Usefulness of Electron-Beam Computed Tomography in Arrhythmogenic Right Ventricular Dysplasia
Relationship to Electrophysiological Abnormalities and Left Ventricular Involvement
Background Electron-beam computed tomography (CT) may be useful for detecting myocardial fat infiltration and diagnosing arrhythmogenic right ventricular dysplasia (ARVD). There are several characteristic electron-beam CT findings of ARVD. However, the incidence, their relation to electrophysiological abnormalities, and the usefulness of electron-beam CT for evaluating left ventricular involvement are unknown. This study aimed to clarify these issues.
Methods and Results Electron-beam CT was performed in 14 patients with ARVD (ARVD group), 16 age- and sex-matched patients with right ventricular enlargement and/or dysfunction without ARVD (RV enlargement group), and 13 control subjects (control group). The incidences of abnormal electron-beam CT findings in the three groups were examined. Furthermore, we examined the endocardial fat–infiltrated areas detected by electron-beam CT (CT-A) and electrophysiologically abnormal areas detected in the mapping study (EPS-A) and compared the relationship between them in the ARVD group. (1) The frequencies of abundant epicardial adipose tissue, low-attenuation trabeculations, scalloping of the right ventricular free wall, and intramyocardial fat deposits were 86%, 71%, 79%, and 50%, respectively, in the ARVD group, whereas these findings were not observed in the RV enlargement and control groups. (2) Three ARVD patients (21%) had adipose tissue involvement of the left ventricle. (3) The relationship between CT-A and EPS-A was as follows: CT-A>EPS-A, 71%; CT-A=EPS-A, 14%; and EPS-A only, 14%.
Conclusions Characteristic electron-beam CT findings are frequently observed only in patients with ARVD. Electron-beam CT is useful for evaluating for left ventricular involvement and can estimate EPS-A.
Arrhythmogenic right ventricular dysplasia is a condition in which the right ventricle is partially or totally replaced with lipomatous or fibrolipomatous tissue. The involved myocardium evokes ventricular arrhythmias of a right ventricular origin, thus inducing syncope and sudden death.1 2 3 4 The morphological abnormalities seen in ARVD mainly affect the right ventricle, but occasionally they are accompanied by left ventricular involvement.2 3 4 5 There are several methods to detect the morphological abnormalities such as two-dimensional echocardiography, cardiac radionuclide angiography, conventional CT, and MRI. However, it is often difficult to make a definite diagnosis of left ventricular involvement with these routine, noninvasive methods currently available.
Recently, electron-beam CT has provided superior temporal resolution to conventional CT6 and enabled better objective evaluation of the ventricle because of its ability to acquire motionless cardiac and cross-sectional imaging. It also has been used to detect any adipose tissue involvement. We have previously reported some characteristic electron-beam CT findings in four patients with ARVD, such as (a) abundant epicardial adipose tissue, (b) conspicuous trabeculations with low attenuation, (c) a scalloped appearance of the right ventricular free wall, and (d) intramyocardial fat deposits.7 However, because we studied a very small number of patients in an unblinded fashion, no conclusion could be drawn regarding the utility of electron-beam CT as a screening modality for ARVD. Furthermore, the frequencies of these findings and their relationship to electrophysiological abnormalities could not be addressed. The aims of this study were (1) to clarify the frequencies of the characteristic electron-beam CT findings and their relationship to any electrophysiological abnormalities and (2) to evaluate the prevalence and characteristics of left ventricular involvement in patients with ARVD.
The study population consisted of 14 patients with ARVD who were referred to the National Cardiovascular Center in Japan from January 1989 through April 1994 because of sustained monomorphic ventricular tachycardia (ARVD group) (Table 1⇓). There were 10 men and 4 women. The mean±SD patient age was 49±12 years (range, 26 to 67 years). Four ARVD patients in this study (patients 1 through 4) were enrolled in a previous study.7 From the results of the initial examinations in these 4 patients with ARVD (patients 1 through 4), we concluded that abundant epicardial adipose tissue, low-attenuation trabeculations, scalloping of the right ventricular free wall, and intramyocardial fat deposits were relatively characteristic electron-beam CT findings in patients with ARVD. Therefore, we used the initial group of patients (patients 1 through 4) as a test set and then prospectively applied these criteria to an experimental set of the next 10 patients (patients 5 through 14). In all cases, the diagnosis of ARVD was established on the basis of (1) local or diffuse wall motion abnormalities exclusively or predominantly affecting the right ventricle and that were unrelated to coronary artery disease, dilated or hypertrophic cardiomyopathy, valvular heart disease, acute myocarditis, or congenital heart disease, (2) abnormal fatty and fibrous infiltration of the right ventricular myocardium on myocardial biopsy, and (3) recurrent sustained monomorphic ventricular tachycardia of right ventricular origin (left bundle branch block configuration) confirmed by electrophysiological studies.4 As a control, 16 age- and sex-matched patients with right ventricular enlargement and/or dysfunction without ARVD (RV enlargement group: 9 men, 7 women; mean age, 43±14 years) and 13 control subjects (control group: 10 men, 3 women; mean age, 46±12 years) also were studied (Table 2⇓). Right ventricular enlargement and/or dysfunction in the RV enlargement group were caused by atrial septal defect in 8 patients, primary pulmonary hypertension in 4 patients, and chronic pulmonary thromboembolism in 4 patients.
