Uterine MRI: A review of technique and diagnosis

Dr. Sydow is a Radiologist, Northside Radiology Associates, Atlanta, GA. At the time this article was written, Dr. Sydow was a radiology Resident at the Hospital of the University of Pennsylvania. Dr. Seigelman is an Associate Professor, Department of Radiology, Hospital of the University of Pennsylvania, Philadelphia, PA.

Magnetic resonance imaging (MRI) is a commonly used technique in the workup of obstetric and gynecologic abnormalities and in the pre- and postprocedural evaluation for uterine artery embolization. This article summarizes the MRI techniques for evaluating the uterus and describes the MRI findings of the most commonly encountered conditions involving the uterus.

MRI techniques

Optimal MRI of the female pelvis and uterus should be performed on a high field-strength MRI system that uses local phased-array coils. Theincreased signal-to-noise ratio provided by the surface coils allows for small field-of-view imaging that results in higher spatial resolution. The typical imaging protocol for the uterus uses both T1-weighted (T1W) and T2-weighted (T2W) images. Axial T1W images are obtained to evaluate the uterine contour, lymph nodes, and bone marrow. We perform T2W imaging in 3 orthogonal planes. One of these planes usually shows the endometrial complex along with the long and short axis of the uterus to good advantage. T2-weighted imaging depicts the zonal anatomy ofthe uterus. Fat-suppressed T1W imaging is used to differentiate between pelvic masses that contain fat and those that contain protein or hemorrhage. Contrast enhancement is used to document the extent of endometrial carcinoma invasion or to detect the presence of necrosis in uterine leiomyomas. Dynamic contrast injection can be used in women who are considering uterine artery embolization (UAE) in order to evaluate the uterine arteries and the potential collateral gonadal arterial supply. 1 In the evaluation of congenital uterine anomalies, sagittal T2W images should be acquired first to determine the long-axis orientation of the uterus. Subsequently, images should then be obtained parallel to the long axis of the uterus in order to show the outer uterine contour. 2

Normal uterine anatomy

T1-weighted images show poor contrast distinction between the endometrium and myometrium. The zonal anatomy of the uterus has a trilaminar appearance on T2W images (Figure 1). 3 The central endometrium has high signal intensity secondary to mucinous rich endometrial glands and stroma. The endometrium varies in thickness with the menstrual cycle and menopausal status. The endometrium may measure upto 14 mm during the secretory phase in menstruating women but is thinned during the follicular phase. 4 Postmenopausal women should have a homogeneous endometrium with a width of <5 mm. 5

The myometrium can be separated into the inner myometrium, also known as the junctional zone, and the outer myometrium (Figure 1). Thejunctional zone contains compact smooth muscle with a paucity of intercellular matrix and has relatively low T2 signal intensity. The normal junctional zone measures <12 mm. 6 The outer myometrium has less compact smooth muscle and increased intercellular matrix and vessels compared with the junctional zone, which results in a higher T2 signal intensity. This zonal anatomy is best visualized during the reproductive years and may be poorly depicted or absent in prepubertal and postmenopausal women. In women taking birth control pills, the junctional zone and endometrium are thinner than in other women and the outer myometrium is higher in signal intensity. 4

