| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Genitourinary Imaging |
1 From the Departments of Nuclear Medicine and Diagnostic Imaging (M.K., K.T., A.K., A.N., T.F.), Radiology (T.K.), and Obstetrics and Gynecology (S.F.), Graduate School of Medicine, Kyoto University, Shogoinkawaharacho 54, Sakyo, Kyoto 606-8507, Japan. Received August 12, 2003; revision requested October 28; final revision received June 16, 2004; accepted June 23. Address correspondence to M.K. (e-mail: makok@kuhp.kyoto-u.ac.jp).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: The university ethics committee approved the study protocol, and all subjects gave written informed consent. Nineteen healthy women were examined during menstrual cycle days 1–3 by using a 1.5-T-magnet unit. Sixty serial half-Fourier rapid acquisition with relaxation enhancement MR images of the uterus were obtained every 3 seconds for 3 minutes and displayed in cine mode. MR images were analyzed in terms of thickness of the inner low-signal-intensity myometrial layer, presence of endometrial distortion, and uterine peristalsis detectability. Pain was assessed by using a four-point scale. For 56 MR imaging cases, the association between MR imaging findings and pain degree was examined with Spearman correlation and Mann-Whitney tests. MR imaging findings in the dysmenorrheic and eumenorrheic subject groups were compared by using Mann-Whitney and 
2 tests. The area of the uterine myometrium in both groups was calculated for quantitative assessment of uterine contraction and was compared between the groups by using the Student t test.
RESULTS: MR imaging findings revealed marked changes during cycle days 1–3. Thickness of the inner low-signal-intensity myometrial layer and endometrial distortion were significantly associated with pain degree (P < .001), while uterine peristalsis was undetectable when pain was severe or moderate. The area of the uterine myometrium significantly decreased during cycle days 1–3 in the dysmenorrheic group, as compared with that in the eumenorrheic group (P = .010).
CONCLUSION: MR imaging features of the uterus on cycle days 1–3 correlated with pain degree. Cine-mode-display MR imaging is a potential tool for evaluating dysmenorrhea.
© RSNA, 2005
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Although ultrasonography (US) has been used to monitor the contractile activity of the uterus, which is depicted as endometrial wavelike movements (11,12), it is not easy to directly visualize the subtle and transient changes in the myometrium during contractile activity at US, which yields limited tissue contrast. Apart from reports on the subtle and rhythmic conductive movements in the subendometrial myometrium and the sustained myometrial contractions that masquerade as leiomyomas during pregnancy, there have been only a small number of studies focusing on myometrial movements (13,14).
In contrast to methods involving the use of intrauterine catheters, magnetic resonance (MR) imaging is noninvasive and does not involve use of ionizing radiation, and it has an important role in the evaluation of the female pelvis. With MR imaging, which yields good soft-tissue contrast, the uterine zonal anatomy that changes during the menstrual cycle can be depicted on T2-weighted images. Furthermore, recent advancements in rapid imaging techniques have enabled the visualization of kinematic changes in the uterus, such as subtle conductions of the subendometrial myometrium that include changes in both configuration and signal intensity (ie, uterine peristalsis) (15). The cine mode display of such MR images obtained at an interval of a few seconds facilitates the evaluation of slow movements. Thus, our aim in this study was to prospectively evaluate uterine contractility during menstruation and its relation to primary dysmenorrhea by using MR imaging with cine mode display.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
The women were initially assigned to dysmenorrheic or eumenorrheic groups according to their own statements regarding menstrual pain; however, final group placements were decided on the basis of responses to a questionnaire about pain during menstrual days 1–3. Each volunteer was offered a free conventional gynecologic MR imaging examination, with a diagnosis made by a radiologist.