Electron-beam CT was performed in all patients in the three groups. Electron-beam CT was performed with a C-100 or C-150 scanner (Imatron) as previously described.7 Volume-mode scanning (scanning time, 100 ms for 512 matrix images) and cinemode scanning (scanning time, 50 ms for 256 matrix images) were performed in all patients after the administration of a nonionic contrast medium (Iopamidol 370; Nippon Schering).8 Initially, we performed serial volume-mode scanning (20 contiguous sections, 6-mm section thickness, 6-mm intervals) at end systole (40% of the RR interval on the ECG) because it was easier to evaluate the ventricular wall. These transverse scans covered the entire heart and were obtained within 40 seconds. A total dose of 60 mL of contrast medium was injected with a mechanical injector. We evaluated the morphological changes of the heart with the volume-mode scans.
After an eight-level localization scan without contrast material was performed, an adjustment was made so that the eight levels could cover the entire heart. The scanner table could turn 25° in a clockwise horizontal direction, and short-axial views of the heart could be obtained. However, because table mobility was limited, the true short-axis orientation was not used in this study. We called the orientations used in our study “near short-axis” orientations. Eight-level (covering 8 cm) near short-axis cinemode scans (10 contiguous images per level) were obtained with administration of a nonionic contrast medium to examine the function of both the left and right ventricles. We used a modified version of Simpson's method to obtain these multisection cinemode scans and then calculated the right and left ventricular end-diastolic volume index and ejection fractions.
The presence and distribution of characteristic electron-beam CT findings, such as (a) abundant epicardial adipose tissue, (b) conspicuous trabeculations with low attenuation, (c) a scalloped appearance of the right ventricular free wall, and (d) intramyocardial fat deposits, were judged by two experienced radiologists familiar with electron-beam CT. Abundant epicardial adipose tissue was defined as increased epicardial adipose tissue surrounding both ventricles. The attenuation value of the epicardial adipose tissue was −65±10 Hounsfield units (HU).9 The adiposed trabeculations and intramyocardial fat deposits could be identified because of their lower attenuation (5 to −17 HU) than intact myocardium. The criteria for identifying endocardial and epicardial borders required looking for a sharp outline between the myocardium and various structures on the outside; if no sharp border was seen, then the middle of the intermediate area was defined as the border. The papillary muscles were cut at their insertion.10
After informed consent had been obtained, an electrophysiological study was performed in all 14 ARVD patients. Endocardial catheter mapping during sinus rhythm was performed with a 6F electrode catheter with a 10-mm interelectrode distance (USCI). The biventricular mapping scheme used in this study is illustrated in Fig 1⇓. The 12 sites in each ventricle were mapped according to this scheme. The catheter sites were verified independently by multiple-plane fluoroscopy in the presence of at least two experienced physicians familiar with this mapping scheme. The catheter was repositioned to another site to confirm the reproducibility of the recordings, and only minimal morphological differences between local electrograms were noted. Stability was ensured by recording from each site for more than 30 seconds. The intracardiac electrograms were recorded at variable gain to achieve the best electrographic definition and accompanied by a 1-mV calibration signal. A 10-mm bipolar fixed-gain signal also was recorded at a 1-cm/mV amplification at each site. The electrograms were filtered at a frequency of 50 to 500 Hz and recorded on a 16-channel Mingograph (Siemens Elema) at a paper speed of 100 or 200 mm/s.