Uterine leiomyoma

Uterine leiomyomas are neoplasms derived from the smooth muscle cells of the myometrium. Uterine leiomyomas are common, occurring in >20% in women over 30 years of age. Most patients with uterine leiomyomas are asymptomatic. Among symptomatic patients, dysmenorrhea and irregular menstrual bleeding are the most common complaints. Leiomyomas may be intramural, subserosal, or submucosal in location. Intramural leiomyomas are centered in the uterine wall and are the most common subtype. A submucosal leiomyoma has a component in the endometrial canal. An intracavitary leiomyoma is almost entirely within the endometrial canal (Figure 2). Subserosal leiomyomas are centered external to the uterus (Figure 3). Nondegenerated uterine leiomyomas have lower T2 signal intensity than the outer myometrium; in addition, they are well circumscribed and can exhibit mass effect. On postcontrast imaging, nondegenerated uterine leiomyomas enhance homogeneously. Uterine leiomyomas can appear heterogeneous in signal intensity on T2W images and show decreased enhancement when they undergo degeneration. Some subsets of degeneration include hemorrhagic, fatty (lipoleiomyoma) (Figure 4), myxoid, hyaline, and cystic. 7 Leiomyosarcoma may have an appearance similar to a degenerated leiomyoma, but, fortunately, leiomyosarcoma is rare. It is difficult to establish a prospective diagnosis of leiomyosarcoma in the absence of metastatic disease. In our experience, degenerated atypical leiomyomas are much more common than leiomyosarcomas.

Subserosal leiomyomas can simulate fibrous or smooth-muscle ovarian masses, such as ovarian fibromas, fibrothecomas, or Brenner tumors. Subserosal leiomyomas can be differentiated from ovarian lesions since they show a normal ipsilateral ovary or a bridging vessel from the uterus to the fibroid. 8 Subserosal fibroids on a stalk <2 cm in width are considered a relative contraindication for UAE, as they may lose their connection to the adjacent uterus. 9 Submucosal fibroids can simulate endometrial masses such as endometrial polyps. Leiomyomas can be separated from polyps by evaluating their signal intensity on T2W images. Leiomyomas usually have low T2 signal intensity and have a stalk that originates within the myometrium, while polyps have heterogeneous T2 signal intensity. Submucosal and intracavitary fibroids are a common cause of infertility or miscarriages by creating an adverse environment for implantation or prohibiting sufficient blood fiow to support a developing embryo.

MRI can be used in the evaluation of patients who undergo UAE for treatment of symptomatic leiomyomas. Preoperatively, MRI can aid in selecting patients who may benefit from UAE. MRI documents the size, location, and vascularity of the fibroids. Leiomyomas that have high signal intensity on T1W images and no enhancement have a poor response to UAE. 10 Detection of a large intracavitary fibroid is a relative contraindication to UAE. Submucosal fibroids may be expelled from the uterus. 11 These women may benefit from hysteroscopic resection rather than UAE. 12 Leiomyomas that show high signal intensity on T2W imaging and homogeneous enhancement have the best responseto UAE. 13 Magnetic resonance angiography (MRA) can evaluate the uterine arteries and identify potential collateral arterial supply from thegonadal vessels that may need to be embolized as well. 1 MR can also identify other potential causes for a woman's signs or symptoms, such as adenomyosis or endometriosis.

MR findings of successful UAE include a decrease in size and enhancement of fibroids, preserved enhancement of the remainder ofthe uterus, and lack of visualization of the uterine arteries (Figure 5). 14 Treated fibroids have increased T1 signal and do not enhance, in keeping with hemorrhagic necrosis (Figure 5). Fibroids that continue to enhance are viable and usually grow in time and may result in recurrent symptoms. 15,16 The angiographic images should be carefully inspected for a possible cause of failure such as uterine artery revascularization or collateral arterial supply from the ovarian arteries (Figure 6).

Uterine adenomyosis

Adenomyosis is the presence of ectopic endometrial tissue in the uterine myometrium with smooth muscle hyperplasia. Adenomyosis may be focal or diffuse, with the latter being more common. On T2W images, adenomyosis appears as a thickening of the low-signal-intensity junctional zone corresponding to the smooth muscle hyperplasia (Figure 7). 17 Thickening of the junctional zone >12 mm is specific for the diagnosis, while a measurement of 8 to 12 mm is indeterminate and a measurement <8 mm excludes the diagnosis with high specificity. 18 One- to 4 mm hyperintense T2 foci within the junctional zone represent the ectopic foci of endometrial tissue and, when present, add specificity to the MR diagnosis (Figure 8). 19 MR can usually differentiate between focal adenomyosis and a leiomyoma. 20 Adenomyosis has poorly defined margins,is often oriented parallel to the endometrial stripe, contains 1- to 5-mm foci of ectopic endometrial glands, and has minimal mass effect on the endometrial canal. Leiomyomas have well-defined margins and mass effect. Adenomyosis can be treated with uterine artery embolization with the expectation that >50% of women will have significant decrease in symptoms after 2 years. 21 Successfully treated adenomyosis shows a decrease in the width of the junctional zone on MRI. 22

