HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) All Versions of this Article: 2271011587v1 227/2/345    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Masui, T. Articles by Sakahara, H. Search for Related Content PubMed PubMed Citation Articles by Masui, T. Articles by Sakahara, H.

Pseudolesions Related to Uterine Contraction: Characterization with Multiphase-Multisection T2-weighted MR Imaging¹

¹ From the Departments of Radiology (T.M., M.K., S.K.) and Pathology (S.S.), Seirei Hamamatsu General Hospital, 2-12-12 Sumiyoshi, Hamamatsu, Shizuoka 430-8558, Japan; Application Research Group, GE Yokogawa Medical Systems, Hino, Japan (A.N.); and Department of Radiology, Hamamatsu University School of Medicine, Japan (H.S.). Received September 24, 2001; revision requested December 10; final revision received September 9, 2002; accepted October 14. Supported in part by a grant from the Japanese Society for Magnetic Resonance in Medicine. Address correspondence to T.M. (e-mail: masui@sis.seirei.or.jp).

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
PURPOSE: To evaluate whether multiphase-multisection T2-weighted magnetic resonance (MR) images help exclude pseudolesions mimicking leiomyoma and adenomyosis on static T2-weighted fast spin-echo (FSE) MR images and to characterize temporal changes in uterine signal intensity related to uterine contraction.

MATERIALS AND METHODS: T2-weighted FSE and multiphase-multisection single-shot FSE (SSFSE) MR imaging were performed in 43 patients who underwent hysterectomy. Each imaging set was evaluated separately by two independent readers, and receiver operating characteristic analysis was performed. In the 43 patients and in 49 other patients suspected of having pelvic abnormality, a combination of signal intensity changes on FSE and SSFSE MR images was classified into five patterns, and temporal low-signal-intensity changes on SSFSE MR images were characterized.

RESULTS: For detection of leiomyoma on FSE and SSFSE MR images, the respective values of the area under the receiver operating characteristic curve were 0.98 and 0.97 for reader 1 and 0.96 and 0.96 for reader 2; for detection of adenomyosis on FSE and SSFSE MR images, the respective values were 0.82 and 0.84 for reader 1 and 0.80 and 0.89 for reader 2 (P > .05). SSFSE MR images helped exclude pseudolesions in 1%–3% cases of leiomyoma and in 3%–4% cases of adenomyosis. Temporal signal intensity changes were observed in 53% of 368 segments. The most frequent shape of temporal low signal intensity was diffuse followed by ill-defined focal type. Characteristic shape of temporal low signal intensities was band- or sticklike, which was observed in as many as 19% of 368 segments.

CONCLUSION: Multiphase-multisection T2-weighted SSFSE MR images do not improve accuracy in detection of leiomyoma and adenomyosis compared with FSE MR images; however, they helped characterize features of temporal low signal intensities in the uterus, which are related to uterine contractions.

© RSNA, 2003

Index terms: Uterus, abnormalities, 854.315, 854.317 • Uterus, MR, 854.121411, 854.121416

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
In magnetic resonance (MR) imaging of the uterus, T2-weighted images are the most useful standard images for detection and characterization of lesions such as leiomyoma and adenomyosis (1–5). However, on T2-weighted images, pseudolesions, which are related to uterine contraction, have been recognized as low-signal-intensity areas in the uterus that mimic leiomyoma or adenomyosis (6). In several sonographic studies, myometrial contractions have been demonstrated (7,8). Similarly, with T2-weighted MR imaging, changes in both signal intensity and thickness of the myometrium and junctional zone, which are related to uterine contraction, have been demonstrated in a kinematic fashion (9). Characterization of the shape of contraction-induced low signal intensities in the uterus and understanding their prevalence might be helpful to increase diagnostic confidence for leiomyoma or adenomyosis on T2-weighted MR images.

Accordingly, the purpose of this study was to evaluate whether multiphase-multisection T2-weighted MR images help exclude pseudolesions that mimic leiomyoma and adenomyosis on static T2-weighted fast spin-echo (FSE) MR images and to characterize temporal signal intensity changes in the uterus related to uterine contraction.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
This study was performed according to the guidelines of our institutional review board, and informed consent was obtained from each patient.

Patients
From August 1999 to January 2000, 249 consecutive female patients underwent pelvic MR imaging for suspected pelvic disease. Of these patients, 92 underwent hysterectomy, a laparoscopic procedure, or myomectomy with endovaginal ultrasonography (US) combined with transabdominal US and/or computed tomography. The remaining 157 patients did not undergo any surgical procedure and were excluded from the study. This study consisted of two parts. In the first part, 43 patients (age range, 35–81 years; mean age, 50.2 years), who underwent hysterectomy after an MR study, were included for the evaluation of lesion detection in the uterus. The presence or absence of leiomyoma and adenomyosis was confirmed histopathologically. The final diagnoses of diseases were based on histopathologic findings following hysterectomy, combined with information of other imaging modalities and clinical data. Final diagnoses were leiomyoma (n = 31), adenomyosis (n = 18), endometrial cancer (n = 9), endometrial polyp (n = 3), cervical cancer (n = 2), ovarian cancer (n = 2), and benign ovarian cyst or tumor (n = 9). Among them, one patient had four conditions, seven had three, 14 had two, and the other 21 patients had one.

