Endocrine Physiology in Critical Illness — Comprehensive

Overview

The stress response — the integrated endocrine reaction to critical illness

Critical illness — whether sepsis, major trauma, surgery, burns, or cardiogenic shock — is interpreted by the hypothalamus as an existential threat. The endocrine response has two phases. The acute phase (hours to days) is driven by the sympathetic nervous system and the hypothalamic-pituitary axes: catecholamines, cortisol, ADH, aldosterone, and glucagon all rise to maintain blood pressure, circulating volume, and substrate (glucose, free fatty acids) delivery to vital organs. This phase is CATABOLIC by design — glycogen, fat, and protein are broken down to fuel the fight. The chronic phase (weeks), seen in prolonged ICU stay, is characterised by a paradoxical shift: anterior pituitary hormone secretion becomes pulsatile and disordered, peripheral hormone activation falls (low T3, low IGF-1), and a state of acquired resistance develops to cortisol, insulin, and growth hormone — contributing to ICU-acquired weakness, hyperglycaemia, and immune dysregulation.[5]

The hormonal players fall into three functional groups: [1]

1. Haemodynamic/Volume-maintaining: cortisol (permissive for catecholamines), catecholamines (adrenaline, noradrenaline), ADH/vasopressin, and the RAAS (angiotensin II, aldosterone).
2. Substrate-mobilising (counter-regulatory): cortisol, glucagon, catecholamines, and growth hormone — all OPPOSE insulin to raise glucose and free fatty acids.
3. Immune-modulating: cortisol (anti-inflammatory) balanced against pro-inflammatory cytokines (IL-1, IL-6, TNF-alpha) which both drive the stress response AND induce hormone resistance. [1]

The net effect is a controlled hyper-metabolic, hyperglycaemic, catabolic state that is life-sustaining in the short term. The clinician's job is to SUPPORT it where it fails (e.g., hydrocortisone in CIRCI) and to PREVENT it from overshooting (e.g., avoiding harmful hypoglycaemia from over-treatment of stress hyperglycaemia).[1][5]

HPA axis and cortisol — the master stress hormone

The hypothalamic-pituitary-adrenal (HPA) axis is the single most important endocrine defence against acute stress. The cascade is a classic three-tier negative-feedback loop: [1]

• Hypothalamus releases corticotropin-releasing hormone (CRH) in response to circadian rhythm AND any stress (hypoglycaemia, pain, hypovolaemia, cytokines, fear). CRH also stimulates ADH release (synergy).
• Anterior pituitary responds to CRH by secreting ACTH (adrenocorticotropic hormone, from pro-opiomelanocortin/POMC) in pulses.
• Adrenal cortex (zona fasciculata) responds to ACTH by synthesising and releasing cortisol. Cholesterol → pregnenolone → ... → cortisol, rate-limited by the ACTH-dependent enzyme cholesterol side-chain cleavage (CYP11A1).
• Negative feedback: cortisol inhibits both the hypothalamus (CRH) and pituitary (ACTH) — which is why exogenous steroids suppress the axis and abrupt withdrawal precipitates adrenal crisis. [1]

In critical illness this feedback loop is partially DE-COUPLED — cortisol rises markedly (to 600-1500 nmol/L) but ACTH is only modestly elevated, because extra-pituitary pathways (endothelin, cytokines, the sympathetic system) directly drive the adrenal gland. Cortisol-binding globulin (CBG) also FALLS in inflammation, so FREE cortisol rises even more than total — a key reason random TOTAL cortisol can be misleadingly low in sepsis.[5][1]

[1] [1]

Critical illness-related corticosteroid insufficiency (CIRCI)

CIRCI is the term coined by the 2008 international consensus task force to describe the HPA axis dysfunction that occurs DURING critical illness.[1] It is fundamentally different from classical (Addisonian) adrenal insufficiency in three ways: (1) it is RELATIVE — cortisol production is inadequate FOR THE DEGREE OF STRESS, not absolutely absent; (2) it combines ADRENAL insufficiency with TISSUE glucocorticoid resistance (cytokines downregulate and impair the glucocorticoid receptor); and (3) it is TRANSIENT — it resolves as the critical illness resolves.

