21.     Post-menopausal bleeding:

There is a good summary at SIGN. 

Most cases end up with no diagnosis or atrophic changes as the explanation of the problem.

It is said that malignancy (mostly endometrial) is present in ~ 10% of cases if the women are not taking HRT.

But I suspect that the real figure is lower.

The figure is ~ 1% for women on HRT – presumably most of the abnormal bleeding in this group is down to the HRT itself.

Technically the bleeding is from the genital tract, so the purist answer must be that it cannot be due to bladder pathology.

In reality, the patient presenting with apparent PMB may have bladder or bowel pathology and this must always be considered in cases of recurrent bleeding.

Likewise ovarian cancer, some 5 - 10% of which is associated with vaginal bleeding.

In question “c”, the answer is “no”, as one can always imagine a scenario in which investigation is not indicated.

She might be ninety nine and terminally ill from some other pathology!

As a rule “always” means a negative answer.

Cervical “erosions” are uncommon after the menopause and not particularly a cause of bleeding – better to use the term “ectropion”.

Anticoagulation increases the risk of bleeding, but not of malignancy.

Tamoxifen is a SERM (see MCQ 9, question 20).

It is used for patients with breast cancer because of its anti-oestrogenic effects on breast tissue.

Perversely, it acts like oestrogen as far as the endometrium is concerned, so benign and malignant change are more common.

This illustrates the fact that they are Selective Oestrogen Receptor Modulators – they affect different oestrogen receptors in different ways.

The same drug might block breast receptors but stimulate those in endometrium and bone.

CIN is a histological diagnosis of intra-epithelial abnormality.

The cervix looks normal until stained at colposcopy and is not more prone to bleed.

Easy bleeding when taking a smear may indicate infection.

The investigation of PMB has been transformed by transvaginal scanning but you need to start with her history.

This should be done with your brain in gear.

What clues are there that point to malignancy?

What clues are there that point to a benign explanation?

What things are there that might impact on her investigation or management?

Risk factors are many and include:

long interval from the menopause,

multiple bleeds,

prolonged or heavy bleeding

and drugs such as unopposed oestrogen or Tamoxifen.

Most people forget membership of a HNPCC family (See MCQ1, question 13) and previous treatment for genital tract malignancy or pre-malignancy.

The usually quoted figures for endometrial malignancy in women with PMB are 10% for those not on HRT and 1% for those on it.

Continuous combined HRT (CCHRT) reduces the risk even more, but few women are on it.

Wells et al, from Sheffield, followed 534 women, including 21 with some degree of endometrial hyperplasia for up to 5 years after taking CCHRT (average duration of usage 4.4 years).

They found no cases of hyperplasia or malignancy during follow-up. BMJ. 2002 Aug 3;325(7358):239.

Anticoagulants might explain the bleeding.

So might the acquisition of an energetic toy boy after many years of abstinence!

Even with an apparent explanation you would still do the basic investigations to exclude malignancy.

Investigation may be complicated by virginal status and reluctance to have vaginal examination, COAD, unstable angina etc.

Examination will show local causes such as advanced cervical malignancy and benign lesions like polyps, atrophic changes etc.

A smear is necessary if not done recently.

TV scanning is the magic.

There is no agreement about the cut-off value, 5 mm, 4 mm or 3 mm.

If the scan is normal and the endometrium is < 5 mm, the risk is reduced by a factor of about ten.

If the endometrium is thin, it becomes reasonable to adopt a “wait and see” policy, so long as the patient is informed and in agreement.

Further bleeding merits fuller investigation and I would normally do a hysteroscopy.

This may change as some advocate repeat tv scanning followed by a scan with saline infused into the endometrial cavity as a contrast medium.

It is claimed that this will sort out those who need hysteroscopy from those who do not.

Some use 4mm. as the cut off point, with an even greater reduction in the risk of malignancy being present.

Some use 3 mm with higher risk patients, e.g. those on tamoxifen.

Remember to check what the ovaries look like – ovarian malignancy can present with post-menopausal bleeding.