Guidelines update: Unscheduled bleeding on HRT

GP and women’s health specialist Dr Toni Hazell provides an overview of new guidance published by the British Menopause Society on the management of unscheduled bleeding on HRT

HRT prescribing has increased drastically over the last few years – from 2021/22 to 2022/23, the number of HRT items prescribed increased by 47%, given to 2.3 million (29%) more women, an increase which continues a trend that started in 2010. With increased use of any drug comes an increase in adverse effects such as unscheduled bleeding with HRT; this guideline is welcome clarification about the management of this symptom.

Unscheduled bleeding after starting HRT is common, and generally only considered a cause for concern if it lasts 3-6 months, or is heavy or starting after a period of amenorrhoea, depending on the type of HRT being used. The main worry for this cohort is the risk of endometrial hyperplasia or cancer, although other causes such as sexually transmitted infection (STI) or genitourinary syndrome of the menopause (which may need a combination of systemic and vaginal HRT to be fully controlled) should also be considered. A good history and relevant examination, including visualisation of the cervix, will help to assess the likelihood of the various differential diagnoses. A temporal connection between starting HRT (or changing the dose) and the bleeding starting is one factor which might indicate that the two are connected.

Unscheduled bleeding in women with HRT was a presentation which sometimes suffered from a lack of an obvious pathway. It isn’t specifically listed in the NICE guideline on suspected cancer, but in the absence of easy access to a swift ultrasound scan, or an alternative pathway, a suspected cancer referral was often the only way to get your patient seen in weeks rather than months. Over the last three years there has been a 43% increase in suspected cancer referrals in this cohort, but only a 2% increase in the diagnosis of endometrial cancers. This is therefore not an efficient use of resources; with widespread reports of missed targets for the start of cancer treatment, targeting suspected cancer referrals more accurately may speed things up for those women who do have cancer.

This new guideline gives us a new way to stratify risk and therefore investigate and refer more appropriately. The guideline was written by the British Menopause Society (BMS), in partnership with other organisations including the RCGP and the Royal College of Obstetricians & Gynaecologists, and Getting It Right First Time (GIRFT). As well as unscheduled bleeding, it also covers the use of high doses of oestrogen and how we should adjust the progestogen dose to protect the endometrium of women using these regimes.

We will have to get used to taking a more structured history than in the past, as the guideline bases management on the presence or absence of six major and seven minor risk factors (see box 1 below), which will all need to be documented. The most common of these are likely to be obesity, diabetes and polycystic ovary syndrome. Use of oestrogen only HRT in a woman who has a uterus is a risk factor – this should never happen and if it is discovered, a significant event analysis should be carried out and systems changes made to prevent recurrence. Other risk factors are less common, including Lynch or Cowden syndromes, the use of sequential HRT for more than five years, and the use of tricycled HRT (where the progestogen is taken for a short period every three months). The latter, with two weeks of a progestogen every three months, is listed as a possible regime by NICE CKS, but this new BMS guideline notes that a 7-10 day course of some of the older progestogens used in HRT, given only every three months, increases the risk of endometrial hyperplasia and cancer, from 1.4% to 11% at three years. Data are lacking for newer progestogens. When used in the private sector, tricycled HRT is often accompanied by six-monthly surveillance scans of the endometrium, but the evidence for this is also unclear.

To use this guideline for risk assessment, you need to know the following, and have access to the flow chart from the guideline:

• How many risk factors the woman has.
• What type of HRT is she using – sequential (sHRT) or continuous combined (ccHRT).
• How long has she been taking HRT and how long has it been since any dose changes.

The first decision point is whether the woman has at least one major or three minor risk factors. A woman fitting into this category might have a BMI of ≥40 as her only risk factor, or a BMI of 30 – 39 as well as both polycystic ovarian syndrome and diabetes. These women should go straight into the urgent suspicion of cancer pathway (USCP) – what most of us would know as a two-week wait referral. In my area, such a referral would mean that the woman is seen at a one-stop clinic which includes an ultrasound scan and so I can send just one referral, but some other USCP pathways might involve the scan being arranged in primary care first; if this is done then, as per the flow chart (Figure 1 in the guideline) the scan result can be used to stratify whether the USCP referral is still needed, using the same criteria which will be discussed in a moment for women with fewer risk factors.

