Uterine fibroids (leiomyomas): Differentiating fibroids from uterine sarcomas

INTRODUCTION — Benign uterine fibroids (leiomyomas) are the most common pelvic neoplasm in females (estimated lifetime risk of 70 percent in White females and 80 percent in Black females) [1-3]. Uterine sarcomas are significantly rarer than leiomyomas and have a poor prognosis. Both present as focal masses in the uterine myometrium. Thus, clinicians who evaluate and manage patients with presumed leiomyomas are presented with a clinical challenge: how to assess if one of the many patients diagnosed with this common condition actually has a rare type of malignancy while avoiding unnecessary hysterectomy or laparotomy and the attendant morbidity of unnecessary surgery for the vast majority of patients with benign uterine masses. While there are differences in the patient populations for the two disorders, there is substantial overlap. (See "Uterine tissue extraction by morcellation: Techniques and clinical issues".)

There are several histologic types of uterine sarcoma. The main type of sarcoma that may resemble a leiomyoma is leiomyosarcoma, which typically presents as a myometrial mass. Endometrial stromal sarcoma may present as an endometrial or myometrial mass.

Historically, hysterectomy was the mainstay of treatment for patients with any uterine mass. Thus, if uterine sarcoma was present and identified postoperatively on pathologic analysis, there was no missed diagnosis, and the treatment received was incomplete but did not worsen prognosis. In current practice, patients with presumed benign leiomyomas may be treated with a variety of uterine-sparing alternatives to hysterectomy, including medication, imaging-based interventions (ie, uterine artery embolization or magnetic resonance-guided focused ultrasound), and conservative surgery (myomectomy, radiofrequency fibroid ablation, or endometrial ablation). Nonexcisional therapies for fibroid-related symptoms (eg, medical treatments, endometrial ablation, interventional radiology procedures) do not produce specimens for pathologic examination. Thus, the diagnosis of malignant neoplasm may be missed. In addition, procedures or techniques that disrupt a uterine mass (ie, myomectomy or hysterectomy, especially when the specimen is morcellated) may disseminate the malignancy, often upstaging the disease. Equally important priorities, however, are to avoid unnecessary surgery for the purpose of excluding the rare sarcoma and also to offer the option of minimally invasive approaches in patients at low risk for sarcoma, many of whom have not completed their childbearing or strongly desire uterine preservation.

Approaches to differentiating benign uterine leiomyomas from uterine sarcoma are reviewed here. General principles of the diagnosis of uterine leiomyomas, the diagnosis and treatment of uterine sarcoma, and benign and atypical smooth muscle neoplasm variants are discussed separately. (See "Uterine fibroids (leiomyomas): Epidemiology, clinical features, diagnosis, and natural history" and "Uterine fibroids (leiomyomas): Variants and smooth muscle tumors of uncertain malignant potential" and "Uterine sarcoma: Classification, epidemiology, clinical manifestations, and diagnosis".)

DIFFERENTIAL DIAGNOSIS OF A UTERINE MASS — The challenge of differentiating a leiomyoma from a uterine sarcoma is one part of the diagnostic process for all patients with a uterine mass. The differential diagnosis of an enlarged uterus includes non-neoplastic conditions and benign and malignant neoplasms:

●Pregnancy

●Leiomyoma

●Uterine adenomyoma or diffuse adenomyosis

●Uterine sarcoma

●Leiomyoma variant

●Uterine carcinosarcoma (considered an epithelial neoplasm)

●Endometrial carcinoma

●Metastatic neoplasm (typically from another reproductive tract primary)

●Hematometra

Endometrial processes that are included in the differential diagnosis, but often do not cause uterine enlargement, include endometrial polyp and endometrial hyperplasia.

The differential diagnosis of an enlarged uterus is discussed in more detail separately. (See "Uterine fibroids (leiomyomas): Epidemiology, clinical features, diagnosis, and natural history", section on 'Differential diagnosis'.)

PREVALENCE OF SARCOMA IN PATIENTS UNDERGOING SURGERY FOR BENIGN INDICATIONS

Presumed benign leiomyomas — While leiomyomas are common, uterine sarcoma is rare (3 to 7/100,000 in the United States population) [4]. The rate of sarcoma in patients with a symptomatic uterine mass (eg, abnormal uterine bleeding, bulk symptoms related to lesion size, infertility, recurrent pregnancy loss) is higher than patients with an asymptomatic uterine mass, though the true prevalence in symptomatic patients is unknown.

The prevalence of sarcoma in patients planning surgery for presumed benign leiomyomas is an active area of investigation. In studies and systematic reviews of patients undergoing hysterectomy or myomectomy for a myometrial mass, the prevalence of sarcoma is approximately 0.20 percent (1 in 500) in most studies or reviews, and estimates of the prevalence range from 0.05 (1 in 2000) to 0.28 percent (1 in 352) depending upon which studies are included or excluded [5-22]. The majority of cases are leiomyosarcoma, a highly aggressive cancer, but there are also reports of cases of endometrial stromal sarcoma, generally a more indolent cancer. Representative studies include:

●A systematic review by the US Food and Drug Administration (FDA) included 23 studies published from 2015 to 2017. The rate of unsuspected uterine sarcoma (any histologic type) in patients undergoing surgery for presumed benign fibroids ranged from 0 to 1.48 percent, and for leiomyosarcoma from 0 to 0.51 percent. Meta-analysis showed the following prevalences: any uterine sarcoma (0.277 percent, 95% CI 0.172-0.445, 1 in 361) and leiomyosarcoma (0.134 percent, 95% CI 0.089-0.202, 1 in 746) [23]. The risk increased with age, and for any uterine sarcoma, approximate rates were the following: 40 to 50 years (0.15 to 0.22 percent), 50 to 60 years (0.58 percent), and 60 years or older (1.5 to 2.8 percent).