In the second part of the study, 92 patients (age range, 15–81 years; mean age, 43.6 years) who underwent the pelvic MR study were included for the characterization of signal intensity changes in the uterus. The final diagnoses of diseases were based on histopathologic findings following hysterectomy (n = 43); findings of endovaginal US, laparoscopy, clinical course, and/or myomectomy (n = 10); or findings of oophorectomy (n = 39). Final diagnoses were leiomyoma (n = 54), adenomyosis (n = 27), endometrial cancer (n = 12), endometrial polyp (n = 3), cervical cancer (n = 2), ovarian cancer (n = 3), benign ovarian cyst or tumor (n = 37), or obstruction of vagina (n = 1). Among them, one patient had four conditions, 10 had three each, 24 had two each, and the other 57 patients had one each.

MR Imaging
MR imaging was performed by using a 1.5-T magnet (Horizon LX Echo Speed; GE Medical Systems, Milwaukee, Wis) with a torso phased-array multicoil (GE Medical Systems). MR imaging was performed in the same way as that described elsewhere (9). After localization, an antiperistaltic drug (7.5 mg of timepidium bromide, Sesden; Tanabe Seiyaku, Osaka, Japan) was administered intravenously to patients who had no contraindications to receiving this drug. Sagittal T2-weighted FSE imaging was performed without breath holding by using the following imaging parameters: repetition time msec/effective echo time msec of 4,000/85, echo train length of 16, section thickness of 5 mm, with a 1-mm intersection gap, receiver bandwidth of 32 kHz, matrix of 256 x 256, with 512 zero filling interpolation, field of view of 26 x 20–30 x 23 cm, and two signals acquired. A chemical shift fat-saturation pulse was applied. Imaging time was 2 minutes 8 seconds for 20 sections.

Then, 10–15 minutes after the start of T2-weighted FSE MR imaging, sagittal multiphase-multisection T2-weighted single-shot FSE (SSFSE) MR imaging with half-Fourier acquisition was performed without breath holding to cover the whole pelvis. Imaging parameters were as follows: 17,000–28,000/94, echo train length of 85, section thickness of 5–6 mm, intersection gap of 1–2 mm, receiver bandwidth of 62.5 kHz, matrix of 256 x 160, field of view of 35 x 21–24.5 cm, and one-half signal acquired. Imaging time for one phase was 17–28 seconds for 20–26 sections, which were obtained from right to left in an interleaved fashion. Fifteen sequential phases of the images were acquired during 5–7 minutes.

Display
SSFSE and FSE images were downloaded onto a workstation (Advantage Windows 3.1; GE Medical Systems). SSFSE MR images were sorted according to location, and at each location, images were sorted according to time. SSFSE MR images were manually paged by a reader using a computer mouse. On FSE and SSFSE MR images, the window level on the display monitor was set at the same level as the signal intensity of the apparently normal myometrium, and the window width was set at twice the window level.

For the evaluation of the uterus, a line was drawn between two points, the center of the internal cervical os and the fundal portion of the uterine cavity. Then, another line was drawn perpendicular to the initial line, bisecting it (Fig 1). The following segments were defined: upper anterior, upper posterior, lower anterior, and lower posterior.

View larger version (32K): [in this window] [in a new window] [Download PPT slide]   Figure 1. Diagram of the uterus in the sagittal plane for the evaluation of the myometrium and junctional zone at four locations. A line was drawn between two points, the center of the internal cervical os (A) and the fundal portion of the uterine cavity (B). Then, a line (A-B) was drawn perpendicular to the initial line, bisecting it. Four segments were defined: upper anterior (UA), upper posterior (UP), lower anterior (LA), and lower posterior (LP).

All qualitative rating evaluations were made separately and independently by two radiologists (T.M.,M.K.) (with 13 and 12 years of experience in abdominal radiology, respectively), who were blinded to clinical symptoms, clinical data, diagnosis, or data of other observers’ evaluations.

First Part of the Study
Lesion detection in the uterus.—FSE and SSFSE MR images were displayed separately. Initially, evaluations of the lesions were performed on FSE MR images; then after an interval of more than 2 weeks, the same evaluations were made on SSFSE MR images to minimize the effects of the previous evaluation of the FSE images.

