| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Radiology, Hôpital Tenon, Assistance
Publique-Hôpitaux de Paris, 4 rue de la Chine, 75020 Paris,
France.
2 Department of Obstetrics and Gynecology, Hôpital Tenon, Assistance
Publique-Hôpitaux de Paris, 75020 Paris, France.
Received April 15, 2002;
accepted after revision September 26, 2002.
Address correspondence to M. Bazot.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Fifty-six consecutive patients referred for hysterectomy prospectively underwent MR imaging. Two fast pulse sequences using a breath-hold technique—true fast imaging with steady-state free precession (FISP) and turbo inversion recovery—and turbo spin-echo T2-weighted images of the pelvis were obtained in each patient. The images were analyzed in a blinded manner and independently by three reviewers with different levels of experience for the accuracy of adenomyosis diagnosis, image quality, anatomic visualization, and image artifacts. The accuracy for the diagnosis of adenomyosis on turbo spin-echo T2-weighted imaging combined with one or two fast pulse sequences was evaluated for each reviewer.
RESULTS. Twenty-four patients (42.9%) had a histologic diagnosis of adenomyosis. The accuracy for the diagnosis of adenomyosis for reviewers 1, 2, and 3 using turbo spin-echo T2-weighted, true FISP, and turbo inversion recovery sequences was 83.9%, 67.8%, 75%; 83.9%, 67.8, 78.5%; and 87.5%, 73.2%, and 75%, respectively. A difference in the accuracy rate was found among the observers for the three sequences (p < 0.001). Whatever the pulse sequence, the accuracy rate was higher for the reviewer with more experience in gynecologic imaging. The combination of turbo spin-echo T2-weighted imaging with at least one rapid sequence increased the accuracy of observers with little experience in gynecology. With turbo inversion recovery sequences, the image quality score was low for the three reviewers compared with turbo spin-echo T2-weighted and true FISP sequences. The combination of turbo spin-echo T2-weighted and true FISP sequences gave the highest image quality scores.
CONCLUSION. Breath-hold T2-weighted sequences optimize the accuracy of MR imaging for the diagnosis of adenomyosis and reduce interobserver variability.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
To improve the diagnosis of adenomyosis, most authors recommend the use of MR imaging, particularly in patients with associated gynecologic disorders [6, 7]. Various MR imaging criteria have been proposed, including the thickness of the junctional zone, the presence of an ill-defined, relatively homogeneous low-signal-intensity area of myometrium, and the presence of high-intensity foci in the myometrium [4, 5, 6]. Nevertheless, the sensitivity and specificity of MR imaging for the diagnosis of adenomyosis range from 77.5% to 89% and from 67% to 92.5% [4, 5, 6]. These discrepancies could be explained by differences in the pathologic criteria used for adenomyosis, MR imaging criteria and techniques, and the gynecologic experience of radiologists.
The aims of this prospective study were to evaluate, in addition to turbo spin-echo T2-weighted sequences, the accuracy of breath-hold fast sequences using turbo inversion recovery and true fast imaging with steady-state free precession (FISP) for the diagnosis of adenomyosis, and intra- and interobserver variability with these methods.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
MR Imaging Technique
MR imaging was performed on a 1.5-T system (Magnetom Vision; Siemens,
Erlangen, Germany) with a phased array body coil. Patients were required to
fast for 3 hr before MR imaging. Antispasmodic drugs were not used.
T2-weighted turbo spin-echo images were acquired in the sagittal and oblique
axial or coronal planes (short axis of the uterus) with the following
parameters: TR/TE, 4500/128; echo-train length, 23; slice thickness, 5 mm;
interslice gap, 10%; rectangular field of view, 280 x 245 mm; and
matrix, 224 x 512. The phaseencoding axis was in the anteroposterior
direction for sagittal images and right-to-left for axial images. Using
abdominal compression, we acquired MR imaging sections every 5 mm with a gap
of 1 mm. These parameters yielded 16 sections in 3 min 58 sec. Respiratory
compensation and fat suppression were not used.
In addition, two breath-hold fast T2-weighted pulse sequences, one true FISP sequence, and one turbo inversion recovery sequence in the sagittal and axial planes were performed. The true FISP sequence was acquired with the following parameters: 6.3/3; flip angle, 70°; slice thickness, 5 mm; rectangular field of view, 320 x 240 mm; and matrix, 256 x 198. These parameters yielded 12 sections in 18 sec. The turbo inversion recovery sequence was acquired with the following parameters: 3905/76; inversion time, 150 msec; echo-train length, 150; slice thickness, 7 mm; rectangular field of view, 320 x 240 mm; and matrix, 256 x 224. These parameters yielded 11 sections in 20 sec.
