Adrenal insufficiency: identification and management

Why the committee made the recommendations

A qualitative review identified studies investigating information and support needs for adults and children with adrenal insufficiency and their carers. There was some evidence addressing the routine management of adrenal insufficiency and support for preventing adrenal crisis, but overall, the evidence was limited because it did not cover all aspects of information and support needs specified in the review questions. It was graded as medium to low quality mainly because of methodological limitations in the studies.

The main themes from the evidence included improving awareness of physiological stress situations that might require an increased dose of hydrocortisone, the need for more information and education throughout a person's treatment, and the value of patient support groups. These themes support the recommendations on giving additional glucocorticoids, signposting people to support groups and networks for their clinical condition, and reviewing information and support needs regularly, including in particular the needs of children and young people as they mature and transition to adulthood. Studies did not specifically address information and support in emergency care during adrenal crisis, or when fasting, travelling or working non-standard hours.

Adults are advised on how to get an NHS Steroid Emergency Card, including people who may develop tertiary adrenal insufficiency and become corticosteroid-dependent. This prompts healthcare professionals to consider adrenal crises in people carrying the card, start appropriate management for planned surgery or invasive procedures, and treat people rapidly in emergency situations. Children with adrenal insufficiency should have a British Society of Paediatric Endocrinology and Diabetes (BSPED) adrenal insufficiency card that provides parents, carers, and healthcare staff with a child's corticosteroid care plan for sick days and emergencies. The committee therefore used their expertise from clinical practice to develop consensus recommendations for these specific areas.

How the recommendations might affect practice

The recommendations are reflective of best practice and are not expected to lead to significant changes.

Return to recommendations

Why the committee made the recommendations

Evidence on the diagnostic accuracy of signs and symptoms associated with adrenal insufficiency was limited, but the symptoms and signs reported in studies were generally in line with the committee's clinical experience. Evidence was identified for low blood pressure, hyperpigmentation for primary adrenal insufficiency, lethargy, salt craving, weight loss, hyponatraemia, hyperkalaemia, nausea, vomiting and diarrhoea. The committee agreed more importance should be placed on the sensitivity of a test for clinical decision making, but none of the signs and symptoms met the agreed thresholds for both sensitivity and specificity. The symptoms and signs of adrenal insufficiency are common to many conditions. One or more persistent and unexplained symptoms, signs or features should raise suspicion of adrenal insufficiency and warrant further investigation. The committee agreed hyperpigmentation is common in people with primary adrenal insufficiency and is the clearest indicator for the condition. However it may not be seen on black or brown skin, in which case clinicians should inspect the buccal mucosa or any surgical scars and ask the person if they have noticed a change in their skin colour. Symptoms and signs particularly seen in children are hypoglycaemia, faltering growth, hypotensive crisis, and differences in sex development. The committee listed the symptoms, signs and features in the order of most to least clinically distinguishing.

The committee made consensus recommendations drawing on their experience of observed symptoms and signs, and knowledge of the risk of adrenal insufficiency associated with some medications and coexisting conditions and comorbidities, such as hypothyroidism and type 1 diabetes.

How the recommendations might affect practice

The recommendations generally reflect current best practice and are not expected to lead to significant changes. They may be useful in particular for non-specialist clinicians in acute areas such as pre-hospital emergency care and emergency departments, and for those doing invasive procedures or surgery. Increased awareness of the possibility of adrenal insufficiency may reduce mortality by enabling early diagnosis and treatment.

Return to recommendations

Why the committee made the recommendations

There was limited evidence for this review question because of small numbers of study participants and diversity between studies, so the committee used their clinical knowledge and experience to make the recommendations.

The committee recommended that people aged 1 year and over with suspected adrenal insufficiency should be offered an 8 am to 9 am serum cortisol test because this is the optimal time for peak cortisol levels, and cortisol tests at other random times should not be done. No reliable evidence was found to support other time frames to accommodate people who work shift patterns, such as night shifts for whom an 8 am to 9 am cortisol test would be difficult. The circadian clock shifts 1 hour in 24 hours and so timings may depend on how many nights have been worked. A healthcare professional would avoid testing straight after a night shift and wait a few days or consider a different diagnostic test, such as a short synacthen test. Due to the variation in diurnal rhythm, for babies under 1 year, serum cortisol levels can be measured at any time of day but paediatric or paediatric endocrinology advice for interpretation of results should be sought. When interpreting 8 am to 9 am serum cortisol results, it is important to take into account clinical context. The committee discussed the difficulties in setting cut-off points because these vary greatly depending on the assay used and only have clinical use if specific to a particular assay. However, the committee agreed that it would be useful for non-specialists to have some guidance on when to refer, providing it is highlighted that the cut-offs are only for use with modern immunoassays and that local guidelines may need to be followed if alternative assays are used.