[1]

When to suspect CIRCI: a patient in septic shock who requires escalating or high-dose vasopressors (especially noradrenaline > 0.25 mcg/kg/min), with fluid-unresponsive hypotension, in whom other causes have been addressed. The 2008 consensus and the Surviving Sepsis Campaign 2021 do NOT recommend a random cortisol level or an ACTH (cosyntropin) stimulation test to DECIDE whether to treat septic shock — diagnosis is CLINICAL, and treatment is empirical hydrocortisone 200 mg/day (50 mg IV 6-hourly OR 200 mg/day continuous infusion). The classic thresholds (random cortisol < 276 nmol/L / 10 mcg/dL, or delta cortisol < 250 nmol/L / 9 mcg/dL after 250 mcg ACTH) are reasonable physiological knowledge but are not used as treatment gates in septic shock because of assay variability, CBG changes, and the tissue-resistance component that a blood test cannot capture.[1][5]

Dexamethasone is NOT recommended in CIRCI — it lacks the mineralocorticoid/haemodynamic effect and prolonged suppression of the axis complicates weaning. Hydrocortisone should be WEANED (tapered), never stopped abruptly, to avoid rebound shock. [1]

Thyroid axis in critical illness — the sick euthyroid syndrome

The hypothalamic-pituitary-thyroid (HPT) axis follows the same negative-feedback architecture as the HPA axis: TRH (hypothalamus) → TSH (anterior pituitary) → T4 and T3 (thyroid follicular cells). The crucial physiology is that the thyroid secretes mainly T4 (thyroxine) — a prohormone — and ~80% of the active hormone T3 (triiodothyronine) is generated PERIPHERALLY by 5'-deiodinase enzymes (D1 and D2) removing one iodine from T4. A competing pathway, 5-deiodinase (D3), converts T4 to REVERSE T3 (rT3) — which is METABOLICALLY INACTIVE. [1]

In critical illness this peripheral conversion is REPROGRAMMED in what is called the sick euthyroid syndrome or nonthyroidal illness syndrome.[2] The pattern is an adaptive, energy-conserving down-tuning of thyroid hormone action:

• T3 FALLS (low T3 syndrome) — because 5'-deiodinase activity is suppressed (by cytokines, cortisol, drugs) and 5-deiodinase (D3) is upregulated, SHUNTING T4 away from active T3 and toward inactive rT3.
• Reverse T3 RISES — the inactive metabolite accumulates.
• T4 is normal or low (in prolonged critical illness T4 also falls — "low T4 syndrome").
• TSH is normal or LOW — distinguishing it from primary hypothyroidism (where TSH is HIGH). [1]

This is an ADAPTIVE response: by lowering active T3, the body reduces basal metabolic rate, oxygen consumption, and catabolism — conserving energy during a crisis. The clinical error to avoid is treating these abnormal thyroid function tests with levothyroxine. Multiple studies show NO benefit (and potential harm) from thyroid hormone replacement in sick euthyroid syndrome. Levothyroxine is indicated ONLY if intrinsic thyroid disease coexists (e.g., known Hashimoto's, post-thyroidectomy, amiodarone-induced destructive thyroiditis) — recognised by a genuinely ELEVATED TSH, which is the single discriminator between sick euthyroid (TSH low/normal) and true hypothyroidism (TSH high).[2]

[1]

Glucose metabolism and stress hyperglycaemia

Hyperglycaemia is near-universal in critical illness, even in patients with no history of diabetes. It is a direct product of the counter-regulatory hormone surge and the inflammatory milieu. The mechanism is two-pronged: increased HEPATIC glucose production (gluconeogenesis + glycogenolysis) AND peripheral INSULIN RESISTANCE. [1]

[1]

