The adrenal cortex produces three functionally and structurally distinct classes of steroid hormones from a common cholesterol precursor: glucocorticoids, mineralocorticoids, and adrenal androgens. Each class is synthesized in a specific cortical zone under distinct regulatory control. Understanding this zonal and enzymatic organization is prerequisite to understanding adrenocortical pharmacology, because the drugs used to treat adrenal excess, adrenal insufficiency, and congenital adrenal hyperplasia (CAH) all act at defined points within this pathway.
The adrenal cortex is organized into three concentric zones that differ in their steroidogenic enzyme expression and therefore in the products they are capable of synthesizing. The outermost zona glomerulosa produces mineralocorticoids, principally aldosterone, under the primary regulatory control of the renin-angiotensin-aldosterone system (RAAS) and serum potassium concentration. The middle zona fasciculata is the largest zone and the primary source of cortisol; it is regulated almost exclusively by adrenocorticotropic hormone (ACTH) from the anterior pituitary. The innermost zona reticularis produces adrenal androgens, principally dehydroepiandrosterone (DHEA) and its sulfate ester DHEA-S (dehydroepiandrosterone sulfate), also under ACTH control.1 The medulla, the innermost structure, is not cortical tissue and produces catecholamines rather than steroid hormones; its pharmacology is covered separately in the autonomic series.
All adrenocortical steroid synthesis begins with cholesterol, which is derived primarily from low-density lipoprotein (LDL) uptake from plasma and, to a lesser extent, from de novo synthesis within the adrenal cell. The first and rate-limiting step in steroidogenesis is the transport of cholesterol from the outer mitochondrial membrane to the inner mitochondrial membrane, where the initial enzymatic conversion occurs. This transport is mediated by the StAR (steroidogenic acute regulatory) protein, whose expression and phosphorylation are acutely upregulated by ACTH through a cyclic adenosine monophosphate (cAMP)-dependent mechanism. The clinical importance of StAR is demonstrated by congenital lipoid adrenal hyperplasia, a rare but severe disorder caused by StAR mutations in which cholesterol accumulates within adrenal cells and all steroid production is profoundly impaired.2
Once cholesterol reaches the inner mitochondrial membrane, the enzyme CYP11A1 (cytochrome P450 11A1), also known as the cholesterol side-chain cleavage enzyme, cleaves the cholesterol side chain to produce pregnenolone. This reaction is the committed step of steroid synthesis and is the point at which ACTH-mediated stimulation of StAR translates into increased steroid output. Pregnenolone then exits the mitochondria and enters the endoplasmic reticulum, where two parallel pathways diverge. In the delta-5 pathway, pregnenolone is converted to 17-hydroxypregnenolone by CYP17A1 (cytochrome P450 17A1), and then to DHEA by the lyase activity of the same CYP17A1 enzyme. In the delta-4 pathway, pregnenolone is converted to progesterone by 3beta-hydroxysteroid dehydrogenase (3beta-HSD), and progesterone is then hydroxylated at the 17-position by CYP17A1 to form 17-hydroxyprogesterone (17-OHP).1
The conversion of 17-OHP to 11-deoxycortisol is catalyzed by CYP21A2 (cytochrome P450 21A2), also called 21-hydroxylase. This reaction is of central clinical importance because CYP21A2 deficiency is the cause of the most common form of CAH, accounting for more than 90% of cases. In 21-hydroxylase deficiency, 17-OHP accumulates and is shunted toward androgen synthesis, producing cortisol deficiency, variable aldosterone deficiency depending on the severity of the mutation, and androgen excess. The 11-deoxycortisol produced by CYP21A2 is then converted to cortisol in the mitochondria by CYP11B1 (cytochrome P450 11B1), also called 11beta-hydroxylase. The parallel mineralocorticoid pathway proceeds from progesterone through 11-deoxycorticosterone (DOC) to aldosterone via sequential hydroxylation and oxidation reactions catalyzed by CYP11B2 (cytochrome P450 11B2), also called aldosterone synthase, which is expressed exclusively in the zona glomerulosa.2
The zona glomerulosa expresses CYP11B2 (aldosterone synthase) but not CYP17A1, making it incapable of producing cortisol or androgens but uniquely capable of producing aldosterone. The zona fasciculata expresses CYP17A1 and CYP11B1 but not CYP11B2, producing cortisol but not aldosterone. The zona reticularis expresses CYP17A1 with high lyase activity and sulfotransferase enzymes to produce DHEA and DHEA-S. This enzyme distribution means that pharmacological inhibition of a specific CYP enzyme can produce zone-selective effects on steroid output, which is directly exploited in the pharmacotherapy of Cushing syndrome using agents such as metyrapone (CYP11B1 inhibitor) and osilodrostat (potent CYP11B1 inhibitor).