People with symptoms of adrenal insufficiency together with hyponatraemia may need discussion with endocrinology especially if cortisol is between 150 nmol/L and 300 nmol/L. They may have developed acute adrenal insufficiency related to other treatments, such as checkpoint inhibitors.

Because of a 'grey area' of clinical suspicion between 150 nmol/L and 300 nmol/L, the committee concluded that if the test result falls between these values a repeat 8 am to 9 am serum cortisol test should be considered. The committee set an upper threshold of 300 nmol/L for re-testing because they agreed that any reading above this would mean adrenal insufficiency is very unlikely.

Serum cortisol tests in people who are taking oral oestrogens are not accurate because cortisol levels are falsely elevated as oestrogen raises levels of cortisol-binding globulin. Therefore, the committee recommended people stop taking it 6 weeks before measuring serum cortisol.

Studies examining salivary cortisol were more recent than those for serum cortisol and used newer assays with greater accuracy. The committee agreed that the use of salivary cortisol and cortisone instead of serum cortisol for first-line testing is an emerging field. Potential benefits would be people being able to do the test themselves at home and without the need for blood tests, but they agreed that further research is needed and therefore made a recommendation for research on initial investigations for people with suspected adrenal insufficiency.

How the recommendations might affect practice

Serum cortisol testing in current practice is not consistently done at the optimal time to diagnose adrenal insufficiency. Therefore, the recommendations will not affect the total number of tests done but may result in a change in practice for some providers. Optimal serum cortisol testing and the use of the recommended referral threshold should minimise costly and unnecessary referrals to secondary care and short synacthen testing.

Return to recommendations

Why the committee made the recommendations

How the recommendations might affect practice

The recommendations on steroid replacement and hyponatraemia reflect current practice and are not expected to lead to significant changes. Current practice on prescribing emergency management kits is variable and the recommendations may lead to a change in practice by some providers.

Return to recommendations

Why the committee made the recommendations

Because only 1 study was identified, which the committee did not think was sufficient to base recommendations on, guidelines from other organisations on the pharmacological management of physiological stress were reviewed. The quality of these guidelines was assessed using the Appraisal of Guidelines for Research and Evaluation (AGREE) II tool. Guidelines that were assessed as high quality and that included recommendations the committee wished to cross-refer to were further assessed using the NICE process for assessing applicability and acceptability. Based on these assessments, the committee either made their own consensus recommendations informed by these guidelines or directly cross-referred to recommendations in external guidelines.

The committee emphasised the importance of having additional supplies of glucocorticoid medication available for periods of physiological stress (covered in the information and support section). They highlighted that some people might find it difficult to obtain additional supplies of glucocorticoids, and health professionals need to be aware of this to prevent adrenal crises.

The frequency and dose of glucocorticoids need adjusting during significant physiological stress, for example, offering an increased dose if a person has a fever or physical trauma. However, there are associated harms of increasing the dose too frequently or for prolonged periods of time because this can lead to symptoms and signs of corticosteroid excess. The committee agreed the duration of increased dosing would vary according to the type of physiological stress and factors related to the individual. If absorption of oral glucocorticoids is difficult because of vomiting or diarrhoea, then an injection of intramuscular or intravenous hydrocortisone may be given. The committee referred to the guideline for the management of glucocorticoids during the peri-operative period by the Association of Anaesthetists, the Royal College of Physicians and the Society for Endocrinology.

The committee agreed the BSPED guideline for children was comprehensive and clearly set out. It had also achieved high scores using the AGREE tool and second-stage NICE assessment. Therefore, the committee agreed not to make their own recommendations but to cross-refer to the BSPED website for recommendations on sick-day dosing for babies, children and young people under 16 years experiencing physiological stress.

As little evidence was found, the committee decided to make a recommendation for research on the use of postoperative glucocorticoids for people with, or at risk of, adrenal insufficiency having inpatient invasive procedures.

