INTRODUCTION — Endometrial polyps are one of the most common etiologies of abnormal genital tract bleeding in both premenopausal and postmenopausal patients (table 1). Unlike polyps of other etiologies (eg, colon), the majority of endometrial polyps are benign with malignancy typically occurring only in high-risk patients (eg, genetic syndrome, postmenopausal).
The pathogenesis, diagnosis, and management of endometrial polyps are reviewed here. General principles of the evaluation of uterine bleeding are discussed separately. (See "Abnormal uterine bleeding in nonpregnant reproductive-age patients: Terminology, evaluation, and approach to diagnosis" and "Approach to the patient with postmenopausal uterine bleeding".)
HISTOPATHOLOGY — Endometrial polyps are localized hyperplastic overgrowths of endometrial glands and stroma around a vascular core that form a sessile or pedunculated projection from the surface of the endometrium (picture 1) [1,2]. Smooth muscle is sometimes present.
Single or multiple polyps may occur and range in diameter from a few millimeters to several centimeters (picture 2) [3]. Polyps can develop anywhere in the endometrial cavity.
PATHOGENESIS — Several molecular mechanisms have been proposed to play a role in the development of endometrial polyps. These include monoclonal endometrial hyperplasia [4], overexpression of endometrial aromatase [5,6], and somatic gene mutations [7,8]. Like uterine leiomyomas, polyps have characteristic cytogenetic rearrangements. Rearrangements in the high-mobility group family of transcription factors appear to play a pathogenic role [7,9,10].
Endometrial polyps express both estrogen and progesterone receptors and these hormones (ie, estrogen, progesterone) may play a role in pathogenesis, especially in postmenopausal women [11,12]. In a subset of patients with polyps (ie, those on tamoxifen therapy), progesterone may serve an antiproliferative function, as it does in normal endometrial tissue [13,14] (see 'Tamoxifen therapy' below). While androgens have been found to cause endometrial atrophy, similar to progestins, data suggest that testosterone does not substitute for progestational activity for polyps [15].
PREVALENCE — The frequency of polyps is difficult to establish, since there are few data and some polyps are asymptomatic. Among clinically recognized polyps, the prevalence appears to rise steadily with increasing age and is higher in postmenopausal compared with premenopausal patients (12 versus 6 percent, respectively) [16]. Among patients undergoing endometrial biopsy or hysterectomy, the prevalence of endometrial polyps is 10 to 24 percent [17,18].
Endometrial polyps are rare among adolescent patients [19].
RISK FACTORS
Tamoxifen — Endometrial polyps are the most common type of endometrial pathology associated with tamoxifen use. Tamoxifen use is also associated with malignant transformation of endometrial polyps (11 percent, one study [20]) and an increase in the overall risk of endometrial cancer [21]. Malignant transformation does not appear to be associated with polyp size or duration of tamoxifen use. This is discussed in detail separately. (See "Abnormal uterine bleeding and uterine pathology in patients on tamoxifen therapy", section on 'Endometrial polyps' and "Endometrial carcinoma: Epidemiology, risk factors, and prevention", section on 'Tamoxifen'.)
Tamoxifen appears to increase the risk of endometrial polyp formation more than other selective estrogen receptor modulators, such as raloxifene. In a randomized trial including over 9400 postmenopausal patients with breast cancer treated with chemoprophylaxis, the incidence of polyps was higher in patients treated with tamoxifen compared with raloxifene (2.1 versus 0.6 percent, relative risk 0.3, 95% CI 0.25-0.35) over the 81-month study period [21].
Obesity — Endometrial polyps appear to be associated with obesity [22-24]. In a retrospective cohort study of 223 patients planning to undergo in vitro fertilization, those with a body mass index ≥30 kg/m2 compared with <30 kg/m2 had a higher rate of polyps (52 versus 15 percent, respectively) [23]. A subsequent cross-sectional study found that postmenopausal patients with polyps compared with postmenopausal patients without polyps had higher rates of obesity (odds ratio [OR] 4.66), elevated glucose levels (OR 2.83), dyslipidemia (OR 7.0), and a diagnosis of metabolic syndrome (OR 2.58) [25].
Other
●Hormone replacement therapy – Postmenopausal hormone therapy may be associated with endometrial polyps, particularly regimens with a high dose of estrogen and/or a progestin with low antiestrogenic activity [22,26].
●Lynch and Cowden syndrome – Patients with Lynch and Cowden syndrome may have an increased incidence of endometrial polyps compared with the general population; these patients are also at an increased risk of endometrial hyperplasia and endometrial carcinoma [27-29]. (See "Lynch syndrome (hereditary nonpolyposis colorectal cancer): Clinical manifestations and diagnosis" and "PTEN hamartoma tumor syndromes, including Cowden syndrome", section on 'Genitourinary'.)