MR Imaging Criteria for Diagnosing Adenomyosis
Adenomyosis was defined as a junctional zone of at least 12 mm or an
ill-defined low-signal-intensity area of myometrium or punctuate
high-intensity myometrial foci
[5]. Adenomyosis was classified
according to its uterine location and size, but no attempt was made to grade
the depth of myometrial involvement in this study.
Accuracy of MR Imaging Diagnosis
MR images were analyzed by three reviewers, radiologists who had different
levels of experience in MR imaging of the female pelvis. Reviewer 1 (an M.D.)
was highly experienced in female pelvic imaging, reviewer 2 (an M.D./Ph.D.)
was highly experienced in upper abdominal and vascular MR imaging, and
reviewer 3 (a fellow) was being trained in MR imaging but had performed more
than 150 MR imaging examinations of the female pelvis over a 6-month period.
The diagnosis of adenomyosis was based on previously described criteria. To
ensure consistency, before reviewing the study images, the three reviewers
were presented with five selected nonstudy cases serving as examples. The
reviewers were informed of the sequence type.
The accuracy rate of each reviewer for the diagnosis of adenomyosis was evaluated, in light of the histologic results, separately for the turbo spinecho T2-weighted, turbo inversion recovery, and true FISP sequences.
For combinations of two MR imaging sequences, the diagnosis of adenomyosis was made when one sequence suggested the diagnosis. For the combination of the three MR imaging sequences, the diagnosis of adenomyosis was made when at least two sequences suggested the diagnosis.
Qualitative Image Analysis
Qualitative analysis was performed by three radiologists who were unaware
of the patients' clinical histories and pathologic results. Initially, sets of
images acquired with each of the three sequences were randomly presented
independently to each of the three radiologists, who assigned image quality
scores of 0, 1, 2, or 3, for poor, moderate, good, and excellent image quality
and anatomic visualization (delineation of the contours and depiction of the
zonal anatomy of the uterus). The presence of image artifacts (respiratory and
peristalsis artifacts, artifacts due to pulsation of large vessels, and
chemical shift artifacts) was scored 0, 1, 2, or 3, representing severe,
moderate, mild, or absent artifacts. For each patient and each sequence, the
image quality, anatomic visualization, and image artifacts scores were
calculated.
Histopathologic Findings
Histopathologic examination was performed by a single pathologist who was
unaware of MR imaging data. Gross and microscopic histopathologic examinations
were performed according to the method of Siegler and Camillien
[8]. Specimens were oriented by
a fixed mark on the anterior uterine wall. Uterus weight, macroscopic
appearance, and associated pathologic abnormalities were recorded. Fundal,
anterior, posterior, right, and left maximal uterine wall thickness were
measured.
Macroscopically, adenomyosis was diagnosed in the presence of an enlarged uterus, a globular and asymmetric uterus, and a dense anarchically fasciculated unlimited myometrium with small cavities (0.5-10 mm). Focal adenomyosis was defined as the presence of adenomyoma (circumscribed nodular lesion) mimicking intramural myoma, or when lesions were restricted to one uterine wall (localized adenomyosis). In other cases, adenomyosis was defined as diffuse lesions.
Block sections were taken from the fundal, anterior, posterior, right, and left uterine walls, and from macroscopically abnormal areas. The number of slides ranged from five to 15, depending on myometrial thickness.
Histopathologic criteria for the diagnosis of adenomyosis included the presence of ectopic endometrial tissue in the myometrium located 2.5 mm beyond the endometrial-myometrial junction. The presence of smooth-muscle cells surrounding ectopic endometrial areas was noted. Adenomyosis was graded according to the depth of myometrial involvement, grades 1, 2, and 3 corresponding respectively to adenomyotic involvement of the inner third (superficial adenomyosis), two thirds, or entire (deep adenomyosis) myometrial thickness. Adenomyosis was also graded as mild, moderate, or severe according to the number of endometrial islets observed (one to three, four to nine, or 10 or more foci, respectively).