Existence or absence of organic lesions such as leiomyoma and adenomyosis was evaluated. Leiomyomas were defined as well-delineated low-signal-intensity lesions with or without high-signal-intensity areas in them. Adenomyoses were described as ill-defined low-signal-intensity lesions with or without high-signal-intensity spots or having focal or diffuse thickening of the junctional zone (>12 mm) (5). On SSFSE MR images, only consistently low-signal-intensity areas were regarded as lesions such as leiomyoma and adenomyosis. Diseases of the endometrium and cervix, such as endometrial thickening, cancer, or cervical cancer, were not evaluated.

When one large lesion was located over two segments, the number of segments that contained the lesions was counted as two.

Histopathologic evaluation.—Hysterectomy specimens were open anteriorly in the sagittal plane and were cut transversely. The pathologists initially reviewed all pathologic specimens with the knowledge of the clinical findings, including those of MR imaging and US. Then, the pathologist further reviewed the specimens to investigate the histopathologic correlation of the MR findings, with information that included the location and extent of the regions of interest. Eighty-five segments in 31 patients contained leiomyoma and 39 segments in 18 patients contained adenomyosis, which were confirmed at histopathologic examination.

Second Part of the Study
Combination pattern of low-signal-intensity areas on FSE and SSFSE MR images.—FSE and SSFSE MR images were displayed side by side at the corresponding location on a monitor. At each location, SSFSE MR images were sorted according to time. Low-signal-intensity areas in the myometrium on FSE and SSFSE MR images were classified into the following five patterns that indicated a combination of presence or absence of a low-signal-intensity area on FSE and SSFSE MR images: pattern A, low-signal-intensity areas in the myometrium on FSE MR images and on all SSFSE MR images; pattern B, low-signal-intensity areas on FSE MR images and on SSFSE MR images obtained at some phases of imaging; pattern C, absent low-signal-intensity areas on FSE MR images but low-signal-intensity areas on SSFSE MR images obtained at some phases of imaging; pattern D, low-signal-intensity areas on FSE MR images but absent low-signal-intensity areas on SSFSE MR images; and pattern E, absent low-signal-intensity areas on FSE MR images but low-signal-intensity areas on all SSFSE MR images.

Shape of the functionally changed signal intensity areas on SSFSE MR images.—The shapes of the low-signal-intensity areas in the myometrium, which were not consistently observed at all phases of SSFSE imaging, were classified into the following four patterns: (a) well-defined focal low-signal-intensity area, which was temporal oval or round focal with sharply delineated margins; (b) ill-defined focal low-signal-intensity area, which was oval or round focal with no definite margin; (c) diffuse low-signal-intensity area without definite margin; and (d) band- or sticklike low-signal-intensity area in any direction.

Thickening and signal intensity changes of the junctional zone on SSFSE MR images.—Changes in thickness of the junctional zone were classified into two patterns: (a) focal or nodular thickening, which showed localized thickening in the junctional zone of each segment of the uterus, and (b) diffuse thickening, which showed approximately one-third or more of the junctional zone in each segment of the uterus. Signal intensity changes in the junctional zone were also evaluated. The evaluated features were similar to those in the previously published article (9).

Statistical Analysis
In the first part of the study, binormal receiver operating characteristic curves, which deal with the fully paired statistics, for each reader and for each MR image in the evaluation of confidence ratings for presence or absence of leiomyoma and adenomyosis were calculated by using a maximum-likelihood estimation with an analysis program (ROCKIT 0.9B; C. E. Metz, University of Chicago, Ill). The area under the receiver operating characteristic curve (A_z) was calculated and used to indicate the overall performance of readers and of FSE and SSFSE MR images (10). The difference between each A_z value was statistically evaluated (11). A two-tailed P value less than .05 was considered to indicate a significant difference. The sensitivities and specificities of readers and of MR images were assessed for lesions that were assigned a confidence rating of 3, 4, or 5 for the presence of and a rating of 1 and 2 for the absence of leiomyoma and adenomyosis.

In both parts of the study, a nonweighted coefficient was calculated to measure the agreement between the two readers in the qualitative rank evaluation, where a rating of 3, 4, and 5 were assigned for presence of and a rating of 1 and 2 for absence of leiomyoma and adenomyosis (12). The level of agreement was defined as follows: = 0–0.20, poor; 0.21–0.40, fair; 0.41–0.60, moderate; 0.61–0.80, good; and 0.81–0.99, excellent (13). All statistical analyses were made by using statistical software programs (Statview version 5, SAS Institute, Cary, NC; or Excel 2001, Microsoft, Redmond, Wash).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
First Part of the Study
Lesion detection in the uterus.—In 43 patients who underwent hysterectomy, there was no significant difference in the performance between the two readers and each MR imaging method (P > .05, Table 1). Excessive degeneration of the leiomyoma did not affect its detection according to the defined criteria of this disease. In two (1%) of 172 segments, low-signal-intensity areas depicted as leiomyoma on FSE MR images were determined to be pseudolesions on SSFSE MR images by reader 1 and in six (3%) segments by reader 2 (Fig 2). For the diagnosis of adenomyosis, sensitivity was lower than that for leiomyoma on both FSE and SSFSE MR images (Table 2).