Regulatory control of adrenal steroid synthesis operates at multiple levels. ACTH, released from anterior pituitary corticotroph cells in response to hypothalamic corticotropin-releasing hormone (CRH), binds to the melanocortin 2 receptor (MC2R) on zona fasciculata and reticularis cells. MC2R is a G protein-coupled receptor linked to Gs, whose activation increases intracellular cAMP, activates protein kinase A (PKA), and acutely increases StAR expression and cholesterol transport. ACTH also has trophic effects that maintain the mass and enzymatic capacity of the zona fasciculata over time, explaining why chronic ACTH deficiency leads to adrenocortical atrophy and why exogenous glucocorticoids, by suppressing ACTH, produce the same atrophic consequence in the zona fasciculata. Aldosterone regulation in the zona glomerulosa is instead driven primarily by angiotensin II (acting through AT1 (angiotensin type 1) receptors and calcium-dependent signaling) and by direct depolarization of glomerulosa cells by elevated serum potassium, both of which increase CYP11B2 expression and activity.3
StAR protein: rate-limiting cholesterol transport to inner mitochondrial membrane; ACTH-driven via cAMP/PKA; deficient in congenital lipoid adrenal hyperplasia.
CYP11A1 (P450scc): cholesterol to pregnenolone; committed step of steroid synthesis.
CYP17A1 (17-hydroxylase/lyase): hydroxylation (toward cortisol) and lyase (toward DHEA) activities; absent in zona glomerulosa, explaining why aldosterone cannot be produced there.
CYP21A2 (21-hydroxylase): 17-OHP to 11-deoxycortisol; deficiency = most common CAH; substrate accumulation drives androgen excess.
CYP11B1 (11beta-hydroxylase): 11-deoxycortisol to cortisol; target of metyrapone and osilodrostat in Cushing syndrome management.
CYP11B2 (aldosterone synthase): zona glomerulosa only; 11-deoxycorticosterone to aldosterone; regulated by angiotensin II and potassium, not ACTH.
Glucocorticoids produce their anti-inflammatory, immunosuppressive, and metabolic effects primarily through the glucocorticoid receptor (GR), a ligand-activated transcription factor that belongs to the nuclear receptor superfamily. The GR mediates both gene activation and gene repression, and the balance between these two modes of action determines the therapeutic and adverse effect profiles of glucocorticoid therapy. Separating transactivation (which drives most metabolic adverse effects) from transrepression (which drives most anti-inflammatory effects) has been a central goal of steroid drug development for decades, though no clinically available agent has yet achieved this separation completely.
The GR is encoded by the NR3C1 (nuclear receptor subfamily 3, group C, member 1) gene and exists in multiple isoforms generated by alternative splicing and alternative translation initiation. The principal functional isoform is GR-alpha (glucocorticoid receptor alpha), which is expressed ubiquitously and mediates the classical genomic actions of glucocorticoids. GR-beta (glucocorticoid receptor beta) is an alternatively spliced isoform that differs in its ligand-binding domain: GR-beta does not bind glucocorticoids and acts as a dominant-negative inhibitor of GR-alpha activity when co-expressed in the same cell. Elevated GR-beta expression in peripheral blood mononuclear cells has been associated with glucocorticoid resistance in asthma and other inflammatory conditions, and may explain some cases of clinically apparent steroid unresponsiveness.4 The unliganded GR-alpha resides in the cytoplasm as part of a multiprotein chaperone complex that maintains the receptor in a conformation competent for ligand binding.
The cytoplasmic chaperone complex includes HSP90 (heat shock protein 90) as its core component, along with HSP70 (heat shock protein 70), the co-chaperone p23, and immunophilins including FKBP51 (FK506-binding protein 51) and FKBP52 (FK506-binding protein 52). HSP90 binds to the ligand-binding domain of GR-alpha and holds it in an open conformation that permits access of the steroid ligand to the hydrophobic binding pocket. FKBP51 and FKBP52 compete for the same binding site on the HSP90 (heat shock protein 90)-GR complex and have opposing effects on receptor function: FKBP52 promotes nuclear translocation while FKBP51 inhibits it. Glucocorticoid binding induces a conformational change in GR-alpha that causes dissociation of FKBP51 and recruitment of FKBP52, promoting cytoskeletal-mediated transport of the ligand-receptor complex to the nucleus.5 The immunosuppressants tacrolimus and cyclosporine, which bind FKBP51 and FKBP52 and cyclophilin respectively, can alter glucocorticoid sensitivity through effects on this chaperone system, a pharmacological interaction relevant in transplant patients receiving both agents.
In the nucleus, the activated GR-alpha homodimerizes and binds to specific DNA (deoxyribonucleic acid) sequences called GRE (glucocorticoid response element) sequences, which consist of two imperfect palindromic half-sites separated by a three-nucleotide spacer. GRE binding recruits coactivator complexes including CREB (cyclic adenosine monophosphate response element-binding protein)-binding protein/p300 (CBP/p300) and SRC-1 (steroid receptor coactivator 1), and the mediator complex, which collectively remodel chromatin and enhance transcription of glucocorticoid-responsive genes. Classical GRE-dependent transactivation drives the expression of numerous genes that contribute to adverse effects: gluconeogenic enzymes (phosphoenolpyruvate carboxykinase, glucose-6-phosphatase), which produce hyperglycemia; lipolytic enzymes; osteocalcin suppression, contributing to osteoporosis; and muscle-specific ubiquitin ligases MuRF1 and MAFbx that drive skeletal muscle atrophy.4 Negative GREs (nGREs) also exist, where GR-alpha binding represses transcription of target genes including the pro-opiomelanocortin (POMC) gene itself, contributing to HPA (hypothalamic-pituitary-adrenal) axis feedback suppression.
