The diagnosis and treatment of endometrial polyps will be familiar to most gynaecologists. However, the aetiology and natural history of these focal intrauterine lesions are yet to be elucidated. This lack of clarity is also true with regard to their clinical significance; whilst endometrial polyps are highly prevalent in all types of abnormal uterine bleeding (AUB), they are also commonly found in women without AUB. These controversies will be discussed along with current thoughts on the diagnosis and treatment of endometrial polyps. Criteria for diagnosis of uterine polyps vary according to the test used, but optimal testing is not yet solidified. Recent data from randomised trials evaluating new and established surgical technologies as well as comparing treatment protocols and settings will be examined.
Introduction
Relationship of endometrial polyps to abnormal uterine bleeding
Definition
Uterine polyps are focal endometrial outgrowths that can occur anywhere within the uterine cavity. They contain a variable amount of glands, stroma and blood vessels, the relative amounts of which influence their visual appearance at hysteroscopy. Polyps may be soft and cystic or firm and fibrous; they may be pedunculated or sessile, single or multiple and vary in size from small – with minimal uterine cavity distortion – to large, filling the whole cavity ( Figure 1 ) .
Most endometrial polyps are diagnosed at imaging by transvaginal pelvic ultrasound (TVUS) or hysteroscopy and confirmation is provided by histological examination of the removed specimen. Thus, in clinical practice, the features that define an endometrial polyp will relate to the diagnostic modality, although agreed, standardised definitions are lacking ( Table 1 ).
| Diagnostic test | Definition |
|---|---|
| Hysteroscopy | A discrete outgrowth of the endometrium, attached by a pedicle, which moves with the flow of the distension medium. Polyps may be pedunculated or sessile, single or multiple and vary in size (the variable amount of glands, stroma and blood vessels that constitute the polyp will influence its macroscopic appearance) . |
| Ultrasound | Non-specific endometrial thickening or a focal mass identified as an echogenic lesion, which disturbs the midline endometrial echo but does not disrupt the interface between the myometrium and endometrium. The lesion is usually oval shaped with a homogeneous texture, although hypoechoic cystic spaces may be seen. Blood flow may be identified within a feeding vessel extending to the polyp on colour-flow Doppler imaging. Saline infusion sonography and 3D ultrasound help delineate the borders of the intra-cavity lesion . |
| Hysterosalpingogram | Filling defects within the uterine cavity |
| Magnetic resonance imaging | Isointense (1 Tesla) or hypointense (2 Tesla) masses within the uterine cavity. Contrast enhancement (1 Tesla) with gadolinium can show either homogeneous or heterogeneous enhancement . |
| Histological sampling a | Endometrial tissue showing at least two of: (i) glandular architectural disarray, (ii) stromal fibrosis and (iii) enlarged, thick-walled stromal blood vessels. Benign ‘simple’ polyps may be covered with functional endometrium (reproductive age) or atrophy (post-reproductive age). They may be described as glandular cystic, fibrous or glandular fibrous according to the relative preponderance of glandular and stromal components. Adenomyomatous polyps contain prominent bundles of smooth muscle within the stroma and adenomatous polyps have many glands with an intensively proliferating epithelium. Hyperplastic polyps have a hyperplastic epithelium with or without cytological atypia. Polyps may more rarely contain intraepithelial carcinomas or be freely malignant . |
a Biopsy may be blind or directed; the oval, polyp shape will be seen if surgically removed en bloc.
Aetiology
The underlying mechanism of uterine polyp formation remains unclear but is believed to be multifactorial . Uterine polyps are thought to start as focal areas of stromal and glandular overgrowth within the endometrium . The effect of hormones on polyp formation is unclear and may differ according to menopausal status. In pre-menopausal women, a decrease in the levels of oestrogen and progesterone receptors within polyp stromal cells may make polyps less sensitive to cyclic hormonal changes . Increased cell longevity may also play a role in the genesis of polyps as a result of inhibition of apoptosis and altered gene expression .
