Endometrial polyp (EP) is a frequently encountered gynaecologic condition that may present with abnormal vaginal bleeding. It can also be asymptomatic and detected incidentally during imaging investigations or hysteroscopy for infertility or other conditions. Most EPs are benign, but premalignancy or even malignancy may occasionally occur.
EPs are hyperplastic mucosal overgrowths projecting from the endometrium consisting of endometrial glands, stromal tissue, and associated vasculature. At times, they may contain smooth muscle components. They appear as pink swellings, which may vary from half to a few centimetres in diameter. EPs may be sessile or pedunculated, single or multiple, and may appear anywhere in the uterine cavity.
Since EPs are often asymptomatic and detected incidentally on ultrasonography, the exact prevalence is difficult to establish. The increasing use of transvaginal ultrasonography (TVUS) in clinical practice has increased the detection rate of EPs. The prevalence of EPs increases gradually with advancing age, and mostly in infertile women, with a prevalence rate ranging from 8–35% as reported in studies on different populations where the definition and method of diagnosis also differ.1
POSSIBLE PATHOGENESIS AND RISK FACTORS
Pathogenesis of EP is likely to be multifactorial. A systematic review suggested that both oestrogen- and non-oestrogen-related pathways are involved in the development of EPs.2
In the normal endometrium, polyps express both oestrogen and progesterone receptors. Oestrogen and progesterone are two key factors controlling the balance between proliferation and secretory changes in the normal endometrial tissue. At the beginning of a normal menstrual cycle, oestrogen is produced by the developing ovarian follicles. The progressive increase in the levels of oestrogen stimulates the growth and proliferation of the endometrium. After ovulation, progesterone, which is antiproliferative, is produced by the corpus luteum to convert the endometrium from the proliferative to the secretory phase. In women with chronic anovulation, or other conditions associated with disturbed oestrogen/progesterone balance, there may be significant increase in endometrial proliferation, which may increase the risk of developing neoplasms, including EPs.
Review of the literature showed that the risk factors for the development of EPs include advancing age, oestrogen therapy, obesity, and tamoxifen therapy. Tamoxifen is a selective oestrogen receptor modulator that has modest proliferative activity in the endometrium. In a study comparing post-menopausal women on different regimens of hormonal therapy, those who received unopposed oestradiol treatment had the highest risk of endometrial proliferation, while those on combined oestrogen-progestogen treatment were at a medium risk, and those who had no treatment were at a lower risk.3 Obese women with body mass index (BMI) ≥30 kg/m2 had a significantly increased EP rates.4
EPs have been found to have a higher expression of aromatase P450 gene and protein, steroidogenic factor-1 protein, and p63 protein than the adjoining endometrium and normal endometrium.5 An increase in aromatase may increase oestrogen production.
All these findings indicate that hormonal factors, especially the oestrogen/progestogen balance, may play an important role in the pathogenesis of EPs.
On the other hand, characteristic cytogenetic rearrangements involving the 6p21, 12q14-15, and 7q22 regions have been reported in EPs.6 Changes in the coding regions for high-mobility group (HMG) family of transcription factors, including HMGI-C and HMGI(Y) genes, which are found at the 12q15 and 6p21 sites, respectively, are more likely to develop EPs.7
Women with EPs typically exhibit abnormal uterine bleeding (AUB). Many polyps are however asymptomatic, only being detected during pelvic imaging or hysteroscopy for other indications such as infertility evaluation. In some cases, prolapsed EPs can be visualized at the external cervical os through vaginal speculum examination.
Abnormal uterine bleeding
AUB is the most frequent presenting symptom in women with EPs, being reported by 67% and 75% among premenopausal and menopausal women, respectively.8
In a prospective study of 1,220 women with AUB, 26% were diagnosed with EP.9 The most common symptom in premenopausal women with EP is intermenstrual bleeding, usually in small amounts. Heavy menstrual bleeding and post-coital bleeding are also reported by some patients with EP. Post-menopausal bleeding is another frequent symptom; some post-menopausal women with EP on hormonal therapy may also have episodes of breakthrough bleeding. Women with AUB who are >40 years of age or at risk of endometrial cancer are recommended to undergo evaluation for endometrial pathologies.
