Abstract
Objective
To explore the views of members of the British Menopause Society on the management of women with unscheduled bleeding on hormone replacement therapy.
Study design
An electronic cross-sectional questionnaire survey.
Main outcome measures
Investigations, treatment options and preferences for the management of women with unscheduled bleeding on hormone replacement therapy.
Results
A total of 91/178 (51%) clinicians investigate patients with unscheduled bleeding within three to six months of starting sequential hormone replacement therapy (seq-HRT) versus 83/178 (47%) for continuous combined hormone replacement therapy (con-HRT). A total of 52/178 (29%) versus 54/178 (30%) would investigate unscheduled bleeding continuing beyond six months while 18/178 (10%) versus 26/178 (15%) would investigate within three months. Assessment is requested as urgent by 88/176 (50%) clinicians, routine by 47/176 (27%) and a two-week-wait-suspected cancer referral by 41/176 (23%). A total of 97/178 (55%) clinicians would continue seq-HRT and refer versus 117/178 (66%) for con-HRT. A total of 46/178 (26%) clinicians would change the progestogen preparation in women with unscheduled bleeding on seq-HRT. For women on con-HRT, 12/178 (7%) clinicians would change to seq-HRT and 8/178 (5%) to the Mirena IUS. The Mirena IUS is the preferred progestogen for 81/178 (45%) of clinicians when prescribing hormone replacement therapy.
Conclusions
There is a varied approach in the practise amongst British Menopause Society members to managing women with unscheduled bleeding on hormone replacement therapy. Further research is needed to determine the optimal assessment pathways for women with unscheduled bleeding on hormone replacement therapy.
Introduction
Unscheduled bleeding on hormone replacement therapy (HRT) is defined as abnormal bleeding on sequential or continuous combined HRT six months post initiation of therapy or after a period of established amenorrhoea. 1 It is common to experience an initial phase of irregular bleeding in the first six months of initiating treatment while persistent bleeding after this period of treatment should always be investigated. 2
Unscheduled bleeding is a common complaint which can lead to up to 50% of women ceasing their HRT within 6–12 months following an episode. 3 With complaints of unscheduled bleeding reported in 38% of women on seq-HRT and 41% of women on con-HRT, 4 it is important for clinicians to reassure women at low risk of pathology 5 whilst appropriately managing those who need further investigations.
The initial investigations of the woman with unscheduled bleeding on HRT include a thorough pelvic examination assessing for atrophic vaginitis, vulval condition, e.g. lichen scleorus/planus and cervical polyps. Commonly a transvaginal ultrasound scan (TVUS) is carried out to assess endometrial thickness (EMT) and morphology and to assess both the uterus and the adnexa for endometrial/ovarian/tubal tumours. 5
The British Gynaecological Cancer Society Uterine Cancer guideline 6 advise that an outpatient endometrial biopsy should be obtained at an endometrial thickness ≥5 mm in women with unscheduled bleeding on HRT. In women on seq-HRT, the EMT varies depending on the phase of the cycle. Persistent changes in timing/severity of a withdrawal bleed six months post initiation warrants investigation with a transvaginal ultrasound, ideally following a withdrawal bleed. An EMT cut-off of 5–8 mm has been quoted in the literature as the indication for endometrial biopsy.6–8 Several guidelines exist to guide clinicians in their investigation and management of this clinical issue.6,9,10 This had led to a varied approach to the management of unscheduled bleeding on HRT amongst practitioners.
The aim of this survey was to assess the current practice amongst British Menopause Society (BMS) members on the management of women who present with unschedueled bleeding on HRT.
Methods
An electronic cross-sectional questionnaire survey was sent to the members of the BMS by email between March and June 2018. The survey assessed a range of areas including when a clinician would investigate a patient with ongoing unschedueled bleeding on seq- or con-HRT and when they would refer them for investigations (routine/urgent/two-week wait); what management plan they would put in place whilst the patient was referred for investigations, e.g. ceasing HRT or changing preparation; at what cut-off an endometrial biopsy would be obtained and their preferred progesterone preparation (Table 1).
Results
A total of 178/659 (27%) questionnaires were returned by BMS members and all of these were analysed and included in the study. Depending on number of responses, the denominators for the above questions vary. Consultant gynaecologists made up the largest group of total responders with 65/173(38%). This was followed by General Practitioners (GP) 51/173 (29%), Specialist Doctors 33/173 (19%), Nurse Specialist 10/173(6%) and Junior Doctors 1/173 (0.5%). A total of 13/173 (7.5%) identified as other (Consultants in Sexual and Reproductive Health, Consultant nurses, retired healthcare professionals). Of all respondents, 65/173 (38%) practised in Primary Care, 56/173 (32%) in specialist menopause services in secondary care and 32/173 (18%) in hospital – general gynaecology. A total of 20/173 (12%) answered as ‘other’.