The anti-inflammatory actions of glucocorticoids depend heavily on a mechanistically distinct mode of GR action called tethered transrepression. Rather than binding directly to DNA, GR-alpha monomers physically interact with and inhibit the activity of pro-inflammatory transcription factors, principally NF-kappaB (nuclear factor kappa-light-chain-enhancer of activated B cells) and AP-1 (activator protein 1). NF-kappaB normally drives transcription of a large array of pro-inflammatory genes including COX-2 (cyclooxygenase-2), iNOS (inducible nitric oxide synthase), multiple interleukin genes, adhesion molecules (ICAM-1, E-selectin), and matrix metalloproteinases. Glucocorticoid-activated GR-alpha binds directly to the p65 subunit of NF-kappaB, preventing its interaction with coactivators and blocking transcription of NF-kappaB target genes without requiring GRE binding by the GR itself.6 Similarly, GR-alpha tethers to and inhibits AP-1, a heterodimeric transcription factor formed by c-Fos and c-Jun subunits that drives expression of matrix metalloproteinases, growth factors, and additional cytokines. Glucocorticoids also stabilize IkappaB (the inhibitory protein that sequesters NF-kappaB in the cytoplasm) by inducing IkappaB gene transcription through a GRE-dependent mechanism, providing a second mechanism of NF-kappaB suppression.
The adverse metabolic effects of glucocorticoids (hyperglycemia, osteoporosis, muscle atrophy, skin thinning, growth suppression) are driven predominantly by GRE-dependent transactivation, while the anti-inflammatory effects are driven predominantly by tethered transrepression of NF-kappaB and AP-1. This mechanistic distinction has motivated decades of research into dissociated glucocorticoid agonists (DIGAs) and selective glucocorticoid receptor modulators (SEGRMs) that favor transrepression over transactivation. Although several compounds have demonstrated this separation in cell-based assays, no agent has achieved it sufficiently in humans to be clinically useful. Budesonide, with its high first-pass metabolism, achieves a partial separation by limiting systemic transactivation while maintaining topical transrepression in bowel mucosa, which is the pharmacokinetic basis of its use in inflammatory bowel disease as an alternative to systemic prednisone.
Non-genomic glucocorticoid effects occur within seconds to minutes of administration, far too rapidly to be explained by gene transcription and translation, which require at least 30 to 60 minutes. Several mechanisms account for these rapid effects. Membrane-associated GRs, structurally related to the classical GR-alpha, couple to Src kinase and phosphatidylinositol 3-kinase (PI3K) signaling through a non-genomic pathway that can rapidly suppress arachidonic acid release from cell membranes through a mechanism independent of lipocortin induction. At high pharmacological concentrations, glucocorticoids interact non-specifically with cell membranes, altering membrane fluidity and ion channel function. Membrane-associated annexin-A1 (lipocortin-1), rapidly translocated to the outer membrane surface by glucocorticoid signaling, inhibits phospholipase A2 and reduces arachidonic acid availability for eicosanoid synthesis by a mechanism that can be demonstrated within minutes in vitro.5 Clinically, the rapid anti-inflammatory effect seen with high-dose intravenous (IV) methylprednisolone in acute spinal cord injury, acute gout, or severe asthma is at least partially attributable to these non-genomic mechanisms, though genomic effects dominate during sustained therapy.
Glucocorticoid signaling is further modulated by post-translational modifications of the GR itself, including phosphorylation, acetylation, ubiquitination, and sumoylation. Phosphorylation of GR-alpha at multiple serine residues by mitogen-activated protein kinase (MAPK) pathways, cyclin-dependent kinases, and other kinases influences the receptor's transcriptional activity, nuclear localization, and protein stability. Pro-inflammatory cytokines, including tumor necrosis factor-alpha (TNF-alpha) and interleukin-1 (IL-1), activate kinase pathways that phosphorylate GR-alpha in ways that favor GRE-independent transrepressive activity while simultaneously reducing GRE-mediated transactivation, creating a context-dependent modulation of glucocorticoid action that varies with the inflammatory milieu. This phosphorylation-dependent regulation of GR activity helps explain why glucocorticoid sensitivity and the balance of their effects can differ significantly between different cell types and inflammatory states, a phenomenon directly relevant to the variable clinical responses seen in complex inflammatory diseases.6
Classical GRE transactivation: GR-alpha homodimer binds GRE sequences; recruits CBP/p300 and SRC-1 coactivators; drives metabolic gene transcription (gluconeogenesis, muscle atrophy, osteocalcin suppression). Onset 30–60 minutes minimum; dominant during sustained therapy.
Tethered transrepression: GR-alpha monomer physically binds p65 subunit of NF-kappaB and c-Fos/c-Jun of AP-1; blocks pro-inflammatory gene transcription without GRE binding. Principal mechanism of anti-inflammatory action. Also induces IkappaB transcription via nGRE, providing dual NF-kappaB suppression.
Non-genomic rapid effects: membrane-associated GR coupled to Src/PI3K; rapid annexin-A1 externalization inhibiting phospholipase A2; direct membrane effects at high concentrations. Onset seconds to minutes. Relevant to high-dose pulse therapy.
GR-beta dominant negative: does not bind ligand; competes with GR-alpha for coactivators and DNA binding sites; elevated expression associated with clinical glucocorticoid resistance.