How the recommendations might affect practice

The recommendations reflect best practice but may not be current practice for all, resulting in changes to practice for some. Although there is a cost associated with providing additional supplies of oral glucocorticoids, this cost is expected to be small, relative to the cost and quality of life impact of an adrenal crisis. These recommendations are therefore not expected to have a significant resource impact.

Return to recommendations

Why the committee made the recommendations

No clinical evidence was identified, so the committee made recommendations to reflect best current clinical practice.

The committee confirmed that all adults with adrenal insufficiency or at risk of adrenal crisis should be provided with information on managing their condition including sick-day rules and crisis management during periods of physiological stress.

A steroid treatment card (blue card) is provided to people prescribed glucocorticoids for other medical conditions. This generally affects people with non-endocrine conditions who are on exogenous corticosteroids where dose and duration could lead to hypothalamic-pituitary-adrenal axis suppression. The card includes guidance on minimising the risks when taking corticosteroids and provides details of the prescriber, drug, dosage and duration of treatment. Education on daily dosing, sick-day rules and crisis management is provided at the time of diagnosis and throughout a person's treatment. In best practice, people are also provided information on the use of patient-held alerts about their condition. This can include medical alert jewellery such as bracelets, and apps or mobile phone medical IDs.

How the recommendations might affect practice

The recommendations for managing periods of physiological stress largely reflect current practice and will therefore not result in a significant change.

Return to recommendations

Why the committee made the recommendations

There was limited evidence for women or people with adrenal insufficiency who are pregnant or planning pregnancy, so the committee made consensus recommendations based on their experience and current practice. Continuing replacement doses of glucocorticoid and mineralocorticoid is essential in pregnancy to prevent adrenal crisis. Normal pregnancy is associated with increases in cortisol and aldosterone that combat the anti-glucocorticoid and anti-mineralocorticoid effects of progesterone. Therefore, continuing replacement doses of glucocorticoid and mineralocorticoid is essential in pregnancy to prevent adrenal crisis. Despite these increases in cortisol and aldosterone, which are more apparent by the third trimester, few people with adrenal insufficiency routinely require increases in their replacement corticosteroid doses. Clinical signs including symptoms of adrenal insufficiency, postural hypotension and hyponatraemia justify increases in replacement doses during the third trimester. Increased doses of hydrocortisone such as stress doses or injections will not harm the baby because hydrocortisone is broken down and inactivated in the placenta. Many people will experience nausea and vomiting in pregnancy and may not be able to keep their medications down. Advice on taking glucocorticoids during periods of pregnancy-related vomiting should be provided. Hyperemesis gravidarum should be managed within a hospital setting because parenteral replacement of increased doses, intravenous fluid replacement and closer monitoring of blood pressure and serum electrolytes are often required and are more suited to an inpatient setting. Glucocorticoid requirements decline after the birth and if replacement doses have been increased in pregnancy, they should be decreased to pre-pregnancy levels providing there are no complications which may require continuation of increased dosing.

How the recommendations might affect practice

The recommendations for managing pregnancy largely reflect current practice and will therefore not result in a significant change. The exception to this is the management of hyperemesis gravidarum in an inpatient setting which is not happening in current practice. The committee noted that deaths have been reported following outpatient management of hyperemesis gravidarum in people with adrenal insufficiency. The committee highlighted the importance of inpatient care and noted that although this is more costly than outpatient care, the population for whom this recommendation would apply is small and therefore this should not result in a significant resource impact.

Return to recommendations

Why the committee made the recommendations

No evidence was found, so the recommendations were made by consensus based on the experience and expert opinion of the committee.

The committee were aware that there is variation in current clinical practice on whether to adjust medication to account for psychological stress. This is partly because of the wide variation in factors and events that could lead to psychological stress, such as a mental health crisis or bereavement, and the variation in what people find stressful and how they react. Periods of psychological stress could also vary between a short-term or single event to many weeks. This variation makes it difficult to determine whether a person would be at risk of adrenal crisis because of psychological stress. The committee agreed that an occasional increase in glucocorticoid dose was unlikely to lead to side effects, but long-term increases were not advised. An adjustment to the dose of glucocorticoid medication has the potential to reduce the risk of harm to a person experiencing an adrenal crisis because of psychological stress. Overall, the committee agreed that a short-term increase in oral glucocorticoids using sick-day dosing for 1 or 2 days could be considered in times of acute and intense psychological or emotional stress. For people experiencing a severe mental health crisis and who cannot take oral glucocorticoids, the committee advised they should be given intramuscular hydrocortisone.