●The presence of more than one polyp and endometriosis may be independent risk factors of recurrent polyps, as suggested by at least one study [30].
NATURAL HISTORY
Continued growth or regression — If left untreated, endometrial polyps may persist, progress, or resolve. In a prospective study including 64 asymptomatic premenopausal patients (mean age 44 years), two saline infusion sonograms (SISs) were performed 2.5 years apart [31]. Of the seven patients (11 percent) diagnosed with polyps on the first examination, four (57 percent) had spontaneous resolution of their polyps at the second scan; seven additional patients developed new polyps over the 2.5-year interval. Polyps larger than 1 cm compared with smaller polyps were less likely to resolve. Hormone use did not appear to affect the natural history of the polyps, but the study sample was small. By contrast, a retrospective study including 112 patients managed expectantly for a mean of 22.5 months found that only 6.3 percent of polyps spontaneously resolved and that no factors were reliably associated with polyp growth and regression [32].
Risk of malignancy — Approximately 95 percent of endometrial polyps are benign [33]. In a systematic review including 17 observational studies (over 10,000 patients), the incidence of polyps that were malignant or hyperplastic was higher in postmenopausal compared with premenopausal patients (5.4 versus 1.7 percent, relative risk [RR] 3.86, 95% CI 2.9-5.1) [34]. The risk of malignancy was also higher in patients with versus without abnormal uterine bleeding (AUB; 4.2 versus 2.2 percent, RR 2.0, 95% CI 1.2-3.1). These characteristics (ie, postmenopausal status, AUB) are also associated with an increased risk of endometrial malignancy without polyps.
The data regarding polyp size and risk of malignancy are less clear. In the systematic review, data were inconsistent regarding whether increased polyp size was associated with malignancy [34]. Other cohort studies have reported that premalignant or malignant histology is associated with polyps greater than 1.3 cm [35] or 1.5 cm in greatest dimension [36,37].
Treatment of malignant endometrial polyps is the same as for endometrial carcinoma. (See "Overview of endometrial carcinoma".)
Effect on fertility and pregnancy — Patients undergoing evaluation for infertility may have a finding of an endometrial polyp on ultrasound or hysteroscopy; the reported prevalence in those undergoing in vitro fertilization is 6 to 8 percent [38,39]. The management of patients with polyps and infertility is discussed below. (See 'Infertility' below and "Evaluation of female infertility".)
Endometrial polyps do not appear to be associated with an increased risk of spontaneous abortion or adverse obstetric outcomes. In studies performed in patients with a recent miscarriage, the prevalence of polyps was the same as in the general population [40,41]. (See 'Prevalence' above.)
CLINICAL PRESENTATION — The most common clinical presentation is abnormal uterine bleeding (AUB). Endometrial polyps may also be asymptomatic and diagnosed incidentally on pelvic imaging performed for another indication, reported on cervical cytology or endometrial biopsy results, or visualized on physical examination.
●Abnormal uterine bleeding – AUB is the most common presenting symptom and occurs in 64 to 88 percent of patients with polyps [16,42,43]. Bleeding due to polyps is referred to as "AUB-P" in the nomenclature recommended by the International Federation of Gynecology and Obstetrics system, also known as the Polyp, Adenomyosis, Leiomyoma, Malignancy and hyperplasia, Coagulopathy, Ovulatory dysfunction, Endometrial dysfunction, Iatrogenic, and "Not otherwise classified" (PALM-COEIN) system [44]. (See "Causes of female genital tract bleeding", section on 'Polyps (AUB-P)'.)
Intermenstrual bleeding is the most frequent bleeding pattern in premenopausal patients with endometrial polyps [45]. The volume of bleeding is usually small and may be just spotting. Some patients experience heavier bleeding between menstrual cycles. (See "Abnormal uterine bleeding in nonpregnant reproductive-age patients: Terminology, evaluation, and approach to diagnosis", section on 'Intermenstrual bleeding'.)
Postmenopausal bleeding is another common presentation; some postmenopausal patients with polyps have breakthrough bleeding during hormonal therapy. (See "Approach to the patient with postmenopausal uterine bleeding".)
Patients with AUB may require evaluation for endometrial cancer. (See "Endometrial carcinoma: Clinical features, diagnosis, prognosis, and screening", section on 'Abnormal uterine bleeding'.)