The hypothalamic-pituitary-adrenal (HPA) axis is a tightly regulated neuroendocrine system that controls cortisol secretion in response to both basal physiological demands and superimposed stressors. Glucocorticoid drugs suppress this axis in a dose- and duration-dependent fashion, with consequences ranging from a blunted stress response to complete adrenocortical atrophy. Understanding HPA axis physiology is therefore inseparable from understanding the risks of exogenous glucocorticoid therapy and the rationale for steroid tapers, timing strategies, and stress-dose supplementation.
The HPA axis begins in the paraventricular nucleus (PVN) of the hypothalamus, where specialized parvocellular neurons synthesize and release CRH (corticotropin-releasing hormone) and, to a lesser extent, AVP (arginine vasopressin) into the hypophyseal portal circulation. CRH acts on CRH receptor type 1 (CRHR1) on anterior pituitary corticotroph cells, stimulating synthesis and secretion of ACTH (adrenocorticotropic hormone) from the precursor protein POMC (pro-opiomelanocortin). POMC is cleaved to produce ACTH, beta-endorphin, and other peptides; only ACTH is relevant to adrenal stimulation. ACTH travels in the systemic circulation to the adrenal cortex, where it binds MC2R (melanocortin 2 receptor) on zona fasciculata and reticularis cells, activating adenylyl cyclase, raising cAMP (cyclic adenosine monophosphate), and activating PKA (protein kinase A), which phosphorylates StAR protein and acutely increases cortisol synthesis as described in Section 1.3
Cortisol secretion follows a pronounced circadian rhythm driven by the suprachiasmatic nucleus (SCN) of the hypothalamus, the central circadian pacemaker. CRH and ACTH pulses are most frequent and of greatest amplitude in the early morning hours, with the cortisol nadir occurring around midnight and the peak cortisol concentration occurring 30 to 60 minutes after awakening, a phenomenon called the cortisol awakening response. Cortisol levels fall progressively through the morning and afternoon, reaching their lowest values in the first half of sleep before the pre-awakening rise begins again. This circadian pattern has direct implications for glucocorticoid dosing: administration of exogenous glucocorticoids in the morning, when endogenous cortisol is already near its peak and the HPA axis is therefore already partially suppressed by negative feedback, produces less cumulative HPA suppression over 24 hours than evening dosing, which occurs when the axis is in its most sensitive rebound phase.7 For this reason, once-daily oral glucocorticoid regimens should be administered in the morning, and the practice of evening dosing, while sometimes used for specific indications such as nocturnal adrenal androgen suppression in CAH (congenital adrenal hyperplasia), carries a substantially greater risk of HPA suppression.
Superimposed on the circadian pattern are ultradian pulses of cortisol secretion that occur approximately every 60 to 90 minutes throughout the day and night. These pulses reflect the pulsatile nature of ACTH release from the pituitary, which in turn reflects pulsatile CRH and AVP release from the PVN. The pulses are not clinically exploited in standard glucocorticoid therapy, but they are relevant to physiological replacement dosing in adrenal insufficiency: the single morning dose of hydrocortisone traditionally used for replacement does not replicate the physiological pulsatile pattern, and newer hydrocortisone formulations designed to mimic the circadian profile more closely are in clinical use in some centers.8
Negative feedback regulation of the HPA axis occurs at three levels. Cortisol feeds back to the anterior pituitary to suppress ACTH secretion directly, and to the hypothalamus to suppress CRH and AVP release. These feedback effects are mediated by GR (glucocorticoid receptor)-alpha binding to nGREs in the POMC gene promoter (pituitary) and the CRH gene promoter (hypothalamus), directly repressing transcription of the respective precursor peptides. A fast feedback mechanism, likely non-genomic, also operates within minutes of a cortisol rise, limiting further ACTH release acutely. This multilevel feedback system explains the pathophysiology of secondary adrenal insufficiency caused by exogenous glucocorticoids: when pharmacological glucocorticoid concentrations persistently exceed physiological cortisol levels, CRH and ACTH secretion are tonically suppressed. The duration and intensity of this suppression depend on the dose, potency, duration of exogenous administration, and the time of day at which glucocorticoids are given.7
Administering the daily oral glucocorticoid dose in the morning (typically 7:00 to 8:00 AM) exploits the normal circadian nadir of HPA axis sensitivity. At this time, endogenous cortisol is already elevated, the axis is in partial feedback inhibition, and the incremental suppression added by an exogenous morning dose is minimized. An identical dose given at 6:00 PM, when endogenous cortisol is low and the HPA axis is beginning its pre-awakening activation, suppresses the nighttime ACTH surge and the morning cortisol peak with substantially greater impact on cumulative 24-hour HPA function. For patients requiring every-other-day regimens, doses should be given on alternate mornings for the same reason. Patients should be counseled about this timing and not simply told to take their steroid "once a day."