View larger version (187K): [in this window] [in a new window] [Download PPT slide]   Figure 2a. Sagittal MR images in a 41-year-old woman with large intramural leiomyoma. (a) T2-weighted FSE MR image (4,000/85) demonstrates a large low-signal-intensity mass (black arrow) with high intensity foci in the anterior wall. Diffuse low-signal-intensity areas (curved arrow) are identified in the posterior segments compressed by the lower lumbar spine and sacrum. A signal void (straight white arrow) posterior to the upper uterus indicates the left common iliac artery. (b, c) T2-weighted SSFSE MR images (20,000/94, echo train length of 85) in different phases of imaging demonstrate a large uterine mass (black arrow), diffuse low-signal-intensity areas (solid curved arrow) in the posterior segments, and signal void indicating the artery (straight white arrow), which are identical to those in a. An ill-defined focal low-signal-intensity area (open curved arrow in c) is indicated in the upper anterior segment near the fundus, which is not observed in b.

View larger version (183K): [in this window] [in a new window] [Download PPT slide]   Figure 2b. Sagittal MR images in a 41-year-old woman with large intramural leiomyoma. (a) T2-weighted FSE MR image (4,000/85) demonstrates a large low-signal-intensity mass (black arrow) with high intensity foci in the anterior wall. Diffuse low-signal-intensity areas (curved arrow) are identified in the posterior segments compressed by the lower lumbar spine and sacrum. A signal void (straight white arrow) posterior to the upper uterus indicates the left common iliac artery. (b, c) T2-weighted SSFSE MR images (20,000/94, echo train length of 85) in different phases of imaging demonstrate a large uterine mass (black arrow), diffuse low-signal-intensity areas (solid curved arrow) in the posterior segments, and signal void indicating the artery (straight white arrow), which are identical to those in a. An ill-defined focal low-signal-intensity area (open curved arrow in c) is indicated in the upper anterior segment near the fundus, which is not observed in b.

View larger version (182K): [in this window] [in a new window] [Download PPT slide]   Figure 2c. Sagittal MR images in a 41-year-old woman with large intramural leiomyoma. (a) T2-weighted FSE MR image (4,000/85) demonstrates a large low-signal-intensity mass (black arrow) with high intensity foci in the anterior wall. Diffuse low-signal-intensity areas (curved arrow) are identified in the posterior segments compressed by the lower lumbar spine and sacrum. A signal void (straight white arrow) posterior to the upper uterus indicates the left common iliac artery. (b, c) T2-weighted SSFSE MR images (20,000/94, echo train length of 85) in different phases of imaging demonstrate a large uterine mass (black arrow), diffuse low-signal-intensity areas (solid curved arrow) in the posterior segments, and signal void indicating the artery (straight white arrow), which are identical to those in a. An ill-defined focal low-signal-intensity area (open curved arrow in c) is indicated in the upper anterior segment near the fundus, which is not observed in b.

View larger version (177K): [in this window] [in a new window] [Download PPT slide]   Figure 3a. Sagittal MR images in a 36-year-old woman with endometriosis. (a) T2-weighted FSE MR image (4,000/85) demonstrates well-defined focal low-signal-intensity area (straight arrow) in the anterior lower segment and irregular low-signal-intensity areas (solid curved arrow), which are continuous to the junctional zone. In the posterior segment, thick bandlike low-signal-intensity areas (open curved arrow) are noted. Fluid in the cul-de-sac is demonstrated. (b, c) T2-weighted SSFSE MR images (20,000/94, echo train length of 85) in different phases of imaging demonstrate effects of uterine contraction. (b) In the anterior segments, ill-defined low-signal-intensity areas (solid curved arrow) are noted in the myometrium. A low-signal-intensity area (open curved arrow) in the posterior segment, which has a different shape from that in a, is noted. An ill-defined focal low-signal-intensity area (straight arrow) appears in the posterior segment. Fluid in the cul-de-sac is also noted. (c) With uterine contraction, a slightly different shape of the uterus is noted compared with that in a and b. Low-signal-intensity areas (solid curved arrow) in the anterior segment have disappeared, indicating contraction-induced low signal intensity. In the upper posterior segment, low-signal-intensity areas (open curved arrow), which have slightly different orientation from that in a, are demonstrated. The ill-defined lower signal intensity (straight arrow) in the posterior segment is recognized as in b.