Electrographic amplitude (in millivolts) was defined as the peak-to-peak deflection measured in the 10-mm variable gain bipolar electrogram. Electrographic durations (in milliseconds) were defined as the time from the earliest electrical activity that deviated from a stable baseline to the onset of the amplification signal decay artifact (an artifact caused by electronic decay of an amplified filtered signal) measured in the fixed-gain bipolar electrogram as previously demonstrated by Cassidy et al.11 Fractionated electrograms were defined as electrograms having a duration >100 ms and multiple rapid low-amplitude (<1 mV) deflections (Fig 2⇓).
Initially, the frequencies of the characteristic electron-beam CT findings described above were examined unblinded in the initial 4 patients of the ARVD group (patients 1 through 4). We then prospectively applied these criteria to an experimental set of the next 10 ARVD patients (patients 5 through 14) and all RV enlargement patients and control subjects. We also evaluated the prevalence and characteristics of any left ventricular involvement in all 14 ARVD patients. Except for the initial 4, all ARVD patients had documented ventricular tachycardia before this diagnostic evaluation. The diagnosis of ARVD, however, was established during this study in all ARVD patients. Therefore, the electron-beam CT findings were analyzed blinded. In the ARVD group, adipose tissue replacement toward the endocardium is considered to be closely related to the genesis of the abnormal endocardial electrograms. Therefore, we compared the area in which the fractionated electrograms were recorded during ventricular mapping with the areas in which findings of (b) conspicuous trabeculations with low attenuation, (c) a scalloped appearance of the right ventricular free wall, or (d) intramyocardial fat deposits were recognized by electron-beam CT. The CT-A was defined as the area in which findings of b, c, or d were recognized by electron-beam CT, and the distribution of the CT-A was examined according to the endocardial mapping scheme (Figs 1 and 3⇑⇓) in each ARVD patient. The EPS-A was defined as the area in which the fractionated electrogram was recorded during the ventricular mapping. CT-A>EPS-A was defined as a CT-A larger but including the EPS-A, whereas CT-A<EPS-A was defined as an EPS-A larger but encompassing the CT-A. CT-A=EPS-A was defined as a CT-A equal to the EPS-A. The CT-A was examined by two experienced radiologists blinded to the results of the EPS-A, and the EPS-A also was evaluated by two experienced physicians blinded to the results of the CT-A.
The cardiac functional parameters obtained from the electron-beam CT, such as the right or left ventricular end-diastolic volume index and the right or left ventricular ejection fraction, were expressed as mean±SD. Statistical analysis was done by one-way ANOVA with the Bonferroni simultaneous multiple comparison method to test the significance of the differences among the means in the three groups. A value of P<.05 was considered statistically significant.
Clinical Findings and Cardiac Functional Parameters
All ARVD patients had a history of palpitations and/or syncope. Nine patients in the ARVD group had cardiomegaly (cardiothoracic ratio >50%). The right ventricular end-diastolic volume index was significantly larger in the ARVD and RV enlargement groups than in the control group. Conversely, the right ventricular ejection fraction was significantly lower in the ARVD and RV enlargement groups. The left ventricular end-diastolic volume index in the RV enlargement group was significantly smaller than in the ARVD and control groups, whereas the left ventricular ejection fraction did not differ significantly among the three groups. See Table 1⇑.
Characteristic Electron-Beam CT Findings and Their Frequencies
Representative electron-beam CT findings in patients of the ARVD group are shown in Figs 4⇓ and 5. In ARVD group patients, the right ventricle was enlarged, and the free wall had a scalloped appearance that was associated with abundant epicardial adipose tissue. Conspicuous trabeculations with low attenuation (5 to −17 HU) and intramyocardial fat deposits also were seen (see Fig 4⇓). In the initial group of 4 ARVD patients, the frequencies of (a) abundant epicardial adipose tissue, (b) low-attenuation trabeculations, (c) scalloping of the right ventricular free wall, and (d) intramyocardial fat deposits were 100%, 100%, 100%, and 50%, respectively. In the prospectively studied 10 ARVD patients, these frequencies were 80%, 60%, 70%, and 40%, respectively. Combining these two groups, the incidence of a, b, c, and d in all 14 ARVD patients was 86%, 71%, 79%, and 50%, respectively (Fig 6⇓). In contrast, none of these findings were observed in any of the RV enlargement patients or control subjects (Fig 7⇓).