The initial management of patients on seq- or con-HRT with ongoing unscheduled bleeding for most of clinicians was to continue HRT whilst referring for further investigations (97/178 (55%) in seq-HRT and 117/178 (66%) in con-HRT). For those on seq-HRT, 46/178 (26%) of clinicians would change the progestogen preparation first. A smaller proportion of 19/178 (11%) would stop HRT and investigate whilst 2/178 (1%) would change the progestogen preparation to a Mirena IUS. For patients on con-HRT, 23/178 (13%) of clinicians would stop HRT whilst investigating the cause of ongoing unscheduled bleeding. A total of 12/178 (7%) would change patients on con-HRT to seq-HRT and 8/178 (5%) to the Mirena IUS.
The majority of clinicians would refer patients for further investigations between 3 and 6 months from onset of unscheduled bleeding (91/178 (51%) in seq-HRT and 83/178 (47%) con-HRT). Nearly one-third would refer after six months in both groups 52/178 (29%) in seq-HRT and 54/178 (30%) in con-HRT. The time for referral following unscheduled bleeding on HRT by clinicians is summarised in Table 2.
The survey questions assessed.
Referral for further investigation following presentation with unscheduled bleeding in patients on HRT.
The preference for the type of progestogen when prescribing HRT is shown in Figure 1. When initially prescribing HRT, the Mirena IUS is prescribed by 81/178 (46%) clinicians as a first option. Oral micronised progesterone prescribed by 46/178 (26%) clinicians was the second choice with oral dydrogesterone by 10/178 (6%), transdermal norethisterone by 7/178 (4%), transdermal levonorgestrel by 7/178 (4%), oral norethisterone by 5/178 (3%), and oral Provera by 4/178 (2%) as the other first choices. Approximately 18/178 (10%) did not have a direct preference but would tailor to patient choice.

Clinicians’ preference of progesterone when prescribing HRT.
Clinicians’ investigations as part of the management of women with ongoing unscheduled bleeding on HRT are shown in Table 3.
Investigation practises of clinicians treating patients who present with unscheduled bleeding on HRT.
Discussion
This survey assessed the views of members of the British Menopause Society on the management of unscheduled bleeding on HRT. This is a challenging clinical problem to manage and may require changes to the progestogen component of the HRT regimen and further investigations including TVUS and endometrial biopsy whilst reassuring a likely anxious patient.
Main findings
The survey revealed that there is a varied approach amongst BMS members in managing women with unscheduled bleeding on HRT.
When presented with ongoing unscheduled bleeding, more than half of clinicians will continue HRT whilst referring the patient for further investigations within three to six months of presentation. Half of these referrals would be through an urgent referral (six weeks), and the remaining through either a routine or a two-week wait/cancer pathway. For those on seq-HRT, nearly a third of clinicians would initially change the progestogen preparation prior to referring for further investigations. For patients on con-HRT, only 1 in 10 clinicians would initially consider switching a patient with unscheduled bleeding to seq-HRT. Although the Mirena IUS did not appear to be the preferred option for managing patients with unscheduled bleeding on HRT, it was the progestogen of choice when commencing HRT. Following a TVUS, two-thirds of clinicians would obtain a biopsy at an endometrial thickness of >5 mm. Approximately a third of clinicians would await six months before further investigating. However, most clinicians would be influenced by the number of unscheduled bleeds, the duration since normal bleeding pattern, the duration of amenorrhoea and the heaviness of unscheduled bleeds when referring patient for further investigations.
Strengths and limitations
This was a subjective analysis of the clinical practices amongst BMS members which provides an insight into the current management of women with unscheduled bleeding on HRT. There was a good response rate for our survey with nearly a third of the cohort completing and returning the survey; the greater proportion being senior clinicians who have years of experience in menopause health.
As questions could be left unanswered whilst completing the survey, a total of five questions were left unanswered by a minority of respondents which left us with uncaptured data. This may either have been secondary to difficulty in interpreting the question, clinicians missing the question or survey fatigue.