Evaluation of Pain
Because the pain associated with primary dysmenorrhea usually begins a few hours before or just after the onset of menstruation and can last as long as 48–72 hours (6), the level of pain experienced during the first 3 days of menstruation was evaluated just after the MR imaging examinations were performed. The evaluations were conducted by one of several examiners (M.K., A.K., A.N., T.F.) by using a questionnaire that was given to each subject. The level of pain was determined by using a four-point rating scale: Grade 0 indicated no; grade 1, mild; grade 2, moderate; and grade 3, severe pain. Severe pain was defined as pain so strong that the subject would have used an analgesic if she had not been recruited into the study, while moderate pain was defined as pain that usually could be endured without analgesics. In our study protocol, to exclude the effect of analgesics that might have contained prostaglandin synthase inhibitors, the subjects were asked to avoid taking analgesics for at least 12 hours before undergoing MR imaging, even if they were experiencing severe pain.
The need for analgesics is a reasonable clinical criterion for judging the severity of pain experienced during ordinary menstrual cycles. Thus, the subjects were classified into the two groups on the basis of whether they reported having or not having severe pain during the first 3 days of menstruation. The dysmenorrheic group consisted of subjects who had severe pain at least once during cycle days 1–3. The eumenorrheic group consisted of subjects whose pain was no more than moderate during this period.
MR Imaging Protocol
MR imaging examinations were performed on the first, second, and third days of menstruation (ie, cycle days 1–3). MR images were obtained by using a 1.5-T-magnet unit (Symphony; Siemens, Erlangen, Germany) and a phased-array coil. Fast spin-echo and half-Fourier rapid acquisition with relaxation enhancement (RARE) techniques were used to obtain T2-weighted MR images. With the patient in quiet respiration, a total of 60 serial MR images were obtained in a midsagittal plane of the uterus along the long axis of the uterine body every 3 seconds for 3 minutes by using the RARE sequence (3000/80 [repetition time msec/echo time msec], 300-mm field of view, 5-mm section thickness, 256 x 256 matrix). Sagittal fast spin-echo T2-weighted MR images were obtained once during each examination. Sagittal spin-echo T1-weighted and transverse fast spin-echo T2-weighted MR images were obtained on cycle day 3, when uterine changes were at a minimum, in all but one subject, in whom images could not be obtained on that day. The images obtained on cycle day 2 were used for this subject. These MR images were evaluated to determine whether there were any findings suggestive of an underlying disease such as endometriosis or uterine myoma. No premedications, including anticholinergic drugs, were given to any subject.
Qualitative Image Analysis
RARE MR images were displayed in cine mode—at 12 times faster than the actual speed—to facilitate evaluation of slow movements, according to findings in a report by Lyons et al (13). The MR images were evaluated by two independent radiologists (M.K., K.T.), who have been reading gynecologic MR images as part of their daily clinical and research practices for 4 and 18 years, respectively. After completing their individual reviews, the radiologists then jointly reviewed the images and reached a consensus. On the basis of reported MR imaging findings related to sustained contractions (16) and subtle myometrial contractions (15,17), the radiologists assessed the images in terms of the following three features (Fig 1), all of which were evaluated in cine mode display: (a) thickness of the inner layer of the myometrium that exhibited low signal intensity—that is, the junctional zone; (b) presence or absence of endometrial distortion during MR imaging; and (c) detectability of uterine peristalsis (ie, conduction of the subendometrial low-signal-intensity area). In cases in which considerable changes that could alter the evaluation categories were observed during the 3-minute image acquisition, the most prominent findings were used for evaluation.
|
Endometrial distortion was evaluated as follows: Grade 0 meant distortion was absent, and grade 1 meant distortion was present. Uterine peristalsis detectability was evaluated as follows: Grade 0 meant peristalsis was detectable, and grade 1 meant peristalsis was undetectable. When uterine peristalsis is not seen, it is hard to determine whether it is actually absent or merely covered by other findings; hence, we used the concept of detectability, not presence or absence, for evaluation purposes. With regard to inner myometrial layer thickness and endometrial distortion, the findings with high grades were considered to be the most prominent, whereas with regard to uterine peristalsis detectability, the findings with lower grades were considered to be the most prominent.
