Endometrial polyps

The endometrial polyp is a common gynaecological lesion associated with abnormal bleeding and infertility. It can also be an asymptom­atic incidental finding during imaging. It is a localized overgrowth of the uterine endometrium that can be sessile or polypoid.

Aetiology

Age, hypertension, obesity, and tamoxifen use are some of the risk factors (1-3). The causation is likely to be multifactorial. Thus in obese women, while an excess of oestrone may play a role, hyper­tension may be a confounding factor. There is a 30-60% prevalence of polyps in women using tamoxifen. In women with infertility, the use of gonadotropins may be associated with the development of polyps. Endometrial polyps are also associated with cervical polyps in 24-27% of cases and the association becomes stronger with advancing age and abnormal vaginal bleeding (4). Atypical glandular cells in a cervical smear are also associated with endo­metrial polyps in 3.4-5% of cases (5). Genetic factors play a role, with altered endometrial proliferation and overgrowth being as­sociated with specific alleles on chromosomes 6 and 12 (6). It has been hypothesized that an excess of endometrial cytokines and metalloproteinases may increase the risk of developing polyps, fi­broids, and adenomyosis, the same mediators that are implicated in intrauterine disease associated with infertility. In postmenopausal women there is an excess of growth regulating protein P63, which is also a marker of the reserve cells of the basalis layer. The latter is thought to be the precursor of polyps (7, 8).

While oestrogen and progesterone are key factors in the prolifer­ation and apoptosis of the endometrium, their role in the aetiology and pathophysiology of polyps is controversial.

Epidemiology

The reported prevalence of polyps is between 7.8% and 34.9% depending on the population studied, the diagnostic tool used, and the definition of polyps. While it is generally thought that polyps are more prevalent in postmenopausal (11.8%) compared to premeno­pausal women, this could simply reflect the fact that any abnormal bleeding in postmenopausal women will be investigated, which is not the case in premenopausal women (10).

Clinical presentation

Approximately 68% of all women with polyps present with abnormal vaginal bleeding (11) and 6-88% of premenopausal women with polyps have abnormal vaginal bleeding in the form of menorrhagia, intermenstrual bleeding, or postcoital bleeding (11). Endometrial polyps account for 39% of all abnormal vaginal bleeding in pre­menopausal women and this is thought to be due to stromal conges­tion leading to venous stasis and apical necrosis.

In contrast, postmenopausal women with polyps are more often symptom free, with approximately 56% presenting with abnormal bleeding (11). Polyps account for only 21- 28% of all vaginal bleeding in postmenopausal women (11). In premenopausal women, polyps are associated with infertility (12). This might be due to mechanical obstruction at the tubal ostium or due to a mechanical or biochem­ical effect on implantation of the developing embryo, possibly due to the excess of intrauterine metalloproteinases and cytokines as­sociated with polyps (7).

Natural history

Polyps can regress spontaneously, with one study reporting a re­gression rate of 27%, and a correlation between size and regres­sion: polyps smaller than 1 cm are more likely to regress than larger ones (13).

While most polyps are benign, some can become hyperplastic with malignant transformation in 0-12.9% of cases (1). The risk is highest in postmenopausal women with symptoms and low in premenopausal women. There is a significant correlation between age, menopausal status, obesity, hypertension, tamoxifen use, and size of the polyp and incidence of malignant transformation (2). In one study the risk of malignancy was similar in symptomatic and asymptomatic patients, suggesting that polyps should be removed whenever identified (14). Ultrasonography may aid in identification of malignancy with a sensitivity of 67-100% and a specificity of 71­89%. The variation in range is dependent upon the thickness of the endometrium used for further invasive testing.

Diagnosis

Transvaginal ultrasound

On transvaginal ultrasound (TVUS), a polyp typically appears as a hyperechoic lesion with regular contours within the uterine lumen, outlining the endometrial wall on which it rests, surrounded by a hyperechoic halo. Cystic spaces within the polyp corresponding to dilated glands filled with proteinaceous material may be seen, or it may appear as a thickening of the endometrial lining or focal mass within the endometrial cavity. However, such appearances are not pathognomonic of endometrial polyps, since submucosal fibroids can also look similar. To minimize false-positive or false-negative results in the premenopausal woman, TVUS should be performed within the first 10 days of the menstrual cycle.

Compared to hysteroscopy and guided biopsy, studies have re­ported that TVUS has a sensitivity of 19-96%, specificity of 53­100%, positive predictive value of 75-100%, and negative predictive value of 87-97%.

There are limited data to substantiate the use of colour or power Doppler in the diagnosis of malignant change or hyperplasia in a polyp. In one study the specificity and the negative predictive value were claimed to be 95% and 94% respectively for identifying a single large feeding vessel by colour flow Doppler in TVUS (16) whereas others have shown limited value in the diagnosis of endo­metrial cancer, with no significant difference in histology of polyps depending on their resistive or pulsatility index.