Endometrial carcinoma

Endometrial carcinoma is the most common gynecologic malignancy. Roughly 75% of endometrial carcinomas occur in postmenopausal women. Excessive estrogen stimulation is the most recognized association with endometrial cancer. Women taking tamoxifen are also at increased risk of endometrial cancer. The most common symptom of endometrial cancer is postmenopausal bleeding. The initial procedure of choice in the evaluation of a woman with suspected endometrial carcinoma is transvaginal sonography or hysteroscopy with biopsy. MR is not used to screen women for endometrial carcinoma. However, in women with biopsy-proven endometrial carcinoma, MRI can be used to determine the depth of myometrial invasion. 23 On dynamic contrast-enhanced MRI, endometrial carcinoma enhances less than the subjacent myometrium. 23 Malignant adenopathy is associated with muscle-invasive endometrial carcinomas. Some surgeons use the information provided by preoperative MRI to determine the need to perform lymphadenectomy. As opposed to evaluating secondary findings of node-positive disease (muscle invasion), it would be ideal to develop MRI techniques to both detect and characterize lymph nodes as benign or malignant in order to guide therapy. 24

When abnormal signal within the endometrial canal invades the junctional zone on T2W or contrast-enhanced images, then endometrial carcinoma is the likely diagnosis. The differential diagnosis of a thickened endometrial stripe without myometrial extension is superficial endometrial carcinoma, endometrial polyps, and endometrial hyperplasia. Polyps may depict a zonal phenomenon with a low T2-signal-intensity fibrous core and outer high-signal-intensity glands. Endometrial hyperplasia often has a “Swiss cheese” configuration on T2W and enhanced imaging, with internal nonenhancing cysts surrounded by enhancing glandular tissue.

Tamoxifen

Tamoxifen is used in the treatment of breast cancer. While tamoxifen exhibits antiestrogenic effects within the breast, it can also show proestrogenic activity within the endometrium. Thus, tamoxifen can produce endometrial thickening on MRI (Figure 8). Pathologically, the endometrial abnormalities associated with tamoxifen use include endometrial atrophy, hyperplasia, polypoid glandular-cystic proliferation, polyps, and endometrial cancer. 25 Endometrial sampling is often needed in these women, especially those with abnormal uterine bleeding.

Mullerian duct anomalies

The incidence of mullerian duct anomalies is approximately 1% but varies by population. In patients with infertility, it is estimated to be 3%. Most mullerian duct anomalies are thought to be sporadic or multifactorial in nature. The most widely quoted classification scheme formullerian duct anomalies is the American Fertility Society classification that has 7 classes (Table 1). 26 A simplified framework for the discussion of mullerian duct defects consists of 3 categories: duct agenesis and hypoplasia, defects of vertical fusion of the ducts with the ascending urogenital sinus, and defects of lateral duct fusion. 27

Patients with mullerian duct anomalies may manifest with primary amenorrhea in the adolescent age group. MRI can evaluate for the presence or absence of the vagina, cervix, or uterus and also depicts any associated renal anomalies (Figure 9). MRI can show the level of obstruction of antegrade menstrual fiow in women with functioning endometrium as well as depict findings of retrograde menstruation such as hematosalpinx or endometriosis. In the reproductive years, women with uterine anomalies may present with spontaneous abortion, premature delivery, and abnormal fetal lie.