Quantitative Image Analysis
For quantitative assessment of uterine contraction involving the entire myometrium, changes in an area of the uterine myometrium during menstrual cycle days 1–3, as seen on midsagittal sections of the uterus, were assessed. Because the spin-echo T2-weighted MR images were too blurred by motion artifacts, RARE MR images were used. One representative static image—the one that showed the most prominent findings—from the 60 consecutive RARE images was used for evaluation because it is considerably more complicated to quantitatively evaluate kinematic images. Areas of uterine corpus, defined as regions of the uterus above the level of the internal os (Fig 2), and the area of the endometrium were calculated on a midsagittal RARE MR image of the uterus. Regions of interest were drawn over the uterine corpus or the endometrium. These areas were then measured, in centimeters squared, by using a satellite console of the MR imaging unit. One radiologist (M.K.) selected the regions of interest.
|
One of the subjects in the dysmenorrheic group did not undergo MR imaging on cycle day 3 and was thus excluded from this part of the study, which involved quantitative image analysis. Therefore, eight subjects from the dysmenorrheic group and 10 subjects from the eumenorrheic group were included.
Relationship between MR Imaging Findings and Pain Degree
The following two investigations were performed at qualitative image analysis: (a) We sought to determine whether a relationship existed between the three MR imaging findings (thickness of the inner low-signal-intensity layer of the myometrium, presence of distortion, and detectability of peristalsis) and the level of pain in 56 MR imaging examinations, and (b) we compared the MR imaging findings in the dysmenorrheic (n = 9) and eumenorrheic (n = 10) subject groups by performing imaging at the time of maximum pain during menstrual cycle days 1–3. For quantitative image analysis, the mean contraction ratio for the dysmenorrheic group (n = 8) was compared with that for the eumenorrheic group (n = 10) to see whether there were any differences between the two values.
Statistical Analyses
To evaluate the consistency of the depiction of each MR imaging finding, agreement between the radiologists regarding interpreted findings was evaluated by using 
correlation coefficients.
The thickness of the inner myometrial layer represented ordinal data, whereas endometrial distortion and peristalsis detectability were considered dichotomous data for qualitative analysis of the MR imaging findings and pain degree. Thus, these two sets of data were analyzed differently. The correlation between inner myometrial layer thickness and pain degree at each MR imaging examination was assessed by using the Spearman correlation test. The relationships between endometrial distortion and pain degree and between peristalsis detectability and pain degree were examined by using the Mann-Whitney test. Comparisons between the dysmenorrheic and eumenorrheic groups in terms of inner myometrial layer thickness also were conducted by using the Mann-Whitney test. Comparisons of endometrial distortion and peristalsis detectability between the two groups were performed by using the 2 test.
With regard to quantitative analyses, the contraction ratios for subjects in both groups were examined by using the two-tailed Student t test to assess differences between the two groups. P < .05 was considered to indicate a significant difference. All statistical analyses were performed by using statistical computer software (SPSS for Windows, version 11.0, SPSS, Chicago, Ill; or Excel 10.0, Microsoft, Redmond, Wash).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
The relationships between pain degree and menstrual cycle day are summarized in Table 1. Nine subjects reported having severe pain on cycle day 1, while only two subjects reported having continued severe pain and nine reported having moderate pain on cycle day 2. No subjects reported having severe pain and 10 reported having no pain on cycle day 3 (Table 1). Fifteen subjects reported that they felt maximum pain on cycle day 1, whereas four experienced maximum pain on cycle day 2. During all three days, the maximum pain was severe for nine subjects, moderate for seven, and mild for three. Thus, nine subjects were assigned to the dysmenorrheic group, and the remaining 10 were assigned to the eumenorrheic group.
|
The correlation coefficients for the two radiologists’ agreement regarding interpreted MR imaging findings were 0.747 for thickness of the inner myometrial layer, 0.913 for endometrial distortion, and 0.845 for peristalsis detectability. Thus, it can be concluded that there was good to excellent interobserver agreement regarding the interpretation of MR imaging findings in the uterus in this study. The MR imaging features of the uterus observed on cycle days 1–3 are shown in Table 2; marked changes were observed during this period. For example, the inner layer of low signal intensity showed prominent thickening on cycle day 1; it even reached full thickness on day 1 in 10 cases. However, the inner layer became thinner, diminishing to less than half the thickness of the myometrium, toward cycle day 3. Endometrial distortion was observed in almost half of the subjects on cycle day 1. The distortion became less frequent toward cycle day 3. In contrast, uterine peristalsis was seen most prominently on cycle day 3. Therefore, the typical MR imaging findings in the uterus on cycle day 1—a thick inner low-signal-intensity layer and endometrial distortion—differed from the typical findings seen on day 3—a thin inner low-signal-intensity layer and peristalsis.