Power Doppler seems to be a more promising technique for the depiction of the vascular network (17) and in one study the sensi­tivity and specificity were reported to be 87% and 85% respectively in identifying a single large feeding vessel as a marker of an endo­metrial polyp compared to a network of multiple or scattered vessels for hyperplasia or malignancy (17). However, the study showed this to be more effective for women in whom the polyp was an incidental finding. Ultimately the only way to confirm or exclude malignancy is histological examination following its removal.

Saline infusion sonography

The use of saline infusion sonography (SIS; also referred to as sonohysterography) increases the sonographic contrast helping in the delineation of size, location, and other features of endometrial polyps. Polyps appear as echogenic intracavitary masses with either a broad base or a thin stalk floating in the fluid. This technique is thought to increase diagnostic accuracy and small polyps missed on grey-scale sonography are picked up on SIS. Differentiating a polyp from a submucosal fibroid can be difficult but examining the echotexture and identifying echogenic endometrium overlying the polyp can be helpful.

A number of studies comparing the diagnostic accuracy of diag­nostic hysteroscopy and SIS showed no significant difference be­tween the two (18). When comparing SIS and hysteroscopy with guided biopsy, the sensitivity, specificity, positive predictive value, and negative predictive value were 58-100%, 35-100%, 70-100%, and 83-100% respectively (18).

When compared with hysteroscopy, SIS has the advantage of allowing the assessment of the myometrium and other pelvic or­gans. It has also been reported to be less painful than diagnostic hysteroscopy when the latter is performed as an outpatient procedure under similar conditions to SIS: both techniques involve insertion of an instrument through the cervix and distending the uterine cavity with fluid. However, these reports are from earlier studies before the advent of present-day hysteroscopes of much smaller diameter (19). The disadvantages of SIS include an inability to give a histological diagnosis, a longer learning curve, and discomfort caused by leakage of fluid or pain by distension with the balloon catheter.

Three-dimensional TVUS and three-dimensional SIS

Three-dimensional (3D) ultrasound can generate multiplanar re­constructed images of the uterus including coronal views and there­fore improve the diagnostic accuracy compared to two-dimensional (2D) ultrasound. Three-dimensional SIS includes addition of saline infusion to 3D ultrasound. However, this technique has been shown to improve diagnostic accuracy only slightly and given the greater expense and less frequent availability of 3D SIS, 2D ultrasound with intrauterine contrast will remain the preferred effective and reliable non-invasive method to diagnose polyps.

Histological diagnosis

Blind biopsy

In contemporary practice, blind dilatation and curettage should no longer be used as a diagnostic technique due to its poor sensi­tivity and negative predictive value compared to hysteroscopy and guided biopsy, which has a specificity and positive predictive value of 100% (20).

Hysteroscopy with guided biopsy

This is considered the gold standard in the diagnosis of endo­metrial polyps (21). The ability to diagnose and remove polyps concurrently makes it superior to diagnostic hysteroscopy alone. Despite the growing popularity of outpatient hysteroscopy, most of the diagnostic hysteroscopies are still performed under gen­eral anaesthesia, particularly if operative hysteroscopy is required. The evidence supports use of outpatient hysteroscopy for diag­nosis with a reported success rate of 92-96% and no difference be­tween premenopausal and postmenopausal women (22). Studies have shown it to be superior both in terms of expense and patient preference.

Flexible hysteroscopy is less painful for patients and allows easier passage through the cervical canal when compared to rigid hysteroscopy, making it more acceptable for office procedures. It is thought to have inferior image quality compared to rigid hysteroscopy as the light and images are transferred through the same fibreoptic bundle. New flexible hysteroscopes with video chips are superior in this respect, although these may be susceptible to breakage, have a limited operative scope, and may be more costly than rigid hysteroscopes. With new, technologically improved narrow scopes, more and more operative hysteroscopies can be per­formed in the outpatient setting. While smaller endometrial polyps can be removed with minimal patient discomfort, polyps larger than the internal cervical os are best removed under general anaesthesia.

The choice of distension medium is an important consideration for patient comfort and diagnostic accuracy in outpatient settings. Normal saline causes less discomfort and less shoulder tip pain when compared with carbon dioxide and therefore produces im­ages which are clear and reliable. Use of paracervical blocks and intrauterine anaesthesia can also be helpful in outpatient operative hysteroscopy.

Complication rates are low in hysteroscopic polypectomy. When compared to hysteroscopic myomectomy, endometrial ablation, and hysteroscopic adhesiolysis the risks of perforation, cervical lacer­ation, infection, and haemorrhage remain low (23).

Management

The management of polyps is guided by the presence of symptoms, desire for future fertility, risk of malignancy, and operator skills. The options are conservative non-surgical, conservative surgical, and radical surgical.

Conservative non-surgical management

While the removal of polyps is associated with a low risk of compli­cations, it is not a completely risk-free procedure and therefore pre­intervention patient counselling is mandatory. The rate of regression of polyps less than 10 mm in size is 27% over 12 months and the risk of malignancy is very low: such polyps can therefore be managed conservatively in asymptomatic patients (13).