Incidental finding during ultrasonography or hysteroscopy
Almost half of women with EPs are asymptomatic. EPs are usually incidentally detected by pelvic ultrasonography, endometrial aspiration, or hysteroscopy performed for other indications such as infertility.
Some EPs may spontaneously regress without any treatment. Various studies have reported that the regression rate ranges from 6–27%. EPs larger than 1.5 cm are less likely to regress.10 There is no reliable factor that can predict polyp growth or regression. Some women with polyps that regressed reported episodes of heavy menstrual bleeding with cramping pain, followed by restoration of normal menstruation.11-12
Influence on fertility and pregnancy
In two large prospective trials including 1,000 and 2,500 infertile women preparing to undergo in vitro fertilization, the reported prevalence of EP was 23% and 32%, respectively.13-14 In a randomized comparative trial of women with EP who had intrauterine insemination (n=215), a significantly higher pregnancy rate was observed in women who underwent polypectomy than hysteroscopic biopsy without polypectomy (63% vs 28%).15 However, not all pregnancies in that series resulted from assisted reproductive technology. Sixty-five percent of pregnancies in the polypectomy group occurred before the first intrauterine insemination cycle. The spontaneous pregnancy rate was 29% in the polypectomy group compared with 3% in the control group. Based on these results, many gynaecologists and some professional bodies recommend polypectomy when EP is detected in women with infertility. However, it was subsequently noted in a Cochrane review that there were internal inconsistencies in the results reported in this randomized trial.16 Therefore, the current evidence is insufficient to draw conclusions regarding the efficacy of routine polypectomy on the outcome of subsequent infertility treatment. More prospective randomized clinical trials will be necessary to confirm the efficacy of hysteroscopic EP removal as a means of managing infertility.
EPs do not appear to increase spontaneous abortion rate, and there was no adverse obstetric event associated with EPs reported.
The oncogenic potential
Most EPs are benign. EPs have low oncogenic potential compared with polyps occurring in other organs (eg, gastrointestinal tract and bladder). In a Canadian retrospective study of 1,027 women with EPs, benign polyps accounted for about 96% of all cases, and premalignant and malignant histology were found in 2.8% and 1.5% of all cases, respectively. EPs in post-menopausal and older women (>60 years of age), and in those with post-menopausal vaginal bleeding, have a significantly higher risk of being malignant.17
In another retrospective study including 1,011 women with EPs, 95.4% of the polyps were reported as benign, 1.3% was hyperplasia without atypia, 0.5% was hyperplasia with atypia, and 1.3% was endometrial cancer. Therefore, rates of premalignant and malignant polyp were 1.8% and 1.3%, respectively, with the highest risk among post-menopausal women with AUB.18
Another multicentre cohort study showed that although the risk of EP oncogenesis is correlated with older age, high BMI, post-menopausal state, presence of AUB, and hypertension, only age and AUB occurrence remained as significant predictors of malignancy.19
It has been shown that EP was the most common endometrial pathology in women who received tamoxifen therapy, with a prevalence rate of up to 30–60%.1 In a retrospective study of 120 premenopausal breast cancer patients who were given tamoxifen, 42% had histopathologically confirmed EP.20
Malignant polyps are more commonly encountered in women using tamoxifen.21 A meta-analysis including four randomized controlled trials (RCTs) concluded that the cumulative risk of endometrial malignancy in women who received prolonged tamoxifen therapy >5 years was increased twofold (from 1.5% to 3.2%) compared with those who received the standard 5-year tamoxifen therapy.22
Ultrasonography is a commonly performed investigation in women with suspected pelvic abnormality or AUB symptoms. TVUS may have better resolution than transabdominal ultrasonography for evaluating intrauterine lesions (Figure 1). However, the accuracy of ultrasonography in detecting intrauterine lesions is variable in premenopausal women, as the endometrial thickness and sonographic morphology changes with the menstrual cycle. Having the ultrasound done during the early proliferative phase (before day 10) offers better visualization of the polyp as the endometrium remains thin around this period.