The committee agreed the BSPED guideline for children was comprehensive and clearly set out. It had also achieved high scores using the AGREE tool and second-stage NICE assessment. Therefore, the committee agreed not to make their own recommendations but to cross-refer to the BSPED website for recommendations on sick-day dosing for babies, children and young people under 16 years experiencing psychological stress.

How the recommendations might affect practice

There is variation in current practice, so these recommendations could lead to a change in practice for some. Given the small additional cost of increasing oral corticosteroids for 1 or 2 days during periods of acute and intense psychological or emotional stress and the potential for avoiding costly and harmful adrenal crisis, this recommendation is not expected to have a significant resource impact. The recommendation for intramuscular hydrocortisone for those experiencing a severe mental health crisis and who are unable to take oral glucocorticoids would apply to a very small proportion of people and therefore would not have a significant resource impact.

Return to recommendations

Why the committee made the recommendations

No evidence was identified for this review. The committee decided to make consensus recommendations to highlight the significance of psychological and emotional stress as a triggering factor for adrenal crisis, and to provide advice on accessing information and support to help reduce stress and avoid an adrenal crisis. The committee acknowledged the importance of patient support groups and organisations in providing information and support, particularly to newly diagnosed people. These groups can promote awareness about exploring adjustments that may be possible within the workplace or educational setting to help people with adrenal insufficiency participate in everyday activities. Recommendations were made to direct people to their specialist clinical team for support and advice on self-management strategies to manage stress and anxiety, and, where needed, onward referral to NHS talking therapies or mental health services.

How the recommendations might affect practice

The recommendations reflect best practice. Where best practice is not currently implemented, the recommendations cover the provision of information which will likely only involve a couple of minutes of extra staff time on top of existing patient contact with healthcare professionals and are not expected to result in any significant change. The recommendation to consider referral or self-referral to NHS talking therapies or mental health services is in line with existing NICE guidance and is considered current practice.

Return to recommendations

Why the committee made the recommendations

Evidence for the risk factors associated with adrenal crisis was very limited and of poor quality, so the committee used their expertise to inform the recommendations and supplement the available evidence.

Evidence available from only 1 study suggested that lower sodium levels are associated with an increased risk of developing adrenal crisis, and hyponatraemia below 135 mmol/L is indicative of adrenal insufficiency and an indicator of the possibility of adrenal crisis. No relevant studies were identified that investigated hyperpigmentation, hypoglycaemia, circulatory shock or collapse, or failure of the condition to respond to initial treatments as risk factors or exposures.

The committee used their clinical experience to specify the features that should raise suspicion of adrenal crisis in people who are critically unwell. They noted that hyperpigmentation was the most indicative feature in people with primary adrenal insufficiency and should raise clinicians' suspicions of an adrenal crisis even in the absence of any other signs or symptoms. However, this may not be seen on black or brown skin, in which case clinicians should inspect the buccal mucosa or any surgical scars and ask the person if they have noticed a change in their skin colour. The committee also agreed that a broader range of clinical signs and symptoms should be highlighted as indicative of adrenal crisis in people who have a known diagnosis or are at high risk of adrenal insufficiency, so that treatment can be delivered as soon as possible. By raising awareness of the most common risk factors, signs, and symptoms, delayed and missed diagnosis of adrenal crisis could be reduced, which could save lives.

As no evidence was found for people taking long-term corticosteroids, a recommendation for research was made for what increases the risk of adrenal crisis and adverse hospital outcomes in this group.

How the recommendations might affect practice

The recommendations reflect current practice, so the committee agreed there would be no change in practice.

Return to recommendations

Why the committee made the recommendations

No research evidence was identified, so existing guidelines on emergency management of adrenal crisis were used to inform the recommendations. The quality of these guidelines was assessed using the AGREE II tool. Guidelines that were assessed as high quality and that included recommendations the committee wanted to cross-refer to were further assessed using NICE's process for assessing applicability and acceptability.

