GPs, as the usual first point of clinical contact, are in the unenviable position of having to assess whether or not a post-menopausal woman who presents with bleeding should be referred to gynaecological services for investigation. The main concern is the possibility of underlying endometrial malignancy.
Difficulties have arisen, when making the initial assessment, as a result of differences in clinical opinion as to what constitutes post-menopausal bleeding (PMB) and, in particular, how much time must have passed since a womanÍs last menstrual bleed before bleeding can be considered to be post-menopausal. In addition, patients may be reluctant to be referred to hospital.
This situation has been compounded by uncertainty regarding the relative risk factors for endometrial cancer for users and non-users of hormone replacement therapy following the increase in HRT use in recent years, and also by the increased use of tamoxifen, which is associated with a higher risk of endometrial cancer, in the management of breast cancer.
In parallel with the difficulties experienced in primary care, newer methods of hospital investigation have superseded dilatation and curettage as the principal means of investigation, creating further clinical uncertainty about the most accurate, acceptable and efficient diagnostic approach. These factors have created an urgent need for the SIGN guideline on this complex, but common, condition.
Defining post-menopausal bleeding and when to refer
PMB represents a common clinical problem in primary care, with GP consultations on the increase and at their highest in women aged 50-59 years (14.3 per 1000 of the population in Scotland in 1999-2000).
Traditionally, PMB has represented an absolute indication for gynaecological investigation because endometrial cancer cannot be ruled out by clinical assessment alone. The rise in HRT use over recent years has caused further complications for GPs as they are increasingly having to assess patients for referral with irregular bleeding associated with HRT use. There is also a wide variation in risk factors for endometrial cancer in different groups of women who present with PMB.
To enable all clinicians to achieve a common understanding of what constitutes PMB in non-users and users of HRT, the guideline has clearly defined PMB for all these patient groups (Box 1, below). When considering when to refer it should be noted that the guideline recommends that the risk of endometrial cancer in non-users of HRT with PMB and in HRT users experiencing abnormal bleeding is sufficient to recommend all patients for investigation.
Box 1: Defining post-menopausal (or abnormal) bleeding
Non-users of HRT
Users of HRT sequential regimens
Users of HRT combined regimens
|All patients falling into any of these categories should be recommended for referral|
Assessing abnormal bleeding in women using HRT
Abnormal bleeding in post-menopausal women using HRT can be caused by any of the following: poor compliance (especially related to omission of progestogens); poor gastrointestinal absorption (of oral preparations) because of gastroenteritis for example; drug interactions; coagulation defects and gynaecological disorders.
Clinical enquiry should, therefore, aim to establish whether the bleeding is abnormal and also to identify any other related symptoms or contributory factors associated with endometrial cancer.
When assessing abnormal bleeding in post-menopausal women using HRT, the questions listed in Box 2 (below) should be asked.
Box 2: Questions to ask when assessing patients on HRT with abnormal bleeding
Women presenting with PMB should undergo a pelvic examination at some stage during their assessment. If referred to a gynaecologist, an examination by the GP is not always necessary.
However, examination by a GP or practice nurse can alter the course of clinical management if it expedites referral on grounds of raised suspicion of a malignancy (including cervical carcinoma) or highlights an obvious cause of bleeding (e.g. cervical polyps). This examination may also present an opportunity to take a routine cervical smear if it is due.
There is uncertainty as to whether or not HRT should be stopped before investigation for PMB.
While continuation of treatment before investigation may depend on the patientÍs wishes and how long she has to wait for investigations, current evidence shows that there is no specific reason for discontinuing it.
Risk factors for endometrial cancer
Background incidence in users and non-users of HRT
The guideline recommends that all clinicians should be aware of the background incidence of endometrial cancer among users and non-users of HRT and in those who present with post-menopausal bleeding.
Endometrial cancer is present in approximately 10% of patients referred with PMB. The absolute risk of endometrial cancer in non-users of HRT who present with post-menopausal bleeding ranges from 5.7 to 11.5%. No evidence was identified which determined whether different patterns of PMB, such as one-off or more frequent bleeds, are more or less likely to be associated with endometrial cancer.
Older HRT regimens that used unopposed oestrogen increase the relative risk of endometrial carcinoma by around six times after 5 years of use. Progestogens are added to HRT regimens to prevent endometrial hyperplasia and cancer: their inclusion reduces the relative risk of endometrial cancer to around 1.5.
In women receiving tamoxifen for treatment or prevention of breast cancer the incidence of endometrial cancer is increased by three to six times. The risk of endometrial cancer rises with both higher doses and increasing duration of tamoxifen use. Treatment beyond 5 years increases risk at least fourfold.
Other risk factors
Hereditary non-polyposis colorectal cancer is one of the best recognised cancer family syndromes. Its inheritance is autosomal dominant and it is characterised by a family aggregation of colorectal cancer in addition to extra-colonic cancers of which endometrial cancer is the most common.