●A meta-analysis including 64 prospective studies (5223 patients) and 70 retrospective studies (24,970 patients) of patients who underwent surgery for presumed benign leiomyomas [5]. The overall rate of leiomyosarcoma was 0.05 (1 in 2000) surgeries; in prospective studies, it was 0.01 percent (1 in 8300) surgeries, and in retrospective studies, it was 0.06 percent (1 in 1700) surgeries. There are several possible explanations for the significantly lower rate in the prospective studies. There may be less selection bias than in retrospective studies. Also, in the prospective study group, more than one-half of the studies were of patients undergoing myomectomy, suggesting these patients may be younger and/or have less clinical suspicion of malignant neoplasm.

●The largest study that evaluated unexpected malignant neoplasm of all types in patients who underwent morcellation was a report from a United States insurance database, including 232,882 patients who underwent minimally invasive (laparoscopic or robotic [24]) hysterectomy, which found that morcellation was performed in 36,470 (15.7 percent) [25]. Among those who had morcellation, there were 99 cases of uterine cancer including sarcomas and endometrial carcinomas (0.27 percent, 1 in 370), 39 uterine neoplasms of uncertain malignant potential, 368 cases of endometrial hyperplasia (most cases were without atypia (non-neoplastic) and thus not thought to have a premalignant phenotype [24]), and 26 other gynecologic malignant neoplasms. The prevalence of malignant neoplasm was 0.34 percent and of all neoplasms was 1.5 percent, although the clinical significance of endometrial hyperplasia without atypia is uncertain. Presence of malignant neoplasm was associated with increasing age; compared with patients age 40 years, the prevalence ratios were: 50 to 54 years, 4.97; 55 to 59, 19.37; 60 to 64, 21.36; and ≥65, 35.97. In absolute numbers, the prevalence of cancer in this series for patients under age 40 was 1/1500 and for patients 40 to 44 was 1/1100. A limitation of the study was that the histology of the uterine malignant neoplasm cases was not reported. These data add support to limiting morcellation to premenopausal patients since 50 years old is approximately the average age of menopause.

●Regarding myomectomy, data from a national United States hospital database of 41,777 patients undergoing myomectomy, including 3220 who underwent myomectomy with electromechanical morcellation, reported that 0.43 percent had a pathologic finding (malignant neoplasm, neoplasms of uncertain malignant potential, or endometrial hyperplasia) and 0.09 percent had uterine cancer [26]. Among all myomectomy patients, increasing age was significantly associated with an increased risk of uterine cancer. The risk of occult uterine cancer stratified by age was <40 years 1 in 2337, 40 to 49 years 1 in 702, 50 to 59 years 1 in 154, and ≥60 years 1 in 31. These data support the US FDA guidance that electromechanical morcellation should not be used in peri- and postmenopausal patients with a uterine tumor.

●In a single-center retrospective study of 4232 patients undergoing hysterectomy, 16 patients (0.38 percent) were diagnosed with uterine sarcoma [27]. The risk of unexpected uterine sarcoma increased with increasing age and uterine size (0.03 versus 15.4 percent for a uterine weight of <250 and ≥2000 grams, respectively).

Pelvic organ prolapse — Uterine sarcoma or other uterine cancer may be discovered in patients after hysterectomy for pelvic organ prolapse. Compared with patients undergoing surgery for leiomyomas, patients undergoing surgery for prolapse are more likely to be perimenopausal or postmenopausal. This is a challenging issue because postmenopausal patients with prolapse do not usually have abnormal uterine bleeding and thus may not undergo endometrial sampling and/or pelvic imaging [28]. However, these patients may be less likely to have morcellation since the treatment of choice for most patients would be a vaginal hysterectomy due to the prolapse and need for concomitant vaginal procedures to repair the prolapse. Additionally, these uteri tend to be small and would be less likely to require scalpel morcellation for removal. In a large study of this issue, among 1196 patients who underwent hysterectomy for prolapse, malignant neoplasm was found postoperatively in 0.3 percent and endometrial hyperplasia in 0.6 percent [29]. There were no specimens with sarcoma. Another study included 640 patients who underwent hysterectomy for prolapse; three cases of malignant neoplasm were identified, and one case was sarcoma [30].

CLINICAL FEATURES — Assessing clinical characteristics is the initial approach to evaluating a patient with a presumed leiomyoma in whom there is some concern of uterine sarcoma. Suspicion of sarcoma may change based upon a patient's risk factors or response to treatment. Unfortunately, the clinical features of benign leiomyomas and uterine sarcomas are often indistinguishable.

Risk factors — Uterine sarcoma is rare and risk factors are not well defined for uterine sarcomas in general or specifically for leiomyosarcoma. The known risk factors are summarized below and discussed in detail separately. (See "Uterine sarcoma: Classification, epidemiology, clinical manifestations, and diagnosis", section on 'Epidemiology'.)

Common to both leiomyomas and sarcoma

Race — Black race is a risk factor for both uterine leiomyomas and uterine sarcoma. The relative risk and incidence of leiomyomas is two- to threefold greater in Black females than in White females. The cumulative incidence of leiomyomas of any size, including very small tumors, by age 50 years is >80 percent for Black females and almost 70 percent for White females [1]. (See "Uterine fibroids (leiomyomas): Epidemiology, clinical features, diagnosis, and natural history", section on 'Prevalence'.)

