Corticosteroids

Glucocorticoids are small and lipophilic, so they diffuse into cells and bind the cytoplasmic glucocorticoid receptor, which is held by heat-shock proteins until hormone binds. The receptor sheds them and the complex moves into the nucleus, where it binds DNA. Transactivation switches on genes such as lipocortin-1, IL-10 and an NF-κB inhibitor; lipocortin-1 decreases phospholipase A2, cutting the arachidonic acid that feeds prostaglandins and leukotrienes. Transrepression blocks transcription factors such as NF-κB and activator protein-1, so COX-2, inducible NO synthase and most inflammatory cytokines (IL-1, IL-2, IL-6, TNF-α, interferon-γ) fall. The anti-inflammatory effects come mainly through transrepression, the metabolic effects through transactivation.

These genomic effects take time. At high doses glucocorticoids also act within seconds to minutes through non-genomic pathways — membrane-associated receptors on cells such as T lymphocytes — which is the basis of high-dose pulse therapy.

Glucocorticoids raise the white count while depleting most cell lines. Neutrophils rise: they detach from the endothelium (demargination), migrate less into tissue, live longer and are released from the marrow. Eosinophils fall through apoptosis and sequestration, and lymphocytes, monocytes and basophils fall too. In lymphoid tissue they inhibit IL-2 signaling and T-cell proliferation, cause T-lymphocyte apoptosis, inhibit NF-κB and mast-cell degranulation, and shrink the thymus, spleen and lymph nodes; macrophages present and phagocytose antigen poorly. B-cell function and immunoglobulin production are affected very little.

The same suppression of cytokines explains why infections are both more likely and harder to see: patients may not mount the usual inflammatory and febrile response, so an infection can go unrecognized early.

Cortisol is a stress hormone that keeps glucose available to the brain and muscle. It induces gluconeogenic enzymes such as PEPCK, increases glycogenolysis, antagonizes insulin (less glycogen synthesis and less GLUT4 glucose uptake) and drives muscle protein catabolism, sending amino acids to the liver as gluconeogenic substrate. The hyperglycemia is dose-dependent and affects postprandial glucose more than fasting glucose, and it is hardest to control in patients who already have diabetes.

The catabolic state also explains myopathy — painless proximal weakness with normal CK and type II fiber atrophy without inflammation — and bone loss: glucocorticoids activate osteoclasts through RANK-ligand, reduce the number and function of osteoblasts and osteocytes, and inhibit calcium absorption in the gut, with trabecular bone lost first, within the first 6–12 months.

Cortisol can activate both glucocorticoid and mineralocorticoid (aldosterone) receptors. 11β-hydroxysteroid dehydrogenase type 2 converts cortisol to inert cortisone and so protects the mineralocorticoid receptor — but when cortisol levels are high and the enzyme is saturated, cortisol activates it, usually with supraphysiologic exogenous doses. Mineralocorticoid action retains sodium and water and wastes potassium, causing edema, hypertension and hypokalemia.

The drugs differ in this balance and in duration. Hydrocortisone (cortisol) is short-acting with roughly equal glucocorticoid and mineralocorticoid effects. Prednisone, prednisolone and methylprednisolone are intermediate-acting with little mineralocorticoid effect; methylprednisolone is about 5 times as potent an anti-inflammatory as hydrocortisone. Dexamethasone is long-acting with the highest glucocorticoid efficacy and very little, if any, mineralocorticoid activity. Prednisone is itself biologically inert — a prodrug converted to prednisolone in the liver. In primary adrenal insufficiency, where aldosterone is lost, hydrocortisone or prednisone supplies some mineralocorticoid effect but dexamethasone does not, so a separate mineralocorticoid (fludrocortisone) is needed in full.

Glucocorticoids feed back on the HPA axis, suppressing CRH and ACTH directly (and indirectly, by suppressing the cytokines that stimulate them). Prolonged ACTH suppression causes functional adrenal atrophy of the cortisol-producing cortex, sparing the outer mineralocorticoid-producing cortex, which is independent of ACTH. Who will be suppressed is hard to predict from dose and duration alone, and inhaled, topical and joint-injected steroids can contribute. Exogenous glucocorticoids are the most common cause of this central (secondary or tertiary) adrenal insufficiency, and because aldosterone stays under renin–angiotensin control, these patients do not have the hyperkalemia or skin hyperpigmentation of primary adrenal insufficiency.

Abrupt withdrawal, or a stressor such as surgery, trauma or infection without a dose increase, can precipitate adrenal crisis: vomiting, abdominal pain, hypotension and shock; hyponatremia or hypoglycemia, especially with hypotension, may point to it. That is why long courses are tapered rather than stopped, why the suppression may last for months — up to 9–12 months after long-term high-dose therapy — and why patients need increased glucocorticoid cover for stress during that time ("sick-day rules"). Giving a single daily dose in the early morning, when the adrenal cortex is most active, suppresses the axis least. Adrenal crisis itself is treated with hydrocortisone.

Glucocorticoid therapy is the most common cause of Cushing syndrome: weight gain with truncal obesity, a dorsocervical fat pad ("buffalo hump") and moon facies, often within the first 2 months, with thin fragile skin, striae, purpura and poor wound healing (reduced collagen synthesis). The same drug causes the classic list of harms, each rising with dose and duration: hyperglycemia, osteoporosis and osteonecrosis, proximal myopathy, infection, hypertension, posterior subcapsular cataracts and painless glaucoma (the optic-nerve damage can be permanent), psychiatric effects (early euphoria or hypomania, later depression, psychosis with high doses), and growth suppression in children, which can also signal adrenal suppression.

The peptic-ulcer risk of a glucocorticoid alone is low, but with an NSAID it rises about 4-fold, which is why patients taking both are given a proton pump inhibitor.

Cerebral edema: dexamethasone is used for cerebral edema; corticosteroids are recommended for raised intracranial pressure from a neoplasm or CNS infection but generally avoided in moderate to severe brain injury. The dexamethasone suppression test: dexamethasone binds the glucocorticoid receptor with high affinity, has minimal mineralocorticoid activity and does not interfere with cortisol measurement, so it is the preferred steroid for testing the HPA axis. In a normal axis it feeds back on CRH and ACTH and cortisol falls; in pathological hypercortisolism the axis is partly or wholly resistant to that feedback. A low-dose test that fails to suppress cortisol is confirmed with a second test; plasma ACTH then separates ACTH-independent (adrenal; ACTH low) from ACTH-dependent disease, and in ACTH-dependent disease a high dose partly or completely suppresses a pituitary source (Cushing disease) but not an ectopic tumor. Patients taking any exogenous corticosteroid are identified first, because iatrogenic hypercortisolism is the most common cause of Cushing syndrome.

The placenta: it breaks down non-fluorinated glucocorticoids such as prednisone and prednisolone, so little reaches the fetus, whereas dexamethasone readily crosses the placenta. Antenatal corticosteroids — intramuscular betamethasone before 37 weeks — promote fetal lung maturity in threatened preterm delivery, and dexamethasone increases surfactant levels; repeated antenatal courses have been linked to lower birth weight and neonatal hypoglycemia. In thyroid storm, a glucocorticoid is part of the sequence because it reduces T4 → T3 conversion (Thyroid Agents page).