Objective
The aim of the present study was to differentiate and specify the subtypes of adenomyosis.
Study Design
Surgically treated adenomyosis (n = 152) was subcategorized retrospectively into 4 subtypes on the basis of magnetic resonance imaging geography. Subtype I (n = 59) consisted of adenomyosis that occurs in the uterine inner layer without affecting the outer structures. Subtype II (n = 51) consisted of adenomyosis that occurs in the uterine outer layer without affecting the inner structures. Subtype III (n = 22) consisted of adenomyosis that occurs solitarily without relationship to structural components. Adenomyosis that did not satisfy these criteria composed subtype IV (n = 20). Stepwise logistic regression analysis was used for specification of the subtypes.
Results
Subtypes I-III were suggested as a product of direct endometrial invasion, endometriotic invasion from the outside, and de novo metaplasia, respectively. Subtype IV was a heterogeneous mixture of far advanced disease.
Conclusion
Adenomyosis appears to consist of 3 distinct subtypes of different causes and an additional subtype of indeterminate cause.
Uterine adenomyosis is a pathologic condition that is characterized by the presence and the growth of endometrial or endometrium-like structures in the myometrium. This has been thought to happen most likely during the fourth and the fifth decades of life and after completion of childbearing activity. However, the recent development of diagnostic tools such as transvaginal sonography and magnetic resonance imaging (MRI) has disclosed that this pathologic condition occurs in women of younger ages.
The clinical symptoms of uterine adenomyosis are dysmenorrhea, menorrhagia, chronic pelvic pain, and subfertility; therefore, it is a target of treatment. Hysterectomy is the definitive treatment; in cases in which adenomyosis is located focally in either 1 of the uterine hemispheres, uterus-preserving surgery, adenomyomectomy, is an alternative surgery that is applicable to those who desire to preserve their reproductive potential. It has been accepted generally that uterine adenomyosis results from the direct invasion of the endometrium into the myometrium. However, our clinical experience has indicated that there are cases in which this classic hypothesis is unverifiable. In such cases, no direct relationship between adenomyosis and the endometrium is proved histologically; rather, the disease appears to be a result of the invasion of endometrium-like structures (presumably endometriosis) from outside of the uterus that disrupt the uterine serosa. Furthermore, it can be recognized that there is another fraction of patients in whom adenomyosis is present in the myometrium completely isolated from the uterine structural components such as the endometrium and the serosa. Such a difference would postulate a new hypothesis that adenomyosis is not a disease of homogeneity but is composed of multiple heterogeneous subtypes.
The aim of the present study was to subcategorize adenomyosis on the basis of MRI geography and to disclose the characteristics that are specific to each subtype of the disease.
Materials and Methods
We investigated 163 patients with adenomyosis who were treated surgically in the Department of Obstetrics and Gynecology, Takanohara Central Hospital, Nara, Japan, during the period from April 2007 through to August 2011. The surgery consisted of hysterectomy (n = 40) and adenomyomectomy (n = 123) that were performed laparoscopically in most cases (n = 155; 95.1%). In 11 of the 163 patients, MRI films were not available because they were taken at different institutions and were returned to the patients after the surgery. These 11 patients were eliminated, and the remaining 152 patients were the actual test subjects. Institutional Review Board approval was not required for this retrospective cohort study because none of the patients’ personal records was identifiable.
Adenomyosis was diagnosed as an ill-demarcated low-/mid-low intensity mass within the asymmetrically bulged uterine wall by T2-weighted MRI. It frequently contained small spots of high intensity, which represented hemorrhagic vesicles, and rough textures. At MRI diagnosis, we focused on which uterine hemisphere the adenomyosis resided in and on the geographic relationship between the adenomyosis and the uterine structural components (such as the endometrium, the junctional zone, the myometrium, and the serosa). The MRI diagnosis was established when the diagnosis by the corresponding author and 2 additional independent coauthors matched. In cases in which the diagnosis did not match, the other coauthors were invited to complete the diagnosis. Accordingly, we retrospectively categorized the patients into the following 4 discrete subtypes. Subtype I (intrinsic) adenomyosis group consisted of patients whose MRI revealed that the adenomyosis had developed in direct connection to the thickened junctional zone and that healthy muscular structures were preserved outside the adenomyosis ( Figure 1 ). Subtype II (extrinsic) adenomyosis group consisted of those patients whose MRI revealed that adenomyosis was located in the outer shell of the uterus, that the junctional zone was kept intact without aberrancy, and healthy muscular structures were preserved in between the adenomyosis and the junctional zone ( Figure 2 ). Subtype III (intramural) adenomyosis group was composed of those patients whose MRI showed adenomyosis residing solitarily without any geographic relationship to the junctional zone or the serosa ( Figure 3 ). Patients who had an MRI that did not meet any of the categorization criteria mentioned earlier (ie, the MRI geography was obscure and indeterminate) went to subtype IV (indeterminate) adenomyosis. Subtypes I-IV consisted of 59, 51, 22, and 20 patients, respectively.