Similarly, Black females have an approximately twofold higher incidence of leiomyosarcomas (but not other types of uterine sarcoma) and carcinosarcoma than White females [4,31].

Sarcoma only

Increasing age and postmenopausal status — Benign leiomyomas are responsive to gonadal steroids (estrogen and progesterone) and develop and grow primarily in patients of reproductive age, with symptoms developing on average as early as the 20s in Black patients and typically in the 30s or 40s in White patients. Leiomyomas typically stabilize or diminish in size following menopause. Postmenopausal estrogen therapy may be associated with modest growth of myomas and/or persistent symptoms but does not appear to induce the development of new leiomyomas. (See "Uterine fibroids (leiomyomas): Epidemiology, clinical features, diagnosis, and natural history", section on 'Prevalence' and "Uterine fibroids (leiomyomas): Epidemiology, clinical features, diagnosis, and natural history", section on 'Patients on hormone therapy'.)

By contrast, increasing age is a significant risk factor for uterine sarcomas. The average age at diagnosis is 60 years; thus, the majority of sarcomas occur after menopause. Young age does not exclude the diagnosis of uterine sarcoma, however, since it has been reported in patients as young as their mid-20s [32]. As an example, in a study of uterine sarcoma cases from 2000 to 2012 in a United States cancer database, the incidence was more than fourfold higher in patients ≥50 years old compared with younger patients (6.4 versus 1.5 per 100,000) [31]. (See "Uterine sarcoma: Classification, epidemiology, clinical manifestations, and diagnosis", section on 'Epidemiology'.)

Thus, for postmenopausal patients, a new or growing uterine mass warrants further evaluation for uterine sarcoma. The level of suspicion may be lower in patients who are on postmenopausal estrogen therapy and have a small increase in the size of a fibroid known to present prior to menopause. In this subgroup, a trial of discontinuing postmenopausal estrogen therapy is an option to see if regression occurs; obtaining imaging (eg, ultrasound, magnetic resonance imaging) and endometrial biopsy may also provide reassurance. (See 'Diagnostic methods' below and "Uterine fibroids (leiomyomas): Epidemiology, clinical features, diagnosis, and natural history", section on 'Patients on hormone therapy'.)

The available evidence supports an increased risk of undetected malignant neoplasm in patients 50 years or older who undergo morcellation of uterine tissue [25]. (See 'Presumed benign leiomyomas' above.)

Tamoxifen — Long-term use of tamoxifen (five years or more) is associated with an increase in risk of developing uterine sarcoma. The absolute risk remains small (approximately 17/100,000 females) [33]. In general, sarcomas present two to five years following the start of tamoxifen therapy and are often at an advanced stage at diagnosis [34,35].

Other risk factors — Other risk factors that are associated with uterine sarcoma include pelvic irradiation, a history of childhood retinoblastoma, and hereditary leiomyomatosis and renal cell carcinoma (HLRCC) syndrome. (See "Uterine sarcoma: Classification, epidemiology, clinical manifestations, and diagnosis", section on 'Pelvic radiation' and "Uterine sarcoma: Classification, epidemiology, clinical manifestations, and diagnosis", section on 'Hereditary conditions'.)

Data are inconclusive regarding parity and time of menarche and menopause as risk factors for sarcoma. For benign leiomyomas, early menarche (<10 years old) is a risk factor (particularly in Black patients), and parity decreases the chances of fibroid formation. (See "Uterine fibroids (leiomyomas): Epidemiology, clinical features, diagnosis, and natural history", section on 'Prevalence'.)

Signs and symptoms — Abnormal uterine bleeding, pelvic pain/pressure, and a pelvic mass are the primary presenting symptoms and signs for both leiomyomas and sarcoma, making it difficult to differentiate between the two on this basis [22,32,36-38]. Some patients with sarcoma present with a foul-smelling vaginal discharge, but this is not a reliable indicator of malignant neoplasm since vaginal discharge is a common gynecologic symptom. (See "Uterine fibroids (leiomyomas): Epidemiology, clinical features, diagnosis, and natural history", section on 'Clinical features' and "Uterine sarcoma: Classification, epidemiology, clinical manifestations, and diagnosis", section on 'Clinical presentation'.)

Clinical manifestations associated with metastatic neoplasm may be found in patients with sarcoma but not with ordinary leiomyomas. Uterine sarcomas spread via intraabdominal, lymphatic, or hematogenous routes. Hematogenous spread is most often to the lungs. Of note, some histologically benign leiomyoma variants also disseminate and must be excluded if widespread disease is found. (See "Uterine fibroids (leiomyomas): Variants and smooth muscle tumors of uncertain malignant potential", section on 'Leiomyomas with disseminated, intravascular, or metastatic growth patterns' and "Treatment and prognosis of uterine leiomyosarcoma" and "Endometrial stromal sarcomas, related tumors, and uterine adenosarcoma", section on 'Evaluation'.)

DIAGNOSTIC METHODS — Uterine sarcoma is a histologic diagnosis based upon pathology evaluation after resection of uterine tissue (myomectomy or hysterectomy). Uterine sarcomas are most commonly diagnosed following surgery for presumed leiomyomas. (See 'Presumed benign leiomyomas' above.)

Preoperative and intraoperative findings are of limited value in estimating the likelihood that a mass is a uterine sarcoma and are more useful in predicting typical leiomyomas. Magnetic resonance imaging (MRI) and endometrial sampling are the most potentially useful techniques.