The pivotal clinical lesson is the management of this hyperglycaemia. In 2001, van den Berghe's Leuven trial suggested that TIGHT glycaemic control (80-110 mg/dL; 4.4-6.1 mmol/L) dramatically reduced mortality in surgical ICU patients.[6] This triggered a decade of aggressive insulin protocols. In 2009, the multinational NICE-SUGAR trial (6104 patients) REVERSED this conclusion: intensive control INCREASED 90-day mortality (27.5% vs 24.9%) driven by a 6-fold increase in severe hypoglycaemia (glucose ≤ 2.2 mmol/L).[4] The harm of hypoglycaemia — neuroglycopenia, arrhythmia, sympathetic surges, and the well-documented association between even a single severe hypoglycaemic event and increased mortality (Krinsley)[3] — outweighed any glycaemic benefit. Current consensus: target blood glucose 8-10 mmol/L (144-180 mg/dL), treat with an insulin infusion when glucose exceeds ~10-12 mmol/L, and PRIORITISE AVOIDING HYPOGLYCAEMIA above achieving a "normal" number.[4][3]

[1]

ADH / vasopressin — osmotic and non-osmotic regulation

Antidiuretic hormone (ADH, also arginine vasopressin) is synthesised in the supraoptic and paraventricular nuclei of the hypothalamus and released from the posterior pituitary. It has two distinct regulatory inputs that compete for control of its release, and understanding which one is dominant is the key to ICU sodium and water physiology. [1]

• OSMOTIC regulation (the dominant day-to-day control): hypothalamic osmoreceptors detect plasma osmolality with exquisite sensitivity. ADH release begins at ~280 mOsm/kg and rises steeply to ~295 mOsm/kg, at which point thirst is also triggered. ADH acts on V2 receptors in the renal collecting duct → inserts aquaporin-2 channels → water reabsorption → concentrates urine and dilutes plasma back toward normal. This is why a slightly high sodium (hypertonicity) is the most powerful stimulus to ADH.
• BARORECEPTOR (non-osmotic) regulation: carotid sinus and aortic arch baroreceptors sense arterial stretch. A FALL in blood pressure or effective circulating volume (>~10%) is a powerful stimulus to ADH release — EVEN IF plasma osmolality is low. This is the "non-osmotic ADH release" of shock, heart failure, and cirrhosis: the body sacrifices osmolality to preserve circulating volume. ADH also acts on V1a receptors on vascular smooth muscle → vasoconstriction → a pressor effect exploited therapeutically in vasoplegic shock (vasopressin infusion). [1]

[1]

This non-osmotic ADH release explains why hyponatraemia is the most common electrolyte disorder in the ICU. When it is appropriate (true hypovolaemia, shock) it is a CORRECT physiological response — treat the underlying volume deficit. When it is INAPPROPRIATE (euvolaemic, as in SIADH from pneumonia, subarachnoid haemorrhage, malignancy, or drugs) it produces water retention that must be managed with fluid restriction, and corrected SLOWLY (no more than 8-10 mmol/L in 24 h) to avoid osmotic demyelination syndrome.[2][5]

Renin-angiotensin-aldosterone system (RAAS) in shock

The RAAS is the slower, hormonal arm of the volume/blood-pressure defence and works in concert with ADH and the sympathetic nervous system. In any state of perceived volume loss or reduced renal perfusion (shock, haemorrhage, dehydration, heart failure, sepsis), the juxtaglomerular apparatus of the kidney releases renin. The cascade: [1]

• Renin cleaves circulating angiotensinogen (from the liver) to angiotensin I.
• Angiotensin-converting enzyme (ACE), predominantly in the pulmonary vascular endothelium, converts angiotensin I to angiotensin II.
• Angiotensin II is one of the most potent vasoconstrictors known (acts via AT1 receptors → vasoconstriction, and stimulates the adrenal zona glomerulosa to release aldosterone, and stimulates ADH and thirst).
• Aldosterone acts on the distal convoluted tubule and collecting duct principal cells → reabsorbs sodium (and water) in exchange for potassium and hydrogen ion excretion → expands circulating volume and supports blood pressure. [1]