The other MRI findings that were adopted for further statistical analysis were high-intensity spots within the adenomyosis on T1- and T2-weighted images, teardrop deformity of the rectum on the transverse sections ( Figure 4 ), and ovarian endometrioma. Surgical profiles were obtained by a review of all the operation DVDs. Among various surgical findings, anterior and posterior cul-de-sac endometriosis and posterior cul-de-sac obliteration were the variables that were incorporated in the statistical analysis. The histologic condition of the surgically excised specimens was included as 1 of the variables. The patient’s basic background aspects were also adopted.
We used stepwise logistic regression analysis (IBM Statistics software; version 10; SPSS Inc, Chicago, IL) with the variables listed in Table 1 to differentiate and specify the subtypes of adenomyosis. Subtype IV was not added to the statistical analysis because this subtype was not of clear definition but consisted of heterogeneous mixtures of advanced cases of subtypes I-III adenomyosis. We thought therefore that the inclusion of subtype IV variables might result in distorted conclusions.
| Variable | Subtype | |||
|---|---|---|---|---|
| I: intrinsic (n = 59) | II: extrinsic (n = 51) | III: intramural (n = 22) | IV: indeterminate (n = 20) | |
| Background aspects | ||||
| Age, y a , b | 38.7 ± 5.6 | 36.9 ± 6.1 | 34.3 ± 4.7 | 38.6 ± 4.0 |
| Gravidity, n (%) b | 34 (57.6) | 18 (35.3) | 7 (31.8) | 9 (45.0) |
| Parity, n (%) | 17 (28.8) | 15 (29.4) | 5 (22.7) | 7 (35.0) |
| Termination of pregnancy: curettage, n (%) c | 19 (32.2) | 4 (7.8) | 2 (9.1) | 1 (5.0) |
| Gynecologic surgery, n (%) | 9 (15.3) | 11 (21.6) | 3 (13.6) | 8 (40.0) |
| Menorrhagia, n (%) | 53 (89.8) | 44 (86.3) | 18 (81.8) | 20 (100.0) |
| Dysmenorrhea, n (%) | 56 (94.9) | 50 (98.0) | 18 (81.8) | 20 (100.0) |
| Dyspareunia, n (%) | 20 (33.9) | 19 (37.3) | 8 (36.4) | 10 (50.0) |
| Dyschezia, n (%) | 16 (27.1) | 21 (41.1) | 4 (18.2) | 6 (30.0) |
| Chronic pelvic pain, n (%) | 18 (30.5) | 19 (37.3) | 7 (31.8) | 8 (40.0) |
| Magnetic resonance imaging, n (%) | ||||
| Anterior wall involvement d | 34 (57.6) | 3 (5.9) | 13 (59.1) | 4 (20.0) |
| Posterior wall involvement d | 41 (69.5) | 49 (96.1) | 12 (54.5) | 20 (100.0) |
| Ovarian endometrioma d | 8 (13.6) | 34 (66.7) | 3 (5.9) | 14 (70.0) |
| T1/T2 high-intensity spots d | 54 (91.5) | 48 (94.1) | 10 (45.5) | 19 (95.0) |
| Teardrop deformity d | 5 (8.5) | 37 (72.5) | 3 (13.6) | 14 (70.0) |
| Surgery and histologic findings | ||||
| Hysterectomy/adenomyomectomy, n/n b | 20/39 | 9/42 | 1/21 | 6/14 |
| Anterior cul-de-sac endometriosis, n (%) | 2 (3.4) | 4 (7.8) | 4 (18.2) | 3 (15.0) |
| Posterior cul-de-sac endometriosis, n (%) d | 15 (25.4) | 47 (92.3) | 6 (27.3) | 17 (85.0) |
| Posterior cul-de-sac obliteration, n (%) d | 4 (6.8) | 49 (96.1) | 2 (9.1) | 13 (65.0) |
| Adenomyosis/adenomatoid, n/n d | 59/0 | 50/1 | 16/6 | 19/1 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