Preoperative evaluation

Pelvic examination — A thorough pelvic examination should be performed in all patients with a pelvic mass. The size, contour, and mobility of the uterus should be noted, along with any other findings (eg, adnexal mass, cervical mass, or vaginal nodules). These findings are helpful to follow changes in the uterus over time and/or to aid surgical planning. A fixed mass is more suggestive of a malignant neoplasm than a mobile mass. However, this is not pathognomonic, since a malignant neoplasm that has not invaded the uterine serosa may be mobile, and a mass associated with endometriosis or pelvic infection may be fixed. Unfortunately, there are no examination findings that can distinguish a leiomyoma from a uterine sarcoma.

Screening and evaluation for gynecologic malignancies — Prior to treatment for presumed leiomyomas, patients should undergo routine screening for cervical neoplasm. (See "Screening for cervical cancer in resource-rich settings".)

Evaluation of abnormal uterine bleeding is of critical importance to exclude endometrial hyperplasia or carcinoma and can be useful in diagnosing sarcomas. A full menstrual history and a history of risk factors for endometrial carcinoma (including Lynch syndrome) should be elicited. Pelvic imaging or endometrial sampling (table 1) should be performed if appropriate. (See "Abnormal uterine bleeding in nonpregnant reproductive-age patients: Terminology, evaluation, and approach to diagnosis" and "Approach to the patient with postmenopausal uterine bleeding".)

Imaging

Choice of imaging modality — Leiomyomas and uterine sarcomas often appear similar on imaging; both are focal masses within the uterus, and both can have central necrosis. Therefore, there is no pelvic imaging modality that can reliably diagnose uterine sarcomas, and because of the low prevalence of sarcomas, it is unlikely that any single test will have a high positive predictive value for sarcoma.

Despite these limitations, we suggest that patients with a newly diagnosed or markedly enlarging pelvic mass undergo a pelvic ultrasound as an initial imaging study. Sonographic evaluation of a uterine mass may identify features suggestive of sarcoma including mixed echogenic and poor echogenic parts, central necrosis, irregular tumor borders, color Doppler findings of irregular vessel distribution, low impedance to flow, and high peak systolic velocity [39,40]. However, many of these characteristics may also be found in benign leiomyomas.

If uterine sarcoma is suspected based on sonographic findings, MRI with gadolinium contrast may aid in assessing the likelihood of malignancy. An MRI that shows a typical fibroid (dark and homogenous in T2-weighted images) has a high negative predictive value [41]. High signal intensity is not a reliable indicator of uterine sarcoma [39]. A consistent finding in leiomyosarcomas is the absence of calcifications [42]. Calcifications suggest that the mass has undergone necrosis. This is most typically seen in fibroids that have outgrown their blood supply (for example, postpartum or after uterine artery embolization) or in postmenopausal patients as blood flow to the uterus regresses. Some data suggest that ill-defined margins are consistent with a sarcoma [43]. Two small studies using different techniques of MRI with gadolinium contrast have reported specificities of 93 to 100 percent and positive predictive values of 53 to 100 percent [44,45]. Diffusion weighted MRI also appears to differentiate ordinary and degenerated leiomyomas from sarcomas and cellular leiomyomas [46]. Finally, the presence of intralesional hemorrhage appears to be suggestive of sarcoma [47,48]. Further study of use of MRI for this purpose is needed.

Neither computed tomography (CT) nor positron emission tomography/CT with fluorodeoxyglucose (FDG) can reliably differentiate between leiomyomas and uterine sarcomas [49,50]. While the FDG uptake is generally high in sarcomas and low in leiomyomas, the uptake varies across individual tumors.

Findings that do NOT reliably predict sarcoma

Rapidly growing uterine mass in premenopausal patients — Historically, the common teaching had been that rapid growth of a uterine mass (eg, increasing by six weeks' gestational size within one year [7]) is a sign of a potential uterine sarcoma. However, it appears that most patients with a rapidly enlarging uterus or uterine mass do not have a sarcoma.

With modern pelvic imaging, there is an increased appreciation that, in patients of reproductive age, leiomyomas both grow and shrink and do so at differing rates within the same patient and even within the same uterus [51-54]. Research on normal leiomyomas shows that growth of up to 138 percent can occur within six months [53]. As an example, in a prospective study, 101 fibroids in 36 patients were evaluated with MRI at three-month intervals for one year [51]. Increase in volume of ≥30 percent in a three-month period was found in 37 myomas; rapid growth was more likely in tumors that were ≤5 cm in diameter. (See "Uterine fibroids (leiomyomas): Epidemiology, clinical features, diagnosis, and natural history", section on 'Natural history'.)

Rapid growth may occur in either a sarcoma or a benign leiomyoma. In addition, it is theoretically possible for a sarcoma to remain indolent for a long period of time and only come to diagnostic attention when a more aggressive phase of disease is entered.

The teaching about rapid growth of a uterine mass being associated with an increased risk of uterine sarcoma was not substantiated in a study of 1332 patients who underwent hysterectomy or myomectomy of presumed uterine leiomyomas [7]. The incidence of uterine sarcoma was low for all patients, with a similar incidence in the 371 patients with a rapidly growing uterus by clinical or ultrasound examination (one uterine sarcoma, 0.27 percent) and in the 961 without rapid uterine growth (two uterine sarcomas, 0.15 percent). In addition, a literature review of 26 studies found that a history of rapid uterine enlargement was documented in only 15 of 580 patients (2.6 percent) with uterine sarcomas [7]. Thus, the great majority of premenopausal patients with a rapidly enlarging uterus or uterine mass do not have a sarcoma.