Left Ventricular Involvement in ARVD
Representative electron-beam CT images demonstrating left ventricular involvement in ARVD patients are shown in Fig 5⇓. Three of the 14 ARVD patients (21%) had adipose tissue involvement of the left ventricle. A wedge-shaped and irregular surface of the left ventricular myocardium and a linear, low-density area (5 to −17 HU) were observed along the endocardium at the apical portion of the left ventricle, indicating direct adipose tissue involvement of the left ventricle (Fig 5⇓, a and b). In this patient (patient 13), cinemode scanning revealed severe hypokinesis of the lateral and apical segments of the left ventricle (data not shown), consistent with the findings of the volume-mode scan. Fig 5⇓ (c and d) shows linear, low-density deposits in the lateral portion of the left ventricular myocardium and in the interventricular septum of the left ventricle and the free wall of the right ventricle, respectively.
Endocardial Sites With Abnormal Electron-Beam CT Findings or Fractionated Electrogram
Twelve of the 14 (86%) ARVD patients had abnormal electron-beam CT findings of either (b) conspicuous trabeculations with low attenuation, (c) a scalloped appearance of the right ventricular free wall, or (d) intramyocardial fat deposits, which indicated adipose tissue replacement toward the ventricular endocardium. Fractionated electrograms were recognized in the right ventricle in all 14 patients and in the left ventricle in 2 (14%) patients in the ARVD group. These two patients (patients 9 and 13) also had abnormal electron-beam CT findings in the corresponding areas of their left ventricle. See Table 3⇓.
Relationship Between CT-A and EPS-A in Patients With ARVD
The relationship between the CT-A and EPS-A in ARVD patients is shown in Table 2⇑ and Fig 8⇓. The CT-A was larger than the EPS-A and tended to encompass the EPS-A in 10 of the 14 patients (71%). The CT-A was nearly equal to the EPS-A in 2 patients (14%). The remaining 2 patients (14%) had EPS-A only. The EPS-A in these patients was located locally at the diaphragmatic portion of the right ventricle.
Usefulness of Electron-Beam CT in Patients With ARVD
In this study, cinemode electron-beam CT detected significantly reduced right ventricular function in ARVD patients. Volume-mode electron-beam CT clearly depicted an enlarged right ventricle with a scalloped surface of the right ventricular free wall, conspicuous trabeculations with low attenuation, intramyocardial fat deposits, and/or abundant epicardial adipose tissue. Furthermore, these abnormal findings were observed in most patients of the ARVD group but were not seen in any of the RV enlargement patients or control subjects, indicating that these findings were characteristic and reasonably specific for ARVD. Therefore, electron-beam CT is a suitable noninvasive examination for diagnosing ARVD.
Pathological abnormalities in ARVD are usually associated with right ventricular morphological changes. The subepicardial layer is most heavily involved, whereas the endocardium and subendocardium are often intact in the early stage. During the later stages of disease, the endocardium and subendocardium are involved and the left ventricle is also involved. We believe that the electron-beam CT findings of conspicuous trabeculations with low attenuation, a scalloped appearance of the right ventricular free wall, or intramyocardial fat deposits indicate adipose tissue replacement toward the endocardium. Therefore, periodic electron-beam CT examination is a suitable, noninvasive technique for evaluating not only the structure and function of both ventricles but also the progression of ARVD.
MRI is also a useful modality for diagnosing ARVD, particularly when there is fatty degeneration in the myocardium and epicardial area.12 13 14 15 However, MRI is more expensive than the previously described noninvasive diagnostic tools including electron-beam CT and requires a prolonged period of acquisition. Unlike MRI, electron-beam CT eliminates respiratory motion and its devastating degrading effects on image sharpness completely because the exposure times of electron-beam CT are on the order of milliseconds; consequently, the patients can totally suspend respiration during the brief period of image acquisition. Millisecond exposures minimize both cardiac and respiratory motion as well as patient movement.16 Furthermore, we thought that electron-beam CT might be superior to MRI in terms of detecting fatty changes of right ventricular endocardial area because MRI might show an abnormally strong signal from the right ventricular cavity as the result of reduced right ventricular performance or blood flow abnormalities caused by altered right ventricular geometry in those with ARVD.17 Therefore, we believe that electron-beam CT is superior to MRI as a screening modality and in diagnosing ARVD.
Evaluation of Left Ventricular Involvement in ARVD
In this study, electron-beam CT disclosed that 3 of 14 ARVD patients (21%) had adipose tissue involvement of the left ventricle. In patients 9 and 13, the electrophysiological study recorded fractionated electrograms at the corresponding areas of the left ventricle, where the volume-mode scan indicated positive findings. Furthermore, in patient 13, cinemode scanning revealed severe hypokinesis of the lateral and apical segments of the left ventricle (data not shown), also consistent with the findings of the volume-mode scan. To our knowledge, a wedge-shaped defect of the left ventricular myocardium as observed in patient 13 has not yet been reported as a finding of left ventricular involvement in patients with ARVD. This patient had regional left ventricular wall motion abnormalities and a fractionated electrogram in the left ventricle, indicating a clinically advanced stage of left ventricular involvement. On the other hand, patients 8 and 9 also had adipose tissue involvement of the left ventricular myocardium without left ventricular wall motion abnormalities, indicating a subclinical stage of left ventricular involvement.
