| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Radiology, Massachusetts General Hospital, 55 Fruit St., Boston,
MA 02114.
2 Present address: Department of Radiology, Cambridge Health Alliance, 1493
Cambridge St., Cambridge, MA 02139.
Received March 22, 2001;
accepted after revision January 8, 2002.
Address correspondence to C. A. Hulka.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Patients' records were evaluated for the diagnosis of adenomyosis determined by hysterectomy and for sonography performed within 2 months of surgery. Seventy-three sonograms were evaluated by sonologists without knowledge of the extent of adenomyosis. Sonographic categories included visualization of the endometrium, presence of a diffuse uterine process, presence of fibroids, and normal findings. Pathologic results included mild, focal, and severe adenomyosis. Histologic and sonographic categories were correlated using the chi-square and Fisher's exact tests.
RESULTS. Forty-six specimens contained mild adenomyosis, 18 contained severe disease, and nine contained focal disease. Forty-one specimens contained fibroids. The endometrium was visualized in 10 cases of severe adenomyosis, seven cases of adenomyoma, and 35 cases of mild disease. Visualization of the endometrium did not relate to the severity of disease (p = 0.6). Of 18 cases of severe disease, 13 sonograms showed a diffuse process. Of nine cases of adenomyomas, no sonograms showed a diffuse process; and of 46 cases of mild disease, nine sonograms showed a diffuse process. A diffuse process was related to the severity of adenomyosis (p < 0.001). When fibroids were present, a diffuse process did not relate to the extent of adenomyosis (p = 0.01).
CONCLUSION. In the absence of focal fibroids, a diffuse uterine process seen on sonography relates to the severity of adenomyosis. Fibroids limit the ability to diagnose the severity of adenomyosis. The visualization of the endometrium does not relate to the severity of adenomyosis.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Other disorders coexist with adenomyosis. Uterine leiomyomas are associated with adenomyosis in 36-50% of cases [1]. Endometrial hyperplasia and carcinoma have also been reported to occur with greater frequency in women with adenomyosis than in women without adenomyosis, raising the question of hormonal influence.
The clinical presentation of adenomyosis is the parous, perimenopausal patient with dysmenorrhea and menorrhagia. Physical examination may reveal an enlarged, tender uterus, but the diagnosis has been difficult to obtain without histologic confirmation. The preoperative diagnosis of adenomyosis remains elusive, although the use of sonography and MR imaging improves the diagnosis, with high sensitivities reported for MR imaging in high-prevalence populations [3,4,5,6].
The evaluation of endovaginal sonography in the diagnosis of adenomyosis has received attention, with small series reporting sensitivities and specificities of up to 87% and 98%, respectively [7,8,9,10]. Because sonography is frequently the initial imaging study in these patients, improving the diagnosis of this disease with sonography is important. The diagnosis of adenomyosis is frequently overlooked, especially in the presence of a fibroid uterus. Management, such as planning for myomectomy for uterine preservation, may be altered if extensive adenomyosis is suspected.
The purpose of this study was to evaluate whether endovaginal sonographic characteristics can indicate the extent of adenomyosis and whether the presence of fibroids alters the ability to predict the extent of adenomyosis on sonography. In addition, we tested whether the visibility of the endometrium is related to the severity of disease.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
All patients underwent transabdominal and endovaginal sonographic examinations; transabdominal probes (2.25, 3.5, 3.75, and 5.0 MHz) and transvaginal probes (5.0 and 7.0 MHz) (Acuson, Mountain View, CA; General Electric Medical Systems, Milwaukee, WI; Diasonics, Miltipas, CA) were used. Multiple levels through the uterus in the sagittal and transverse planes were imaged. Any focal fibroids or abnormality was imaged and measured in multiple planes. The endometrium was measured in the sagittal plane from echogenic interface to echogenic interface.