Epidemiology
Prevalence
The prevalence of uterine polyps in a general adult female population without abnormal uterine bleeding (AUB) is generally estimated to be approximately 10–15% . Uterine polyps were detected incidentally during TVUS in 12% of pre-menopausal women and in 6–11% of infertile women without AUB . In asymptomatic post-menopausal women undergoing TVUS, a prevalence of polyp of 13% was found and 16% during hysteroscopy. Risk factors for uterine polyp development are thought to include obesity, late menopause and the use of the partial oestrogen agonist tamoxifen . The role of hormone replacement therapy (HRT) on polyp formation is unclear, with some studies supporting an association and others not .
Although uterine polyps may be an incidental finding, they appear to be more prevalent in women undergoing investigation with high-resolution pelvic US or hysteroscopy for AUB. The prevalence of polyps found to be associated with AUB varies according to the criteria used to define a polyp, the diagnostic test used and the type of population studied. In general, the prevalence of endometrial polyps is considered to be between 20% and 30% . Uterine polyps affect pre-menopausal and post-menopausal women with AUB, and indeed, the prevalence may be increased after the menopause .
The high frequency in which uterine polyps are reported in women of reproductive age and the likely causative association between uterine polyps and AUB have been documented in the International Federation of Gynecology and Obstetrics (FIGO) classification system for causes of AUB during the reproductive years. This nomenclature is based on the acronym ‘PALM–COEIN’ with the ‘P’ denoting a ‘polyp’, i.e. describing AUB associated with the presence of uterine polyps (AUB-P) .
Natural history
Most uterine polyps will persist if left untreated although small polyps may spontaneously regress . In a small study of asymptomatic pre-menopausal women, 27% of polyps naturally regressed after 1 year of follow-up be smaller, comparing with an earlier case series .
The majority of uterine polyps are benign. Estimates of the prevalence of hyperplasia and cancer vary across study populations. A systematic review of observational studies reported endometrial hyperplasia (EH), without atypia, rates between 0.2% and 23.8% in polyps . The prevalence of pre-malignant atypical endometrial hyperplasia (AEH) appears to be lower with estimates in primary studies ranging from 1% to 3% .
The prevalence of endometrial polyp cancer has been reported within the range 0.5–3% .
Risk factors for malignancy within uterine polyps include AUB, increasing age, post-menopausal status, obesity, diabetes , an increased polyp size and tamoxifen administration . Another systematic review reported the prevalence of EH or cancer within polyps in women with symptomatic bleeding as 4.2% (195/4697) compared with 2.2% (85/3941) for those without bleeding (relative risk (RR) 1.97; 95% confidence interval (CI) 1.24–3.14). The risk of pre-malignancy or malignancy within a uterine polyp was higher in symptomatic post-menopausal women (5.4%, 214/3946) compared with 1.7% (68/3997) in reproductive-aged women (RR 3.86, 95% CI 2.92–5.11).
Abnormal uterine bleeding
The high prevalence of uterine polyps in women with all forms of AUB, namely heavy menstrual bleeding (HMB), intermenstrual bleeding (IMB) and post-menopausal bleeding (PMB), has become increasingly apparent with the widespread adoption of high-resolution pelvic US and greater access to outpatient hysteroscopy (OPH). In addition to this observation of high prevalence in women with AUB, causality has also been inferred from the effect of removing polyps on bleeding symptoms. Collated evidence from systematic reviews have found that alleviation or improvement in AUB symptoms is generally observed following polypectomy (range 75–100%) at follow-up intervals of between 2 and 52 months. However, these studies were generally uncontrolled small series. Moreover, reliable stratification of treatment outcome according to type of abnormal bleeding was not possible in these reviews. Recent randomised controlled trials (RCTs) have provided additional data on treatment outcomes in AUB following polypectomy, and this is discussed in more detail in the treatment section of this chapter.