The precise mechanisms underlying left ventricular involvement in ARVD are still unclear. Manyari et al3 have suggested three possible mechanisms that could account for this left ventricular involvement. First, the left ventricular myocardium may be involved by the same process of dysplasia that affects the right ventricle. Second, left ventricular dysfunction may arise secondary to an enlarged and poorly functioning right ventricle. Third, recurrent ventricular tachycardia may produce recurrent episodes of hypotension that lead to subendocardial hypoxia and subclinical left ventricular damage.
Various noninvasive imaging techniques have been used to diagnose left ventricular involvement in patients with ARVD, including echocardiography, cardiac radionuclide angiography, and conventional CT.3 18 19 However, it is impossible to directly detect adipose tissue involvement with these conventional methods and to make a definite diagnosis of subclinical left ventricular abnormality without any left ventricular wall motion abnormality as in patients 8 and 9.
Invasive cardiac cineangiography is still the standard technique for diagnosis of ARVD.5 20 However, a severely dilated right ventricle as a result of ARVD may affect left ventricular wall motion even without left ventricular involvement.3 Therefore, left ventriculography may not be sufficiently informative for the diagnosis of left ventricular involvement in those with ARVD. Electron-beam CT can detect adipose tissue involvement directly as well as any morphological abnormalities of the left ventricle. Therefore, electron-beam CT is a suitable noninvasive technique for evaluating left ventricular involvement and even may detect a subclinical form of ARVD in those without left ventricular wall motion abnormalities.
Relationship Between CT-A and EPS-A
Fractionated electrograms are thought to be linked to the slow conduction area of a reentrant circuit and are related to sustained monomorphic ventricular tachycardia in patients with organic heart disease.21 In ARVD, fractionated electrograms are considered to originate from endocardial thin bundles of myofibrils separated by fat infiltration. Endocardial fatty changes, particularly in the right ventricle, cannot be detected easily by noninvasive imaging techniques because of its thin wall. Electron-beam CT is a useful technique for detecting endocardial adipose tissue of the right ventricle. Therefore, we investigated the relationship between CT-A and EPS-A. CT-A included or equaled the EPS-A in most ARVD patients, indicating that the area recorded by the fractionated electrogram can be estimated by electron-beam CT. In the 2 patients in whom only EPS-A was recognized, the EPS-A was located locally at the diaphragmatic portion of the right ventricle. Therefore, electron-beam CT might not detect adipose tissue involvement in this region, although we performed serial volume-mode scanning with 6-mm section thickness at end systole.
We reported that abundant epicardial adipose tissue was one of the characteristic electron-beam CT findings in ARVD. However, a distinction between the ARVD patients and the RV enlargement patients or control subjects was difficult to make if it was based on the presence of abundant epicardial fat tissue alone. It is the possible that adjacent fat may become more prominent and thus give the appearance of abundant epicardial fat when the right ventricle thins, even in control subjects. Increased epicardial fat tissue in the right ventricle was detected in some patients with RV enlargement and in some control subjects, as shown Fig 7a⇓. In these patients, the increased epicardial fat tissue was located mainly at the right ventricular apex. However, prominent and abundant epicardial fat tissue surrounding both ventricles, as shown in Figs 4 and 5⇑⇑, was not observed in any of the RV enlargement patients or control subjects. The right ventricle was not dilated and wall motion was normal in control subjects. Furthermore, no other abnormal findings, such as low-attenuation trabeculations, scalloping of the right ventricular free wall, or intramyocardial fat deposits, were observed in the RV enlargement patients or control subjects. Therefore, the diagnosis of ARVD should not be made if only abundant epicardial adipose tissue is present. In addition, the diagnosis should be based on the right ventricular morphology and function as well as the other characteristic electron-beam CT findings.