Absence of the mullerian ducts (agenesis and hypoplasia)

Mayer-Rokitansky-Kuster-Hauser syndrome represents failure of vaginal development with varying degrees of cervical and uterine agenesis/hypoplasia. In 90% of cases, there is complete uterine agenesis. In the remaining 10% of cases, there may be an obstructed or small rudimentary uterus. The ovaries are usually normal. Mayer-Rokitansky-Kuster-Hauser syndrome is the most common cause of amenorrhea in women with breast development, refiecting the normal ovarian development with this condition. A rudimentary uterus may contain functioning endometrial tissue. MRI can be used to detect associated abnormalities such as endometriosis, renal anomalies, and collecting system abnormalities (Figure 9). 28 The differential diagnosis of primary amenorrhea with normal breast development and an absent uterus includes androgen insensitivity syndrome; affected individuals have rudimentary testes and absent ovaries on MR.

Unicornuate uterus

Unicornuate uterus is a type of uterine hypoplasia in which there is incomplete or absent development of 1 mullerian duct and represents 20% of mullerian duct anomalies. The abnormal uterine horn can be completely absent (one third of cases) or rudimentary (two thirds of cases). Of those patients with a rudimentary horn, 50% do not contain an endometrial canal and 50% have an endometrial-lined cavity. The endometrial lined cavity within a rudimentary horn communicates with the contralateral horn in approximately one third of women. 27 A unicornuate uterus is usually an incidental finding unless a cavitary rudimentary horn is present. These women can present with nonviable pregnancy, endometriosis, or ectopic pregnancy. Therefore, rudimentary horns that contain endometrium, whether they are communicating or noncommunicating, are usually removed. Among all of the mullerian duct anomalies, unicornuate uteri have the highest association of renal abnormalities, present in up to 40% of these women. The renal anomaly is ipsilateral to the abnormal horn. MRI shows a banana-shapeduterus. 29 When the endometrium is absent, the horn is of low T2 signal intensity, with loss of normal zonal anatomy. When the endometrium is present, zonal anatomy may be preserved.

Disorders of vertical fusion (vaginal septum)

A transverse vaginal septum consists of a band of fibrous connective tissue with vascular and muscular components. It is the most commondisorder of vertical fusion. A transverse vaginal septum can occur anywhere along the vagina, but most commonly occurs at the junction of the upper and middle third. Patients typically present with primary amenorrhea. In patients with a uterus and functioning endometrium, hematocolpos can be depicted with MR with less severe dilation of the endometrial cavity. 27 This finding is secondary to the decreased compliance of the more muscular myometrium. A transverse vaginal septum can be present in conjunction with other mullerian duct anomalies. The most common association is a uterus didelphys and complex duplication anomalies. The differential diagnosis for primary amenorrhea with a present uterus includes an imperforate hymen, which can also cause hematocolpos. Imperforate hymen is not a mullerian anomaly, and distinction from a transverse vaginal septum can be difficult although the treatment is similar.

Disorders of lateral fusion

Disorders of lateral fusions result in varying degrees of duplication of the uterus and cervix. Possibilities include didelphys, septate uterus, and bicornuate uterus. With a septate uterus (Figure 10), a fibromuscular central septum is incompletely resorbed after mullerian duct fusion. The septate uterus represents approximately 55% of mullerian duct anomalies. Septate uteri may result in first trimester miscarriages caused by implantation on the septum. Reproductive outcome improves after hysteroscopic resection of the septum. In uterine didelphys (Figure 11), there is complete separation of distinct left and right uteri and accounts for 5% of mullerian duct anomalies. The cervix is usually duplicated as well. Women with uterine didelphys have normal or nearly normal fertility. The bicornuate uterus is caused by incomplete fusion of the uterovaginal horns at the level of the fundus and represents approximately 10% of mullerian duct anomalies.