The mechanism by which polyps, arising from the endometrium, precipitate these different forms of AUB remains uncertain, and may be linked to the equally unclear underlying aetiology of polyp formation. Thus, factors such as altered responses to oestrogen and progesterone compared with the background endometrium, inflammatory changes and disturbed angiogenesis may be involved, but these possible explanations remain speculative. Attempts are being made to produce a practical and reproducible sub-categorisation system of the AUB-P category within the PALM–COEIN nomenclature . By developing such a system based on features such as the number, size, shape, location, composition and vascularity of polyps, it is expected that questions pertaining to the significance of uterine polyps found in association with particular AUB presentations may be more readily answered.
Diagnosis of endometrial polyps in AUB
In women with AUB in whom endometrial polyps are suspected, there are three main modes of investigation.
Ultrasound
Initial assessment of endometrial disease is often conducted by radiological imaging. US scan (USS) is usually the first imaging modality and has the advantage of providing information on the size of the uterine cavity and other pathology including leiomyomas. It is non-invasive and therefore generally acceptable to the majority of patients with minimal discomfort. With plain TVUS, a polyp appears as endometrial thickening or with the more typical appearance of a hyperechoic lesion within the uterine lumen with a regular contour and surrounded by a thin hyperechoic halo . TVUS is more accurate when performed in the proliferative phase of the menstrual cycle . The accuracy of TVUS in diagnosing uterine polyps varies across test accuracy studies with accuracy estimates ranging from sensitivities of 19–96%; specificities between 53% and 100%; positive predictive values (PPVs) between 75% and 100%; and negative predictive values (NPVs) of 87–97% when compared with hysteroscopy and guided biopsy . In general, TVUS appears to have a good degree of accuracy when performed with high-resolution equipment by proficient practitioners. Polyp size should be assessed at the time of US as this can provide vital information, useful in aiding management. An increase in the diameter of the polyp appears to correlate with risk of malignancy with smaller polyps being more likely to resolve spontaneously .
Further enhancement with colour-flow Doppler or power Doppler can be used to improve diagnostic accuracy when using TVUS to assess the endometrial cavity. Colour-flow Doppler is useful in demonstrating the single feeding vessel observed along with endometrial polyps and power Doppler of the vascularity has been shown to improve diagnostic accuracy when assessing endometrial polyps in both asymptomatic and symptomatic women . There is, however, limited evidence that either power Doppler or colour-flow Doppler increases the diagnosis of malignancy or hyperplasia within polyps and therefore histological diagnosis is still necessary following detection by TVUS. Power Doppler has been reported to be more accurate than colour flow for demonstrating vascular networks in one study assessing post-menopausal women with abnormal bleeding and thickened endometrium on baseline US .
The addition of intrauterine contrast by saline infusion sonography (SIS) or gel installation sonography (GIS) may be used to improve diagnostic accuracy compared with TVUS alone. By using SIS, the fluid allows better contrast within the endometrial cavity allowing for delineation of the base or stalk of the polyp and improving detection of smaller polyps, which may have been missed by TVUS alone . In contrast to SIS, there is currently limited data on the accuracy of GIS in diagnosing uterine polyps.
When comparing SIS with hysteroscopy, SIS had a sensitivity of 58–100%, specificity 35–100%, PPV 70–100% and NPV of 83–100% . A systematic accuracy review using hysteroscopy with or without biopsy or hysterectomy as reference standards found that the accuracy of SIS in the diagnosis of endometrial polyps was lower than that for diagnosis of other uterine cavity abnormalities such as submucous fibroids (SMFs). The pooled sensitivity was 0.86 (95% CI 0.81–0.91), the pooled specificity was 0.81 (95% CI 0.72–0.88), and the likelihood ratios (LRs) were 5.23 (95% CI 3.98–6.90) and 0.12 (95% CI 0.08–0.17), respectively, consistent with a moderately accurate test for detecting and excluding polyps . SIS simultaneously allows the assessment of other pelvic structures including the adnexa and myometrium as well as the uterine cavity and tubal patency, and therefore provides a useful adjunct to TVUS, especially when assessing patient’s pre-operatively to decide on the procedure most likely to benefit the patient. The main disadvantage of SIS and GIS is the increased level of operator training needed compared with TVUS alone as well as patient discomfort during longer examination.
