| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Department of Radiology, Seirei Hamamatsu General Hospital, 2-12-12 Sumiyoshi,
Hamamatsu, Shizuoka, 430-8558 Japan.
2
Department of Radiology, Hamamatsu University School of Medicine, 3600
Handa-cho, Hamamatsu, Shizuoka, Japan.
3
Application Research Group, General Electric Yokogawa Medical Systems, 4-7-127
Asahigaoka, Hino, Tokyo, Japan.
Received July 25, 2000;
accepted after revision December 14, 2000.
Address correspondence to M. Katayama.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Before surgery, 52 women with gynecologic disorders underwent half-Fourier acquisition single-shot fast spin-echo imaging with multiphase and multislice acquisitions. Images were displayed in a cine mode, and the motion of each organ against adjacent organs was evaluated by two radiologists who were unaware of the patients' histories or of the findings of their clinical examinations or surgeries. Findings from MR imaging were classified into three types relative to the adjacent organs: type 1, sliding, defined as organs moving 1 cm or more; type 2, fine motion, defined as organs moving less than 1 cm; or type 3, no motion. Type 2 was further subdivided into two groups: type 2-A, independent fine movement, and type 2-B, synchronous fine movement. All MR imaging findings were verified by laparotomy or laparoscopy. Peristalsis at the rectum, colon, and small intestine were also evaluated.
RESULTS. A total of 317 interfaces were evaluated. For findings of type 1 (n = 8 interfaces) and type 2-A (n = 245) on kinematic MR imaging, the negative predictive values for adhesions were 100% and 95.5%, respectively. Findings of type 2-B (n = 52) and type 3 (n = 12) with no adhesions were observed in 40.4% and 66.7%, respectively. When type 1 and type 2-A were regarded as negative findings of adhesions, and type 2-B and type 3 as positive, sensitivity, specificity, and accuracy were 72.5%, 87.4%, and 85.4%, respectively. Peristalsis was observed in 69.2% of patients at the rectum, 86.5% at the colon, and 100% at the small intestine.
CONCLUSION. Multiphase and multislice MR imaging with kinematic display may provide new information about the presence of pelvic adhesions.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
A half-Fourier acquisition single-shot fast spin-echo sequence enabled us to obtain images with a particular amount of tissue contrast within a second per slice. By acquiring multiphase and multislice images with such a sequence, we obtained kinematic information that we hoped would improve our recognition of pelvic adhesions. At sonography using a kinematic approach, a sliding sign—two adjacent organs moving differently during respiration—was apparently helpful in localizing abdominal adhesions [10,11,12]. The purpose of this study was to assess the feasibility of kinematic MR imaging of the pelvis and to evaluate the detectability of pelvic adhesions using this technique.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
At surgery, the 33 uterine lesions found were diagnosed as 23 uterine fibroids, five uterine adenomyosis, and five malignant uterine tumors. In one patient, the findings of uterine fibroid and adenomyosis overlapped. In addition, 37 adnexal masses—34 of which were benign ovarian tumors and three of which were malignant ovarian tumors—12 cases of endometriosis, and one case of pelvic inflammatory disease were verified at surgery. One patient had a history of right oophorectomy.
Adhesions to the uterus were confirmed in four patients at the rectum, six patients at the colon, and four patients at the small intestine. Adhesions to the uterus and ovary were confirmed in 10 patients at the right ovary and 11 patients at the left ovary. Adhesions to the ovary and intestine were verified in 11 patients at the right ovary and six patients at the left ovary.