By contrast, postmenopausal patients who have a uterine mass that is new or is growing at either a slow or rapid pace should be evaluated for malignancy. The level of suspicion for uterine sarcoma is lower in patients who are on postmenopausal estrogen therapy and have a small increase in the size of a presumed fibroid known to have been present prior to menopause. (See 'Increasing age and postmenopausal status' above.)

Large or solitary uterine mass — Retrospective studies have reported that a sarcoma is often the largest (or the only) mass within a uterus, averaging 7 to 9 cm in diameter [32,36,55,56]. However, leiomyomas may also be singular and may be of any size. Although data are limited, large uterine size (in excess of 20 gestational weeks) has also not been shown to be associated with sarcoma risk [32,57]. As a result, these features are not useful in identifying malignancy.

Endometrial sampling — Since endometrial biopsy is a minimally invasive procedure and is already indicated in many patients with abnormal uterine bleeding to exclude endometrial neoplasia, we suggest endometrial sampling in patients in whom sarcoma is suspected or for whom the planned procedure includes intraperitoneal morcellation since morcellation disrupts the specimen and potentially disseminates malignant neoplasm. (See "Uterine tissue extraction by morcellation: Techniques and clinical issues", section on 'Tissue dissemination and worsened prognosis'.)

However, endometrial sampling yields a preoperative diagnosis in only 33 to 68 percent of patients with uterine sarcomas [58,59]. The method of endometrial sampling (office endometrial biopsy or dilation and curettage) does not appear to impact the sensitivity of the test.

There are few data regarding use of endometrial sampling in patients with sarcoma. Representative studies include:

●The largest study includes data from the Canadian Task Force III database [60]. Among patients with leiomyosarcoma who underwent endometrial sampling before surgery (n = 68), the sensitivity for a diagnosis of features of a malignant smooth muscle neoplasms was 52 percent (leiomyosarcoma 35 percent and other features suspicious for malignancy 16 percent). There was no significant difference in the performance of the test between office endometrial biopsy and dilation and curettage.

●In a small study, among patients with leiomyosarcoma, three of eight (38 percent) who underwent preoperative endometrial sampling had a correct diagnosis [9].

●In another small study, the sensitivity of endometrial sampling for a diagnosis of leiomyosarcoma was four of six participants (67 percent) and for a diagnosis of endometrial stromal sarcoma was two of six participants (33 percent). Two additional patients with leiomyosarcoma and two with endometrial stromal sarcoma had preoperative pathology results reported as malignant neoplasm but an incorrect histologic diagnosis (including endometrial carcinoma, stromal sarcoma, and other sarcoma [61]. Thus, these patients would have been managed surgically as appropriate for a malignancy. There was no significant difference in the performance of the test between office endometrial biopsy and dilation and curettage.

Other biopsy techniques — Infrequently, both benign leiomyomas and uterine sarcomas will prolapse through the cervix and can be biopsied [7]. (See "Uterine fibroids (leiomyomas): Prolapsed fibroids".)

Use of minimally invasive needle biopsy of a uterine mass (guided by pelvic imaging or laparoscopy) has been proposed but is not typically performed. Limitations of this method are that the accurate diagnosis of sarcoma requires sampling of multiple sites and that the procedure may spill malignant cells within the peritoneal cavity.

Other tests — There are no laboratory tests that have been found to help differentiate uterine leiomyomas from sarcoma. Some reports have mentioned use of lactate dehydrogenase (LDH), LDH isozyme 3, or cancer antigen 125 [62]. In one retrospective study of 2254 patients with uterine masses (43 [2 percent] of whom were diagnosed with uterine sarcoma), LDH subtypes predicted uterine sarcomas with 100 percent sensitivity and 99.6 percent specificity; however, replicate studies in different populations have not been published [63].

Intraoperative evaluation — During myomectomy or hysterectomy, some characteristics of a uterine mass may raise suspicion of malignancy. Frozen section analysis cannot definitively diagnose or exclude uterine sarcoma.

Gross characteristics of the mass — During excisional surgery (ie, myomectomy), uterine sarcomas may appear or feel different than leiomyomas. Potential features of sarcomas compared with leiomyomas include (picture 1) [64-66]:

●Loss of the typical whorl pattern.

●Homogeneous texture.

●Yellow color.

●Soft consistency.

●Absence of a bulging surface when the capsule is incised.

●Ill-defined margins – The mass may be difficult to excise, although this may also be true of an adenomyoma or certain leiomyomas (eg, due to degenerative change or prior treatment with a gonadotropin-releasing hormone agonist or antagonist).

Detection of such characteristics may raise the suspicion of a sarcoma but should not be the sole reason for proceeding with an unplanned hysterectomy. These differences are subtle, and their identification relies upon adequate surgical experience with ordinary leiomyomas.

Frozen section — Frozen section analysis is not reliable for excluding uterine sarcoma [9,36,67]. Multiple areas must be sampled to obtain an accurate diagnosis, while frozen section analysis typically depends upon a limited tissue sample. Thus, there is a high likelihood of a false-negative result even if a sarcoma is present. Only a definitive diagnosis of sarcoma on frozen section should influence surgical decisions (eg, whether to perform hysterectomy or staging). (See 'Should unplanned hysterectomy be performed based on intraoperative findings?' below.)

Postoperative pathology evaluation — Criteria are important to the diagnosis of uterine sarcomas: mitotic index, cellular atypia, cellularity, and geographic areas of coagulative necrosis separated from viable neoplasm [66,68]. The more features exhibited by the lesion, the more likely it is to have clinically aggressive behavior [66,68]. There are a number of pathologic entities in which lesions display some, but not all, of the facets of malignant neoplasm. There are also reports of molecular-based diagnostics for leiomyosarcomas, but there is no current clinical consensus [69-71]. (See "Uterine fibroids (leiomyomas): Variants and smooth muscle tumors of uncertain malignant potential", section on 'Classification'.)