The HPA axis responds to physiological stress with a dramatic increase in cortisol secretion that can increase plasma cortisol from a basal level of 10 to 20 micrograms per deciliter to peak values of 60 to 80 micrograms per deciliter or higher during major surgery or critical illness. This stress response is mediated by neural inputs to the PVN from the amygdala, hippocampus, brain stem, and ascending noradrenergic pathways that override the normal circadian and feedback constraints on CRH and ACTH secretion. The physiological cortisol stress response serves multiple functions: it mobilizes glucose through gluconeogenesis to fuel the stress response, modulates immune activation to prevent it from causing collateral tissue damage, and sensitizes cardiovascular tissues to the vasopressor effects of catecholamines. In patients whose HPA axis has been suppressed by exogenous glucocorticoids, this stress response is impaired or absent, creating the risk of adrenal crisis during physiological stressors. Stress-dose glucocorticoid supplementation in such patients is discussed in Module 2 in the context of perioperative management.3
Doses associated with HPA suppression: prednisone equivalent greater than 20 mg/day for more than 3 weeks; any dose greater than 40 mg/day for more than 1 week. Doses less than 5 mg/day prednisone equivalent rarely cause clinically significant suppression.
Duration threshold: suppression is unlikely after less than 3 weeks of any dose; becomes near-certain after more than 4 weeks of doses exceeding the physiological equivalent (approximately 7.5 mg/day prednisone).
Timing effect: evening doses suppress more than morning doses at the same daily dose; divided doses suppress more than single morning doses.
Route considerations: inhaled corticosteroids (ICS) at standard doses rarely suppress the HPA axis in adults; high-dose ICS (fluticasone propionate greater than 500 micrograms/day) has been associated with measurable HPA suppression, particularly in children.
Individual variability: some patients show measurable suppression at lower doses and shorter durations than expected; clinical vigilance is required whenever systemic glucocorticoids are used beyond a few days.
The clinical glucocorticoids in routine use span a wide range of pharmacokinetic and pharmacodynamic properties. Understanding these differences is necessary for rational agent selection: the choice between hydrocortisone, prednisone, methylprednisolone, dexamethasone, and budesonide is not arbitrary but is grounded in their distinct potency ratios, plasma half-lives, biologic durations of action, protein binding characteristics, and routes of elimination. These properties determine the appropriate clinical context for each agent and the specific monitoring requirements they carry.
Hydrocortisone is the pharmaceutical form of endogenous cortisol and is the reference compound against which all other glucocorticoids are compared. It has an oral bioavailability of approximately 75 to 80% and is well absorbed from the gastrointestinal tract. In plasma, approximately 90% of cortisol is bound to proteins: 70 to 75% to CBG (corticosteroid-binding globulin, also called transcortin) and 15 to 20% to albumin, with only 5 to 10% existing as free, biologically active drug. CBG has high affinity but low capacity for cortisol; once CBG is saturated at plasma cortisol concentrations of approximately 25 to 30 micrograms per deciliter, additional cortisol binds only to albumin, which has lower affinity but vastly higher capacity. This saturable binding has pharmacological consequences: at pharmacological doses, the free fraction increases disproportionately, amplifying glucocorticoid effects at tissues beyond what total plasma concentration would predict. Hydrocortisone has a plasma half-life of approximately 60 to 90 minutes but a biologic duration of action of 8 to 12 hours due to the persistence of GR (glucocorticoid receptor)-mediated transcriptional changes after plasma concentrations have declined.9 This dissociation between plasma half-life and biologic effect duration is characteristic of all glucocorticoids and is clinically important: it means that once-daily or twice-daily dosing is pharmacologically sufficient even for drugs with short plasma half-lives.
Prednisone is an inactive prodrug that requires hepatic conversion to prednisolone by 11beta-hydroxysteroid dehydrogenase type 1 (11beta-HSD1) for biological activity. In patients with normal hepatic function, this conversion is rapid and nearly complete, making prednisone and prednisolone clinically interchangeable for most purposes. However, in patients with severe hepatic insufficiency, impaired 11beta-HSD1 activity can reduce prednisolone generation from prednisone, and prednisolone itself is preferred in this setting to bypass the activation step. Prednisolone has an oral bioavailability of approximately 80 to 90% and a plasma half-life of 2 to 3 hours, with a biologic duration of action of 18 to 36 hours. Both prednisone and prednisolone have moderate mineralocorticoid activity relative to hydrocortisone, approximately 0.8-fold on a milligram-equivalent basis, which becomes clinically relevant at high doses with risk of sodium retention, hypertension, and hypokalemia. The anti-inflammatory potency of prednisone/prednisolone relative to hydrocortisone is approximately 4:1 on a milligram basis.9
Methylprednisolone has an anti-inflammatory potency approximately 5-fold that of hydrocortisone and negligible mineralocorticoid activity, making it the preferred agent when sodium retention must be minimized. Its oral bioavailability is approximately 85% and its plasma half-life is 2 to 3 hours. Intravenous (IV) methylprednisolone is widely used for pulse dosing in acute exacerbations of multiple sclerosis, acute spinal cord injury protocols, lupus nephritis flares, and acute rejection episodes after organ transplantation. IV pulse doses of 500 to 1000 mg produce extremely high peak plasma concentrations that saturate CBG and albumin completely, driving large free fractions and activating non-genomic in addition to genomic mechanisms, which may contribute to the rapid effects seen within hours of infusion. The biologic duration of action of methylprednisolone is similar to that of prednisone/prednisolone at 18 to 36 hours.10
