| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Cardiac Imaging |
1 From the Departments of Radiology (O.V., P. Legmann), Cardiology (J.V., D.D., S.W.), and Biostatistics (J.C.), Université René Descartes, Hôpital Cochin, 27 rue du Fg Saint Jacques, 75014 Paris, France; InParys, St Cloud, France (A.L.); Service Hospitalier Frederic Joliot, Orsay, France (P.C.); General Electric Medical Systems, Buc, France (C.A.); and Myology Institute, Hôpital Salpétrière, Paris, France (P.C., P. Laforet). Supported by an AFM grant. Received June 1, 2001; revision requested July 12; revision received September 24; accepted November 12. Address correspondence to O.V. (e-mail: olivier.vignaux@cch.ap-hop-paris.fr).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Conventional T1-weighted single-shot black-blood fast spin-echo and gradient-echo MR imaging of the heart was prospectively performed in 32 patients with myotonic dystrophy who required EP testing. Patients were divided into two groups according to EP test results: (a) inducible (n = 15), indicating inducible ventricular tachyarrhythmias, and (b) noninducible (n = 17). Morphologic and functional MR data were analyzed by two independent investigators. Nonparametric statistical methods and 
statistics were used.
RESULTS: No morphologic or functional abnormalities of the RV wall were observed in noninducible patients. Increased signal intensity of the RV wall, indicative of fatty replacement, was identified in 13 inducible patients. Myocardial thinning of the RV was observed in six inducible patients. An overlap of morphologically abnormal areas and areas of hypo- or dyskinesis were present in 11 inducible patients. RV outflow tract diameter was larger and RV ejection fraction was smaller in inducible patients than in noninducible patients, although differences were not significant. Interobserver agreement for MR findings was good (increased signal intensity: = 0.87, P > .30 [pairwise Wilcoxon signed rank test]; myocardial thinning: = 0.87, P > .30; hypo- or dyskinesis: = 1.00, P > .99). There was a strong relationship between MR abnormalities and inducibility during EP testing (increased signal intensity, P < .001; myocardial thinning, P < .01; hypo- or dyskinesis, P < .01).
CONCLUSION: The relationship between MR morphologic and functional RV abnormalities and EP testing suggests potential for the use of MR imaging as a noninvasive method to estimate the individual risk of arrhythmia in patients with myotonic dystrophy.
© RSNA, 2002
Index terms: Heart, arrhythmia, 523.7119 • Heart, function, 523.7119 • Heart, ventricles, 523.7119 • Myocardium, diseases, 511.1935
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Magnetic resonance (MR) imaging has shown promise in depicting focal or diffuse fatty infiltration and dysfunction of the right ventricle (RV) in patients with arrhythmogenic RV cardiomyopathy (5–7). T1-weighted spin-echo pulse sequences that are sensitive to the presence of fat are useful in assessing the adipose substitution of RV myocardium, which is an important landmark of the disease. While perception of the presence of fatty infiltration may be subjective, gradient-echo pulse sequences are also used to assess global and local kinetics and to quantify RV function (8,9).
Results of endomyocardial biopsies in patients with myotonic dystrophy have shown fatty infiltration and fibrosis of the myocardium, as well as myofibrillar degeneration, mitochondrial abnormalities, and focal myocarditis (10). MR abnormalities suggesting fatty infiltration in the RV have been reported in patients with myotonic dystrophy (11). The purpose of our study was to prospectively determine whether a relationship exists between MR abnormalities of the RV and EP test results in patients with myotonic dystrophy.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Together, two operators performed the intracardiac EP tests and programmed ventricular stimulation according to standard techniques as previously described (3). These operators were blinded to the results of MR imaging. Programmed stimulation was performed with three or fewer extrastimuli at a strength equal to twice the end-diastolic threshold delivered during spontaneous rhythm and after paced trains from two RV sites (apex and pulmonary outflow tract). According to the results of invasive EP testing (12), patients were divided into two groups: (a) inducible, indicating patients with ventricular tachyarrhythmias that were inducible with ventricular stimulation (three or more consecutive ventricular complexes), and (b) noninducible, indicating patients without inducible ventricular tachyarrhythmias. Tachyarrhythmia was defined as sustained if it lasted more than 30 seconds or if it was hemodynamically unstable.
MR Imaging
MR imaging was performed before EP testing (mean time interval, 3 days; range, 0–6 days) with a 1.5-T imager (Signa LX; GE Medical Systems, Milwaukee, Wis) and a torso phased-array coil. ECG triggering was performed with standard software. No gating problems were encountered as a result of arrhythmias. After the patient was positioned headfirst in a supine position in the center of the magnet, three-plane gradient-echo scout images were acquired.