Three-dimensional (3D) USS has been used for assessing the endometrium but showed limited improvement in diagnosis compared with 2D TVUS , although this is slightly improved with the addition of intrauterine contrast. One study reported high accuracy of 3D USS in diagnosing uterine polyps with a derived sensitivity of 100%, specificity 99%, PPV 99% and NPV of 100% compared with hysteroscopy and guided biopsy .
Hysteroscopy
The gold standard investigation for diagnosis of endometrial polyps is hysteroscopy and guided biopsy. Hysteroscopy has the advantage of allowing the practitioner to directly visualise the endometrium and simultaneously removing any polyps for histological diagnosis, thus saving the patient from returning for treatment. It also has the advantage of detecting other endometrial pathology such as SMFs, which may also be treated simultaneously or assessed for further treatment at a later operation. The overall complication rate for hysteroscopy is small but increased compared with USS imaging alone with the overall risk of serious complications, 2 in 1000 women . A systematic review comparing hysteroscopy with hysteroscopically directed biopsy or hysterectomy showed a high degree of accuracy with a pooled sensitivity of 0.94 (95% CI 0.92–0.96) and specificity of 0.92 (95% CI 0.91–0.94). The corresponding positive and negative LRs were 12.9 (95% CI 8.0–20.9) and 0.09 (95% CI 0.06–0.14), respectively. The ability to exclude a polyp was higher in women after the menopause, which may reflect easier visual discrimination of a focal endometrial lesion from the thin inactive background endometrium (pre-menopausal women LR+ 33.5, 95% CI 8.2–136.0 and LR− 0.16, 95% CI 0.09–0.28; post-menopausal women LR+ 12.0 (95% CI 4.0–35.8) and LR− 0.04 (95% CI 0.01–0.26).
Although hysteroscopy has high accuracy for the diagnosis of endometrial polyps, the test involves a subjective assessment of the size and features of the polyp and directed biopsy should therefore be performed, even if the polyp appears benign and is not causing symptoms. Diagnostic hysteroscopy missed EH in 0.9% of patients in one large RCT and had a lower sensitivity and PPV than hysteroscopy with directed biopsy . However, a large systematic review and meta-analysis of hysteroscopy has demonstrated the high accuracy of the test in malignant and pre-malignant endometrial disease, although this did not specifically relate to cancer or hyperplasia within a uterine polyp .
Hysteroscopy is frequently performed in the outpatient setting with greater patient satisfaction with the outpatient procedure and similarly high success rates of 92–96% . Care should be taken handling the equipment used in OPH. Flexible hysteroscopy may be used owing to the reduction in patient discomfort over the use of rigid scopes; however, two prospective studies have revealed reduced accuracy when assessing for endometrial polyps with flexible hysteroscopy giving a sensitivity of 74% and specificity of 90% .
Endometrial biopsy
Pelvic examination of women with AUB may provide an opportunity for blind sampling of the endometrium using miniature aspiration devices most being based on the Pipelle ® biopsy prototype or traditional dilatation and curettage (D&C). These blind techniques fail to sample a significant proportion of the endometrial cavity and so not surprisingly for focal pathologies such as uterine polyps, blind biopsy has low accuracy compared with hysteroscopy and guided biopsy . This technique will frequently miss polyps particularly if small or pedunculated and makes histological diagnosis more difficult because of the tissue fragmentation. Histology is also unable to confirm that the polyp has been removed at the base and therefore blind biopsy should not be used as a diagnostic test when investigating for endometrial polyps.