Interpretation
Persistent unscheduled bleeding on HRT beyond six months warrants investigation with ultrasound scan and endometrial biopsy if clinically indicated. 11 Counselling patients who take seq- or con-HRT about unscheduled bleeding in the initial six months is crucial. Women need to be informed to report unscheduled vaginal bleeding at the three-month review appointment or promptly if it occurs after the first three months. 9 This will allow the clinicians to enquire about the nature of the bleed, assess any compliance issues and offer reassurance.
Patients on seq-HRT have an expected withdrawal bleed at the end of the progestogen phase of their HRT which should not be unduly heavy or unpredictable – if such, investigation would be warranted. Women who are amenorrhoeic after six months of HRT have a 90% chance of remaining amenorrhoeic. 12 Therefore, women with ongoing unscheduled bleeding need to be referred for investigations six months post initiation of therapy or after established amenorrhoea. 10
Endometrial cancer is estimated to be present in 3–10% of women on no HRT who present with post-menopausal bleeding. 6 In women on HRT, the risk has been estimated to be 1–2% as reported in previous studies. 2 , 5 The Million Women Study 13 reported a reduced risk of endometrial cancer in those on con-HRT, and no difference in those on seq-HRT as compared to women on no HRT.
The management of unscheduled bleeding on HRT compromises of (1) excluding underling pathology; (2) ceasing the troublesome bleeding by modifying the progestogen component of the HRT regimen and (3) reassuring patients who have a low risk of disease. Patient compliance has been shown to decrease significantly with persistent bleeding and invasive investigations with up to 70% of patients choosing to discontinue treatment. 3
Continuing or ceasing HRT and referral pathways
Studies have shown that endometrial pathology is low in women who present with unscheduled bleeding on HRT, for both those on seq- as well as con-HRT regimens. In patients with abnormal TVUS results (EMT ≥5 mm or polyps), results in our recent retrospective study revealed that the risk of atypical hyperplasia/endometrial cancer was low (2% and 1% for seq-HRT versus con-HRT). 5
As the risk of malignancy is low in women who present with unscheduled bleeding, it would be reasonable to continue HRT while referring them for further investigations. Any hormonal-induced changes in the endometrium secondary to HRT use can be interpreted by the pathologist given sufficient information. 7 If HRT is ceased and depending on which referral pathway the patient is on, they may be waiting for several weeks prior to further investigations, hence leading to recurrence of postmenopausal symptoms. Furthermore, ceasing HRT may cause alarm to patients and stopping of therapy altogether. There is currently no consensus as to which pathway to refer these patients on, unlike women who present with postmenopausal bleeding under the as a two-week wait.14,15
Investigations for unscheduled bleeding on HRT
Initial investigations of the woman with unscheduled bleeding on HRT includes: a pelvic examination; vulval examination for signs of vulval conditions such as urogenital atrophy, lichen sclerosus/planus or vulval cancer; speculum examination to rule out cervical polyps or cervical cancer and to obtain a cervical smear test if indicated as well as a genital infection screen if the history suggests this. Further investigation when warranted includes a transvaginal ultrasound scan (TVUS) to assess endometrial thickness (EMT) and morphology, the uterus for pathology and the adnexa for ovarian/tubal tumours. 5
In women who present with post-menopausal bleeding, a normal EMT has been agreed at a cut-off of <4 mm on TVUS. An outpatient endometrial biopsy is obtained in women with unscheduled bleeding on con-HRT when EMT ≥5 mm. 15 A careful assessment of the morphology of the endometrium and the timing of a withdrawal bleed is made in those patients on seq-HRT. 4
Endometrial biopsy and the histopathological outcomes
The EMT threshold for which to obtain an endometrial biopsy in women with unscheduled bleeding on HRT is not well defined. The probability of malignancy with an EMT <5 mm is less than 1% in women with post-menopausal bleeding. 16 As there have been no large prospective studies to determine a cut-off EMT in women with unscheduled bleeding on HRT, an EMT of <5 mm can be used to reassure and safely discharge women on both seq- and con-HRT. The most common pathology found in patients with unscheduled bleeding on HRT is endometrial polyps. Patients on con-HRT have a significantly higher proportion of insufficient samples on histology. 5
Choice of progestogen in HRT
Breakthrough bleeding on HRT commonly occurs after initial commencement, during the changeover from seq-HRT to con-HRT and after three to five years on the con-HRT due to sloughing of atrophic endometrium. 14
Postmenopausal women with an intact uterus using estrogen therapy should receive a progestogen for endometrial protection. Most data suggest that micronised progesterone is equally effective as synthetic progestogens at protecting the endometrium from the proliferative effects of estrogen. 10