●Incidental finding
•On imaging – Endometrial polyps are often identified incidentally on a pelvic ultrasound or hysteroscopy performed for another indication. (See 'Pelvic imaging' below.)
•On cervical cytology – While cervical cytology is not a useful method for diagnosing endometrial polyps, studies have shown an association between the finding of benign endometrial cells on liquid cervical cytology testing and both benign and malignant endometrial neoplasms. In a retrospective chart review including 440 patients (age 40 years or older) with a cervical cytology report that included the presence of endometrial cells and follow-up endometrial sampling, 12 percent of patients had endometrial polyps alone and 2 percent had polyps with a coexistent diagnosis (ie, hyperplasia, endometritis) [46]. Among patients with polyps alone, 72 percent were asymptomatic. The evaluation of endometrial cells on cervical cytology is discussed separately. (See "Cervical cancer screening: The cytology and human papillomavirus report", section on 'Benign-appearing endometrial cells'.)
•On endometrial biopsy – In some patients, an endometrial biopsy may show evidence of an endometrial polyp. In such cases, polypectomy should be performed, if indicated, since the entire polyp may not have been removed with the endometrial biopsy. (See 'Polypectomy' below and 'Management' below.)
•Prolapsed polyp – An endometrial polyp may prolapse and be visible at the external cervical os at the time of speculum examination. Prolapsed polyps may be symptomatic or asymptomatic. (See 'Physical examination' below.)
DIAGNOSTIC EVALUATION
History — The medical history should include questions about symptoms (eg, abnormal uterine bleeding [AUB]) and risk factors associated with endometrial polyps. (See 'Clinical presentation' above and 'Risk factors' above.)
Physical examination — In the absence of a prolapsed polyp, there are no physical examination findings associated with an endometrial polyp. A prolapsed polyp can be visualized during a speculum examination, typically as a globular, friable, pedunculated lesion protruding from the external cervical os.
A polypoid lesion at the external cervical os is most commonly a cervical polyp (picture 3), but may be a prolapsed endometrial polyp or leiomyoma. In general, a cervical polyp is identified by visualizing or palpating a stalk originating from the endocervical canal, while the stalk of an endometrial polyp originates from the uterine cavity. Prolapsed leiomyomas typically have a firm consistency, while polyps are soft and friable. (See "Benign cervical lesions and congenital anomalies of the cervix", section on 'Polyps' and "Uterine fibroids (leiomyomas): Prolapsed fibroids", section on 'Incidental finding on pelvic examination'.)
Pelvic imaging — Patients with a suspected endometrial polyp are typically evaluated with pelvic imaging or hysteroscopy.
●Transvaginal ultrasound – Transvaginal ultrasound (TVUS) is the first-line imaging study of choice for evaluation of patients with AUB. This modality is effective at characterizing uterine and adnexal lesions and is less expensive than other modalities.
●Sonohysteroscopy or diagnostic hysteroscopy – We suggest sonohysterography, also referred to as saline infusion sonogram (SIS (image 1)), or diagnostic hysteroscopy for patients with any of the following:
•An uncertain finding on TVUS alone.
•Postmenopausal patients with a thickened endometrium on TVUS.
In a randomized trial including 200 postmenopausal patients with bleeding, a thickened endometrium (>4 mm), and a negative endometrial biopsy, two patients had undiagnosed endometrial cancer and one patient had endometrial hyperplasia with atypia in polyps [47]. The authors concluded that either hysteroscopy or SIS to evaluate for structural lesions in postmenopausal patients with thickened endometrium is warranted.
•Premenopausal patients in whom removal is not indicated and expectant management is planned. (See 'Premenopausal patients' below.)
Three-dimensional SIS does not appear to confer significant benefit over two-dimensional SIS [48].
A systematic review including over 5000 patients reported a similar performance for the diagnosis of polyps with TVUS, SIS, and hysteroscopy (sensitivity: 91, 95, and 90 percent; specificity: 90, 92, and 93 percent, respectively) [16]. The shape of the lesion is often better visualized with either SIS or hysteroscopy compared with TVUS alone. While SIS compared with hysteroscopy allows for visualization of the adnexa and is thus useful in patients with suspected adnexal pathology, diagnostic hysteroscopy allows direct visualization of the lesion and in many cases simultaneous removal. (See "Saline infusion sonohysterography" and "Overview of hysteroscopy", section on 'Procedure'.)
Magnetic resonance imaging does not appear to be a sensitive diagnostic test for polyps [49].
DIAGNOSIS — The diagnosis of an endometrial polyp is a histologic diagnosis based on the evaluation of the specimen, usually collected at time of polypectomy. Histologic evaluation can also exclude malignancy.