The estimated lifetime risk of developing endometrial cancer in women carrying these mutations is between 42% and 60%. Importantly, in contrast to sporadic endometrial cancer, women from such affected families usually develop endometrial cancer pre-menopausally.
The evidence on other risk factors is less robust, but it suggests that there are potential risk groups, i.e. obese patients with diabetes, women with hypertension, and those with a history of hyperoestrogenism (endogenous or exogenous). Examples of the latter include women with early menarche and late menopause. Current evidence is, however, insufficient to quantify this risk as a guide to referral practice.
While the section of the guideline dealing with investigative techniques will be of more direct relevance to gynaecologists and radiography staff, there are some important points which may be of interest to GPs.
Perhaps of most interest will be the recommendation that, where sufficient local skills and expertise exist, transvaginal ultrasound (TVUS) is the first line of investigation to identify which women with post-menopausal bleeding are at higher risk of endometrial cancer.
TVUS and the measurement of endometrial thickness can considerably enhance the level of risk assessment compared with pre-test; an endometrial thickness of <=3mm can exclude endometrial cancer in women who have never used HRT, or have not used any form of HRT for 1 year or more, or who are using combined HRT. In this group the estimated pre-test risk of cancer is 10% and TVUS investigation produces post-test risks of 0.6-0.8% if the endometrial thickness is <=3mm, and 20-22% if the thickness is >3mm.
Similarly, an endometrial thickness of <=5mm can exclude endometrial cancer in women using sequential combined HRT (or who have used it within the past year) with unscheduled bleeding. In this group the estimated risk before TVUS investigation is 1-1.5%; after TVUS the risk is 2-5% if the endometrial thickness is >5mm and 0.1-0.2% if it is <=5mm.
In addition to being more effective and accurate, TVUS may be more acceptable than other methods of investigation to a wider number of women, particularly the elderly, because it is relatively non-invasive.
Linked to this new recommendation regarding TVUS is the recommendation that dilatation and curettage should no longer be the first-line method of investigating PMB in most cases. The guidelines do not recommend a specific course of action if the post-test risk of cancer is high, except that some form of endometrial sampling is essential and best coupled with visualisation of the endometrium by hysteroscopy.
Figure 1 (below) shows the guidelineÍs Quick Reference Guide.
|Figure 1: Front of the Quick Reference Guide containing key points of the guideline|
|Figure 1 (continued): Reverse of the Quick Reference Guide containing key points of the guideline|
|© Scottish Intercollegiate Guidelines Network 2002|
Those interested in reading more about other investigative techniques, such as transabdominal ultrasound, endometrial biopsy and hysteroscopy, or who require further information on the sequencing of investigations and the investigation of women using tamoxifen, should consult the full guideline, available on the SIGN website.
Guideline development and strength of the evidence
Searches were restricted to systematic reviews, meta-analyses, randomised controlled trials, and longitudinal studies. Internet searches were carried out on the websites of the Canadian Practice Guidelines Infobase, the UK Health Technology Assessment Programme, the US National Guideline Clearinghouse and the US National Institutes of Health.
Searches were also carried out on the search engines Northern Light and OMNI, and all suitable links followed up. Database searches were carried out on Cochrane Library, Embase (1985 ¿ May 1999), Healthstar (1975 ¿ May 1999), and Medline (1966 ¿ May 1999).
While there have been numerous publications on the topic of PMB, our review of the literature has shown that there is a paucity of evidence on how best to address it. The literature often fails to differentiate the post-menopausal from the peri-menopausal or the patient with dysfunctional uterine bleeding. Similarly, investigative techniques in many reviews have been applied to a variety of gynaecological problems and the number of cases has often been small, tending to represent local interest, skills or referral patterns.
Consequently, techniques that may be useful and which are currently widely used have an inadequate evidence base to support their recommendations in a guideline. The large quantity of poorly conducted research available stresses the need for future studies to employ a more rigorous methodology.
We are confident that we have produced a guideline that will provide much needed clarity for all clinicians involved in assessing PMB. In particular, we hope that the guideline will increase awareness of the incidence of endometrial cancer in users and non-users of HRT, and believe that the risk in these groups is sufficient to recommend referring all patients who present with PMB to the local gynaecology services for investigation.
- Scottish Intercollegiate Guidelines Network. SIGN 50: A Guideline DevelopersÍ Handbook. Edinburgh: SIGN, 2001.
- Harbour R, Miller J. A new system for grading recommendations in evidence based guidelines. Br Med J 2001; 323: 334-6.
SIGN guideline no 61: Investigation of Post-Menopausal Bleeding is available on the SIGN website: www.sign.ac.uk and can be downloaded free of charge.