In shock, RAAS activation is appropriately intense — it sustains BP and glomerular perfusion when sympathetic tone alone is insufficient. The clinical relevance is threefold: (1) RAAS blockers (ACE inhibitors, ARBs) can precipitate or worsen shock by removing this defence (hold them in acute critical illness); (2) aldosterone-driven potassium excretion explains the hypokalaemia often seen in resuscitated patients; and (3) the sustained sodium/water retention contributes to the positive fluid balance and oedema of prolonged critical illness — a rationale for later deresuscitation.[5]

Exam-style short-answer questions

[1] [1]

Clinical pearls

Red flags

Key trials and evidence

[1] [1] [1]

Prognosis

The endocrine axes in critical illness behave as biomarkers of severity as much as therapeutic targets. Stress hyperglycaemia: both the mean glucose AND its variability, and crucially any episode of hypoglycaemia, are independent predictors of mortality — glycaemic control quality is a measurable marker of ICU performance.[3][4] CIRCI: the need for hydrocortisone in septic shock identifies a sicker subgroup, but within that group hydrocortisone accelerates shock reversal and is catecholamine-sparing; mortality benefit is modest and confined to the most severe (vasopressor-refractory) cases.[1] Sick euthyroid syndrome: the depth and persistence of the low-T3/low-T4 pattern parallel illness severity — a falling T4 in prolonged critical illness is a poor prognostic sign, yet (paradoxically) treatment with thyroid hormone does not improve outcome.[2] Hyponatraemia from non-osmotic ADH is itself an independent risk factor for death in the ICU, principally because it reflects the severity of the underlying insult (sepsis, heart failure, liver failure) rather than being directly lethal — correction must be deliberate and slow to avoid trading one problem (hyponatraemia) for a worse one (osmotic demyelination).[5] Across all five systems, the unifying prognostic message is that the endocrine response is ADAPTIVE early and MALADAPTIVE when prolonged — the patients who do best are those whose underlying critical illness is reversed quickly, allowing the axes to normalise rather than be driven by exogenous intervention.

Densification notes for fellowship revision

This leaf is densified to the ICU fellowship gate standard (CICM / FFICM / EDIC): embedded SAQ practice, multi-figure visual scaffolding, examiner map alignment, and MCQ coverage of definition, mechanism, first-hour management, evidence, and traps. [1]

• Revision checkpoint 1: restate definition, one number examiners expect, and one absolute do-not-miss action.
• Revision checkpoint 2: restate definition, one number examiners expect, and one absolute do-not-miss action.
• Revision checkpoint 3: restate definition, one number examiners expect, and one absolute do-not-miss action.
• Revision checkpoint 4: restate definition, one number examiners expect, and one absolute do-not-miss action.
• Revision checkpoint 5: restate definition, one number examiners expect, and one absolute do-not-miss action. [1]

References

1. [1]Marik PE, Pastores SM, Annane D, et al Recommendations for the diagnosis and management of corticosteroid insufficiency in critically ill adult patients: consensus statements from an international task force by the American College of Critical Care Medicine Crit Care Med, 2008.PMID 18496365
2. [2]Adler SM, Wartofsky L The nonthyroidal illness syndrome Endocrinol Metab Clin North Am, 2007.PMID 17673123
3. [3]Krinsley JS, Egi M, Kiss A, et al Diabetic status and the relation of the three domains of glycemic control to mortality in critically ill patients: an international multicenter cohort study Crit Care, 2013.PMID 23452622
4. [4]NICE-SUGAR Study Investigators; Finfer S, Chittock DR, Su SY, et al Intensive versus conventional glucose control in critically ill patients N Engl J Med, 2009.PMID 19318384
5. [5]Cooper MS, Stewart PM Corticosteroid insufficiency in acutely ill patients N Engl J Med, 2003.PMID 12594318
6. [6]van den Berghe G, Wouters P, Weekers F, et al Intensive insulin therapy in critically ill patients N Engl J Med, 2001.PMID 11794168