Statistical Analysis
Sensitivity, specificity, and positive and negative predictive values were
calculated for each pulse sequence for each observer. Statistical analysis was
performed using the chi-square, Kruskall-Wallis, and Mann-Whitney tests, as
appropriate. Values for p of less than 0.05 were considered to denote
significance.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Fifteen (62.5%) of the 24 patients had diffuse adenomyosis, including two patients with associated adenomyoma. Nine (37.5%) of the 24 patients were diagnosed with focal adenomyosis, including two women with an adenomyoma and seven with localized adenomyosis.
The adenomyosis was grade 1 in six patients, grade 2 in nine, and grade 3 in nine, respectively. The degree of adenomyosis was minimal in nine patients, moderate in seven, and severe in eight.
Adenomyotic uteri were associated with other pelvic disorders in 20 patients (83.3%). Of these patients, 17 (85%) had uterine myomas.
MR Imaging Diagnosis of Adenomyosis
The sensitivity, specificity, and positive and negative predictive values
of turbo spin-echo T2-weighted imaging, turbo inversion recovery, and true
FISP for the diagnosis of adenomyosis are given in
Table 1 for each observer.
|
On turbo spin-echo T2-weighted imaging, the numbers of false-negative findings were six, 12, and 12, respectively, for reviewers 1, 2, and 3. The numbers of false-positive findings were three, six, and two, respectively, for reviewers 1, 2, and 3. The accuracy rates for the diagnosis of adenomyosis differed significantly among the three reviewers (p < 0.0001). The accuracy was higher for reviewer 1 than for reviewers 2 and 3 (p < 0.01) and higher for reviewer 3 than for reviewer 2 (p < 0.01).
On true FISP sequences, the number of false-negative findings was seven for each reviewer.
The numbers of false-positive findings were two, 10, and five, respectively, for reviewers 1, 2, and 3. For the diagnosis of adenomyosis, the accuracy rate differed significantly among the three reviewers (p < 0.0001). The accuracy rate was higher for reviewer 1 than for reviewers 2 and 3 (p < 0.01) and higher for reviewer 3 than for reviewer 2 (p < 0.01).
On the turbo inversion recovery sequence, the numbers of false-negative findings were seven, 12, and 11, respectively, for reviewers 1, 2, and 3. The respective numbers of false-positive findings were three, five, and three. Accuracy for the diagnosis of adenomyosis differed significantly among the three reviewers (p < 0.0001). Accuracy was significantly higher for reviewer 1 than for reviewers 2 and 3 (p < 0.001); no difference was found between reviewers 2 and 3.
No intraobserver variability was observed according to the type of MR imaging sequence.
Accuracy for the Diagnosis of Adenomyosis with Combinations of the
Three MR Imaging Sequences
The sensitivity, specificity, and positive and negative predictive values
of turbo spin-echo T2-weighted combined with turbo inversion recovery
sequences, of turbo spin-echo T2-weighted combined with true FISP images, and
of turbo spin-echo T2-weighted combined with turbo inversion recovery and true
FISP sequences, are given in Tables
2 and
3 for each reviewer.
|
|
Compared with turbo spin-echo T2-weighted imaging alone, the combination of turbo spin-echo T2-weighted and turbo inversion recovery imaging, and the combination of turbo spin-echo T2-weighted and true FISP imaging, increased the accuracy of reviewer 3 (Table 2).
Compared with turbo spin-echo T2-weighted imaging alone, the combination of the three MR imaging sequences increased the accuracy of reviewer 2 but not that of reviewers 1 and 3.
Quality Scores of Three MR Imaging Sequences
The image quality, anatomic visualization, and image artifacts obtained
with turbo spin-echo T2-weighted, turbo inversion recovery, and true FISP
sequences are given in Table 4
for each reviewer.
|
With turbo spin-echo T2-weighted images, image quality (p = 0.037) and anatomic visualization (p = 0.01) scores differed among the three reviewers. No difference was found for image artifacts among the reviewers (Figs. 1A and 2A).
|
|
With turbo inversion recovery sequences, image quality differed among the three reviewers (p = 0.0001). Anatomic visualization also differed among the three reviewers (p = 0.0002). No difference in image artifacts was noted among the three reviewers (Figs. 1B and 2B).
|
|
With true FISP sequences, image quality (p = 0.0001), anatomic visualization (p = 0.0001), and image artifacts (p = 0.0001) differed among the three reviewers (Figs. 1C and 2C).
