Practical guidance for stopping glucocorticoids

[Music] Welcome to the Australian Prescriber Podcast. An independent, no-nonsense podcast for busy health professionals.

Glucocorticoids are a key medicine in many clinical situations for many clinicians. As a rheumatologist myself, I'm very aware how often they can bail us out of difficult situations. But unfortunately, they're often easy to start but harder to stop, especially if our patients have been on them for a while. That's why I'm particularly delighted that Australian Prescriber have an excellent article out now on practical guidance for stopping glucocorticoids from Dr Faran Khalili and Professor Morton Burt, 2 endocrinologists from Flinders Medical Centre and Flinders University in Adelaide. They both joined me today here on the Australian Prescriber Podcast to talk through their findings. Professor Burt, Dr Khalili, welcome and thanks for making the time today.

FK: Good morning, David.

MB: Thanks, David.

So Faran, maybe you can detail out how common the situation this is to see both in endocrinology clinic, but maybe more broadly in the community.

FK: Yeah, David, as I'm sure you're aware and you've experienced it in your own practice, glucocorticoids are prescribed for a myriad of autoimmune and inflammatory conditions that cross over multiple specialties, especially in primary care as well. So it's certainly not uncommon that we see patients that are treated for a protracted period of time with glucocorticoids, firstly to get disease control and then obviously to taper it as the underlying pathology will allow. A lot of these conditions, which glucocorticoids are prescribed, require a duration of therapy of glucocorticoids, which may exceed more than 4 weeks. Some might be for a week, such as exacerbations of COPD [chronic obstructive pulmonary disease], but really having an understanding and an awareness of when patients may be developing hypothalamic pituitary adrenal axis suppression is very valuable for a clinician to know, well, once we've got disease control and our taper is going as planned, what are the extra considerations we need to have from the HPA [hypothalamic–pituitary–adrenal] axis perspective?

In terms of how commonly we see these patients in clinic, they really only come to our attention if something has been identified such as prolonged HPA axis suppression or if they've presented to hospital with an adrenal crisis because of a rapid taper without awakening of the HPA axis. And as a result, we probably come into the piece more towards the end of this process, which is why it's really important to have a proactive approach in how we evaluate these patients, so that things can be done safely to maintain disease control. But also ensure, from a hormone standpoint, that the appropriate taper with appropriate testing and evaluation has taken place.

I'm sure the community burden is enormous, and even though you might only see the pointy end in endocrinology clinic, this is the kind of thing that really affects a lot of clinicians every day. Maybe I could just tease you out though, on the 3 to 4 week question there, why in particular is that important in this case?

FK: Yeah, because there are some disease processes that may only require a period of less than 3 to 4 weeks of glucocorticoid therapy. And although there isn't quite detailed randomised controlled trials looking at this, from the limited evidence that we do have, it doesn't appear to demonstrate that you will get HPA axis suppression with less than 3 to 4 weeks of glucocorticoid therapy. The reason why that is important to know is that in those patients that may have been prescribed glucocorticoid therapy for less than 3 to 4 weeks, then this protracted taper may not need to take place and the glucocorticoid can be stopped abruptly. And of course with education to the patient about symptoms and safety netting as appropriate, but it will enable to potentially bypass protracted exposure and taper to glucocorticoids.

So let's address the hypothalamic pituitary adrenal axis elephant in the room. We can't get away without talking about physiology and relative pharmacological properties here. But Morton, I've been told your descriptions of this are poetic. So why don't we talk a little bit to our audience about the HPA axis and why it's important and how that works here.

MB: Thanks, David. The pituitary gland makes ACTH [adrenocorticotropic hormone] and that stimulates the adrenal gland to make cortisol, which then binds to a receptor and underlies a myriad of effects within the body. So the human body makes somewhere between 10 and 20 mg of cortisol a day, which equates to 2 and a half to 5 mg of prednisolone a day. So if you take prednisolone at a dose above 5 mg a day, as Faran says, for somewhere between 3 and 4 weeks, the body initially responds to that by switching off ACTH production by the pituitary gland, but during prolonged exposure, that leads to atrophy of the adrenal gland.