Diagnosis requires examination of multiple sites in the mass [65]. Features of the gross appearance of the tissue, including the color, consistency, and variegation of the incised surface, can help guide sampling of tissue for microscopic examination.

CLINICAL ISSUES

Do techniques that disrupt a mass disseminate tissue and worsen prognosis? — Surgical techniques such as myomectomy, intraperitoneal morcellation, and supracervical hysterectomy disrupt a uterine specimen. Infrequently, an occult uterine cancer is diagnosed postoperatively and, in such cases, there is a concern that malignant cells may have been disseminated or remain in uterine or cervical tissue that was not removed. In addition, dissemination of benign leiomyomatous tissue may result in benign tissue implants that require further treatment [72]. (See "Uterine fibroids (leiomyomas): Variants and smooth muscle tumors of uncertain malignant potential" and "Uterine tissue extraction by morcellation: Techniques and clinical issues".)

Disruption of the specimen appears to worsen prognosis of uterine cancer, but there are few data, and the comparative risk of the specific surgical methods (electromechanical morcellation, scalpel morcellation, supracervical hysterectomy, or myomectomy) is uncertain [17,73-76].

Morcellation — Morcellation of uterine tissue (cutting tissue into pieces to remove through a small incision) is performed in some procedures to allow removal of a large tissue block through a small laparoscopic or vaginal incision. This may be performed using an electromechanical morcellator device, as is commonly done during laparoscopy, or with a scalpel, commonly done during vaginal hysterectomy. The risk of dissemination of tissue and other risks during morcellation is discussed in detail separately. (See "Uterine tissue extraction by morcellation: Techniques and clinical issues".)

Supracervical hysterectomy — Supracervical hysterectomy is typically performed either via laparotomy or laparoscopy. With laparotomy, the specimen is not usually morcellated, so the only tissue disruption is the incision to amputate the uterine cervix. With laparoscopy, the specimen is typically morcellated, which likely results in greater tissue dissemination. (See "Uterine tissue extraction by morcellation: Techniques and clinical issues".)

There are few data about the risk of neoplasm dissemination and worsened prognosis with supracervical hysterectomy. The only study to address this issue included 12 patients with endometrial carcinoma or uterine sarcoma who underwent supracervical hysterectomy without morcellation; 9 of 12 underwent subsequent surgery for restaging, and none were upstaged [77]. The study also included patients who underwent morcellation for hysterectomy or myomectomy; four of five underwent restaging surgery, and two of these four were upstaged.

Myomectomy — Myomectomy is a uterine-sparing procedure in which only the leiomyomas are removed. Intraperitoneal myomectomy is performed via laparotomy or laparoscopy. With laparotomy, the specimen(s) are not usually morcellated, but there may be multiple fibroids, and there is usually substantial exposure of myometrial tissue. With laparoscopy, the specimen is typically morcellated. Smooth muscle cells have been found in pelvic washing after myomectomy and morcellation [78].

There are few data about the risk of neoplasm dissemination and worsened prognosis with myomectomy. No studies have examined patients who underwent myomectomy without morcellation. One retrospective cohort study of 30 patients with uterine sarcoma diagnosed postoperatively who underwent myomectomy with morcellation (electromechanical or scalpel morcellation) were compared with patients who underwent total hysterectomy [79]. At five years, there were no significant differences in recurrence-free survival (24 versus 46 percent) or overall survival rate (38 versus 43 percent), but the trend was toward a lower rate with myomectomy combined with morcellation.

Should hysterectomy be performed to exclude uterine sarcoma? — For most premenopausal patients with presumed uterine leiomyomas, whether asymptomatic or symptomatic, we recommend not performing hysterectomy for the sole purpose of excluding malignant neoplasm. Reasonable exceptions to this are patients with endometrial sampling and/or magnetic resonance imaging (MRI) results that strongly suggest sarcoma, those with thoracic imaging consistent with lung metastases, those with constitutional symptoms suggestive of malignancy, or those with multiple risk factors for uterine sarcoma.

The best clinical estimate of the likelihood of uterine sarcoma depends upon a constellation of factors. Unfortunately, preoperative and intraoperative findings are of limited value in diagnosing a uterine sarcoma. Factors that may raise a suspicion of sarcoma include symptoms, risk factors, failure of response to prior therapy, and the findings of MRI and endometrial sampling. (See 'Risk factors' above and 'Choice of imaging modality' above and 'Patients on medical therapy' below and 'Endometrial sampling' above.)

In the rare cases in which a patient presents with a uterine mass and symptoms of lung lesions (eg, dyspnea), evidence of lung metastases on thoracic imaging greatly increases the likelihood of a malignant process. However, benign metastasizing leiomyomas is a benign condition that may also present with lung lesions. (See "Uterine fibroids (leiomyomas): Variants and smooth muscle tumors of uncertain malignant potential", section on 'Metastasizing leiomyoma'.)

Asymptomatic patients — Hysterectomy should not be performed in most patients with asymptomatic presumed leiomyomas for the sole purpose of excluding a uterine sarcoma.

The most important factor in such decisions is the rarity of uterine sarcoma. The likelihood of finding sarcoma in patients with a preoperative diagnosis of leiomyomas has been consistently reported in observational studies of 1000 or more patients as approximately 0.2 percent [7-16]. These data are from symptomatic patients, so the risk is likely much lower in asymptomatic patients. This risk is much lower than the risk of severe complications associated with hysterectomy (3.5 to 11.0 percent) for benign disease found in studies of 1000 to 30,000 hysterectomies [80,81]. This recommendation is in agreement with the American College of Obstetricians and Gynecologists, which advises that hysterectomy be reserved for patients with symptomatic fibroids in whom definitive management is desired and not to rule out malignancy [82].