Treatment of endometrial polyps in AUB
Expectant management
The surgical treatment of uterine polyps is excision or ‘polypectomy’. Uterine polypectomy is one of the most common procedures in contemporary gynaecological practice. Surgical removal aims to treat associated symptoms such as AUB or sub-fertility and to obtain tissue for histological examination. A UK national survey and two subsequent Dutch surveys confirmed that the vast majority of gynaecologists advocated surgical removal of polyps from the uterus after diagnosis. However, the need to universally remove uterine polyps may be questioned in light of the observations that polyps are found incidentally in approximately 5–15% of women , most are benign and some may regress spontaneously . Two RCTs have addressed this issue, randomising women with AUB and uterine polyps to expectant management or surgical removal . One of these RCTs randomised 150 pre-menopausal women with uterine polyps, of which 60% had AUB symptoms. Overall, reduction in periodic blood loss was not demonstrated at 6 months following surgical excision but IMB symptoms were significantly improved .
Another RCT attempted to determine the significance of uterine polyps on the risk of recurrent PMB by randomising women with PMB to either polypectomy or expectant management. However, because of the lack of equipoise of both patients and their clinicians hindered recruitment, the trial could not be completed. The authors then redesigned their RCT to overcome the observed reluctance to leave a hysteroscopically detected uterine polyp in situ . Women with PMB and a thickened endometrium on TVUS and a subsequently benign endometrial biopsy were randomised to undergo further testing with a hysteroscopy without further investigation. Women allocated to hysteroscopy who were found to have a uterine polyp had it duly removed. The prevalence of the same polyp was assumed in the expectant group, and hence, the impact of polypectomy on recurrent PMB at 12 months could be assessed. Nearly, one in five women experienced recurrent PMB over the year, but differences in the prevalence of recurrent PMB were not observed between the groups. Thus, expectant management on symptomatic grounds seems a viable option as opposed to hysteroscopic polypectomy. The RCT was however underpowered. A highly interesting finding was a 6% incidence of atypical hyperplasia or cancer in the hysteroscopically removed polyps. Thus, hysteroscopic polypectomy appears to be indicated to aid diagnosis of serious endometrial disease but not to alleviate bleeding symptoms.
Medical management
Evidence supporting the use of medical treatment of uterine polyps is lacking, although hormonal treatments are widely used to treat menstrual problems of which some will be associated with uterine polyps. Pre-treatment with gonadotrophin-releasing hormone analogues (GnRH-as) before hysteroscopic resection of focal pathologies in pre-menopausal women has been reported, but the costs and morbidity of this intervention is unjustifiable given the relative simplicity and success of removing uterine polyps as opposed to more technically challenging submucosal fibroids. Rather than using medical therapies to directly treat polyps, medical treatments have been examined to prevent their formation. For example, the use of the levonorgestrel-releasing intrauterine system (LNG-IUS) in women taking tamoxifen may reduce the incidence of endometrial polyps .
Surgical management
The surgical removal of intrauterine polyps can be achieved blindly or under direct hysteroscopic vision.
Blind polypectomy
Blind removal of uterine polyps utilising ‘D&C’ under general anaesthetic or avulsion with polyp forceps has yet to be fully consigned to its rightful place in the history books. National surveys from the UK and the Netherlands , albeit from a decade or so ago, still indicate that such practices are common place. The technique involves wide dilatation of the cervix and the use of standard surgical polypectomy forceps to explore the uterine cavity. These approaches can be associated with potential uterine trauma and visceral trauma . Incomplete removal of polyps is also are well recognised . Most gynaecologists would perform a hysteroscopy beforehand to locate the polyp to direct blind avulsion of the lesion followed by curettage .
Hysteroscopic uterine polypectomy
Advances in hysteroscopic technology have enabled polyps to be removed under direct vision using fine mechanical and electrosurgical equipment, which are passed down a 5- or 7-French working channel of a rigid, continuous-flow-operating hysteroscope , and more recently, the introduction of bespoke tissue removal systems (TRSs) . All these techniques require hysteroscopic visualisation within the uterine cavity, excision of the polyp tissue from the uterine wall and retrieval from the uterine cavity. Traditional approaches to insert the hysteroscope have entailed the use of a vaginal speculum, cervical instrumentation and local anaesthesia. However, vaginoscopic approaches are increasingly being adopted to avoid the use of vaginal instrumentation, thereby minimising pain and allowing the operator greater degrees of movement externally to facilitate manipulations within the uterine cavity during the surgical procedure .