Dexamethasone has the highest anti-inflammatory potency among routinely used systemic glucocorticoids, approximately 25 to 30 times that of hydrocortisone, and essentially no mineralocorticoid activity. Its oral bioavailability is approximately 78% and its plasma half-life is 3 to 5 hours, but its biologic duration of action is 36 to 54 hours, making it a long-acting agent. The absence of mineralocorticoid activity and long duration make dexamethasone particularly suitable for cerebral edema management, anti-emetic use, fetal lung maturation (antenatal dexamethasone), diagnostic testing (overnight and two-day dexamethasone suppression tests for Cushing syndrome), and treatment of COVID-19 (coronavirus disease 2019)-related ARDS (the RECOVERY (Randomized Evaluation of COVID-19 Therapy) trial protocol used dexamethasone 6 mg daily for up to 10 days). Dexamethasone is also the preferred agent for spinal cord compression and CNS (central nervous system) lymphoma-related edema because of its high potency and absence of sodium-retaining effects. Because of its long biologic half-life, dexamethasone causes more sustained HPA (hypothalamic-pituitary-adrenal) axis suppression per dose than shorter-acting agents, and is not appropriate for chronic anti-inflammatory use if HPA suppression is a concern.7
Budesonide is a high-potency glucocorticoid with approximately 200-fold greater topical anti-inflammatory activity than hydrocortisone. When given orally (in the controlled-ileal-release or extended-release formulations used for Crohn disease and microscopic colitis) or inhaled, budesonide achieves high local tissue concentrations in the gut mucosa or airway. It is then absorbed systemically but undergoes approximately 85 to 90% first-pass hepatic metabolism to inactive 16-alpha-hydroxyprednisolone and 6-beta-hydroxybudesonide metabolites, limiting systemic bioavailability to approximately 10 to 15%. This pharmacokinetic profile achieves the dissociation between local anti-inflammatory transrepression and systemic transactivation-mediated adverse effects that has eluded true receptor-level DIGRA development. HPA suppression is significantly less with budesonide than with equipotent doses of prednisone, though not absent at high doses or with prolonged use. This is the pharmacokinetic basis of preferring budesonide over prednisone for induction of remission in mild-to-moderate ileocecal Crohn disease and for maintenance in microscopic colitis.
The volume of distribution (VD) of glucocorticoids reflects their lipophilic nature and extensive tissue penetration. Hydrocortisone has a VD of approximately 40 liters in adults, prednisolone approximately 50 liters, and dexamethasone approximately 70 liters, all substantially greater than plasma volume and reflecting significant tissue uptake. This tissue distribution contributes to the biologic duration of action exceeding the plasma half-life: glucocorticoids stored in tissue compartments slowly re-equilibrate with plasma as plasma concentrations fall, but their genomic effects persist as long as GR-mediated transcription remains altered. The elimination of glucocorticoids occurs primarily through hepatic metabolism, predominantly by CYP3A4 (cytochrome P450 3A4), with renal excretion of the polar metabolites. CYP3A4 inducers (rifampin, phenytoin, carbamazepine, phenobarbital) increase glucocorticoid metabolism and can reduce plasma concentrations by 50 to 75%, potentially causing loss of therapeutic effect or adrenal crisis in dependent patients. CYP3A4 inhibitors (ketoconazole, itraconazole, ritonavir and other HIV (human immunodeficiency virus) protease inhibitors, clarithromycin) increase glucocorticoid plasma concentrations and can cause iatrogenic Cushing syndrome even with standard doses, particularly with inhaled corticosteroids (ICS) in patients on ritonavir-boosted antiretroviral regimens.11
Hydrocortisone: anti-inflammatory potency 1x; mineralocorticoid activity 1x; equivalent dose 20 mg; plasma half-life 60–90 min; biologic duration 8–12 h. Use: physiological replacement, stress dosing, topical preparations, short-term IV.
Prednisone/Prednisolone: anti-inflammatory 4x; mineralocorticoid 0.8x; equivalent dose 5 mg; plasma half-life 2–3 h; biologic duration 18–36 h. Use: most oral anti-inflammatory indications; prednisone requires hepatic activation to prednisolone.
Methylprednisolone: anti-inflammatory 5x; mineralocorticoid negligible; equivalent dose 4 mg; plasma half-life 2–3 h; biologic duration 18–36 h. Use: IV pulse therapy; preferred when sodium retention must be avoided.
Dexamethasone: anti-inflammatory 25–30x; mineralocorticoid none; equivalent dose 0.75 mg; plasma half-life 3–5 h; biologic duration 36–54 h. Use: cerebral edema, fetal lung maturation, ARDS (RECOVERY trial), diagnostic suppression testing, anti-emetic. Avoid for chronic therapy due to sustained HPA suppression.
Budesonide: topical anti-inflammatory approximately 200x; low systemic bioavailability due to 85–90% first-pass hepatic metabolism. Use: inhaled (asthma, COPD), oral controlled-release (Crohn disease, microscopic colitis), intranasal (allergic rhinitis). Preferred over prednisone when steroid-sparing is the goal and local drug delivery is feasible.
Glucocorticoid-induced adrenal insufficiency is the most common cause of adrenal insufficiency encountered in clinical practice, far exceeding primary Addison disease in prevalence. It arises from HPA (hypothalamic-pituitary-adrenal) axis suppression by exogenous glucocorticoids and manifests as an impaired cortisol stress response rather than, in most cases, frank glucocorticoid deficiency at rest. Because the condition is iatrogenic, often unrecognized, and potentially life-threatening when a physiologically stressed patient cannot mount an adequate cortisol response, its pharmacological basis, assessment, and clinical management must be understood by all prescribers of glucocorticoids.