MR Imaging
All MR examinations were performed using a 1.5-T magnet (Horizon LX;
General Electric Medical Systems, Milwaukee, WI) with a torso phased array
multicoil scanner. Multiphase and multislice MR imaging was performed using a
half-Fourier acquisition single-shot fast spin-echo sequence in only the
sagittal plane with the following parameters: TR effective range/effective TE,
16-26/94; matrix size, 256 x 160; rectangular field of view, 35 x
21.0-24.5 cm; number of excitations, 0.5; echo train length, 85; echo spacing,
4.8 msec; receiver bandwidth, 62.5 Hz; slice thickness, 5-6 mm; interslice
gap, 1-2 mm. Chemical-selective fat saturation was not applied. The
acquisition time per slice was approximately 1 sec. One set of 15-25 images
(mean, 19.5) per one phase covered the entire pelvis from the right to the
left, and 10-20 phase acquisitions (mean, 14.6) were repeated with free
respiration. The total acquisition time was 4-7 min for 200-380 images (mean,
283.5).
Evaluations
All images were reviewed on a workstation (Advantage; General Electric
Medical Systems) and sorted according to location. At each location, images
were sorted according to time. Then all images were sequentially and
repeatedly displayed on a monitor in the cine mode at a frame rate of 10
frames per sec.
We classified the patterns of movements of the organs—ovary, uterus, small intestine, colon, and rectum—against other organs into three types: type 1, sliding, defined as organs moving 1 cm or more; type 2, fine motion, defined as organs moving less than 1 cm; or type 3, no motion. Type 2 was further subdivided into two groups: type 2-A, independent fine movement, defined as motion independent of that of the adjacent organ, and type 2-B, synchronous fine movement, defined as motion in accordance with that of the adjacent organ.
Based on these classifications, movements of the pelvic organs were interpreted independently by two radiologists who were unaware of the patient's history or clinical and surgical findings. Differences in interpretation by the radiologists were resolved by consensus. All the adherent interfaces were retrospectively correlated with surgical findings. In addition, we determined the presence or absence of peristalsis of the rectum, colon, and small intestine. We defined peristalsis of the intestine on kinematic MR imaging as wavelike contractive motions of the intestinal wall visible on the display.
Exclusion criteria
We applied three exclusion criteria to our data. First, single-shot fast
spin-echo images on which target organs were not identifiable were excluded.
The second exclusion factor was whether images of the interfaces of the target
organs had been obtained in left-to-right—sided dimensions. Because
kinematic imaging was performed in the sagittal plane, evaluations of the
adhesions over slices of different locations were not considered accurate, and
thus, the third excluding factor was whether the images of the target organs
had apparently been obtained from separate locations.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The 13 interfaces eliminated from the study because the images were obtained in left-to-right—sided dimensions included one interface between the right ovary and the colon, four between the right ovary and the uterus, one between the left ovary and the colon, and seven between the left ovary and the uterus. With surgical confirmation, one interface between the left ovary and the uterus proved to be adhesive.
The 12 interfaces eliminated because of apparent separation included four interfaces between the small intestine and the uterus, two between the right ovary and the colon, one between the right ovary and the uterus, and five between the left ovary and the colon. There were no adhesions found in this category.
MR Findings
Of the total 364 interfaces, 317 (87.1%) in 52 patients were evaluated.
Table 1 summarizes the
relationship between the findings of kinematic MR imaging and surgical
results. Type 1 findings of sliding were present in eight interfaces (2.5%).
Type 2-A findings of independent fine movement were present in 245 interfaces
(77.3%), and type 2-B findings of synchronous fine movement were present in 52
interfaces (16.4%). Type 3 findings of no motion were present in 12 interfaces
(3.8%).
|
Eight interfaces in five patients showed the findings of type 1, sliding, (Fig. 1A,1B) without adhesions. Type 1 movement was found in one interface between the rectum and uterus, two interfaces between the colon and uterus, three interfaces between the small intestine and uterus, one interface between the right ovary and intestine, and one interface between the left ovary and intestine. In these cases, the movements of the intestine were prominent. In the interfaces of the ovaries to the uterus, no sliding signs were detected.
|
|
Of the findings showing type 2-A, independent fine movements, 234 of 245 interfaces revealed no adhesions. At surgery, 11 interfaces in 10 patients that had been interpreted on MR imaging as showing no adhesions were found to have been false-negatives. The false-negatives were attributable to difficulty in locating exactly the adherent interfaces because of large coexisting tumors (n = 4), and the poor soft-tissue contrast between the target organs (n = 7).