|
|
2 test results (P = .096). There also was no significant difference between the two groups with regard to peristalsis detectability (P = .937) (Table 4). Finally, contractions were evaluated quantitatively by using selected RARE MR images. The contraction ratio was significantly larger in the dysmenorrheic group (P = .010) (Table 5). Representative images are presented in Figures 3 and 4.
|
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
The MR imaging findings in the uterus displayed in cine mode in this study revealed that marked changes occur during the menstrual period. The inner low-signal-intensity myometrial layer showed prominent thickening on cycle day 1 and gradually reduced in thickness toward cycle day 3. Endometrial distortion was observed in almost half of the subjects on cycle day 1, with the endometrium gradually restoring its lentiform shape on cycle day 3. These findings were associated with the degree of pain perceived by each subject. Another finding that seems to be associated with dysmenorrheic pain is reduced size of the myometrium in the uterine corpus. Comparisons of MR imaging findings between the dysmenorrheic and eumenorrheic groups revealed significant differences in the thickness of the inner low-signal-intensity layer and in the contraction ratio in the area of the myometrium in the uterine corpus, although no significant difference in endometrial distortion was noted.
The transient decreases in myometrial signal intensity on T2-weighted MR images were explained by the effect of squeezing blood from the myometrium (16). The low signal intensity of the junctional zone—that is, the inner low-signal-intensity myometrial layer—can also be explained by decreased water content (19). Thus, the thicker inner low-signal-intensity zone seems to reflect lower blood content in this zone due to the squeezing effect of the contractions. Decreased uterine corpus size and severe endometrial distortion also directly reflect the effects of myometrial contractility, which has a role in the discharge of menstrual blood from the cavity. All three MR imaging findings (inner myometrial layer thickness, presence or absence of endometrial distortion, and uterine peristalsis detectability) were prominent in the dysmenorrheic group and seemed to directly relate to the degree of myometrial contraction.
The production of endometrial prostaglandin increases the contractility of the myometrium (20), and high-amplitude contractions result in decreases in uterine blood flow (21,22). Pulkkinen (9) suggested that diminished uterine blood flow and anoxic pain are the underlying pathophysiologic mechanisms of primary dysmenorrhea. A report on the measurement of blood flow in the uterine artery with Doppler US also indicated that uterine artery resistance was significantly higher in subjects with dysmenorrhea than in those without it and thus supported the idea that ischemia is a cause of dysmenorrhea (23). MR imaging, in addition to having a well-established role in investigations of the cause of secondary dysmenorrhea caused by a variety of lesions, could be a new tool for evaluating degrees of pain and aid in investigations of the underlying mechanisms of primary dysmenorrhea.
Static MR images depict the sum of subtle and transient movements of the uterus, so it is reasonable to expect that findings seen on cine mode images will also be identifiable on conventional static MR images. Serial changes in uterine zonal appearances throughout the menstrual cycle have been studied. With regard to the thickness of the junctional zone, no significant differences between the follicular and secretory phases (24,25) and/or in the accelerated growth from cycle days 8–16 have been reported (26). However, although the uterus during menstruation seems substantially different from the uterus during other phases of menstrual cycles such as the follicular phase or luteal phase, to our knowledge there have been no specific reports on the size of the uterine corpus or the thickness of the junctional zone during the menstrual period (ie, cycle days 1–3). Our study results show that the area of myometrium in the uterine corpus is significantly smaller when subjects have maximum pain (ie, on cycle days 1 or 2) as compared with this area when subjects have no or mild pain.