Suppression of the HPA axis by exogenous glucocorticoids follows predictable dose-response and time-course relationships but with substantial interindividual variability. At doses below the physiological cortisol equivalent of approximately 5 to 7.5 mg prednisone per day, HPA suppression is generally not clinically significant because the exogenous dose does not substantially exceed what the axis would produce endogenously. At doses between 7.5 and 20 mg prednisone per day given for more than 3 weeks, partial HPA suppression is common; the axis retains basal function but the stress response is blunted. At doses exceeding 20 mg prednisone per day, or any dose given for more than 4 to 6 weeks, substantial HPA suppression with impaired stress response is the rule rather than the exception. Duration is as important as dose: the same total cumulative dose given over a shorter period causes less suppression than the same dose given over a prolonged course, because the period of persistent GR (glucocorticoid receptor)-mediated POMC (pro-opiomelanocortin) and CRH (corticotropin-releasing hormone) gene repression must be long enough to produce adrenocortical atrophy.12 Recovery of HPA axis function after glucocorticoid cessation is also variable, typically requiring weeks to months, with full recovery of adrenocortical reserve sometimes taking 6 to 12 months after prolonged high-dose therapy.
Assessment of residual HPA axis function after glucocorticoid therapy is most reliably performed with dynamic testing rather than basal cortisol measurement alone. A morning plasma cortisol greater than 18 micrograms per deciliter (approximately 500 nmol per liter) measured at 8:00 to 9:00 in the morning generally indicates adequate HPA axis recovery and low risk of adrenal crisis. A morning cortisol less than 3 micrograms per deciliter indicates persistent severe suppression. Values between these thresholds require dynamic testing, most practically the low-dose short Synacthen test (LDSST), in which 1 microgram of synthetic ACTH (tetracosactide) is given intravenously and cortisol is measured at 30 minutes; a peak response greater than 18 micrograms per deciliter is considered a normal response. The insulin tolerance test (ITT), in which insulin-induced hypoglycemia is used to activate the full HPA axis stress response, remains the gold standard but is rarely used in routine clinical practice because of its risks in older or cardiac patients. When dynamic testing is not available, clinical judgment based on dose history and symptoms is the practical guide.8
The rationale for glucocorticoid tapering rather than abrupt cessation after prolonged therapy serves two distinct purposes that are often conflated. The first purpose is prevention of adrenal crisis: abrupt cessation of glucocorticoids in a patient with HPA axis suppression removes exogenous cortisol without the axis being capable of restoring endogenous cortisol production rapidly, potentially precipitating adrenal insufficiency (AI) particularly during intercurrent illness or stress. The second purpose is management of the underlying disease for which the glucocorticoid was prescribed: many inflammatory conditions will flare if the anti-inflammatory dose is reduced too rapidly, and the taper rate is constrained by disease activity rather than purely by HPA considerations. These two purposes require different taper strategies and should be separated in clinical reasoning. Tapering for HPA recovery is most critical when the total duration of therapy exceeds 3 to 4 weeks and the dose has been above physiological replacement level; below these thresholds, tapering can often be compressed or omitted.12
Adrenal crisis is a life-threatening emergency caused by acute glucocorticoid deficiency, most commonly precipitated by physiological stress (infection, surgery, trauma, vomiting preventing oral steroid ingestion) in a patient with either primary adrenal insufficiency or glucocorticoid-induced HPA suppression. Clinical features include severe hypotension or shock disproportionate to apparent cause, nausea, vomiting, abdominal pain, fever, confusion, and hyponatremia with or without hyperkalemia (the latter more prominent in primary adrenal insufficiency with concurrent mineralocorticoid deficiency). Treatment is immediate: hydrocortisone 100 mg IV bolus followed by continuous infusion at 200 mg over 24 hours (or 50 mg IV every 6 hours); aggressive IV saline resuscitation; and management of the precipitating cause. Mineralocorticoid replacement is not required during acute treatment because the high doses of hydrocortisone provide sufficient mineralocorticoid activity at these concentrations. Diagnosis should not delay treatment: if adrenal crisis is suspected clinically, give hydrocortisone immediately; a random cortisol and ACTH sample drawn before the dose allows retrospective confirmation.
