finite element model can predict the severity of retrosternal adhesions prior to redo cardiac surgery J. Thorac. Cardiovasc. Surg., April 1, 2009; 137(4): 957 - 962. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Belloni, F. De Cobelli, A. Esposito, R. Mellone, G. Perseghin, T. Canu, and A. Del Maschio MRI of Cardiomyopathy Am. J. Roentgenol., December 1, 2008; 191(6): 1702 - 1710. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Dellegrottaglie, J. Sanz, M. Poon, J. F. Viles-Gonzalez, R. Sulica, M. Goyenechea, F. Macaluso, V. Fuster, and S. Rajagopalan Pulmonary Hypertension: Accuracy of Detection with Left Ventricular Septal-to-Free Wall Curvature Ratio Measured at Cardiac MR Radiology, April 1, 2007; 243(1): 63 - 69. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. K. Eltzschig, D. J. Sugarbaker, T. W. Felbinger, C. M. Boltwood Jr., D. R. Rosing, J. Sagrista-Sauleda, J. Angel, J. Soler-Soler, and E. W. Hancock Effusive-Constrictive Pericarditis N. Engl. J. Med., May 27, 2004; 350(22): 2310 - 2312. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
= LVFW, 
= septum.
