Adrenal Insufficiency: Review of Clinical Outcomes With Current Glucocorticoid Replacement Therapy

Gudmundur Johannsson; Alberto Falorni; Stanko Skrtic; Hans Lennernäs; Marcus Quinkler; John P. Monson; Paul M.Stewart


Clin Endocrinol. 2015;82(1):2-11. 

In This Article

Abstract and Introduction


Glucocorticoid replacement therapy in patients with adrenal insufficiency (AI), whether primary (Addison's disease) or secondary (due to hypopituitarism), has been established for some 50 years. The current standard treatment regimen involves twice- or thrice-daily dosing with a glucocorticoid, most commonly oral hydrocortisone. Based on previous small-scale studies and clinical perception, life expectancy with conventional glucocorticoid replacement therapy has been considered normal, with a low incidence of adverse events. Data from the past 10–15 years, however, have shown that morbidity remains high and life expectancy is reduced. The increased morbidity and decreased life expectancy appear to be due to both increased exposure to cortisol and insufficient cortisol coverage during infections and other stress-related events. This is thought to reflect a failure of treatment to replicate the natural circadian rhythm of cortisol release, together with a failure to identify and deliver individualized cortisol exposure and to manage patients adequately when increased doses are required. The resulting over- or under-treatment may result in Cushing-like symptoms or adrenal crisis, respectively. This review summarizes the morbidity and mortality seen in patients receiving the current standard of care for AI and suggests areas for improvement in glucocorticoid replacement therapy.


The eponymous term Addison's disease was introduced following the classic description by Dr Thomas Addison in 1849 of a 'remarkable form of anaemia' attributed to a disease of the 'supra-renal capsules' (for review, see Bishop).[1] It is now understood that the disease is due to cortisol and mineralocorticoid deficiency resulting from a variety of conditions affecting the adrenal glands (Table 1). Before the availability of synthetic glucocorticoid replacement therapy, the disease resulted in a 2-year mortality above 80% and the death of all patients within 5 years of diagnosis.[2]

A synthetic mineralocorticoid, desoxycorticosterone acetate, was introduced for the treatment of Addison's disease (primary adrenal insufficiency; PAI) in 1938. Although this improved the outcome for patients,[1] the major breakthrough in treatment occurred only after hydrocortisone (the pharmaceutical equivalent of natural cortisol) was synthesized some 10 years later (for review, see Hillier).[3]

Previous radioisotope studies to assess cortisol production in humans (for review, see Cope and Black)[4] led to a mean daily dose of hydrocortisone exceeding 30 mg for the treatment of adrenal insufficiency (AI). Recent studies of cortisol production, however, using stable isotope and deconvolution analysis, have estimated much lower rates of cortisol production in healthy individuals (approximately 10–20 mg/day) than those previously reported, leading to a corresponding reduction in the daily dose of replacement glucocorticoid.

Hydrocortisone has conventionally been administered in two or three daily doses, with the largest dose being given in the morning, followed by a reduced dose at midday and a third dose, if required, in the late afternoon or early evening. Such a dosing regimen does not, however, mimic the physiological circadian rhythm of cortisol secretion.[5,6,7,8,9] The resulting nonphysiological plasma cortisol profile and the higher than physiological total daily doses, together with inadequate treatment during intercurrent illness, may explain why PAI and secondary adrenal insufficiency (SAI) continue to be associated with considerable morbidity and reduced life expectancy.[10,11,12,13]

Here, we present a narrative review of the characteristics of AI and the morbidity and mortality associated with current glucocorticoid replacement therapy, in order to increase the understanding of the disease and to promote further discussion of the effectiveness of current treatments. In reviewing the outcome of glucocorticoid replacement, we acknowledge that PAI and SAI are two different disease entities and that confounding factors are important, such as other pituitary hormone deficiencies and their management in patients with SAI and the management of mineralocorticoid insufficiency and androgen deficiency in PAI. A detailed description and discussion of these factors is outside the scope of this review.