Differentiating between the lateral fusion anomalies has important implications for treatment decisions. For instance, a septate uterus isoften treated with hysteroscopic resection of the septum, while women with bicornuate uteri are treated with a transabdominal metroplasty if they have a history of repeated pregnancy loss. The key in imaging lateral fusion anomalies is the evaluation of the external uterinefundal contour. The best imaging plane for evaluation of the fundal contour is one that passes through the long axis of the uterus. In a normal or septate uterus, the outer contour is convex, fiat, or <10 mm ofconcavity. The myometrial fundal indentation is smooth and broad, and the signal intensity of this region is isointense to normal myometrium. The outer contour concavity of a bicornuate uterus or uterine didelphys, to the contrary, should be >10 mm. Two measurements that may be helpful include the intercornual distance and theintercornual angle. In a septate uterus, the intercornual distance is <4 cm and the intercornual angle is <60°, as compared with the bicornuate uterus, in which these values are >4 cm and >60°, respectively. 30,31 Lateral fusion anomalies may also result in vaginal duplication. Vertical vaginal septa are present in 75% of women with didelphys, 25% with bicornuate, and 5% with septate uteri.

Conclusion

Because of its superb soft tissue contrast and direct multiplanar capabilities, MRI can detect and characterize normal uterine anatomy and focal and diffuse uterine conditions. In patients with primary amenorrhea, MRI can be used to confirm the absence or presence of the uterus as well as evaluate for any associated urogenital abnormalities such as renal agenesis. In patients with infertility, MRI can confirmthe presence and extent of a septate uterus and define the fibrous and muscular components. In patients with pelvic pain, MRI is more sensitive and specific than ultrasound for the findings of adenomyosis. In patients with known leiomyomas, MRI is used in the preoperative evaluation for uterine artery embolization. In oncology, MRI can be used to stage women with known endometrial carcinoma. Understanding the appearance of the common pathologic entities involving the uterus along with the strengths and limitations of MRI will help radiologists be more confident in their approach to pelvic MR interpretation.