The hard-copy sonograms of these 73 patients were reviewed independently and retrospectively by two sonologists who were experienced in gynecologic sonography. At the time of analysis, the reviewers did not know the final diagnosis regarding the extent of adenomyosis. The sonograms were evaluated for four specific sonographic categories that included visualization versus nonvisualization of the endometrium, presence or absence of a diffuse uterine process without focal abnormalities (homogeneous versus heterogeneous echotexture of the myometrium), the presence or absence of fibroids, and normal findings. The reviewers assessed the endometrium for visualization and not whether distortion, thickening, or other abnormal features of the endometrium were present, because fibroids or other endometrial disorders could confound the results. Visualization of the endometrium was selected for study because several anecdotal cases of severe adenomyosis before this study showed poor visualization of the endometrium on sonography. Fibroids were defined as focal masses in the serosa, muscle, or submucosal regions of the myometrium that were usually hypoechoic in echotexture, although some could be hyperechoic or heterogeneous with or without calcifications. If sonographic findings were not normal, the sonogram was assessed for each of the first three categories. If the reviewer suspected a focal adenomyoma, that was stated during collection of the data. Any discrepant readings were resolved by discussion between the reviewers for a consensus.
The pathologic results for each patient were placed into one of three categories regarding the extent of adenomyosis. These pathologic categories are those used in our institution. The histologic diagnosis of adenomyosis was determined by the presence of subendometrial glands at least one-half to one low-power field (2-3 mm) below the endometrial junction with the myometrium. Mild adenomyosis (category 1) was determined when only microscopic foci of adenomyosis were present or only the inner one third of the myometrium was involved with disease. Focal disease (category 2) included those specimens with focal adenomyomas. Severe or diffuse adenomyosis (category 3) was present when the specimen showed disease extending into the outer two thirds of the myometrium and up to gross involvement of the entire uterus. Other histologic findings were recorded, such as the presence of fibroids, endometrial carcinoma, endometrial polyps, endometrial hyperplasia, and metastatic lesions.
The three histologic categories of mild, focal, and severe adenomyosis were further categorized for the presence or absence of fibroids (total of six categories) and were correlated with the four sonographic categories of visualization versus nonvisualization of the endometrium, diffuse versus nondiffuse process in the uterus, presence or absence of fibroids, and a normal sonographic appearance of the uterus.
The chi-square and Fisher's exact tests were applied, with and without controlling for the presence of fibroids, for the visualization of the endometrium and for the presence of a diffuse sonographic process.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Of the nine specimens with focal adenomyomas, one case was a polypoid adenomyoma extending into the endometrium, and one was a paratubal adenomyoma extending into the right adnexal region in a patient with severe adenomyosis and coexisting fibroids.
Other diagnoses of the uterus included one diagnosis of endometrial carcinoma with severe adenomyosis and coexisting fibroids; one diagnosis of metastatic adenocarcinoma to the myometrium of unknown cause with microscopic foci of adenomyosis; and four diagnoses of polyps, two of which were in the setting of mild adenomyosis, one of which was in the setting of severe adenomyosis, and one of which was in the setting of mild adenomyosis and coexisting fibroids. Other extrauterine diagnoses included one case of endometriosis in the setting of severe adenomyosis and one diagnosis of metastatic adenocarcinoma of gastrointestinal origin to the fallopian tube.
Sonographic Results and Correlation with Histologic Results
Endometrial visualization and the effect of fibroids.—The
endometrium was visualized in 52 cases and was not visualized in 21 cases
(Table 1). In three cases,
discrepant interpretations were resolved by consensus review. Of the 18 cases
with severe adenomyosis, the endometrium was visualized on 10 sonograms. Of
the 11 cases of severe adenomyosis with coexisting fibroids, the endometrium
was visualized in six.
|
Of the nine cases with focal adenomyosis, the endometrium was visualized in seven cases, two of which contained fibroids and five of which did not contain fibroids. Of the two remaining cases of focal adenomyomas that did not show the endometrium, no fibroids were present.
Of the 46 cases with mild adenomyosis, the endometrium was seen in 35 cases and was not seen in 11 cases. In 28 of these 46 mild cases, fibroids were present, and in only seven of these cases did the sonograms not show the endometrium.