For women who don’t have one major or three minor risk factors, the bleeding needs to be considered in more detail. If it has persisted for more than six months after starting HRT or more than three months after a change of dose, or is heavy, then an urgent transvaginal ultrasound (TVS) should be arranged. This should also be done if she has two minor risk factors, eg, a woman with a BMI of 30 – 39 who has diabetes or polycystic ovarian syndrome but not both. This scan needs to be done within six weeks (commissioning GPs might want to check with their local radiology teams as to whether this can be done) and the next steps will depend on the endometrial thickness (ET). Those taking ccHRT should have a thinner endometrium; an ET of >4 mm should therefore prompt concern, and an USCP referral. For those taking sHRT, an USCP referral only needs to be done if the ET if >7 mm. If the scan is reassuring, HRT should be optimised, and an endometrial assessment (biopsy and/or hysteroscopy) arranged if the bleeding continues for six more months. This strategy would also be used for those who don’t have either two minor risk factors, or bleeding which raises concerns as discussed above – optimise HRT and arrange a TVS if bleeding is ongoing at six months.

What does ‘optimise HRT’ mean in real life? Table 6 in the guideline lays out a raft of recommendations for reducing unscheduled bleeding on HRT. Some are obvious and would apply to other medicines, such as checking adherence. More specific recommendations are set out in box 2 below; consult the guideline for the full list. 

Women being referred for an HRT related adverse effect often ask if they can continue their HRT whilst waiting to be seen. In favour of continuing is the fact that sudden cessation may cause significant menopausal symptoms, not helpful at a time of already heightened stress. On the side of stopping is the fact that the referral is being done due to a suspicion of a hormone related cancer, which makes stopping prescribed hormones seem sensible.

Medicine is an art, not a science, and the guideline doesn’t give us a definite answer on this, suggesting only that we discuss the pros and cons of worsened menopausal symptoms versus extra exposure to hormones. A common sense approach might be that if a woman has been on HRT for many years, an extra few weeks of hormones seems a small extra addition and maybe unlikely to contribute much to her risk, whereas an extra few weeks in a woman who has only been on HRT for six months is proportionally a greater addition to her total oestrogen load. Continuing HRT won’t affect the accuracy of any endometrial assessment by biopsy or hysteroscopy, but the presence of a LNG-IUD may make TVS assessment of ET harder and lower the threshold for needing endometrial assessment. The same goes for women who refuse a TVS; a trans-abdominal ultrasound is less accurate at measuring ET and so further assessment of the endometrium might be needed.

For women with unscheduled bleeding who have no risk factors and do not want an ultrasound, the guideline suggests weaning off HRT over six weeks and offering non-hormonal options instead. If the bleeding stops, and a resumption of HRT is not wanted, no further action is needed. If bleeding continues without the HRT, an USCP referral will likely be needed. If the bleeding stops without HRT, but the woman wants to resume taking HRT, the guideline should be consulted for the best way to start HRT to reduce the risk of bleeding (eg, start at a low dose, use a LNG-IUD) and the pathway followed again if bleeding restarts.

The guideline also discusses the risks of endometrial cancer in women who take HRT and do not have unscheduled bleeding. The relative risk of endometrial cancer for various uses of HRT is set out below; ccHRT users have a lower risk because the endometrium atrophies under continuous suppression. Women who use sHRT should switch to ccHRT after a maximum of five years, or at the age of 54 (whichever comes first) – this switch can be discussed from 12-18 months after starting HRT. Data are largely based on older progestogens such as norethisterone and medroxyprogesterone acetate – we do not yet have data for newer preparations such as micronised progesterone. The 52 mg LNG-IUD gives protection against all doses of oestrogen – only the Mirena brand has this licence, which is for four years, however FSRH guidance and this BMS guidance say that any brand of 52 mg LNG-IUD can be used off-licence for this indication for five years. The other brands of 52 mg LNG-IUD currently available in the UK are Levosert and Benilexa – the lower dose LNG-IUDs, Kyleena and Jaydess should not be used for this indication.

Table 1. Risk of endometrial cancer with HRT use

The guideline is clear that there are few data for the use of high doses of oestrogen but takes the pragmatic view that the progestogen dose should be proportional to the oestrogen dose. They list the oestrogen doses of various preparations which are considered to be ultra-low, low, standard, moderate or high and advise increasing the progestogen dose accordingly. See Tables 2 and 3 in the guidelines for more details on this.

This is a massive guideline – 47 pages long – which introduces extra complexity in the way that we manage HRT. For those who regularly prescribe HRT it is a must-have on your desktop, and I expect to be consulting it regularly in the future.

Dr Toni Hazell is a GP in north London and GPSI in women’s health