REFERENCES

  1. Kroencke TJ, Scheurig C, Kluner C, et al. Uterine fibroids: Contrast-enhanced MR angiography to predict ovarian artery supply initial experience. Radiology . 2006;241:181-189.
  2. Pellerito JS, McCarthy SM, Doyle MB, et al. Diagnosis of uterine anomalies: Relative accuracy of MR imaging, endovaginal sonography, and hysterosalpingography. Radiology. 1992;183:795-800.
  3. Togashi K, Nakai A, Sugimura K. Anatomy and physiology of the female pelvis: MR imaging revisited. J Magn Reson Imaging. 2001;13:842-849.
  4. Chaudhry S, Reinhold C, Guermazi A, et al. Benign and malignant diseases of the endometrium. Top Magn Reson Imaging . 2003;14:339-357.
  5. Imaoka I, Sugimura K, Masui T, et al. Abnormal uterine cavity: Differential diagnosis with MR imaging. Magn Reson Imaging. 1999;17:1445-1455.
  6. 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.
  7. Murase E, Siegelman ES, Outwater EK, et al. Uterine leiomyomas: Histopathologic features, MR imaging findings, differential diagnosis, and treatment. RadioGraphics .1999;19:1179-1197.
  8. Kim JC, Kim SS, Park JY. "Bridging vascular sign" in the MR diagnosis of exophytic uterine leiomyoma. J Comput Assist Tomogr . 2000;24:57-60.
  9. Katsumori T,Akazawa K, Mihara T. Uterine artery embolization for pedunculated subserosal fibroids. AJR Am J Roentgenol . 2005;184:399-402.
  10. Burn PR, McCall JM, Chinn RJ, et al. Uterine fibroleiomyoma: MR imaging appearances before and after embolization of uterine arteries. Radiology . 2000;214:729-734.
  11. Abbara S, Spies JB, Scialli AR, et al. Transcervical expulsion of a fibroid as a result of uterine artery embolization for leiomyomata. J Vasc Interv Radiol. 1999;10:409-411. Commentin: J Vasc Interv Radiol . 1999;10:1135.
  12. Hovsepian DM, Siskin GP, Bonn J, et al. Quality improvement guidelines for uterine artery embolization for symptomatic leiomyomata. J Vasc Interv Radiol . 2004;15:535-541.
  13. deSouza NM, Williams AD. Uterine arterial embolization for leiomyomas: Perfusion and volume changes at MR imaging and relation to clinical outcome. Radiology .2002;222:367-374.
  14. Kitamura Y,Ascher S, Cooper C, et al. Imaging manifestations of complications associated with uterine artery embolization. RadioGraphics . 2005; 25(suppl 1):S119-S132.
  15. Pelage JP, Guaou NG, Jha RC, et al. Uterine fibroid tumors: Long-term MR imaging outcome after embolization. Radiology. 2004;230:803-809.
  16. Marret H, Alonso AM, Cottier JP, et al. Leiomyoma recurrence after uterine artery embolization. J Vasc Interv Radiol . 2003;14:1395-1399.
  17. Outwater EK, Siegelman ES, Van Deerlin V. Ad enomyosis: Current concepts and imaging considerations. AJR Am J Roentgenol . 1998;170:437-441.
  18. Reinhold C, Tafazoli F, Mehio A, et al. Uterine adenomyosis: Endovaginal US and MR imaging features with histopathologic correlation. Radio-Graphics . 1999;19 SpecNo:S147-S160.
  19. Togashi K, Nishimura K, Itoh K, et al. Adenomyosis: Diagnosis with MR imaging. Radiology . 1988;166 (1 Pt 1):111-114.
  20. Togashi K, Ozasa H, Konishi I, et al. Enlarged uterus: Differentiation between adenomyosis and leiomyoma with MR imaging. Radiology . 1989; 171:531-534.
  21. Pelage JP, Jacob D, Fazel A, et al. Midterm results of uterine artery embolization for symptomatic adenomyosis: Initial experience. Radiology . 2005;234:948-953. Commentin: Radiology . 2005; 236:1111-1112; author reply 1112.
  22. Kim MD, Kim S, Kim NK, et al. Long-term results of uterine artery embolization for symptomatic adenomyosis. AJR Am J Roentgenology . 2007;188:176-181.
  23. Manfredi R, Gui B, Maresca G, et al. Endometrial cancer: Magnetic resonance imaging. Abdom Imaging . 2005;30:626-636.
  24. Harisinghani MG, Barentsz J, Hahn PF, et al. Noninvasive detection of clinically occult lymph-node metastases in prostate cancer. N Engl J Med . 2003;348:2491-2499. Erratumin: N Engl J Med . 2003;349:1010.
  25. Ascher SM, Johnson JC, Barnes WA, et al. MR imaging appearance of the uterus in postmenopausal women receiving tamoxifen therapy for breast cancer: Histologic correlation. Radiology . 1996;200:105-110.
  26. The American Fertility Society classifications of adnexal adhesions, distal tubal obstruction, tubal occlusion secondary to tubal ligation, tubal pregnancies, mullerian anomalies and intrauterine adhesions. Fertil Steril. 1988;49:944-955. Comment in: Fertil Steril . 1989;51:199-201.
  27. Troiano RN, McCarthy SM. Mullerian duct anomalies: Imaging and clinical issues. Radiology . 2004;233:19-34.
  28. Fedele L, Dorta M, Brioschi D, et al. Magnetic resonance imaging in Mayer-Rokitansky-Kuster-Hauser syndrome. Obstet Gynecol . 1990;76:593-596.
  29. Fedele L, Bianchi S, Agnoli B, et al. Urinary tract anomalies associated with unicornuate uterus. J Urol . 1996;155:847-848.
  30. Carrington BM, Hricak H, Nuruddin RN, et al. Mullerian duct anomalies: MR imaging evaluation. Radiology . 1990;176:715-720.
  31. Fedele L, Dorta M, Brioschi D, et al. Magnetic resonance evaluation of double uteri. Obstet Gynecol . 1989;74:844-847.
© Anderson Publishing, Ltd. 2024 All rights reserved. Reproduction in whole or part without express written permission Is strictly prohibited.