The 3 essential aspects of treatment are giving hydrocortisone and fluids and ensuring that the person is rapidly transported to hospital. The former 2 were included in all of the guidelines that were reviewed on the emergency management of adrenal crises. The committee highlighted that if an adrenal crisis is suspected, treatment should be given immediately by anyone, including the person, their families and their carers. All adult guidelines recommended immediate intravenous administration of hydrocortisone with a further dose of hydrocortisone over the following 24 hours. The committee agreed with these recommendations but acknowledged the guidelines covered treatment in hospital so only mention the intravenous route. The committee decided to recommend either intravenous or intramuscular routes to enable anyone to administer the medication. The committee acknowledged doses used for emergency management are as stated in the BNF.

The committee emphasised the importance of giving parenteral fluids, noting that deaths can occur even if hydrocortisone is given but fluids are not. Guidelines suggested various protocols for sodium chloride infusion, but they all agreed that after an initial 1 litre infusion, sodium chloride should be continued for 24 hours or until the patient is stable. The committee agreed that the main aim should be to give the initial dose of fluids as soon as possible, ideally within 30 minutes but acknowledged that how this is delivered depends on the hospital setting.

In considering the balance of benefits and harms of administering a high dose of hydrocortisone in an emergency, the committee highlighted that hydrocortisone is a life-saving replacement therapy in such situations and it has no toxic dose. Therefore, they made strong recommendations for immediate administration of hydrocortisone and an additional consensus recommendation to reassure that there is no risk of an overdose.

Monitoring was a key feature in all of the external guideline recommendations and most commonly included monitoring of cardiac and haemodynamic parameters. Some recommendations also included transfer to intensive care if necessary. Therefore, the committee made a recommendation to highlight the importance of caring for a person in a high-observation area with frequent monitoring. The committee made a consensus recommendation to offer oral glucocorticoids at a higher dose than usual until any underlying cause has resolved and the person is haemodynamically stable because it is important to ensure that the dose is adequate for recovery and for preventing a relapse back into a crisis.

To prevent deterioration of the person's condition, aid in recovery and help prevent further crises, a consensus recommendation was made to highlight the need for referral to a specialist endocrine team for ongoing clinical advice and support throughout admission and during the hospital stay, and for identifying and treating any underlying cause of adrenal crisis.

The committee agreed the BSPED guideline for children was comprehensive and clearly set out. It had also achieved high scores using the AGREE tool and second-stage NICE assessment. Therefore, the committee agreed not to make their own recommendations but to cross-refer to the BSPED recommendations for emergency management of adrenal insufficiency in babies, children and young people.

How the recommendations might affect practice

The recommendations about emergency hospital treatment for people experiencing an adrenal crisis are reflective of current practice.

Return to recommendations

Why the committee made the recommendations

No evidence was identified for ongoing care and monitoring of people with adrenal insufficiency, including those who are receiving end of life care, so the recommendations were made by expert knowledge and consensus of the committee.

The frequency of clinical reviews should vary depending on the person's needs as well as the type of adrenal insufficiency they have. People with newly diagnosed primary adrenal insufficiency may need more intensive monitoring initially until the healthcare professional is sure that a person understands the condition and how to manage it, or if the person has symptomatic adrenal insufficiency requiring more clinical management. However, adults with adrenal insufficiency who are confident with self-management and have stable clinical needs may need less frequent monitoring. The method of follow up and monitoring would also differ according to individual needs, with face-to-face appointments more suitable for some people and telephone or video consultations for others.

It is important to monitor signs and symptoms of under- or over-replacement of glucocorticoids. Under-replacement of glucocorticoids may cause weight loss, nausea, and fatigue. It is important to investigate whether these broad, non-specific symptoms can be attributed to under-replacement of glucocorticoids or have other causes. For example, short-term fatigue may occur while a person adjusts to the treatment and would not need a change in the dosage, but sudden-onset fatigue or fatigue that significantly affects activities of daily living should not be ignored. Signs and symptoms indicating over-replacement of glucocorticoids, particularly in people on higher than standard doses, may include unexplained weight gain, new or worsening diabetes or hypertension.

An important part of reviews is to make sure that people with adrenal insufficiency understand the importance of adhering to their medication, how to avoid having an adrenal crisis and knowing what to do in emergency situations.