The most frequent interfaces were of type 2-B findings, with the synchronous fine movement occurring between the ovaries and the uterus (n = 27). In 31 of 52 interfaces, the findings of synchronous fine movement revealed adhesions that were confirmed at surgery.
No interface showed type 3 findings of no movement between the small intestine and the uterus. The most frequent interfaces showing no motion were between the rectum and uterus (n = 5). In four of 12 interfaces, the findings of no movement revealed the existence of adhesions.
Findings of sliding and independent fine movement had high negative predictive values for the diagnosis of adhesions, 100% and 95.5%, respectively. Those of synchronous fine movement and no motion showed the existence of adhesions in 40.4% and 66.7% of the interfaces. When findings of sliding and independent fine movement and those of synchronous fine movement and no motion were regarded as negative and positive findings of adhesions, respectively, negative predictive value was 95.7%, and positive predictive value was 54.7%; sensitivity, specificity, and accuracy were 72.5%, 87.4%, and 85.4%, respectively.
Peristalsis
The wavelike movements of the intestine were recognized on the display.
Peristalsis was observed in 36 patients (69.2%) at the rectum, 45 patients
(86.5%) at the colon, and 52 patients (100%) at the small intestine.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
From multiphase and multislice MR imaging, we obtained about 15 sets of 20 sagittal images in approximately 5 min. Multislice imaging enabled us to evaluate anatomic relations and multiorgan adhesions.
Other studies have reported that viscera slide was 2-3 cm of longitudinal movement at the abdominal wall during normal respiration [11, 12]. Respiratory movement was postulated to be the main source of the motions of intrapelvic organs, and it has been previously suggested that the sliding sign on sonography is helpful in localizing an abdominal mass [10,11,12]. Spontaneous viscera slide is evident with longitudinal scanning of the abdomen. A common sonographic observation suggests that the intraperitoneal and retroperitoneal organs, such as the liver and right kidney, move differently during respiration. Respiratory movement may be one reason to consider using kinematic evaluations when imaging the pelvis, although the motions in this region might be less affected by respiration than those in the upper abdomen.
For the present study, we classified motion patterns of the target organs in the pelvis into four categories: sliding (type 1), independent fine movement (type 2-A), synchronous fine movement (type 2-B), and no motion (type 3). Some authors have reported that at sonography, during normal and exaggerated respiration, the extent of spontaneous viscera slide was determined to be 1 cm or more [11, 12]. In our study, the sliding sign was present in only 2.5% of the interfaces, although none of these interfaces showed adhesions. Fine movements of the pelvic organs were more frequently observed. Thus, because the motions of the organs in the pelvis are less frequent than in those in the upper abdomen, we subcategorized the fine movements into two patterns, independent and synchronous. In 234 of 245 interfaces with the finding of independent fine movements, no adhesions were observed even though motion between organs was less than 1 cm. Of the 52 interfaces with the finding of synchronous fine movement, 31 interfaces had adhesions at surgery. Thus, this finding could not be the definite criterion determining whether the adhesions were present or not. In eight of 12 interfaces with the finding of no motion, adhesions were not revealed at laparotomy or laparoscopy; however, this finding could be also observed in the patients with a so-called frozen pelvis. These two categories might be nonspecific findings of adhesions on this kinematic MR imaging.
Ten ovaries (9.7%) could not be detected on half-Fourier acquisition single-shot fast spinecho images. In two of these 10 ovaries, severe adhesions were confirmed at surgery. It has been reported that with conventional MR imaging, normal ovaries can be identified in 85% of women of reproductive age and at a lower rate in postmenopausal women [22].