The EPS-A were examined according to the scheme in Fig 1⇑. The catheter sites were verified independently by multiple-plane fluoroscopy and the 12 sites in each ventricle were completely examined. On the other hand, the distribution of the CT-A also was examined according to this scheme. In the right ventricle, each site corresponding to the scheme was determined without difficulty. However, in the left ventricle, the identification of the sites that occupied the left ventricular anterior wall (sites 4 through 6) was relatively difficult. Electron-beam CT has the capability of producing high-resolution 3-mm or 1.5-mm images.22 23 In this study, we used 6-mm contiguous images. By using high-resolution 3-mm or 1.5-mm images, the determination of the CT-A might be made more easily and, furthermore, the diagnosis of left ventricular involvement and minor fatty changes of the myocardium might be made more precisely.
Selected Abbreviations and Acronyms
|ARVD||=||arrythmogenic right ventricular dysplasia|
|CT-A||=||endocardial fat–infiltrated areas detected by electron-beam CT|
|EPS-A||=||electrophysiologically abnormal areas detected by mapping study|
|MRI||=||magnetic resonance imaging|
This work was supported in part by a grant from the Japan Cardiovascular Research Foundation (to W.S.) from Bayer Cardiovascular Disease Research Scholarship, Osaka, Japan.
Presented in part at the 67th Scientific Session of the American Heart Association, Dallas, Tex, November 16, 1994.
- Received July 18, 1995.
- Revision received January 16, 1996.
- Accepted January 22, 1996.
- Copyright © 1996 by American Heart Association
Marcus FI, Fontaine GH, Guiraudon G, Frank R, Laurenceau JL, Malergue C, Grosgogeau Y. Right ventricular dysplasia: a report of 24 adult cases. Circulation. 1982;65:384-398.
Manyari DE, Klein GJ, Gulamhusein S, Boughner D, Guiraudon GM, Wyse G, Mitchell B, Kostuk W. Arrhythmogenic right ventricular tachycardia: a generalized cardiomyopathy? Circulation. 1983;68:251-257.
Shimizu W, Ohe T, Nagata S, Shimomura K. Arrhythmogenic right ventricular dysplasia: clinical features, diagnosis, treatment, and prognosis. In: Sekiguchi M, Olsen EGJ, eds. Cardiomyopathy Update 5. Tokyo, Japan: University of Tokyo Press; 1994:295-310.
Boyd DP, Farmer DW. Cardiac computed tomography. In: Collin SM, Skorton DJ, eds. Cardiac Imaging and Imaging Processing. New York, NY: McGraw-Hill; 1986:68-87.
Reiter SJ, Rumberger JA, Feiring AJ, Stanford W, Marcus MI. Precision of measurements of right and left ventricular volume by cine computed tomography. Circulation. 1986;74:890-900.
Wegner OH. Techniques of computed tomography. In: Wegner OH, ed. Whole Body Computed Tomography. Boston, Mass: Blackwell Scientific Publications; 1993:3-10.
Lessick J, Sideman S, Azhari H, Marcus M, Grenadier E, Beyar R. Regional three-dimensional geometry and function of left ventricles with fibrous aneurysms: a cine computed tomography study. Circulation. 1991;84:1072-1086.
Cassidy DM, Vassallo JA, Buxton AE, Doherty JU, Marchlinski FE, Josephson ME. The value of catheter mapping during sinus rhythm to localize site of origin of ventricular tachycardia. Circulation. 1984;69:1103-1110.
Klersy C, Raisaro A, Salerno JA, Montemartini C, Campani R. Arrhythmogenic right and left ventricular disease: evaluation by computed tomography and nuclear magnetic resonance imaging. Eur Heart J. 1989;10(suppl D):33-36.
Sotozono K, Imahara S, Masuda H, Akashi K, Kamegai M, Miyake F, Murayama M, Sugai J. Detection of fatty tissue in the myocardium by using computed tomography in a patient with arrhythmogenic right ventricular dysplasia. Heart Vessels. 1990;5:59-61.
Blomstrom-Lundqvist C, Beckman-Suurkura M, Wallentin I, Jonsson R, Olsson SB. Ventricular dimensions and wall motion assessed by echocardiography in patients with arrhythmogenic right ventricular dysplasia. Eur Heart J. 1988;9:1291-1302.
Stanford W, Galvin JR, Weiss RM, Hajduczok ZD, Skorton DJ. Ultrafast computed tomography in cardiac imaging: a review. Semin Ultrasound CT MR. 1991:12;45-60.
Marcus ML, Weiss RM. Evaluation of cardiac structure and function with ultrafast computed tomography. In: Marcus ML, Skorton DJ, Schelbelt HR, Wolf GL, eds. Cardiac Imaging. Philadelphia, Pa: WB Saunders Co; 1991:669-681.