Another notable finding was the marked thickening of the inner low-signal-intensity myometrial layer, which even reached full thickness, on T2-weighted MR images. On T2-weighted MR images, thickening of the junctional zone of more than 12 mm has been considered by some investigators to be predictive of a diagnosis of adenomyosis (27). In contrast, our MR imaging findings in healthy subjects during the menstrual period revealed that the inner low-signal-intensity myometrial layer on T2-weighted images is considerably thicker than has been reported for normal junctional zones (27). If the greater than 12-mm thickness had been directly applied to the MR imaging findings in our study, the uteri of more than half of the subjects on cycle day 1 would have fulfilled the criteria for the diagnosis of adenomyosis. Endometrial distortion was also frequently observed during the menstrual period and could have masqueraded as an endometrial deformation caused by adenomyosis. However, neither thickening of the inner low-signal-intensity myometrial layer nor endometrial distortion was consistently observed on cycle day 3; thus, adenomyosis was excluded. It can be concluded that any diagnosis of adenomyosis should be made with care, including possibly avoiding assessment during the menstrual phase, especially on cycle days 1 and 2.
On cycle day 3, the inner myometrial layer tended to be thinner and less endometrial distortion was observed. Uterine peristalsis in the usual cervicofundal direction was frequently observed on day 3, but these peristaltic waves tended to be undetectable on cycle days 1 and 2. Thus, on cycle day 3, morphologic and physiologic features seem to return to their more usual forms.
Our study had a number of limitations. First, the study population was small—19 subjects in total—and only 9 subjects reported having severe pain. An additional study with more subjects and a control group is needed to better validate our results. Second, the assessment of pain degree with questionnaires might have been too subjective. It is difficult to objectively assess levels of pain, but in many studies of dysmenorrhea, scale or objective point systems have been used to evaluate pain. Therefore, we believed it was reasonable to use similar methods in this study. A third point is that the three MR imaging findings that we used in our evaluation might not necessarily be independent of each other.
A fourth limitation could be that dysmenorrhea secondary to an underlying pelvic abnormality could not be definitively ruled out. None of our volunteers underwent physical or laparoscopic examinations to rule out tiny endometrial foci. However, we checked the questionnaire responses to determine whether any subjects reported having had persistent menstrual pain since their teenage years (not progressive as in case of secondary dysmenorrhea) and examined the MR images to confirm the presence or absence of the underlying pelvic abnormalities causing dysmenorrhea, such as adenomyosis or submucosal leiomyoma. Thus, it seems reasonable to assume that the study population consisted of healthy subjects without pelvic abnormalities. However, forms of endometriosis other than general endometrioma, such as extraovarian endometriosis, are difficult to evaluate with MR imaging. At present, we have no effective noninvasive way to exclude endometriosis, and the possible presence of minimal endometriosis in our study population might have been another limitation.
In conclusion, we believe that this was the first study in which MR imaging findings of the uterus, including thickening of the inner low-signal-intensity myometrial layer, endometrial distortion, and reduced size of the uterine corpus, were identified on menstrual cycle days 1 and 2. These findings differ markedly from those previously reported and are closely related to the degree of menstrual pain. All of the findings were considered to reflect the effects of sustained contractions, which cause blood to be squeezed out of the myometrium and might cause ischemic changes. Although this was a preliminary study, its results indicate that MR imaging is a potential tool for evaluating dysmenorrhea and uterine contractility.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Authors stated no financial relationship to disclose.
Author contributions: Guarantors of integrity of entire study, M.K., K.T.; study concepts, M.K., K.T.; study design, M.K.; literature research, M.K.; clinical studies, T.K., S.F.; data acquisition, M.K., A.K., A.N., T.F.; data analysis/interpretation, M.K., K.T.; statistical analysis, M.K.; manuscript preparation and editing, M.K.; manuscript definition of intellectual content, revision/review, and final version approval, K.T.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  P. Royo and J. L. Alcazar Three-Dimensional Power Doppler Assessment of Uterine Vascularization in Women With Primary Dysmenorrhea J. Ultrasound Med., July 1, 2008; 27(7): 1003 - 1010. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 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]
|
|
|
|
 |
|
 A. Kido, K. Togashi, M. Kataoka, Y. Maetani, A. Nakai, M. L. Kataoka, T. Koyama, and S. Fujii The effect of oral contraceptives on uterine contractility and menstrual pain: an assessment with cine MR imaging Hum. Reprod., July 1, 2007; 22(7): 2066 - 2071. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