For children, the committee agreed that appointments with the specialist team should be at least every 6 months, but as for adults, should be adjusted according to individual needs. An annual face-to-face hospital appointment should be offered to measure the height and weight of children to ensure their condition is being well managed. The committee noted more frequent monitoring may be needed during periods of rapid growth when dosages of medication may need to be changed, during transition of care to adult services to facilitate a smooth handover, or if there are concerns with medicines adherence or whether the child and their family or carers are able to safely self-manage the condition.

The committee agreed that for people receiving end of life care, decisions on withdrawing active treatment should be made as part of shared decision making. This does not mean withdrawing glucocorticoids but may include changes to how medication is given, such as by injection rather than orally. The committee agreed to cross-refer to the recommendations in NICE's guidelines on end of life care for adults and end of life care for infants, children and young people with life-limiting conditions for general principles of care appropriate for people with adrenal insufficiency.

How the recommendations might affect practice

The recommendations reflect current practice and are not expected to lead to significant changes.

Return to recommendations

Why the committee made the recommendations

The evidence available was very limited. Withdrawal interventions varied between the studies, and only 1 study covered children. Many studies looked at the withdrawal of oral prednisolone, and although this is not licensed for use in the UK, the committee agreed it was relevant because withdrawal strategies for other medicines would be similar. The committee decided by consensus that outcomes specifying adrenal insufficiency as an adverse event, or those which could indicate corticosteroid withdrawal syndrome or be indicative of adrenal suppression, would aid decision making. These included nausea, hyponatraemia, diarrhoea, vomiting, lethargy, malaise, anorexia and myalgia. The committee used their consensus opinion to formulate the recommendations. They discussed that in clinical practice, decisions around tapering are rarely straightforward and are decided on a case-by-case basis through assessment of individual needs. The evidence suggests that rapid tapering regimes do not lead to an increase in adverse events or incidence of adrenal insufficiency, but because of the limited evidence available, the committee were not confident in the results reported. However, they recognised the need to provide generalised guidance for non-endocrine specialist clinicians and agreed that starting with a tapering regimen that involves the following could be trialled: taking the physiological equivalent dose every other day for 2 weeks, then twice a week for 2 weeks, then stopping. The committee reasoned that this is roughly the equivalent to halving the dose for 2 weeks and then halving it again. They agreed that it is simple for people to understand and follow and has been widely used in clinical practice so there should not be any safety concerns. Glucocorticoids can also be tapered more slowly. The committee highlighted that if there are any symptoms of adrenal insufficiency or any uncertainty, glucocorticoid should be reverted to a physiological equivalent dose and consideration given to contacting an endocrine specialist.

The committee discussed the practice of switching to different types of glucocorticoids while tapering. People aged 16 or over should not be routinely switched from prednisolone to hydrocortisone because there is no evidence to support this. The committee noted this is happening in current practice despite the lack of evidence of benefit. For babies, children and young people under 16 years this can be considered to enable doses to be reduced to a non-suppressive dose more easily. The committee agreed that if people aged 16 and over are taking dexamethasone for a longer duration and have any difficulty while tapering, then clinicians should consider switching to prednisolone. For babies, children and young people under 16 years, hydrocortisone may be considered instead. This is because of dexamethasone being significantly more potent and having a longer half-life, so it is difficult to give a corticosteroid-free period over 24 hours, which is not enough for the hypothalamic-pituitary-adrenal axis to recover.

The committee highlighted to consider investigations to exclude adrenal insufficiency only when a slow tapering regimen has been attempted and the person has developed signs and symptoms of adrenal insufficiency.

The committee noted that there is an increased chance of difficulties withdrawing glucocorticoids for people using multiple glucocorticoid preparations simultaneously, using high-dose inhaled glucocorticoids, or for those people who had intra-articular or intramuscular glucocorticoid injections in the previous 2 months, or who had treatment with strong cytochrome P450 3A4 inhibitors along with glucocorticoids.

The committee decided to make a recommendation for research on glucocorticoid withdrawal because very little evidence was found and there is uncertainty around how best to withdraw glucocorticoids, which can lead to overtreatment and an increased risk of adrenal insufficiency.

How the recommendations might affect practice

The recommendations on tapering regimens reflect current practice and the committee agreed there should be no significant change in practice. Because of the uncertainty in the population size and to minimise the resource impact to the NHS, the recommendations for testing in this population were restricted to those who develop signs and symptoms after trying a slow taper, as opposed to everyone withdrawing from long-term glucocorticoids and a weaker 'consider' recommendation was made.

Return to recommendations