For women with uncertain findings, sonohysterography (SHG), also termed as saline infusion sonogram, can be helpful.
A systematic review including 1,645 SHG procedures reported that the sensitivity, specificity, and positive and negative likelihood ratios were 88%, 94%, 20.93, and 0.15, respectively, in detecting lesions in the uterine cavity among subfertile women prior to assisted reproduction treatments. SHG also had good accuracy for the evaluation of all intrauterine abnormalities, with an area under the summary receiver operating characteristic (sROC) curve of 0.97.23
A more recent meta-analysis showed a similar area under the sROC curve (0.97) of two-dimensional SHG in diagnosing EP.24 Although both SHG and hysteroscopy provide visualization of the intrauterine lesions, SHG has the advantage of detecting not only EPs but also myometrial or adnexal abnormalities. Both two- and three-dimensional SHG are similarly accurate. A Cochrane review showed no significant difference in summary sensitivity and specificity between the two- and three-dimensional SHG procedures in diagnosing focal pathology in the uterine cavity.25
Hysteroscopic and histologic evaluation
Hysteroscopy is the standard approach to diagnose diseases of the endometrium, as it allows for the direct viewing of uterine focal lesions and for targeted biopsy.
The main aim of histologic evaluation is to exclude malignancy and premalignancy. The histopathological specimen is usually collected either by blind biopsy (eg, Pipelle sampling and curettage) or hysteroscopic-guided biopsy.
In a multicentre RCT of women suffering from post-menopausal bleeding and having endometrial thickening of >4 mm but being negative for blind endometrial sampling, EP was detected in 51% of those randomized to hysteroscopy (n=98). Of these, 3% had endometrial hyperplasia and 3% eventually developed endometrial cancer.26 Therefore, the diagnosis of malignancy or premalignancy may be missed if only blind biopsy is done in such scenario.
With the improvements in hysteroscopic techniques, hysteroscopic evaluation and some minor operative procedures can be performed using hysteroscopes with smaller diameters in the office setting.
For EP, the differential diagnoses include submucous fibroid and endometrial neoplasia or hyperplasia that has focal protrusion into the uterine cavity. It is usually possible to differentiate between fibroids and polyps with ultrasonography because of the different echotexture. Fibroids and polyps typically have different appearances under hysteroscopy. Polyps often present with a beefy red appearance or as pink swellings, and are soft when touched with an instrument. On the other hand, fibroids are usually firm and have whiter appearance with blood vessels on the surface. In certain situations, visual differentiation between the two lesions may be challenging and the final determination requires histopathology.
Cervical polyp and prolapsed fibroid may sometimes be visualized on speculum examination. The key differential feature between prolapsed endometrial and cervical polyps is the origin of its stalk. If the origin cannot be visualized, ultrasonography or hysteroscopy may help. A prolapsed leiomyoma may be distinguished as it has a different appearance and consistency.
Since EP may spontaneously regress, for asymptomatic young women (<40 years old) with small EP (<10 mm in diameter), conservative management can be safely adopted. In two studies, approximately 25% of polyps regressed in 12 months and small polyps (<10 mm in diameter) were less likely to be malignant.11-12 In those women who opt for expectant treatment, repeat TVUS or SHG is recommended within 6 months. There is no evidence that medication could help treat EP.
Hysteroscopic polypectomy is the standard treatment for EPs. Polypectomy aims to alleviate symptoms and rule out malignancy.
Common situations where surgical removal of EP is indicated include (i) women with AUB, (ii) EPs >15 mm in diameter, (iii) infertility, (iv) recurrent EP, and (v) additional risk factors for endometrial malignancy (eg, diabetes mellitus, hypertension, and obesity), as well as (vi) women on tamoxifen and (vii) post-menopausal women.27
Various methods available for polyp removal at hysteroscopy include the use of grasping forceps, microscissors, electrosurgical loop (resectoscope with monopolar or bipolar loops), and morcellation system. Blind curettage is not the preferred method for excision of polyps where hysteroscopy is available. Alternatively, to remove polyp by curettage after hysteroscopic visualization, hysteroscopy should be reintroduced immediately to confirm complete removal of the polyp.