Medical treatment may have some role in the management of polyps. Gonadotropin-r eleasing hormone (GnRH) agonists have been shown to cause temporary regression of polyps and can be used as a treatment adjunct before polypectomy. However, the cost and side effects of such treatment need to be compared with simple al­ternative extirpative treatment without the use of such medications. A variety of progesterone preparations including norethisterone, medroxyprogesterone, and tibolone have been used in the context of hormone replacement therapy in postmenopausal women and tibolone, which has the highest androgenic activity, is thought to cause regression of polyps. Hysteroscopic examination at 3 years after treatment revealed a low risk of recurrence after use of these preparations. However, these treatments are not without side effects, and high-quality studies are required to further establish their place.

In a randomized controlled trial of the levonorgestrel-releasing intrauterine system (LNG-IUS) compared with observation, a re­duced rate of polyp recurrence was shown in the LNG-IUS group. In a 4.5-year study observation period, eight cases of polyp occur­rence were seen in the observation group compared to three in the LNG-IUS group. Out of these three, one woman did not have an IUS inserted and in other two it was taken out after 1 year due to side effects. Reduction in endometrial thickness due to progesterone suppression is thought to contribute to the regression or reduced development of polyps (24).

Conservative surgical treatment

Blind dilatation and curettage has been used as a treatment for endo­metrial polyps for many years. A survey of practice in the United Kingdom carried out in 2002 showed that 2% of gynaecologists used this technique and 51% used blind curettage after hysteroscopy (21, 25). Evidence suggests that this technique has a high complication rate with a perforation rate of approximately1 in 100 and an infec­tion rate of approximately 1 in 200 (21). Studies suggest that with blind curettage alone, the rate of polyp removal is only 4% which increases to 41% if a polyp removal forceps is also used. The rate of incomplete removal is also high (26). TVUS-guided polypectomy has been suggested as an alternative in order to improve the rate of removal of polyps, however this has received little enthusiasm (27). Hysteroscopic resection of polyp is safe with a low complication rate, is widely available, and can be performed in the outpatient setting, and therefore should replace blind methods of polyp removal.

Hysteroscopic polypectomy

This is a safe and effective method for polyp removal which allows rapid recovery and can be sometimes be performed in an outpatient setting (28). There are various techniques of polypectomy depending on the type of instrument used. This is dependent on availability, ex­pense, surgical expertise, and also the size and location of polyps. Large and sessile polyps are best removed with a resectoscope, an electrosurgical loop fitted to the hysteroscope, while smaller polyps are best removed either by scissors or polyp forceps under direct hysteroscopic vision (28). Hysteroscopic resection carries more complications, probably due to the greater cervical dilatation re­quired in these cases. However, the polyp recurrence rate is nearly zero after use of the resectoscope compared with about 15% with grasping forceps (28).

Other instruments that may be considered include the bipolar Versapoint, which requires less cervical dilatation and uses normal saline instead of glycine thus reducing the potential risk of postop­erative hyponatraemia (11). The hysteroscopic morcellator removes the polyp chips while resecting, thus reducing the operating time, fluid loss, and movement through the cervix. Such techniques are, however, expensive and not readily available and the outcomes are not significantly different from other methods of hysteroscopic removal.

Radical surgical treatment

Hysterectomy is a definitive treatment for endometrial polyps, guaranteeing no recurrence. However, it can only be justified in the presence of other pathology such as symptomatic fibroids, given the significant morbidity associated with such a radical approach.

Outcome of treatment

The outcome of treatment is generally good with reduction or ces­sation in abnormal vaginal bleeding. The risk of intrauterine adhe­sion formation after polypectomy is low, as the myometrium is not damaged and the endometrium has excellent regenerative capacity. In women undergoing polypectomy as a treatment of subfertility, reported postoperative pregnancy rates vary between 43% and 80% with improvement seen in both chances of natural and assisted con­ception (28). In a class 1 study, polypectomy before intrauterine in­semination significantly increased subsequent pregnancy rates (29). The rate of pregnancy in the study group was 51%, and of these, 65% had a spontaneous conception before the first intrauterine insem­ination, whereas all pregnancies in the control group were obtained during the fertility treatment (29). There is a lack of consensus over the size of the polyps that may affect fertility, with data suggesting that removal of a polyp less than 2 cm in size does not improve fertility (30).

Conclusion

Endometrial polyps are a common gynaecological condition whose prevalence increases with age. They are rarely associated with malig­nancy. They can be associated with both subfertility and abnormal uterine bleeding. Non-invasive techniques such as grey-scale TVUS give a reliable diagnosis and diagnostic enhancement can be achieved by the use of contrast medium. In the management of polyps, hysteroscopic resection is safe and effective and allows histo­logical examination. Blind techniques should be avoided because of the high incidence of incomplete resection and complications such as perforation. Polypectomy is an effective treatment of infertility although evidence from randomized controlled trials to demon­strate improvement in in vitro fertilization outcome is still needed. Conservative medical treatment is a viable option pending definitive surgical treatment. Radical surgical treatment such as hysterectomy is unnecessary in the treatment of polyps.