View larger version (178K): [in this window] [in a new window] [Download PPT slide]   Figure 3b. Sagittal MR images in a 36-year-old woman with endometriosis. (a) T2-weighted FSE MR image (4,000/85) demonstrates well-defined focal low-signal-intensity area (straight arrow) in the anterior lower segment and irregular low-signal-intensity areas (solid curved arrow), which are continuous to the junctional zone. In the posterior segment, thick bandlike low-signal-intensity areas (open curved arrow) are noted. Fluid in the cul-de-sac is demonstrated. (b, c) T2-weighted SSFSE MR images (20,000/94, echo train length of 85) in different phases of imaging demonstrate effects of uterine contraction. (b) In the anterior segments, ill-defined low-signal-intensity areas (solid curved arrow) are noted in the myometrium. A low-signal-intensity area (open curved arrow) in the posterior segment, which has a different shape from that in a, is noted. An ill-defined focal low-signal-intensity area (straight arrow) appears in the posterior segment. Fluid in the cul-de-sac is also noted. (c) With uterine contraction, a slightly different shape of the uterus is noted compared with that in a and b. Low-signal-intensity areas (solid curved arrow) in the anterior segment have disappeared, indicating contraction-induced low signal intensity. In the upper posterior segment, low-signal-intensity areas (open curved arrow), which have slightly different orientation from that in a, are demonstrated. The ill-defined lower signal intensity (straight arrow) in the posterior segment is recognized as in b.

View larger version (182K): [in this window] [in a new window] [Download PPT slide]   Figure 3c. Sagittal MR images in a 36-year-old woman with endometriosis. (a) T2-weighted FSE MR image (4,000/85) demonstrates well-defined focal low-signal-intensity area (straight arrow) in the anterior lower segment and irregular low-signal-intensity areas (solid curved arrow), which are continuous to the junctional zone. In the posterior segment, thick bandlike low-signal-intensity areas (open curved arrow) are noted. Fluid in the cul-de-sac is demonstrated. (b, c) T2-weighted SSFSE MR images (20,000/94, echo train length of 85) in different phases of imaging demonstrate effects of uterine contraction. (b) In the anterior segments, ill-defined low-signal-intensity areas (solid curved arrow) are noted in the myometrium. A low-signal-intensity area (open curved arrow) in the posterior segment, which has a different shape from that in a, is noted. An ill-defined focal low-signal-intensity area (straight arrow) appears in the posterior segment. Fluid in the cul-de-sac is also noted. (c) With uterine contraction, a slightly different shape of the uterus is noted compared with that in a and b. Low-signal-intensity areas (solid curved arrow) in the anterior segment have disappeared, indicating contraction-induced low signal intensity. In the upper posterior segment, low-signal-intensity areas (open curved arrow), which have slightly different orientation from that in a, are demonstrated. The ill-defined lower signal intensity (straight arrow) in the posterior segment is recognized as in b.

View larger version (182K): [in this window] [in a new window] [Download PPT slide]   Figure 4a. Sagittal MR images in a 51-year-old woman with multiple leiomyoma and endometrial carcinoma. (a) T2-weighted FSE MR image (4,000/85) shows leiomyoma in the myometrium as low signal intensity (large straight arrows). Diffuse thickening of the endometrium (small arrow) indicates endometrial carcinoma. Fluid was identified in the cul-de-sac (curved arrow). (b, c) T2-weighted SSFSE MR images (22,000/95, echo train length of 85) in different phases of imaging demonstrate the effect of uterine contraction. The shape and signal intensity of the uterus are almost identical to those in a. Leiomyoma (large white arrows), endometrial carcinoma (small arrow), and fluid (black arrow) in the cul-de-sac are also seen. With uterine contraction, a well-defined focal low-signal-intensity area (open curved arrow in c) at the junctional zone of the anterior wall is noted, mimicking leiomyoma.

View larger version (176K): [in this window] [in a new window] [Download PPT slide]   Figure 4b. Sagittal MR images in a 51-year-old woman with multiple leiomyoma and endometrial carcinoma. (a) T2-weighted FSE MR image (4,000/85) shows leiomyoma in the myometrium as low signal intensity (large straight arrows). Diffuse thickening of the endometrium (small arrow) indicates endometrial carcinoma. Fluid was identified in the cul-de-sac (curved arrow). (b, c) T2-weighted SSFSE MR images (22,000/95, echo train length of 85) in different phases of imaging demonstrate the effect of uterine contraction. The shape and signal intensity of the uterus are almost identical to those in a. Leiomyoma (large white arrows), endometrial carcinoma (small arrow), and fluid (black arrow) in the cul-de-sac are also seen. With uterine contraction, a well-defined focal low-signal-intensity area (open curved arrow in c) at the junctional zone of the anterior wall is noted, mimicking leiomyoma.