DIFFERENTIAL DIAGNOSIS — The differential diagnosis of an endometrial polyp includes other structural lesions of the uterine cavity.
●Intracavitary leiomyomas – Intracavitary leiomyomas can often be differentiated from an endometrial polyp based on their appearance on ultrasound (leiomyomas appear hypoechoic with shadowing versus polyps which appear hyperechoic and homogeneous) and pattern of Doppler flow (leiomyomas demonstrate peripheral flow versus the appearance of a single feeding vessel in polyps). In addition, polyps and fibroids typically have different appearances when visualized with hysteroscopy; polyps often have a beefy red appearance, are usually thinner and less likely to be sessile, and are soft and friable when touched with an instrument, and a dilated gland can sometimes be visualized. By contrast, myomas are firm and are mainly white in color with surface blood vessels. Final determination is made with histology. (See "Uterine fibroids (leiomyomas): Epidemiology, clinical features, diagnosis, and natural history", section on 'Diagnostic evaluation'.)
●Endometrial hyperplasia or carcinoma – In contrast to endometrial hyperplasia or neoplasia, polyps are generally well demarcated at hysteroscopy; however, only histology can exclude malignancy. (See "Endometrial carcinoma: Clinical features, diagnosis, prognosis, and screening", section on 'Ultrasound'.)
●The differential diagnosis of a prolapsed endometrial polyp includes a cervical polyp and a prolapsed leiomyoma. The approach to differentiating between these lesions is discussed above. (See 'Physical examination' above.)
MANAGEMENT
Postmenopausal patients — For postmenopausal patients, we recommend removal of all endometrial polyps, regardless of symptoms (table 2). The risk of malignancy in a polyp is highest in postmenopausal patients and the risk of complications associated with polypectomy is low. (See 'Risk of malignancy' above and 'Polypectomy' below.)
Premenopausal patients — Premenopausal patients are managed based on symptoms (ie, abnormal uterine bleeding [AUB]) and other indications for removal (eg, risk factors for endometrial cancer, large polyp size (table 2)).
Patients in whom removal is not indicated are managed expectantly; there are no studies regarding the need for continued surveillance. In our practice, we do not perform further surveillance in these patients.
With abnormal bleeding — Premenopausal patients with symptomatic polyps (ie, AUB) require removal. The goal of polypectomy is both relief of symptoms and detection of malignancy, since symptomatic compared with asymptomatic polyps are more likely to be malignant. (See 'Risk of malignancy' above.)
With other indications for removal — We also suggest removal for patients with the following:
●Risk factors for endometrial hyperplasia or carcinoma (table 3). (See "Endometrial carcinoma: Epidemiology, risk factors, and prevention", section on 'Risk factors'.)
●Polyp >1.5 cm in diameter – Studies report that polyps >1.5 cm in diameter are associated with an increased risk of malignancy or hyperplasia, although data are inconsistent regarding polyp size and lower thresholds have been reported. (See 'Risk of malignancy' above.)
●Multiple polyps – In our experience, multiple polyps are unlikely to regress and are likely to become symptomatic.
●Prolapsed polyps – Prolapsed polyps are unlikely to regress, are likely to become symptomatic, and can typically be removed easily in an outpatient setting.
Special populations
Infertility — Most clinicians perform polypectomy in patients with infertility, although data regarding the impact of polyp removal on fertility are limited and evidence is insufficient to make a recommendation.
A systematic review based on limited data concluded that removing polyps was beneficial in infertile patients [50]; this conclusion was based primarily on a single randomized trial (204 patients) that showed a higher pregnancy rate in patients undergoing intrauterine insemination who underwent polyp removal compared with hysteroscopy alone (63 versus 28 percent, respectively) [51,52].
Recurrent polyps — In rare cases, endometrial polyps recur after removal. In such cases, care should be taken to completely remove the polyp(s) in a repeat polypectomy procedure; however, there are no data regarding the management of recurrent endometrial polyps.
The formation of recurrent polyps may be prevented by use of the levonorgestrel-releasing intrauterine device (Mirena, Liletta; LNG 52), given its reported efficacy in reducing recurrent polyps in patients receiving tamoxifen treatment [13] (see 'Tamoxifen therapy' below). Endometrial ablation may also be an option for patients who have completed childbearing, however it is less preferred as it does not provide contraception. (See "Overview of endometrial ablation".)