Given these data, the risk of a missed diagnosis of sarcoma in a small number of patients does not appear to outweigh exposing significantly more patients to the risks of unnecessary hysterectomy. In our practice, the primary exceptions we make to this are for patients in whom endometrial sampling and/or MRI strongly suggests sarcoma or those with thoracic imaging consistent with lung metastases. Hysterectomy is also a reasonable option for patients with multiple risk factors for uterine sarcoma. (See 'Risk factors' above and 'Choice of imaging modality' above and 'Endometrial sampling' above.)

Following the detection of presumed uterine fibroids in an asymptomatic patient, there are no high-quality data to guide follow-up. We order an initial imaging study (usually an ultrasound) to confirm that a pelvic mass is consistent with a fibroid rather than with other etiologies (eg, ovarian mass). After this initial evaluation, we perform annual pelvic examinations. If the characteristics of the uterus change or symptoms develop, we proceed with further evaluation and patient counseling regarding treatment options. (See "Uterine fibroids (leiomyomas): Treatment overview", section on 'Role of expectant management'.)

Symptomatic patients — For patients with symptomatic presumed leiomyomas, hysterectomy is one of several treatment options. The best treatment option should be chosen based upon the patient's symptoms, findings on evaluation, plans for future pregnancies, and patient preferences.

The advantage of hysterectomy is that it prevents the formation of new fibroids and also can treat concomitant problems such as adenomyosis, endometriosis, and cervical neoplasia. The disadvantages are loss of fertility, risk of surgical complications, the time needed for recovery, and risk of long-term morbidity and earlier menopause, especially if the ovaries are removed at the time of hysterectomy [83-85]. In some patients, the question may arise of whether to choose hysterectomy in order to exclude malignancy rather than a conservative treatment. (See "Elective oophorectomy or ovarian conservation at the time of hysterectomy".)

As with patients with asymptomatic presumed leiomyomas, for most with a symptomatic uterine mass, we recommend not choosing hysterectomy rather than other treatments based solely upon the goal of excluding malignant neoplasm. Reasonable exceptions to this are patients with endometrial sampling and/or MRI results that strongly suggest sarcoma, thoracic imaging consistent with lung metastases, or multiple risk factors for uterine sarcoma.

On the other hand, for patients deciding between several options for treatment of symptomatic fibroids, excluding malignant neoplasm may be a potential benefit of hysterectomy. As noted above, uterine sarcoma is rare, but the prevalence is greater in symptomatic than asymptomatic patients. Nonetheless, the risk of a complication of hysterectomy exceeds the risk of finding an undiagnosed malignant neoplasm. In addition, patients who fail to respond to conservative therapy may have an increased risk of sarcoma and also may require definitive treatment. Thus, excluding malignancy may be one factor that influences the decision regarding whether to choose treatment with hysterectomy. (See 'Risk factors' above and 'Choice of imaging modality' above and 'Patients on medical therapy' below and 'Endometrial sampling' above.)

Should unplanned hysterectomy be performed based on intraoperative findings? — During myomectomy or other pelvic surgery, a surgeon may encounter findings suggestive of a uterine sarcoma. Intraoperative findings rarely indicate the need for an unplanned hysterectomy, and sending pathology samples for evaluation and planning a potential second surgery pending the results is usually prudent, especially for patients who desire future childbearing. It is prudent to include, at the time of preoperative consent, discussion of plans in the event of diagnosing a malignant neoplasm intraoperatively based on the updated guidance from the US Food and Drug Administration. Gross characteristics of a tumor may be suggestive of sarcoma but are not diagnostic. Frozen section analysis is often inconclusive. (See 'Intraoperative evaluation' above.)

The only indications for hysterectomy are a definitive frozen section diagnosis of sarcoma and/or gross evidence of metastases. One note of caution is that patients with apparent intraabdominal metastases may have a primary neoplasm at another site or leiomyomatosis peritonealis disseminata, which is a rare and benign condition. Unplanned hysterectomy should not be performed in patients of reproductive age without a pathologic diagnosis of sarcoma and a documented preoperative discussion. Additionally, because oophorectomy has not been shown to influence prognosis for leiomyosarcomas, ovarian preservation may be carried out in the case of unexpected intraoperative findings of sarcoma [86]. (See "Uterine fibroids (leiomyomas): Variants and smooth muscle tumors of uncertain malignant potential", section on 'Disseminated peritoneal leiomyomatosis'.)

Staging for an incidentally discovered intraabdominal malignancy is best performed by a surgeon experienced in these procedures. (See "Hysterectomy: Abdominal (open) route", section on 'Malignancy'.)

Do leiomyomas progress to sarcoma? — In general, it appears that uterine sarcomas do not arise from benign leiomyomas, with rare exceptions. The consensus from genetic studies has been that most sarcomas arise independently [87]. Sarcomas typically have complex karyotypes and aneuploidy, while leiomyomas have characteristic rearrangements, many of which are shared by other benign neoplasms [88]. Molecular studies suggest that a small percentage of sarcomas may arise from fumarate hydratase mutations in patients with hereditary leiomyomatosis and renal cell carcinoma (HLRCC) syndrome [89]. Intermediate forms between these two patterns had not been described.