Morphologic examination.—All data acquisitions were gated in diastole to minimize artifacts resulting from cardiac motion. Transverse multisection images were acquired to image the whole heart from the vascular pedicle to the diaphragm, with a T1-weighted non–breath-hold conventional spin-echo sequence and with the breath-hold black-blood single-shot fast spin-echo sequence previously described (13). Multisection long-axis and short-axis images of the left ventricle were also acquired with the breath-hold black-blood single-shot fast spin-echo sequence. Section thickness (5 mm) and intersection gap (0 mm) were kept constant for all sequences.
The following parameters were used: (a) ECG-triggered T1-weighted non–breath-hold spin-echo sequence, with 680–960/30 (repetition time msec/echo time msec); matrix, 192 x 256 (phase encoding x frequency encoding); rectangular field of view with a maximum dimension of 380–440 mm; four signals acquired; imaging time range, 4 minutes, 48 seconds to 5 minutes, 54 seconds (mean, 5 minutes, 21 seconds); or (b) ECG-triggered breath-hold black-blood single-shot fast spin-echo sequence, with a repetition time of more than two R-R intervals per 26-msec echo time; matrix, 128 x 256 after half-Fourier processing; rectangular 250 x 125-mm field of view; inversion time, -600 msec; refocusing flip angle, 170°; echo train length, 40; 0.5 signal acquired; and mean imaging time, 21 seconds (range, 17–26 seconds). Two ECG periods were allowed between section acquisitions to allow inverted blood to recover.
Functional examination.—Gradient-echo multisection double-angulated left ventricular short- and long-axis (four-chamber views) sequences were performed with the following parameters: segmented k-space gradient-echo (FASTCARD) sequence, with repetition time dependent on the heartbeat interval; minimum echo time; matrix, 128 x 256; rectangular field of view with a maximum dimension of 380–440 mm; flip angle, 20°; section thickness, 10 mm; number of views per segment, six; mean imaging time, 17 seconds (range, 13–21 seconds); and one section per breath hold.
Image Analysis
Morphologic analysis (spin-echo images).—All spin-echo images were analyzed independently by two experienced observers (2–4 years of experience in cardiac MR imaging) who were unaware of the clinical and ECG findings and EP test results. The images were randomized when presented to the observers. Four sites in the RV (subtricuspid valve area, apex, outflow, and free wall) were scored on the basis of a qualitative evaluation of signal intensity, and the images were graded as follows: normal, indicating medium homogeneous signal intensity of the myocardial wall; increased, indicating hyperintense area with signal intensity close to pericardial or subcutaneous fat, extending from epicardium to myocardium or nonvisualized myocardial layers; or equivocal, indicating equivocal findings. RV wall thickness was evaluated on transverse left ventricular long- and short-axis images as being either normal or decreased (focal or global). Thinning was diagnosed if walls were thinner than 3 mm.
Functional analysis (gradient-echo images).— Functional data were analyzed for all patients at the same time by the same experienced observers with dedicated software (MASS Analysis Plus V0.4; Medis, Leiden, the Netherlands). Regional wall motion of the RV was semiquantitatively assessed with cine-loop simulation in both the true-long and short-left ventricular axes. Both observers independently assigned scores ranging from 1 to 3 (1 = normokinetic, 2 = hypokinetic, and 3 = akinetic or dyskinetic) to MR findings. Right end-diastolic volumes and ejection fractions were calculated on short-axis images. Furthermore, we measured the maximal diameter of the RV outflow tract.
Statistical Analysis
Data are reported as mean ± SD (quantitative variables) or as numbers and percentages (qualitative variables). As a result of the unusual distribution of most variables and the small numbers of patients in the groups of interest, nonparametric statistical methods (Wilcoxon signed rank test, Fisher exact test, and 
2 test) were used to examine relationships between variables. Two-tailed P values were used. P values of less than .05 were considered to indicate a significant difference.
The concordance or agreement of observers for MR findings (assessed on a three-category ordinal scale: normal [normokinetic], increased [hypokinetic], or equivocal [akinetic]) was further evaluated by means of statistics (14) and pairwise Wilcoxon signed rank tests. The statistics are parameters of agreement that take chance agreement into account. The statistic scores can vary from +1 (perfect agreement) to 0 (observed agreement equivalent to agreement observed by chance) or even -1 (agreement lower than that expected with chance alone). According to Landis and Koch (15), values between 0.81 and 1.00 indicate almost perfect agreement; 0.61–0.80, substantial agreement; 0.41–0.60, moderate agreement; 0.21–0.40, fair agreement; and 0.00–0.20, poor agreement.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
MR Imaging Results
The morphologic and functional MR imaging findings in the 15 inducible patients are presented in Tables 1 and 2.
|
|
= 1.00 for the free wall and outflow tract, = 0.89 for the apex, and = 0.87 for the subtricuspid valve area; pairwise Wilcoxon signed rank test results were P > .30 for all three parameters). There was a strong relationship between inducibility during EP testing, increased signal intensity of the RV (P < .001), and myocardial thinning of the RV (P < .01).