The consequence of that is that then if you stop your glucocorticoid therapy rapidly, there is the potential that your body's not making sufficient cortisol and that then can precipitate all the consequences that can get from adrenal insufficiency. And as you alluded to in your first question, this is a very common problem. Data from Europe and the United States of America indicate that up to 1% of the adult population at any one time are taking exogenous glucocorticoid therapy. And despite all the new treatments we've got for rheumatoid arthritis and other conditions, what data we have suggests that the prevalence of use is actually increasing rather than decreasing. So it's a very important clinical problem.

Maybe you can talk us through as well some of the pharmacological glucocorticoids that we do use and their relative properties.

MB: So really in Australia, there's probably 3 main glucocorticoids that are taken orally in clinical practice, they can all be used interchangeably. They all act through the same receptor, but in clinical practice, hydrocortisone is most commonly prescribed to treat cortisol deficiency in patients with adrenal and pituitary disease. We tend to use that because that's really identical to the natural hormone cortisol. It has the shortest half-life and it more commonly mimics normal physiology. So the half-life in blood is only up to 2 hours for hydrocortisone, but its biological action is longer than that, lasting up to 12 hours. Prednisolone is most commonly used as an anti-inflammatory agent to treat a range of inflammatory diseases such as rheumatoid arthritis. It has a slightly longer half-life in serum and a duration of action of 24 hours to 36 hours. And dexamethasone is predominantly used in cancer patients to reduce oedema, although there are other indications and it has the longest half-life and duration of action.

So with that information, understanding those relative properties, knowing that I think a lot of the time in clinical practice, in general practice and across a number of different specialties, often we're using prednisolone, prednisone, sometimes dexamethasone, but knowing that we have to convert it to hydrocortisone. Maybe we can talk a little bit about how we might approach a clinical situation where we've got someone who's been on glucocorticoids. Faran, how do we run our way down a therapeutic algorithm as far as getting through and stopping glucocorticoids?

FK: Yeah, sure, David. So essentially, the first priority is to get disease control for whatever the indication for glucocorticoids is, appropriate disease control will obviously be essential, and I'm sure you can speak to that as a rheumatologist yourself. Now, once appropriate disease control has been achieved and a decision to taper the steroids is clinically appropriate, the thing to be aware of is what can happen during a glucocorticoid taper. Because if we have the blinkers on and we're aware of the 3 possible things that could occur, then we're able to address them if they do present. So at a level of principle, the things to identify is again, will the disease remain in control? Is there a risk of adrenal insufficiency and then glucocorticoid withdrawal syndrome. But in general, once disease control has been achieved, then the glucocorticoid can be tapered to a physiological dose and different pathologies may have set weaning plans such as, for example, giant cell arthritis will have a particular wean.

But in general, if a pathology doesn't have a particular wean, then the taper can occur to a physiological dose. And as you're trying to taper this to a physiological dose, and it may be able to be tapered rapidly, if the underlying pathology allows, you are aiming to avoid relapse of the underlying disease. And if at any point you are getting a disease flare that warrants an increase in the glucocorticoid, then that would need to take place. Once the patient has been weaned down to a physiological dose of steroid, a clinician is then faced with 2 potential options to evaluate the HPA axis, either from a symptom standpoint or an ongoing gradual taper with monitoring of symptoms. So, option one, once the patient is on a physiological dose, and I should also preface by saying you would want them to either be on prednisolone or hydrocortisone by that time once they're physiological, given the half-life properties that Morton was talking about.

Option one is you could taper, for example, prednisolone by one mg each month, once they're down to let's say a dose of 5 mg. You've provided education to the patient on symptoms of adrenal insufficiency, such as increasing fatigue, lethargy, postural dizziness. And if with this slow taper of one mg per month and the patient remains asymptomatic with respect to symptoms of adrenal insufficiency, you can continue that slow taper to cessation. So that's option 1. Option 2, which some clinicians may pursue is trying to assess this chemically. So once the patient is tapered down to a physiological dose of steroid, the morning cortisol level, and when we say a morning cortisol level, it's ideally as close to 8:00 A.M. as we can get. And using the cortisol to help guide whether the HPA axis is suppressed or if there's concern that there is HPA axis suppression.