For all cases, visualization of the endometrium did not relate to the severity of adenomyosis (p=0.6). The presence or absence of fibroids did not alter this lack of relation between endometrial visualization and severity of adenomyosis (p = 0.74 in the absence of fibroids and p = 0.28 in the presence of fibroids).
Diffuse uterine process and the effect of fibroids.—The reviewers scored 22 of the 73 sonograms as showing a diffuse uterine process when the uterus was enlarged and heterogeneous in echotexture (Table 2 and Fig. 1A,1B). Discrepant interpretations in four cases were resolved by consensus.
|
|
|
Of the 18 cases of severe adenomyosis, 13 sonograms (72%) were graded as showing a diffuse uterine process, of which six had coexisting fibroids (Fig. 2). Of the remaining five cases of severe adenomyosis that were graded as not showing a diffuse uterine process on sonography (but as having fibroids), all five cases had coexisting fibroids on pathologic correlation.
|
Of the nine cases of focal adenomyosis, none of the reviewers graded the sonograms as showing a diffuse process, and two of these cases contained coexisting fibroids (Fig. 3).
|
Of the 46 cases of mild adenomyosis, the reviewers graded nine sonograms (20%) as showing a diffuse uterine process, of which eight contained coexisting fibroids.
For all cases, a sonographically determined diffuse uterine process was related to the severity of adenomyosis at histologic examination (p < 0.001). When fibroids were not present, a sonographically determined diffuse uterine process also related to the severity of adenomyosis (p < 0.001). However, when fibroids were present, a sonographically diffuse uterine process did not relate to the extent of adenomyosis determined by histologic examination (p = 0.1). In other words, fibroids confounded the sonographic diagnosis of severe adenomyosis.
Fibroids and pathologic correlation.—The reviewers recorded the presence of fibroids on 38 sonograms and did not visualize focal adenomyomas on any of the sonograms. Of these 38 cases, histologic examination revealed that 27 of the surgical specimens contained fibroids, two specimens contained fibroids and adenomyomas, five specimens contained adenomyomas, one specimen contained metastatic adenocarcinoma, and three specimens showed no focal abnormality. The sensitivity and specificity for diagnosing fibroids in this selected population of patients were both 63%.
Normal sonographic findings and pathologic correlation.—Of the 10 sonograms that the reviewers considered to indicate normal findings, two patients were diagnosed with adenomyomas, one measuring 5 mm and one measuring 7 mm; three patients were diagnosed with small (<1 cm) fibroids and minimal adenomyosis; and five patients were diagnosed with minimal adenomyosis. Thus, the sonograms did not detect small adenomyomas, small fibroids, or mild adenomyosis.
Cystic spaces and pathologic correlation.—In three sonograms, the reviewers commented on cystic spaces in the myometrium that correlated with the pathologic findings of severe adenomyosis in these specimens (Fig. 4). Histologic examination showed fluid-filled dilated endometrial glands that correlated with the cystic spaces seen on sonography.
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Early articles described the findings of cystic spaces in the myometrium and posterior wall thickening on transabdominal pelvic sonography as suggestive of adenomyosis [11,12,13,14]. More recent articles using transvaginal sonography similarly describe cystic spaces and include the finding of heterogeneous zones in the myometrium as indicative of adenomyosis [7,8,9, 10, 15]. Sensitivities and specificities of endovaginal sonography for the diagnosis of adenomyosis are in the range of 80-87% and 74-96%, respectively, although the study by Fedele et al. [7] did not include patients with suspected fibroids and Atzori et al. [10] specified the use of cystic spaces for their diagnostic criteria. As in the study by Reinhold et al. [9], we found the most common sonographic finding in patients with diffuse adenomyosis to be a uterus with a diffuse heterogeneous echotexture. Myometrial cystic spaces were seen in only three cases in our study. These spaces correlated with fluid-filled dilated endometrial glands in the myometrium on histologic examination of the surgical specimen.