Sick-day rules are the practical clinical tool for preventing adrenal crisis in patients with known HPA suppression or primary adrenal insufficiency who are managed on ongoing glucocorticoid replacement. The standard rule is to double or triple the daily glucocorticoid dose during any illness associated with fever greater than 38 degrees Celsius, vomiting, diarrhea, or any surgical procedure or trauma. If the patient is unable to take oral medication due to vomiting, parenteral hydrocortisone (100 mg IM or IV) must be administered; this is why patients with adrenal insufficiency are instructed to keep an emergency hydrocortisone injection kit and are trained in its self-administration, and why medical alert jewelry is recommended. These rules apply equally to patients with primary adrenal insufficiency and to patients with glucocorticoid-induced secondary adrenal insufficiency who are in the process of tapering, because both groups lack the ability to mount an adequate stress response independently. Prescribers must provide explicit written sick-day instructions to every patient on more than 3 to 4 weeks of pharmacological glucocorticoid therapy, as failure to do so is a preventable cause of adrenal crisis-related mortality.3
The question of whether patients on inhaled corticosteroids (ICS) require sick-day rules or HPA assessment is more nuanced. Standard-dose ICS in adults (fluticasone propionate up to 500 micrograms per day or equivalent) produces minimal systemic bioavailability and HPA suppression in most adults and does not require systematic HPA monitoring. However, high-dose ICS (fluticasone propionate greater than 500 micrograms per day, ciclesonide, budesonide in higher dose ranges), particularly when combined with intranasal steroids or topical potent steroids applied to large body surface areas, can produce measurable HPA suppression, most prominently in children. Patients receiving high-dose ICS who develop features of Cushing syndrome or who have poor growth velocity in children warrant morning cortisol testing. An additional consideration is the pharmacokinetic drug interaction between ICS and CYP3A4 (cytochrome P450 3A4) inhibitors: patients on ritonavir-boosted antiretroviral therapy who receive fluticasone-containing ICS can develop iatrogenic Cushing syndrome and HPA suppression because ritonavir markedly inhibits the CYP3A4-mediated first-pass and systemic metabolism of fluticasone, raising systemic fluticasone concentrations 350-fold or more. Substitution with beclomethasone, which is not a CYP3A4 substrate, is the standard approach in this population.11
Low risk (no taper needed, no testing needed): systemic glucocorticoids for less than 3 weeks at any dose; prednisone equivalent less than 5 mg/day for any duration; ICS at standard doses in adults; single-joint corticosteroid injections.
Moderate risk (taper prudent; consider HPA testing if symptomatic): prednisone 7.5–20 mg/day for 3–8 weeks; high-dose ICS in children or combined topical plus ICS use; epidural or extensive soft tissue corticosteroid injections (systemic absorption can be significant).
High risk (taper required; HPA testing before discontinuation if feasible): prednisone greater than 20 mg/day for more than 3 weeks; any systemic glucocorticoid for more than 3 months; Cushingoid features present regardless of dose.
Testing threshold: morning cortisol at 8–9 AM after holding the glucocorticoid dose the night before. If greater than 18 micrograms/dL: axis likely recovered. If 3–18 micrograms/dL: LDSST (low-dose short Synacthen test) recommended. If less than 3 micrograms/dL: continued replacement required.
Sick-day rules: provide written instructions to all patients in the moderate and high-risk categories. Double or triple oral dose during febrile illness; use IM/IV hydrocortisone 100 mg if unable to take oral medication.
Miller WL, Auchus RJ. The molecular biology, biochemistry, and physiology of human steroidogenesis and its disorders. Endocr Rev. 2011;32(1):81–151.
doi:10.1210/er.2010-0013Stocco DM. StAR protein and the regulation of steroid hormone biosynthesis. Annu Rev Physiol. 2001;63:193–213.
doi:10.1146/annurev.physiol.63.1.193Chrousos GP. Adrenocorticosteroids and adrenocortical antagonists. In: Katzung BG, ed. Basic and Clinical Pharmacology. 14th ed. New York, NY: McGraw-Hill; 2018:687–710.
Barnes PJ. Glucocorticosteroids: current and future directions. Br J Pharmacol. 2011;163(1):29–43.
doi:10.1111/j.1476-5381.2010.01199.xStahn C, Buttgereit F. Genomic and nongenomic effects of glucocorticoids. Nat Clin Pract Rheumatol. 2008;4(10):525–533.
doi:10.1038/ncprheum0898Rhen T, Cidlowski JA. Antiinflammatory action of glucocorticoids — new mechanisms for old drugs. N Engl J Med. 2005;353(16):1711–1723.
doi:10.1056/NEJMra050541Debono M, Ghobadi C, Rostami-Hodjegan A, et al. Modified-release hydrocortisone to provide circadian cortisol profiles. J Clin Endocrinol Metab. 2009;94(5):1548–1554.
doi:10.1210/jc.2008-2380Crowley RK, Sherlock M, Agha A, et al. Clinical insights into adrenal insufficiency: low-dose ACTH stimulation test versus insulin tolerance test. Clin Endocrinol (Oxf). 2010;73(4):461–466.
doi:10.1111/j.1365-2265.2010.03829.xLiu D, Ahmet A, Ward L, et al. A practical guide to the monitoring and management of the complications of systemic corticosteroid therapy. Allergy Asthma Clin Immunol. 2013;9(1):30.
doi:10.1186/1710-1492-9-30Czock D, Keller F, Rasche FM, Haussler U. Pharmacokinetics and pharmacodynamics of systemically administered glucocorticoids. Clin Pharmacokinet. 2005;44(1):61–98.
doi:10.2165/00003088-200544010-00003Clevenbergh P, Geleziunas R, Deeks SG, et al. Ritonavir-boosted lopinavir as the first lopinavir-based antiretroviral regimen in treatment-naive patients: a case of pharmacokinetic interactions when fluticasone nasal spray is coadministered. Clin Pharmacol Ther. 2002;72(2):236–240. [Ritonavir-fluticasone CYP3A4 interaction reference; see also Kedem E et al., Drug Des Devel Ther. 2010;4:281–285.]
Henzen C, Suter A, Lerch E, et al. Suppression and recovery of adrenal response after short-term, high-dose glucocorticoid treatment. Lancet. 2000;355(9203):542–545.
doi:10.1016/S0140-6736(99)06290-X