So generally, the thresholds that we give as a guide is that if the morning cortisol level when tested before the glucocorticoid dose in the morning is less than 150 nanomol/L, that would suggest HPA axis suppression, and then you should continue that physiological dose of steroid and retest the cortisol axes in a few months' time. If your morning cortisol level was between 150 to 300 nanomol/L, the suggestion is to continue that physiological dose and retest in a few weeks because you may be getting close to the HPA axis awakening. And if your morning cortisol level is above 300, the likelihood of HPA axis suppression is quite low. Obviously, the higher the cortisol level, the more likely it is that the HPA axis is not suppressed, but above 300, from the more recent guidelines, would suggest that you can stop the glucocorticoid without further tapering.

So I guess part of the real difficulty here, especially once you're getting below those physiological doses of a prednisolone, prednisone, 5 mg a day, seems to be trying to differentiate all the different things that can go wrong, glucocorticoid withdrawal syndrome versus adrenal insufficiency versus a flare of the underlying disease that we're trying to control. Morton, can you talk me through a little bit about the differences between these? I think we probably sometimes get those confused, but really that's a key part of the matter, isn't it?

MB: Yeah, you're right, and a lot of the symptoms overlap. So I think the first thing to consider is that what dose of prednisolone the symptoms are occurring. Because if you're on more than 5 mg of prednisolone a day, that's a supraphysiologic dose for most patients. So these won't be symptoms of adrenal insufficiency unless the patient has an intercurrent illness, such as a major infection, at which case these sorts of things can develop. So above 5 mg a day, you're really considering whether this is a flare of the underlying disease or whether it's a glucocorticoid withdrawal symptom, and there's no test you can do reliably to distinguish that in terms of endocrine testing, you're really looking at whether these are features you'd expect of the underlying disease or whether this is glucocorticoid withdrawal symptoms, which are symptoms that commonly develop when prednisolone is tapered too quickly. If your prednisolone dose is below 5 mg a day, then cortisol deficiency is also an important factor to consider and it's in that setting where testing the HPA axis by measuring a morning cortisol in the first instance can be a benefit.

Of course, if you're a visual learner, there's some beautiful tables and a beautiful figure in the article, the key differences for you to follow along and to go back over through that. Maybe though, Faran, you can talk through a little bit about what we see in terms of the other side of things, how we manage sick days, why that matters, and how we might even get to the point of adrenal crisis.

FK: Yeah, it's a really important note, David, with sick day management and understanding what could potentially precipitate an adrenal crisis. So again, coming back to the underlying physiology to make sense of it all. We know that during times of physiological stress, such as an illness or the flu or anything that puts a physiological stress on the body, that generally in patients that are not taking glucocorticoids, their cortisol concentrations will increase during times of physiological stress. In patients that are on long-term glucocorticoids, because the HPA axis may be suppressed, the increase in glucocorticoids, which the body would usually make during times of physiological stress, now needs to be controlled by an increased dose of glucocorticoid. Now what's really important to note is the current dose of glucocorticoid. So for example, in patients that are taking 5 mg or a physiological dose of glucocorticoid, during times of mild illness, the suggestion would generally be to double the dose to 10 mg of prednisolone until that physiological stress has resolved. And in times of more significant physiological stress, than a triple physiological to 15 mg would be indicated in that case.

The thing to be aware of and a question that often we get asked in endocrinology is, if my patient is on 10 mg of prednisolone and they have a mild physiological stress, do we need to double the dose? And the important thing to note is we're always coming back to what is the dose equivalent of prednisolone that the patient is on. So if a patient is on 10 mg of prednisolone and they have a mild physiological stress, you don't need to double that dose to 20. But if they were to have a more moderate to severe physiological stress, then increasing the dose to triple physiological dose, which would be 15 mg, could very well be indicated. Now the reason why sick day management is so important is that, if this is not followed and if there is a physiological stress, which would generally require a higher dose of glucocorticoid, that can then precipitate an adrenal crisis, and this can be characterised by hypotension, nausea, vomiting, hypoglycaemia, and they can have significant hemodynamic instability which can result in hospital admission for appropriate and aggressive management.