Histologic studies have found rare examples that appear consistent with progression from a leiomyoma to sarcoma [90]. The most compelling evidence comes from a rare subgroup of leiomyomas with cellular or atypical histology based on data from two clinical and two molecular genetic studies [91-94]. In a clinical study, 3 of 18 subjects with cellular or atypical variants died of their disease [91]. There was a longer latency between diagnosis and death in these patients compared with those with leiomyosarcoma (>6 years versus median of 2.1 years). There are no data regarding how often cellular or atypical histology are associated with such abnormal clinical behavior. (See "Uterine fibroids (leiomyomas): Variants and smooth muscle tumors of uncertain malignant potential", section on 'Variants of leiomyomas'.)

CLINICAL APPROACH — The approach to patients with presumed benign leiomyomas in different clinical categories is discussed in this section. All patients with a uterine mass should be screened for premenopausal status, gynecologic malignancies, and sarcoma risk factors and undergo appropriate testing based on this screening. The patient's plans for future childbearing and the full range of fibroid treatment options should be discussed.

Patients managed expectantly — Unless risk factors are identified, patients may be managed expectantly within guidelines for routine care unless new symptoms develop.

Patients on medical therapy — Patients should be reevaluated for possible sarcoma if they fail to respond to medical therapy, new symptoms appear, or symptoms worsen while on medical therapy or at least annually, with frequency of visit determined based on the particular medical therapy they are on.

Failure to respond to treatment (eg, gonadotropin-releasing hormone [GnRH] agonist therapy) has preceded a diagnosis of malignancy in multiple reports [95-101]. However, such treatment failures are not definitive evidence of malignant neoplasm, and at least one case report describes initial shrinkage of a leiomyosarcoma with GnRH therapy [101]. No data are available for GnRH antagonist combination therapy, which are also used for medical treatment of fibroid-related heavy menstrual bleeding. (See "Uterine fibroids (leiomyomas): Treatment overview", section on 'GnRH analogs'.)

Patients planned for surgery — The possibility of finding an unanticipated malignant neoplasm should be discussed, and, only if future childbearing is not desired, discussion can take place regarding proceeding with hysterectomy if suspicious frozen section pathology is found.

Patients planned for imaging-based procedures — Most patients in this category should have screening with magnetic resonance imaging and should be counseled to go in the direction of surgical therapy if suspicious findings are present.

In one case report, a suspected sarcoma was noted on imaging prior to magnetic resonance-guided focused ultrasound (MRgFUS) surgery, which led to expedited hysterectomy rather than the MRgFUS procedure [47].

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Uterine fibroids (leiomyomas)".)

SUMMARY AND RECOMMENDATIONS

●Clinical significance – Benign uterine fibroids (leiomyomas) are the most common pelvic neoplasm in females (lifetime risk 70 to 80 percent). Uterine sarcoma is rare (3 to 7/100,000 in the United States population). The likelihood of finding a sarcoma in a population of patients with an intrauterine mass, some of whom have risk factors for uterine sarcomas, is approximately 0.05 to 0.28 percent. (See 'Introduction' above and 'Presumed benign leiomyomas' above.)

●Natural history – Leiomyomas do not appear to progress to sarcoma, with the possible exception of rare atypical or cellular variants. (See 'Do leiomyomas progress to sarcoma?' above.)

●Findings that do not reliably predict sarcoma

●Clinical manifestations are not useful to distinguish between leiomyomas and uterine sarcomas, since both typically present with abnormal uterine bleeding, pelvic pain/pressure, and a pelvic mass. (See 'Signs and symptoms' above.)

•A rapidly enlarging uterine mass is not a reliable sign of a uterine sarcoma in premenopausal patients. After menopause, on the other hand, a new or growing uterine mass warrants further evaluation. Postmenopausal hormone therapy may cause a small increase in the size of an existing leiomyoma. (See 'Rapidly growing uterine mass in premenopausal patients' above and 'Increasing age and postmenopausal status' above.)

•Large uterine size (in excess of 20 gestational weeks) has not been shown to be associated with increased uterine sarcoma risk in most studies. (See 'Large or solitary uterine mass' above.)

●Endometrial sampling – Endometrial sampling may detect uterine sarcoma in some patients. We suggest endometrial sampling for those with abnormal uterine bleeding or a uterine mass and who have signs, symptoms, risk factors, or other findings that raise suspicion of uterine sarcoma or endometrial carcinoma or for whom the planned procedure includes intraperitoneal morcellation. (See 'Endometrial sampling' above.)

●Imaging

•There is no pelvic imaging modality that can reliably differentiate between benign leiomyomas and uterine sarcomas. We suggest that patients with a pelvic mass that is newly diagnosed or has changed markedly in size or mobility undergo a pelvic ultrasound as an initial imaging study rather than no imaging or imaging with another modality. (See 'Imaging' above.)

•Magnetic resonance imaging (MRI) with gadolinium contrast may be helpful in patients in whom there is a suspicion of sarcoma; however, it does not provide a definitive diagnosis. For patients in whom uterine sarcoma is suspected based on characteristics of the patient or mass or on ultrasound findings, we suggest follow-up imaging with MRI rather than imaging with ultrasound alone or with other modalities. (See 'Imaging' above.)

●Indications for hysterectomy – For most patients with presumed uterine leiomyomas, whether asymptomatic or symptomatic, we recommend not performing hysterectomy for the sole purpose of excluding malignant neoplasm (Grade 1B). Reasonable exceptions to this are patients with endometrial sampling and/or MRI results that strongly suggest sarcoma, those with thoracic imaging consistent with lung metastases, or those with multiple risk factors for uterine sarcoma. (See 'Asymptomatic patients' above and 'Symptomatic patients' above.)