A multicentre RCT including 121 women with EP compared hysteroscopic morcellation with electrosurgical resection. The median procedure time was shorter with hysteroscopic morcellation compared with electrical resection (5 minutes and 28 seconds vs 10 minutes and 12 seconds). Women who underwent hysteroscopic morcellation achieved a higher complete EP removal compared with those who had electrosurgical resection (98% vs 83%). Although both procedures do not require general anaesthesia or conscious sedation, majority of patients found the procedure to be tolerable. In addition, the complication rates were quite low.28
Based on the information available in the literature, the management of EP is outlined in Figure 2.
Occasionally, EPs that have prolapsed through the external cervical os, which is visible upon speculum examination, can be directly removed with forceps. Prolapsed polyps may present with symptoms including post-coital bleeding or increased vaginal discharge. There may be chance of recurrence in women after direct removal of a prolapsed polyp, particularly if performed without dilation of the cervix and visualization for complete removal.
Recurrence of polyps
Recurrence rates of EP following resection are unknown. A previous study of 256 women demonstrated that after diagnostic hysteroscopy and blind polypectomy, 7 women (3.9%) had recurrence at a mean follow-up of 37 months.29 If the EP recurs after removal, repeat hysteroscopic polypectomy is needed.
In a retrospective study involving 240 women with endometrial polyps, resectoscopic polypectomy was found to be more effective than the grasping forceps method in preventing polyp recurrence.30
Women treated with tamoxifen therapy
EP is the most common endometrial pathology observed among tamoxifen users.31 In tamoxifen users with AUB or suspicious finding on ultrasonography, diagnostic hysteroscopy is recommended. The prophylactic effect of the levonorgestrel-releasing intrauterine system (LNG-IUS) against endometrial polyp formation among women on tamoxifen has been demonstrated. A meta-analysis comprising three RCTs (n=359) showed that LNG-IUS reduced rates of endometrial hyperplasia and EP in breast cancer patients on tamoxifen compared with the control group, although this was linked to increased rates of abnormal vaginal bleeding or spotting. Moreover, women treated with LNG-IUS showed no increased breast cancer recurrence rates or breast cancer-related deaths, although larger long-term RCTs are necessary to validate these findings in a larger breast cancer patient population being treated with tamoxifen.32 It should be noted that the current guidelines do not recommend the use of LNG-IUS in women with a history of breast cancer.33-34
EP is a common gynaecological condition that increases in prevalence with advancing age, and is highly prevalent in infertile women. Approximately 95% of EPs are benign. It may present with AUB or as incidental finding on TVUS. For women with suspected EP on ultrasonography alone or who are candidates for expectant management, SHG is recommended. Hysteroscopic polypectomy is the most effective method for diagnosing and treating EPs. Women who are taking tamoxifen have an elevated risk of developing EPs and potential malignancy.
About the authors
Dr Xian Li is a Resident Doctor from the Department of Obstetrics & Gynaecology, The University of Hong Kong – Shenzhen Hospital, Shenzhen China. Conflict of interest: None.
Dr Hang Wun Raymond Li is Associate Professor from the Department of Obstetrics & Gynaecology, The University of Hong Kong, Hong Kong and Honorary Consultant from the Department of Obstetrics & Gynaecology, The University of Hong Kong – Shenzhen Hospital, Shenzhen China. Conflict of interest: None.
Dr Shui Hua Lin is a Resident Doctor from the Department of Medical Imaging, The University of Hong Kong – Shenzhen Hospital, Shenzhen China. Conflict of interest: None.
Professor Pak-Chung Ho is an Emeritus Professor from the Department of Obstetrics & Gynaecology, The University of Hong Kong – Shenzhen Hospital, Shenzhen China. Conflict of interest: None.