In terms of kinematic information of the pelvis viscera, our study found the peristaltic motions. Physiologically, slow-wave activity that produces ring contractions in the small intestine occurs at the rate of about seven cycles per minute [23]. Although the colon also exhibits ring contractions, they differ from those in the small intestine. Peristalsis of the rectum is usually a faint motion, and major motion at the rectum is only seen on defecation [23, 24]. Our study supports this theory in that peristalsis could be observed in 69.2% of patients at the rectum, 86.5% of patients at the colon, and 100% of patients at the small intestine on kinematic MR imaging.
The primary advantage of sonography is its allowance of real-time observation, but the evaluation in the pelvis is dependent on the patient's body habitus. Assessing adhesions of the intestine is sometimes difficult because ultrasonic waves are interrupted by intestinal gas. Further-more, scanning a large field, such as the whole pelvis, is difficult with sonography. When information obtained from kinematic MR imaging is similar to that of sonography, MR imaging may provide more useful and reproducible findings. MR imaging can cover a larger field of view than sonography, and the quality of MR imaging is less dependent on the operator's skills or patient's body habitus. Thus, for the patients with obesity or gas-filled intestines, kinematic MR imaging would be more useful to diagnose pelvic adhesions than sonography, and we speculate that kinematic MR imaging would be more available than sonography for the patients who are suspected of having extensive disease in the pelvis, such as pelvic inflammatory disease and endometriosis.
In our study, we evaluated adhesions only in the sagittal plane and could not evaluate images of 13 interfaces obtained in left-to-right-sided dimensions. In the future, studies of further MR imaging in additional planes (transverse, coronal, and oblique) could provide more information. In this study, we also performed the kinematic MR imaging with free respiration. In kinematic imaging with deep respiration, the sliding phenomenon would be enhanced, and the sensitivity and specificity of the technique might be increased.
Unfortunately we did not measure the actual time and cost for the data acquisition. If the performance of the MR imaging units and computer processors improves in the near future, the filmless diagnosis, or diagnosis from the display, will become widely used. Then the value of kinematic MR imaging will be enhanced.
Kinematic information obtained in this study can be distributed only in a cine display. In most patients with findings of fine movements, static images did not provide useful information.
In conclusion, kinematic MR imaging provided information of respiratory and peristaltic motions in the pelvis and clues for evaluations of adhesions. Although our results show that the sliding sign provided by this imaging has a high negative predictive value for adhesions, this finding was rarely encountered in this study and is uncommon in the pelvis because of the minimal respiratory motion in the pelvis. If independent fine movements are commonly found in the pelvis, this finding may correlate with the absence of pelvic adhesions.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  C. Del Frate, R. Girometti, M. Pittino, G. Del Frate, M. Bazzocchi, and C. Zuiani Deep Retroperitoneal Pelvic Endometriosis: MR Imaging Appearance with Laparoscopic Correlation RadioGraphics, November 1, 2006; 26(6): 1705 - 1718. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. L. Kataoka, K. Togashi, T. Yamaoka, T. Koyama, H. Ueda, H. Kobayashi, M. Rahman, T. Higuchi, and S. Fujii Posterior Cul-de-Sac Obliteration Associated with Endometriosis: MR Imaging Evaluation Radiology, March 1, 2005; 234(3): 815 - 823. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Fujiwara, K. Togashi, T. Yamaoka, A. Nakai, A. Kido, S. Nishio, T. Yamamoto, H. Kitagaki, and S. Fujii Kinematics of the Uterus: Cine Mode MR Imaging RadioGraphics, January 1, 2004; 24(1): e19 - e19. [Abstract] [Full Text]
|
|
|
|
|
|
I. Imaoka, A. Wada, M. Matsuo, M. Yoshida, H. Kitagaki, and K. Sugimura MR Imaging of Disorders Associated with Female Infertility: Use in Diagnosis, Treatment, and Management RadioGraphics, November 1, 2003; 23(6): 1401 - 1421. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