View larger version (179K): [in this window] [in a new window] [Download PPT slide]   Figure 4c. Sagittal MR images in a 51-year-old woman with multiple leiomyoma and endometrial carcinoma. (a) T2-weighted FSE MR image (4,000/85) shows leiomyoma in the myometrium as low signal intensity (large straight arrows). Diffuse thickening of the endometrium (small arrow) indicates endometrial carcinoma. Fluid was identified in the cul-de-sac (curved arrow). (b, c) T2-weighted SSFSE MR images (22,000/95, echo train length of 85) in different phases of imaging demonstrate the effect of uterine contraction. The shape and signal intensity of the uterus are almost identical to those in a. Leiomyoma (large white arrows), endometrial carcinoma (small arrow), and fluid (black arrow) in the cul-de-sac are also seen. With uterine contraction, a well-defined focal low-signal-intensity area (open curved arrow in c) at the junctional zone of the anterior wall is noted, mimicking leiomyoma.

View larger version (173K): [in this window] [in a new window] [Download PPT slide]   Figure 5a. Sagittal MR images in a 37-year-old woman with leiomyoma and uterine septum. (a) T2-weighted FSE MR image (4,000/85) demonstrates two discrete bandlike low-signal-intensity areas (arrows) in the junctional zone in the anterior wall. Fluid is noted in the cul-de-sac. (b, c) T2-weighted SSFSE MR images (20,000/94, echo train length of 85) in different phases of imaging demonstrate a different number and shape of bandlike low signal intensities (arrows).

View larger version (165K): [in this window] [in a new window] [Download PPT slide]   Figure 5b. Sagittal MR images in a 37-year-old woman with leiomyoma and uterine septum. (a) T2-weighted FSE MR image (4,000/85) demonstrates two discrete bandlike low-signal-intensity areas (arrows) in the junctional zone in the anterior wall. Fluid is noted in the cul-de-sac. (b, c) T2-weighted SSFSE MR images (20,000/94, echo train length of 85) in different phases of imaging demonstrate a different number and shape of bandlike low signal intensities (arrows).

View larger version (170K): [in this window] [in a new window] [Download PPT slide]   Figure 5c. Sagittal MR images in a 37-year-old woman with leiomyoma and uterine septum. (a) T2-weighted FSE MR image (4,000/85) demonstrates two discrete bandlike low-signal-intensity areas (arrows) in the junctional zone in the anterior wall. Fluid is noted in the cul-de-sac. (b, c) T2-weighted SSFSE MR images (20,000/94, echo train length of 85) in different phases of imaging demonstrate a different number and shape of bandlike low signal intensities (arrows).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

When low-signal-intensity areas in the uterus related to uterine contraction are falsely recognized as leiomyoma or adenomyosis on static FSE MR images, at least the following conditions are required: the low-signal-intensity areas in the myometrium should remain as such during the imaging time, especially during filling of the center of k space, and they should have a shape similar to that of organic lesions. Otherwise, even when low-signal-intensity areas are identified on T2-weighted MR images, they can be recognized as nonorganic regions.

Currently, we have not concluded that multiphase T2-weighted SSFSE MR images improve the accuracy of the detection of leiomyoma and adenomyosis as compared with FSE MR images because the currently available ROCKIT program developed by Metz did not show a significant difference between two the types of T2-weighted imaging. This approach may be acceptable in the current study, although it does have a limitation of dealing with the clustering problem for the comparisons of the A_z values estimated in the parametric paradigm. Use of other statistical methods, which may deal with the clustering problem appropriately in the nonparametric paradigm, might be the option (16).

By characterizing the signal intensity changes, the pattern recognition of the temporal low-signal-intensity areas on T2-weighted MR images might be helpful in the diagnosis of leiomyoma and adenomyosis. On multiphase SSFSE MR images, various types of low-signal-intensity areas were detected in the uterus. Currently, however, the recognition of these functional low-signal-intensity areas on multiphase SSFSE MR images did not statistically improve the accuracy of detection of leiomyoma and adenomyosis, compared with static FSE MR images. One of the reasons might be the relatively low incidence of pseudolesions as leiomyoma or adenomyosis, which were revealed only on multiphase SSFSE MR images (identified in two to seven of 172 segments or 1%–4% by reader 1 or 2). We do not have to obtain multiphase SSFSE MR images in all patients, and when the pseudolesions related to uterine contraction are strongly suspected, we can additionally perform multiphase SSFSE MR imaging. Even on both FSE and multiphase SSFSE MR images, normal regions adjacent to the large leiomyoma or the area under the promontorium in the enlarged uterus, which showed low signal intensity, were recognized as adenomyosis in five to nine (3%–5%) of 172 segments by reader 1 or 2. Compression of the normal myometrium by an object may functionally change blood volume in the tissue or bulk of the extracellular space, and the effects of compression by the object may remain for at least the acquisition times of both FSE and SSFSE MR images. These regions were not necessarily related to uterine contraction. However, the mechanisms for making nonorganic low signal intensity in the uterus by compression might be similar to those made by uterine contraction.