One focus of this study was to evaluate the effect of the presence of fibroids on the sonographic diagnosis of adenomyosis, because the gynecology literature shows that 60-80% of patients diagnosed with adenomyosis show other pelvic processes, including fibroids, endometriosis, endometrial polyps, and endometrial hyperplasia [2]. In our study, fibroids were present in 56% of all specimens, including 61% of those with severe adenomyosis, 22% of those with adenomyomas, and 61% of those with only mild adenomyosis. When fibroids were not present on sonography (and their absence was confirmed histologically), a sonographically diffuse uterine process correlated significantly with severe adenomyosis. However, the sonographic presence of histologically proven fibroids confounded the reviewers' ability to diagnose severe adenomyosis on the sonograms.
The cause of the association of other pelvic processes and adenomyosis is not clear. In particular, endometriosis is a distinctly different disease process from adenomyosis. In our study, in addition to fibroids, six other uterine processes were present, including four polyps, one metastatic adenocarcinoma to the myometrium, and one endometrial carcinoma. Coexisting extrauterine diagnoses included endometriosis and metastatic adenocarcinoma to the fallopian tube. No significant correlation was seen between severe adenomyosis and endometrial carcinoma or any other diagnosis than fibroids in our study.
The sonographic distinction of adenomyoma from fibroids has been described by Fedele et al. [7]: the presence of a heterogeneous focal nodule with indistinct margins and cystic spaces is suggestive of adenomyomas. In our study, we could not distinguish adenomyomas from fibroids because some fibroids had indistinct margins and no adenomyomas with cystic spaces were seen on sonography or histologic examination. Thus, the sensitivity for diagnosing adenomyomas was poor in our study. Two of the nine adenomyomas in our study were unusual in that one was a polypoid lesion extending into the endometrium and the other was a paratubal adenomyoma, both rare entities. The sensitivity and specificity for the diagnosis of fibroids in this series were both 63%. Because the number of adenomyomas in this study was relatively small, further work in distinguishing these two entities will be of interest.
The strengths of this study are the determination that fibroids limit the sonologist's ability to diagnosis severe adenomyosis and the determination that the most common sonographic finding of severe adenomyosis is a diffusely heterogeneous uterus. Myometrial cystic spaces are helpful in the diagnosis of adenomyosis but need not be present for a successful diagnosis of severe disease.
The retrospective review of hard-copy sonograms presents limitations because real-time endovaginal scanning may reveal the presence and extent of adenomyosis better than hardcopy review alone [9]. In addition, knowledge of patients' symptoms while scanning may aid in the diagnosis. A prospective study could accurately determine the sensitivity, specificity, and positive and negative predictive values for the diagnosis of severe adenomyosis.
Although many authors have described the appearance of adenomyosis on MR imaging, the first imaging test usually obtained for any woman's pelvic symptoms is pelvic sonography [16]. Thus, it is important to consider the diagnosis of severe adenomyosis given the sonographic appearance of a diffuse heterogeneous uterine process, especially in the absence of discrete fibroids. Although an early study by Ascher et al. [15] found MR imaging to be more sensitive than endovaginal sonography for the diagnosis of adenomyosis, a larger study by Reinhold et al. [16] found similar accuracies for these two modalities.
In summary, the diagnosis of severe adenomyosis can be made when the sonographic features of a heterogeneous uterus with or without cystic spaces are present. Fibroids confound the ability to predict the severe form of adenomyosis.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  A. L. Spielmann, C. Keogh, B. B. Forster, M. L. Martin, and L. S. Machan Comparison of MRI and sonography in the preliminary evaluation for fibroid embolization. Am. J. Roentgenol., December 1, 2006; 187(6): 1499 - 1504. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Chopra, A. S. Lev-Toaff, F. Ors, and D. Bergin Adenomyosis:common and uncommon manifestations on sonography and magnetic resonance imaging. J. Ultrasound Med., May 1, 2006; 25(5): 617 - 627. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