Tamoxifen therapy — For patients on tamoxifen therapy, use of the LNG 52 may decrease the incidence of endometrial polyps [14,53,54]. Further study is needed, however, to determine whether such treatment results in a decrease in endometrial carcinoma in general or malignant transformation in polyps and whether use of the LNG 52, particularly in patients with progesterone receptor-positive breast cancer, increases the risk of breast cancer recurrence [55,56]. This is discussed in more detail separately. (See "Abnormal uterine bleeding and uterine pathology in patients on tamoxifen therapy", section on 'Progestins for prevention and treatment'.)
POLYPECTOMY
Procedure — Polypectomy under hysteroscopic guidance is the treatment of choice for most endometrial polyps. Hysteroscopic instruments that may be used to remove a polyp include grasping forceps, microscissors, electrosurgical loop (ie, resectoscope), morcellator, or a bipolar electrosurgical probe [57-59].
Hysteroscopic visualization of the polyp is preferred over blind curettage as the latter may miss small polyps and other structural abnormalities [57,60,61]. Some surgeons visualize the polyp via hysteroscopy and then remove it using a blind approach (eg, using Randall polyp forceps or a Kelly clamp) [62]. If this approach is used, the hysteroscope should be used again after polypectomy to confirm complete removal of the polyp.
Rarely, an endometrial polyp prolapses through the cervix and can be removed vaginally (see 'Physical examination' above). The procedure for removal of a prolapsed endometrial polyp is the same as for a prolapsed leiomyoma. (See "Uterine fibroids (leiomyomas): Prolapsed fibroids", section on 'Vaginal myomectomy'.)
Efficacy — For patients with symptomatic polyps, polypectomy results in improvement of symptoms in 75 to 100 percent of patients, based on studies with follow-up intervals of 2 to 52 months [63].
Complications — Complications of hysteroscopic polypectomy are infrequent, and the risks are the same as for other hysteroscopic procedures. (See "Overview of hysteroscopy", section on 'Complications'.)
Patients who undergo removal of a prolapsed polyp without dilation of the cervix and visualization of complete removal should be counseled about the potential for recurrence. The risk of recurrence in this situation is not known.
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: Abnormal uterine bleeding".)
SUMMARY AND RECOMMENDATIONS
●Histopathology – Endometrial polyps are hyperplastic overgrowths of endometrial glands and stroma that form a projection from the surface of the endometrium (lining of the uterus); they are one of the most common causes of abnormal genital tract bleeding in both premenopausal and postmenopausal patients (table 1). (See 'Introduction' above and 'Histopathology' above.)
●Prevalence – Among clinically recognized polyps, the prevalence is higher in postmenopausal compared with premenopausal patients. Among patients undergoing endometrial biopsy or hysterectomy, the prevalence of endometrial polyps is 10 to 24 percent. (See 'Prevalence' above.)
●Risk of malignancy – Almost all (approximately 95 percent) of endometrial polyps are benign, but malignancy occurs in some patients. Endometrial polyps are more likely to be malignant in patients who are postmenopausal and those who present with bleeding. Treatment of malignant endometrial polyps is the same as for endometrial carcinoma. (See 'Risk of malignancy' above and "Overview of endometrial carcinoma".)
●Presentation – While most endometrial polyps cause abnormal uterine bleeding (AUB), polyps may also be asymptomatic or present incidentally (eg, on pelvic imaging performed for another indication, reported on cervical cytology results, visualized on physical examination). (See 'Clinical presentation' above.)
●Diagnostic evaluation – Patients with a suspected endometrial polyp are typically evaluated with pelvic imaging or hysteroscopy.
•Transvaginal ultrasound (TVUS) is the first-line imaging study of choice for evaluation of patients with AUB. (See 'Pelvic imaging' above.)
•Sonohysteroscopy (saline infusion sonogram [SIS]) or diagnostic hysteroscopy may be used for additional imaging in postmenopausal patients with a thickened endometrial stripe and premenopausal patients with an uncertain finding on ultrasound alone or who are candidates for expectant management. (See 'Pelvic imaging' above.)
●Diagnosis – The diagnosis of an endometrial polyp is a histologic diagnosis based upon the evaluation of the specimen after it has been removed. Histologic evaluation can also exclude malignancy. (See 'Diagnosis' above.)
●Management (table 2)
•For postmenopausal patients, we recommend removal of all endometrial polyps as the risk of malignancy in such patients is high and the risk of complications associated with polypectomy is low. (See 'Management' above and 'Risk of malignancy' above.)
•For premenopausal patients, symptomatic polyps require removal. We also suggest removal of asymptomatic polyps in premenopausal patients with risk factors for endometrial hyperplasia or cancer. Polypectomy is also a reasonable option for patients with polyps that are >1.5 cm, multiple, or prolapsed, or for patients who are infertile. (See 'Management' above.)
