• glucocorticoid
• adrenal suppression
• adrenal crisis
• Cushing's

Glucocorticoids (GCs) are the mainstay of treatment for many conditions in childhood and their use in a wide variety of diseases means that many clinicians will prescribe steroids to and/or manage children on GC therapy. However, all clinicians should be alert to the possibility that GCs can have significant side effects, notably adrenal suppression. Chronic GC treatment, through negative feedback on the hypothalamus/pituitary, leads to a reduction in hypothalamic corticotrophin-releasing hormone and pituitary adrenocorticotropic hormone (ACTH) production. Over weeks to months this can result in the progressive atrophy of cortisol-producing cells in the adrenal gland, an inability to produce normal levels of cortisol in response to ACTH stimulation and secondary adrenal insufficiency (figure 1). Patients with adrenal suppression may remain well if their normal GC requirements are provided by the GC treatment itself; however, in some cases, this is inadequate and symptoms of adrenal insufficiency may occur.1 Furthermore, in the presence of adrenal atrophy, acute withdrawal of treatment (whether deliberate cessation of therapy or an inability to tolerate usual treatment) or the addition of an acute stressor, for example, an intercurrent illness, in individuals who are unable to mount an appropriate cortisol response will lead to symptoms of adrenal insufficiency and potentially result in adrenal crisis. Importantly, adrenal suppression can persist for some time, so that full hypothalamic–pituitary–adrenal axis recovery after stopping GC therapy may take more than a year.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Chronic glucocorticoid (GC) therapy can lead to adrenal suppression. Chronic GC treatment leads to a reduction in corticotrophin-releasing hormone (CRH) production by the hypothalamus and adrenocorticotropic hormone (ACTH) production by the pituitary. Over time this can result in adrenal atrophy. In individuals who are unable to mount an appropriate cortisol response, abrupt cessation of GC therapy, an inability to tolerate usual treatment or exposure to an acute stressor, for example, an intercurrent illness can lead to symptoms of adrenal insufficiency (which may be non-specific) and potentially result in adrenal crisis. HPA, hypothalamic–pituitary–adrenal.

While adrenal suppression has serious and potentially life-threatening consequences, the signs and symptoms can be insidious and non-specific, including fatigue, nausea/vomiting/diarrhoea/abdominal pain, muscle weakness, weight loss, hypoglycaemia and mood changes, so that adrenal suppression is under-recognised. Notably, the hyperpigmentation seen in Addison's disease is not present because of ACTH suppression. Although adrenal aldosterone production is maintained in secondary adrenal failure, hyponatraemia can still occur because of impaired free water clearance with GC deficiency. Adrenal crisis is a medical emergency that can present with many non-specific features—severe lethargy, hypoglycaemia, hypotension, decreased/altered conscious level, seizures, fever, severe vomiting and diarrhoea—and can be fatal.2 ,3

Goldbloom et al3 report on a study aimed at determining the national incidence of paediatric (<18 years) GC-induced symptomatic adrenal suppression in Canada using a national paediatric surveillance programme. This impressive undertaking used a prospective surveillance system involving >2500 paediatricians and achieved an 80% mean response rate. Forty-six children with symptomatic adrenal suppression were identified in the 2-year period, giving an incidence of 0.35/100 000 total paediatric population.

The presenting features of the children in this study are instructive. Thirty-five per cent presented with growth failure alone. Although the mechanisms accounting for growth failure may include excess GC exposure, GC deficiency and/or undertreatment of the underlying condition (and are therefore not specific for adrenal suppression), the fact that more than one-third of the children in this study presented with growth failure together with biochemical evidence of adrenal suppression clearly highlights the need for growth monitoring in children treated with GCs. A further 28% complained of non-specific symptoms alone, including fatigue, lethargy, nausea, anorexia, vomiting, abdominal pain and myalgia. Importantly, a number of the children exhibited features of Cushing's syndrome. Children with features of Cushing's syndrome have clearly been exposed to excess exogenous GC for some time and are therefore at considerable risk of adrenal suppression and an inability to mount a cortisol response when GCs are acutely withdrawn. Six patients (13%) presented with adrenal crisis—defined as an acute critical illness out of proportion in severity to the current illness and manifested by any of the following: hypotension/shock, decreased level of consciousness, lethargy, unexplained hypoglycaemia or hyponatraemia, seizure or death.1 ,2 In Goldbloom et al's study, one child presenting in adrenal crisis died 3 months later, reportedly from complications of Cushing's syndrome.