|
= 1.00 for the free wall, outflow tract, apex, and subtricuspid valve area; pairwise Wilcoxon signed rank test, P > .99). There was a strong relationship between inducibility during EP testing and contraction abnormality of the RV (P < .01).
|
|
Clinical Correlation
Table 3 summarizes the clinical characteristics of patients with myotonic dystrophy according to EP testing and morphologic MR imaging results. Patients with RV myocardial thinning were older and mean disease duration was longer than those of noninducible patients and those of inducible patients with increased signal intensity without myocardial thinning. These differences were not significant, however (P = .5 and P = .6, respectively). No significant difference among severity of disease (degree of cytosine-thymine-guanine triplet expansion) and EP testing and MR imaging results was found (P > .5).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
On T1-weighted MR images, fatty replacement is characterized by areas of increased signal intensity when compared with that of the normal myocardium. In this study, 13 of 15 patients with myotonic dystrophy and inducible ventricular tachyarrhythmias had fatty replacement of the RV myocardium. Such fatty replacement of the RV occurs in more than 50% of normal hearts in the elderly (16,17). In a morphologic and morphometric analysis of the RV myocardium, Burke et al (18) found intramyocardial fat to be common in normal hearts, although less abundant than in the fibrofatty and fatty forms of arrhythmogenic RV cardiomyopathy. Our patient population was relatively young (mean age, 42.2 years ± 7.5). In patients with pure fatty replacement, mean age was lower (Table 3). Furthermore, among the three older patients (>60 years) with inducible ventricular tachyarrhythmias, two had myocardial thinning associated with increased signal intensity, and one had myocardial thinning without increased signal intensity.
In this series of patients with inducible ventricular tachyarrhythmias, six patients had RV myocardial thinning. This myocardial thinning may indicate fibrous replacement and has been demonstrated in typical fibrofatty arrhythmogenic RV cardiomyopathy and in the Uhl anomaly (19,20). Decreased signal intensity on T1-weighted MR images of fibrous tissue may compensate for the increase in signal intensity due to fatty infiltration, and fat deposits may therefore be indistinguishable from the normal myocardium. In our study, inducible patients with RV myocardial thinning included two patients without increased signal intensity on T1-weighted MR images. An additional problem may be found in evaluating the signal intensity of thin RV walls. This is particularly true at the apex and below the tricuspid valve area. These areas cannot always be sharply demarcated from the atrioventricular sulcus, which is usually rich in adipose tissue (Table 3).
In our study, patients with RV myocardial thinning were older than and mean disease duration was comparable to that of other patients with myotonic dystrophy, although the difference did not reach significance. This finding is in keeping with the hypothesis of Fontaine et al (19), who suggested that the fatty pattern found in patients who had died suddenly may represent an early stage of the disease, preceding the development of fibrotic tissue.
In addition to morphologic characterization on spin-echo images, the use of MR imaging for analysis of RV function has been well validated (8,9). While the presence of fatty infiltration may be the object of individual interpretations, its association with kinetic abnormalities probably increases the specificity of the diagnosis (5–7). In our study, contraction abnormalities were observed in nine of the 13 patients with fatty infiltration. Furthermore, myocardial thinning associated with hypo- or dyskinesis may suggest fibrous replacement.
The other important observation made in our study was the absence of a cardiac MR abnormality in the 17 patients without inducible ventricular arrhythmia. Therefore, the induction of ventricular tachyarrhythmias, even polymorphic ones, may be more specific than previously reported with other diseases, since MR imaging indicates the presence of anatomic abnormalities only in patients with inducible ventricular tachyarrhythmias.
In conclusion, these preliminary results suggest that MR imaging may be used as a noninvasive method to estimate the potential for inducible ventricular tachyarrhythmias in patients with myotonic dystrophy, whose arrhythmic risk differs from that expected in an unselected population. Long-term follow-up of such patients will indicate both the relationship between these MR imaging findings and spontaneous ventricular arrhythmic events and the risk of sudden cardiac death in patients with myotonic dystrophy.
|
FOOTNOTES |
|---|
Author contributions: Guarantors of integrity of entire study, O.V., D.D.; study concepts and design, O.V., D.D.; literature research, O.V., D.D.; clinical studies, O.V., A.L., J.V.; data acquisition, O.V., A.L., J.V., P. Legmann; data analysis/interpretation, O.V., A.L., D.D.; statistical analysis, O.V., J.C.; manuscript preparation, O.V., S.W., P. Legmann, D.D.; manuscript definition of intellectual content, O.V., P.C., C.A., D.D.; manuscript editing, O.V., P. Legmann; manuscript revision/review, P.C., C.A., S.W., P. Laforet, J.C.; manuscript final version approval, O.V., P. Legmann, D.D., J.C.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  F. Muntoni Cardiac Complications of Childhood Myopathies J Child Neurol, March 1, 2003; 18(3): 191 - 202. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