In five (3%) of 172 segments, leiomyoma diagnosed with FSE and SSFSE MR images was adenomyosis, which had a clear margin on both FSE and SSFSE MR images. Thus, the multiphase kinematic approach did not necessarily increase the diagnostic ability of differentiation between leiomyoma and adenomyosis.

Multiphase SSFSE MR imaging took 1 second for one image acquisition. However, FSE MR images, for which acquisition takes at least 2 minutes, did not demonstrate functional low-signal-intensity areas often, since they provide summation and average of the signal intensity in the uterus and thickness of each uterine structure (9). When single-phase SSFSE MR images instead of FSE MR images are routinely used for the evaluation of the uterus, they will demonstrate leiomyoma and adenomyosis, as well as temporal low-signal-intensity areas, which may not be identified on FSE MR images. Thus, we should cautiously interpret low-signal-intensity areas on SSFSE MR images as not due to the different contrast resolution but as due to a different temporal resolution. Because in this study SSFSE MR images were acquired in multiphase fashion, most functional low-signal-intensity areas in the uterus could be recognized as temporal low signal intensity.

For an understanding of the incidence of functionally changed low-signal-intensity areas on SSFSE MR images, the pattern of the combination of findings on FSE and SSFSE MR images was evaluated. The number of segments with low signal intensities on both FSE and SSFSE MR images was partly dependent on the population in which diseases such as leiomyoma and adenomyosis were found. Patterns B and C, which stand for a combination of temporal low-signal-intensity areas on FSE and SSFSE MR images, were recognized in up to 53% of all segments of the uterus, and the discrepancy of findings on FSE and SSFSE MR images was indicated. Although these areas did not always suggest pathologic lesions, they could be recognized potentially as false-positive findings.

Shape and features of functionally changed low-signal-intensity areas were categorized into four patterns: well defined focal, ill defined focal, diffuse, and band- or sticklike. Diffuse low-signal-intensity areas were the most frequently encountered (72%–75%). They may potentially mimic adenomyosis. On the other hand, well-defined low-signal-intensity areas were the least frequently observed (7%). These areas may potentially be recognized as leiomyoma. Band- or sticklike low-signal-intensity areas were characteristic of contraction-induced low-signal-intensity areas (16%–19%). The direction of the band- or sticklike structure was random; thus, some of this type of contraction-induced low signal intensity could be seen as small round or oval signal intensity areas. However, because of the multisection image acquisition, these could be identified as sticklike structures.

The junctional zone in the uterus changed in signal intensity and thickness over time, as was previously reported (9). The junctional zone thickening has been recognized as one of the criteria for the diagnosis of adenomyosis (5). A junctional zone of 12 mm or thicker might be acceptable as the criterion for adenomyosis (5). FSE MR images provided the average thickness of the junctional zone in approximately 2–3 minutes (9). Various thicknesses of the junctional zone in the uterus could be seen on multiphase SSFSE MR images. Then mean thickness of the junctional zone might be one of the criteria for the evaluation of the junctional zone feature. As a consequence, thickness of the junctional zone on static FSE MR images could be applied as one of the criteria for adenomyosis.

For the evaluation of accuracy in the diagnosis of leiomyoma and adenomyosis, we included patients who underwent hysterectomy. They had severe uterine disease; thus, a population bias could be encountered and can be one of the limitations of the study. Stage of menstrual cycle was another limitation, since US has depicted different contraction frequencies of the inner third of the uterus during a menstrual cycle (17,18), which might affect the incidence of the temporal low-signal-intensity areas on multiphase SSFSE MR images.

In conclusion, multiphase-multisection T2-weighted SSFSE MR images help characterize the features of temporal low signal intensities in the uterus, which are related to uterine contractions, they do not improve the accuracy in the detection of leiomyoma and adenomyosis compared with FSE MR images.

	ACKNOWLEDGMENTS

 
The authors thank Mitsuru Ikeda, MD, Department of Medical Information and Medical Records, Nagoya University Hospital, for consultation regarding statistical analysis.