Mechanical
Uterine polyps can be removed by excision or avulsion using a variety of fine mechanical instruments in isolation or combination. Such ancillary instruments include reusable scissors, biopsy cups, forceps and disposable polyp snares ( Figure 2 ). The safety, feasibility and efficiency of such approaches have been well reported . However, fragility of the instruments preclude the excision or avulsion of large or fibrous polyps and bleeding may occur . The limited degree of movement of these instruments can also limit feasibility. The diameter of the cervical canal relative to the excised polyp creates challenges to removing pathology as polyp tissue may slip from the small-diameter-grasping forceps when attempting retrieval down the endo-cervical canal. In such circumstances, the operator will have to consider blind dilatation of the cervix to allow further attempts at hysteroscopic retrieval with grasping forceps or the use of polyp snares. Recourse to blind retrieval of a detached polyp using standard polyp forceps should not be attempted routinely. In the authors’ opinion, blind retrieval in this way should only be attempted where histological examination of a suspicious-looking polyp is considered necessary because the risks of uterine and intra-abdominal trauma are not insignificant, especially if the procedure is to be performed under general anaesthesia . In general, hysteroscopic polypectomy using small-diameter mechanical instruments should be limited to smaller, glandular polyps.
Electrosurgical
The application of electricity has enhanced the cutting potential of hysteroscopic instruments, thereby overcoming the limitations of fine mechanical technologies, namely the removal of larger and more fibrous polyps. Large-diameter hysteroscopic resectoscopes can be used under general anaesthesia or conscious sedation. The resecting loop can be used to remove the polyp in strips with repeated passes of the cutting loop or en bloc by cutting the base of the polyp where it adjoins the uterine side wall. However, although the use of formal resectoscopes is quick and effective, a greater degree of specialised hysteroscopic skills is required . Cervical and uterine trauma can result from the need for blind cervical dilatation along with complications occurring from fluid overload and inadvertent electrosurgical injury . Moreover, given the fully intra-cavity position and generally soft nature of most polyps, the resectoscope is a somewhat ‘overpowered’ technology and in the authors’ opinion, these technologies are better used for the removal of firmer and more deeply sited SMFs. The use of a 5.3-mm (16 Fr gauge) mini-resectoscope has been described for use in the inpatient and outpatient setting, but this technology has not been widely adopted .
Advances in hysteroscopic electrosurgical technologies include improvement in visualisation and image quality, the development of safer bipolar systems and miniaturisation of equipment such that uterine trauma can be minimised and procedures performed without the need for general anaesthesia or sedation. A miniature bipolar electrosurgical system ( Figure 3 ) has been developed (Versapoint, Gynecare, Ethicon, Somerville, NJ, USA) to excise polyps, and the safety, acceptability and feasibility of this approach has been reported . The Versapoint ® bipolar electrodes are for single use but reusable electrodes are available, e.g., the bipolar dissection electrode (Karl Storz, Tuttlingen, Germany). However, as with small-diameter mechanical instruments, retrieval of larger or fibrous tissue specimens from the uterine cavity, especially where the diameter of the endo-cervical canal is restricted, remains problematic. Mechanical instruments such as hysteroscopic graspers, biopsy cups and snares are required to retrieve the specimen with the limitations as outlined in the preceding section. In general, the authors favour the efficient en-bloc electrosurgical removal of a polyp by detaching it directly from its base but where larger or fibrous polyps are identified, and then piecemeal removal by cutting several strips of tissue may be preferable because difficulty with retrieval from the uterine cavity can be anticipated. However, such an approach is more cumbersome and usually results in a longer intrauterine operating time, which can be disadvantageous, especially in an outpatient setting .
