Practice observed

Six years ago I had an email from a colleague which began

‘I hope that you don’t mind me picking your brains once more about HRT?’

The answer to this is, no I don’t. One of the ways we manage the mismatch between NHS demand and clinic availability is that I am funded to the tune of one session a month to respond to queries from my local colleagues. It is educational for the colleagues involved, improves the service available to patients and means that only those patients who really need to be, are seen face to face. Remuneration may not truly reflect the time involved but recognises both the importance of professional support and that commissioning should take account of what providing this type of service actually requires. It means that I do not feel too put upon.

The email continued: ‘I have a 42 year old lady with Turners syndrome who has osteoporosis. She has been on raloxifene for four years but it is not improving her bone density enough. The rheumatologists have recommended oestrogen replacement rather than raloxifene and not a bisphosphonate. I am unclear why. She has been on HRT previously but is nervous to try again due to lethargy, mood swings and unpleasant period related experience before’.

Education does make a difference. This GP had recognised that the different cyclical combinations this patient had previously tried all contained norethisterone as the progestogen component. While she thought that a continuous regimen might be the best option, she was not sure which progestogen to choose or whether to consider a levonorgestrel intrauterine system.

With good understanding such as this, we can work together. My return email summarised the available data which show that estrogen is more effective and therefore preferred to the selective estrogen receptor modulator. I also outlined principle of reserving bisphosphonates for later use. The data to support limiting the duration of use were just beginning to emerge at that time. I did not think that a rudimentary uterus would be likely to accommodate a levonorgestrel intrauterine system and gave a list of alternatives. As a young woman who already had a problem, we were aiming for an optimal effect:

To try another oral combination but with an alternative class of progestogen and suggested estradiol 1 mg + dydrogesterone 5 mg (Femoston Conti®, Abbott Laboratories) to start at one but with the potential to increase (off license) to two daily (=2 mg estradiol) if tolerated for maximal bone response.

To try Tibolone 2.5 mg (Livial®, MSD) one daily for a completely different side effect profile and low risk of bleeding.

Transdermal combined patches delivering estradiol (50 µg/24 h) + norethisterone on the basis this would avoid gastrointestinal and hepatic first-pass effects.

In real life, things do not always go according to plan and somehow this was never enacted as the woman’s father orchestrated a referral from another GP and she ended up in the clinic anyway. My referral warned me about his overprotective nature and advised that the woman should be allowed to speak.

Sarah had no classical estrogen deficiency symptoms, and this was unsurprising as she had not undergone spontaneous menarche. She has been treated for two to three years in her teens (though we do not know what with) to allow development of secondary sexual characteristics but she had disliked the bleeding that resulted and stopped. She had had no spontaneous bleeding at any point. In addition to she was hypothyroid (on replacement therapy) and hypertensive (triple agents needed to control – felodipine, lisinopril and furosemide).

Going back through her history revealed that she had no hormonal therapy for over 10 years until 1995 when she had a bone density assessment which resulted in the prescription of estradiol valerate 1 mg + cyclical norgestrel (Cyclo – progynova 1 mg, now discontinued) for the next six years. She then was prescribed a monthly cycle estradiol 2 mg + norethisterone 1 mg combination (to deliver a higher estradiol dose). On the basis of mood effects from the progestogen, this was subsequently changed to a three monthly cycle regimen with estradiol 2 mg daily and norethisterone 1 mg for two weeks out of 12. She had disliked the protracted bleeding associated with this and had been changed to Raloxifene 60 mg daily.

Sarah demonstrates the potential for early onset cardiovascular disease as well as poor bone health that result from the hypo-estrogenic state of Turners Syndrome. My assessment was that the potential for interaction with thyroid replacement and established arterial disease meant that non-oral delivery would be preferred.

I discussed the options at length and allowed Sarah to go away and consider what she would be prepared to try as she was very hesitant. Six weeks later, she rang me to say that she would try patches.

We started with a combination patch delivering estradiol 50 µg/24 h + 170 µg/24 h norethisterone (Evorel Conti®, Jansen-Cilag Ltd) but cut into quarters. This would maintain the relative proportions of the constituent hormones. These were to be changed twice a week with the aim of titrating the size up. The logic of this was to avoid physiological effects of sudden reintroduction. While this approach is unlicensed, there is a 14 µg/24 h patch licensed for bone protection in the US so on an empirical basis even this small dose should do something.

Six weeks later, I spoke to Sarah again and she reported that the patches were not sticking adequately. We therefore changed tactic to use Estradiol 0.1% gel (Sandrena®, Orion Corporation) 0.5 mg sachets to use one daily with micronised progesterone 100 mg (Utrogestan®, Marlborough Pharmaceuticals) to be taken at night for opposition. This was tolerated. At four months, she was well and had no bleeding, breast tenderness or mood change. Gradually, we increased to one and half and then two sachets daily. At that level, she had some minimal bleeding and so we dropped back to one and a half (0.5 mg + 0.25 mg) daily and the bleeding petered out. This appeared to be the maximal dose to avoid the most unwelcome of effects for her which is bleeding. The importance of using it at all was emphasised when she fell off a train and broke a bone in her foot. That we were using only half the recommended treatment dose we thought might be mitigated by her small stature. But would it work?

Five years after the previous DEXA scan, it was repeated. Sarah had had low-dose estrogen therapy for four of those years.

My interpretation is that after the 1994 assessment, Sarah took cyclical HRT for about six years. This was at 1–2 mg estradiol daily and resulted in the significant improvement in the years to 2001. She then switched to Raloxifene which is less potent (but did not affect mood or trigger bleeding). There followed a reduction in the years 2001–2007 – though this could have been worse in its absence. The change in machine does make interpretation of change less robust.

I was asked in 2008 to try to come up with an HRT regimen that was effective, tolerated and represented least risk of interacting with the other health issues. We have gradually come to the point of using estradiol gel at 0.75 mg daily (0.5 mg mane and 0.25 mg at night) which just avoids bleeding along with continuous micronised natural progesterone 100 mg at night.

I am happy that the result above is satisfactory – there had been no decrease in L spine in the previous five years and possibly a small increase. While there had been a decrease in L hip, this may well be related to both age-related change and lack of weight bearing activity – which had been compromised. The two readings are now consistent and indeed both are no worse than those of 20 years ago. This is an achievement.

This may be a single example but illustrates how good estrogen is at bone protection. It suggests that low-level replacement when young might have prevented a bone problem from developing. Might it also have been cardioprotective?

How To Cite

Gray S. Practice observed. Post Reproductive Health 2014; 20(2): 80–82. OPEN IN VIEWER