|
|
With true FISP sequences, scores for image artifacts and image quality were higher for the reviewers with low experience in gynecologic imaging than for turbo inversion recovery or turbo spin-echo sequences. High-quality scores were obtained for image quality and anatomic visualization with both turbo spin-echo T2-weighted and true FISP sequences. Image artifacts were higher with true FISP than with turbo spin-echo T2-weighted imaging. The combination of turbo spin-echo T2-weighted and true FISP sequences gave the highest image quality scores.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
In our study, the sensitivity and specificity of turbo spin-echo T2-weighted sequences for the diagnosis of adenomyosis was 50-75% and 81.2-90.9%, respectively. Our results are partly in agreement with those of previous reports of the sensitivity and specificity of T2-weighted or turbo spin-echo T2-weighted sequences (77.5-89% and 67-92.5%, respectively) [4, 5, 6]. In our experience, the sensitivity of turbo spin-echo T2-weighted sequences for the diagnosis of adenomyosis is relatively low. This wide range of values could be explained by interobserver variability. Our results show that the main factor influencing accuracy was the radiologist's experience in MR imaging of the female pelvis and the MR imaging sequence used. Indeed, the accuracy of reviewer 2 (highly experienced, but not in gynecologic imaging) was lower with turbo spin-echo T2-weighted or true FISP sequences than with turbo inversion recovery sequences.
Radiologists such as reviewer 2 who specialize in upper abdominal MR imaging, prefer rapid MR imaging sequences. As noted in other publications [13, 14], our data suggest that reviewer experience is an important component in achieving high accuracy in the MR imaging diagnosis of adenomyosis. Multicoil high-resolution turbo spin-echo MR imaging is considered the gold standard for investigating the female pelvis [15, 16, 17]. Turbo spin-echo T2-weighted imaging suffers from the long acquisition time and image degradation caused by motion artifacts. Turbo inversion recovery and true FISP sequences eliminate motion artifacts, improve patient acceptance, and are more rapid [12]. In our study, intraobserver accuracy for the diagnosis of adenomyosis was similar whether turbo spin-echo T2-weighted, turbo inversion recovery, or true FISP sequences were used. Recently, the ultrafast half-Fourier single-shot turbo spin-echo sequence has been suggested for female pelvic imaging [18, 19]. Further studies are necessary to evaluate turbo inversion recovery and true FISP sequences compared with the ultrafast half-Fourier single-shot turbo spin-echo sequence.
With turbo spin-echo T2-weighted imaging, we found a difference among the three reviewers in image quality and anatomic visualization but not in the intensity of artifacts. Although turbo spin-echo T2-weighted imaging is associated with high image quality, the incidence of artifacts is high. Our data are in keeping with previous reports that the main limitations of turbo spin-echo T2-weighted imaging in both upper abdominal and pelvic imaging are pulsation artifacts and the magnetic heterogeneity caused by intestinal gas and respiratory motion [13, 15, 16, 20, 21, 22]. Various methods have been recommended to avoid these artifacts, such as abdominal compression and the IV administration of glucagon. Respiratory triggering of the MR acquisition (not available in our unit at the time of this study) requires a longer examination time [23]. Another way of reducing artifacts is to use fast breath-hold MR imaging sequences. In our study, turbo inversion recovery received poor image quality scores for radiologists experienced in gynecologic imaging. In contrast, despite interobserver variability in subjective image quality and artifact intensity, the imaging quality of true FISP appeared to improve the diagnosis of adenomyosis.
We also evaluated the diagnostic value of combining various MR imaging sequences. The combination of turbo spin-echo T2-weighted and turbo inversion recovery sequences, compared with turbo spin-echo T2-weighted sequences alone, increased the accuracy rate of only the reviewer with extensive gynecologic imaging experience. Similar results were obtained with the combination of turbo spin-echo T2-weighted and true FISP sequences. However, with the reduction in motion artifacts on true FISP sequences compared with turbo spin-echo T2-weighted sequences, our results suggest that the best option for the diagnosis of adenomyosis is a combination of turbo spin-echo T2-weighted and true FISP sequences. Further studies are necessary to validate these preliminary results.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  G. Dubernard, M. Piketty, R. Rouzier, S. Houry, M. Bazot, and E. Darai Quality of life after laparoscopic colorectal resection for endometriosis Hum. Reprod., May 1, 2006; 21(5): 1243 - 1247. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Bazot, E. Darai, R. Hourani, I. Thomassin, A. Cortez, S. Uzan, and J.-N. Buy Deep Pelvic Endometriosis: MR Imaging for Diagnosis and Prediction of Extension of Disease Radiology, August 1, 2004; 232(2): 379 - 389. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