Although GC administration by any route has the potential to result in adrenal suppression (including topical, intranasal, inhaled, oral, intravenous, intramuscular and intra-articular routes), Goldbloom et al's report that 80% of children with symptomatic adrenal insufficiency were receiving GCs by inhaler alone or inhaler in combination with other routes. Sixty-five per cent of the children had asthma and a further 15% had asthma plus another condition treated with GCs. Four of the six children with adrenal crisis were being treated for asthma. The most commonly used inhaled GC was fluticasone, with most children receiving doses of ∼500 µg/day or ≤500 µg fluticasone or equivalent together with minimal additional oral GCs. The risks of adrenal suppression with inhaled GCs have previously been highlighted, for example, a UK survey of adult endocrinologists and paediatricians identified a number of cases of ‘acute adrenal crisis’ in individuals with asthma treated with inhaled GCs, and the relatively low response rate (24%) in this study may have resulted in an underestimation of the true incidence.1 A more recent study reported that 37% of children with asthma had a peak cortisol level below the accepted threshold for biochemical normality following a low dose short synacthen test and suggested that children with asthma taking any preparation of inhaled GCs at any dose may be at risk of adrenal suppression4—that is, it is not only children on ‘excessive’ doses of inhaled GCs who are at risk.

The fact that most of the children were receiving treatment for asthma is important since in Canada, asthma affects ∼13% of children (http://www.asthma.ca) and in the UK, 1.1 million children (1 in 11) currently receive treatment for asthma (http://www.asthma.org.uk) and many of these children will be managed by their general practitioner rather than in secondary care. Nevertheless, despite the majority of children in the study having asthma, it is not only inhaled GCs that pose a risk, the remaining nine children were being treated with intravenous plus oral GCs or oral alone for a variety of conditions, including malignancy, eosinophilic oesophagitis, rheumatological conditions or bronchopulmonary dysplasia.

What this study does not tell us is how common adrenal suppression is in the at-risk population—that is, the incidence of symptomatic adrenal suppression in children who were prescribed GCs during this time. Mortimer et al5 attempted to address this in a large population-based study of all age groups by cross-referencing adrenal suppression as a diagnosis with all electronic prescriptions. Although they found that oral or inhaled corticosteroid therapy resulted in a dose-related increased risk of adrenal insufficiency and that the absolute risk was small, no incidence was reported.

This study highlights (again) that clinicians in any discipline must be aware of the risk of adrenal suppression in children receiving any form of GC. The high proportion of children with growth failure underlines the importance of close monitoring of growth. However, clinicians must be aware that many children with adrenal suppression can present with non-specific symptoms alone and consider undertaking assessment of adrenal function or instigate proactive screening in those at risk. The presence of other systemic side effects of GCs such as Cushing's syndrome and secondary diabetes indicates significant systemic absorption and should prompt rapid evaluation for adrenal suppression. Those working in emergency care need to be alert to the possibility of adrenal suppression in children on GC therapy presenting with non-specific symptoms indicative of adrenal suppression or adrenal crisis—indeed, Todd et al found that adrenal crisis was not correctly identified at presentation in 11% of cases.1 The introduction of alert schemes on electronic healthcare systems may be one way of facilitating this.6 As the authors of this study state, appropriate use of GCs and the awareness of adrenal suppression have the potential to reduce morbidity and mortality in children, although further studies are required to assess the utility and effectiveness of screening.