There is a vast combination of HRT preparations in use. However, there is no dictation as to which type, the dosage, route or duration and an individualised plan, is required for women on HRT. Progestogens bind not only to the progesterone receptor but also to the glucocorticoid, mineralocorticoid and androgen receptors. This can cause side effects such as adverse glucose metabolism, fluid retention, acne, hirsutism and weight gain.11,17
Levonorgestrel (LNG) is a potent progestogen and can either be administered orally, transdermally or in the form of the intrauterine system (Mirena). It has the benefit of having minimal glucocorticoid and anti-mineralcorticoid activity which avoids fluid retention found with some progestogen preparations. 17 Moreover, more than 90% of women on HRT with the Mirena IUS become amenorrhoeic. The benefit of reduced systemic side effects and improved compliance with an intrauterine system makes the vaginal route of progestogen administration favourable. 11
Oral micronised progesterone and dydrogesterone have an anti-mineralcorticoid effect which allows for a better side effect profile. 18 The risk of endometrial hyperplasia with oral micronised progesterone was assessed by Stute et al. In this systematic review of 40 studies, micronised progesterone was found to offer endometrial protection for up to five years if taken orally 12–14 days/month at a dose of 200 mg/day or vaginally 45 mg/day at 4%. On the other hand, transdermal micronised progesterone does not provide sufficient endometrial protection. 19
Norethisterone (NET) is a progestogen which is a testosterone derivative. It has weak androgenic effects with minimal mineralcorticoid or glucocorticoid activity. It is a metabolite of norethisterone acetate (NETA) when taking orally. 18 Oral medoxyprogesterone (Provera) was assessed in the 2012 Cochrane systematic review which recommended a minimum dose of medroxyprogesterone acetate (MPA) of 1.5 mg/day in a continuous combined HRT regimen or NETA in a dose of 1 mg/day. It also concluded that NETA given in a dose of 0.1 mg or above provides adequate endometrial protection within a continuous combined HRT regimen. 20
Unscheduled bleeding on sequential HRT with negative investigations
In perimenopausal women or those who have had their last menstrual period less than one year ago, seq-HRT helps balancing endogenous and exogenous progestogens. 11 Progestogens induce secretory transformation in the endometrium, mimicking the activity of endogenous progesterone in the luteal phase of the menstrual cycle and providing protection from hyperplasia and endometrial cancer. 19 In any cycle, the commencement of the luteal phase is consistent, either at day 15–17 for a 21-day cycle or at day 21 for a 35-day cycle, and the progestogen phase with HRT should coincide with this. 7 If endogenous progesterone is not suppressed in these patients, bleeding may occur bi-monthly. Changing or increasing the dose of progestogen or changing to the Mirena IUS may resolve this issue.7,11,12,21
Equally an unstable endometrium due to lack of stromal formation can lead to spotting. Therefore, an increase in the estrogen dose may help resolve the unscheduled bleeding. 7
If the withdrawal bleed is heavy or erratic, doubling the dose of progestogen or increasing the duration to 21 days can help. 22 Poor compliance can be reduced by an estrogen and progestogen in a single preparation.
Unschedueled bleeding on continuous combined HRT with no pathology
In women who have commenced con-HRT and experience ongoing unscheduled bleeding, the dose of progestogen could be increased, and this results in controlling the bleeding in a significant proportion of women. If they continue to experience ongoing unscheduled bleeding, the HRT regimen could be changed to a cyclical intake of progestogen.7,11,14,21
For cyclical HRT regimens, increasing the progestogen dose or duration of progestogen intake (progestogen for 14 days a month or for 21 days out of a 28-day HRT intake cycle) can be considered.
Conclusion
The management of unscheduled bleeding on HRT is varied amongst BMS members. It is reasonable to continue HRT while referring women with unscheduled bleeding for further investigations. The risk of endometrial cancer in women with unscheduled bleeding on HRT is significantly lower than in women with postmenopausal bleeding not on HRT. This is especially true for those who had not been experiencing bleeding before commencing HRT and who are taking progestogen. Further research is needed to determine the optimal assessment and referral pathways for women with unscheduled bleeding on HRT.
Footnotes
Authors’ contributions
SAI and HH conceived the study. SAI researched literature and wrote the first draft of the manuscript. HH and NP reviewed and edited the manuscript. All authors approved the final version of the manuscript.
Acknowledgements
The authors wish to thank the British Menopause Society for distributing the electronic survey to their members.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