So what kind of level of physiological stress are we talking about here? I think as generalists, we sometimes get lost in the weeds a little bit there.

FK: Yeah, yeah, sure. So when we're looking at physiological stress, what we would characterise as, for example, a mild illness is, again, you might have a temperature between 37 and 38. You might have a respiratory virus that you may have caught from a family member, you're still able to carry out your day-to-day functioning. You're not bed bound and unable to maintain your oral intake, etc. More moderate to severe stress, these are the patients that might have a temperature of more than 38 degrees, so quite febrile. They might have significant shortness of breath that's requiring them to be admitted to hospital or they've got significant diaphoresis and physiologically, they're obviously a greater toll is being taken.

The only part that I haven't prefaced, which I probably should have, is if the patient's unable to take an oral intake. So for example, if they have a gastro illness, which they're vomiting a lot with or have significant diarrhoea, the question then is are they even absorbing the glucocorticoid that they are taking, even if they have appropriately stress dosed? And that's why the education's really important. Because we may have a patient that has appropriately, let's say doubled or tripled their dose of physiological prednisolone, but they still develop hypotension and potentially an adrenal crisis because they may not just be absorbing that dose.

It sounds like there are a lot of really good reasons to be thinking about endocrinological testing, differentiating in some situations, those 3 different scenarios - glucocorticoid withdrawal syndrome, adrenal insufficiency, underlying disease flare - as well as trying to pre-empt some of the issues that we've talked about. Morton, do you think that we do this well in practice across the board and if we don't do it well, why don't you think we do it well?

MB: It's a good question, David. I think we see the bad end of the spectrum in endocrinology, and I think there are probably many patients who their doses are being successfully weaned. But I do think it is a problem that's underappreciated by doctors. And it's not uncommon to see patients who have been hospitalised for an intercurrent illness who are not feeling well, they're not recovering quickly, they perhaps have hyponatremia, and then you can rapidly improve their clinical situation by increasing their prednisolone dose. So I think endocrinologists, we would educate all our patients with adrenal insufficiency with regard to this. The Endocrine Society of Australia has got written instructions that we give them [patients] with regards to how they're to adjust their glucocorticoid dose during an intercurrent illness, but I think that's probably not done so routinely in other clinical areas. Now, perhaps in someone who's only taking a course of glucocorticoids for a few weeks to months, that may be overdoing it, but in the patients such as patients with polymyalgia rheumatica who are needing prednisolone for months to perhaps years, then I think that's an important component of practice that could potentially be improved.

Finally, I might just ask you, when do you as endocrinologists want to see these patients? When are you looking to get these referrals for people who have been on glucocorticoid therapy for some time?

MB: We're always happy to see anyone that you are worried about, but I think if you're able to wean your patient's therapy and they remain well, then there's no indication for an endocrinologist. The patients to refer would be patients who you are unable to wean. Patients who are developing symptoms during the weaning process would potentially be a group to assess, particularly if you think those symptoms are relating to adrenal insufficiency rather than, for example, a flare of the underlying disease where of course, the other clinician probably has greater expertise. And patients that have a low cortisol level during wean, if you do measure that the morning cortisol is very low, that would potentially be a group that we could see. We can do further tests such as a Synacthen test where one gives an injection of synthetic ACTH and measures the cortisol response to that. That's a common thing that we would do.

Well, Faran, Morton, it's been a beautiful primer on how to navigate this tricky area. Hopefully it means that we start to do that better right across the board. Thank you so much for joining us today on the Australian Prescriber Podcast.

MB: Thanks for having us.

FK: Thanks, David.

[Music]

The views of the guests and the hosts on this podcast are their own and may not represent Therapeutic Guidelines or Australian Prescriber. Faran Khalili has received honoraria from Boehringer, Novo Nordisk and AstraZeneca for lectures on sodium-glucose co-transporter 2 inhibitors in diabetes and glucagon-like peptide-1 agonist therapy for weight management. Morton has no conflicts to declare, and I am on the Drug Utilisation Subcommittee of the PBAC [Pharmaceutical Benefits Advisory Committee]. Thank you once again for joining us today on the Australian Prescriber Podcast.