	FOOTNOTES

 
Abbreviations: A_z = area under receiver operating characteristic curve, FSE = fast spin echo, SSFSE = single shot FSE

Author contributions: Guarantors of integrity of entire study, T.M., H.S.; study concepts, T.M.; study design, T.M., A.N.; literature research, T.M., M.K., S.K.; clinical studies, T.M., M.K., S.K.; data acquisition, T.M., M.K., S.K., S.S.; data analysis/interpretation, T.M., S.S., A.N.; statistical analysis, T.M.; manuscript preparation, T.M., A.N.; manuscript revision/review, H.S., S.S.; manuscript definition of intellectual content, editing, and final version approval, T.M., H.S.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1. Ascher SM, Arnold LL, Patt RH, et al. Adenomyosis: prospective comparison of MR imaging and transvaginal sonography. Radiology 1994; 190:803-806.[Abstract/Free Full Text]
2. Hricak H, Tscholakoff D, Heinrichs L, et al. Uterine leiomyomas: correlation of MR, histopathologic findings, and symptoms. Radiology 1986; 158:385-391.[Abstract/Free Full Text]
3. Mark AS, Hricak H, Heinrichs LW, et al. Adenomyosis and leiomyoma: differential diagnosis with MR imaging. Radiology 1987; 163:527-529.[Abstract/Free Full Text]
4. Outwater EK, Siegelman ES, Van Deerlin V. Adenomyosis: current concepts and imaging considerations. AJR Am J Roentgenol 1998; 170:437-441.[Free Full Text]
5. Reinhold C, McCarthy S, Bret PM, et al. Diffuse adenomyosis: comparison of endovaginal US and MR imaging with histopathologic correlation. Radiology 1996; 199:151-158.[Abstract/Free Full Text]
6. Togashi K, Kawakami S, Kimura I, et al. Uterine contractions: possible diagnostic pitfall at MR imaging. J Magn Reson Imaging 1993; 3:889-893.[Medline]
7. IJland MM, Evers JL, Dunselman GA, et al. Subendometrial contractions in the nonpregnant uterus: an ultrasound study. Eur J Obstet Gynecol Reprod Biol 1996; 70:23-24.[CrossRef][Medline]
8. de Vries K, Lyons EA, Ballard G, et al. Contractions of the inner third of the myometrium. Am J Obstet Gynecol 1990; 162:679-682.[Medline]
9. Masui T, Katayama M, Kobayashi S, et al. Changes in myometrial and junctional zone thickness and signal intensity: demonstration with kinematic T2-weighted MR imaging. Radiology 2001; 221:75-85.[Abstract/Free Full Text]
10. Metz CE. Some practical issues of experimental design and data analysis in radiological ROC studies. Invest Radiol 1989; 24:234-245.[Medline]
11. Metz CE, Herman BA, Roe CA. Statistical comparison of two ROC-curve estimates obtained from partially-paired datasets. Med Decis Making 1998; 18:110-121.[Abstract/Free Full Text]
12. Henley J. Standard error of the kappa statistic. Psychol Bull 1987; 102:315-321.[CrossRef]
13. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977; 33:159-174.[CrossRef][Medline]
14. Togashi K, Ozasa H, Konishi I, et al. Enlarged uterus: differentiation between adenomyosis and leiomyoma with MR imaging. Radiology 1989; 171:531-534.[Abstract/Free Full Text]
15. Togashi K, Kawakami S, Kimura I, et al. Sustained uterine contractions: a cause of hypointense myometrial bulging. Radiology 1993; 187:707-710.[Abstract/Free Full Text]
16. Obuchowski NA. Nonparametric analysis of clustered ROC curve data. Biometrics 1997; 53:567-578.[CrossRef][Medline]
17. Lyons EA, Taylor PJ, Zheng XH, et al. Characterization of subendometrial myometrial contractions throughout the menstrual cycle in normal fertile women. Fertil Steril 1991; 55:771-774.[Medline]
18. Lesny P, Killick SR, Tetlow RL, et al. Uterine junctional zone contractions during assisted reproduction cycles. Hum Reprod Update 1998; 4:440-445.[Abstract/Free Full Text]

This article has been cited by other articles:

				A. Nakai, K. Togashi, K. Kosaka, A. Kido, M. Kataoka, T. Koyama, and S. Fujii Do Anticholinergic Agents Suppress Uterine Peristalsis and Sporadic Myometrial Contractions at Cine MR Imaging? Radiology, December 19, 2007; (2007) 2461062091. [Abstract] [Full Text]

				K. Tamai, K. Togashi, T. Ito, N. Morisawa, T. Fujiwara, and T. Koyama MR Imaging Findings of Adenomyosis: Correlation with Histopathologic Features and Diagnostic Pitfalls RadioGraphics, January 1, 2005; 25(1): 21 - 40. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) All Versions of this Article: 2271011587v1 227/2/345    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Masui, T. Articles by Sakahara, H. Search for Related Content PubMed PubMed Citation Articles by Masui, T. Articles by Sakahara, H.