Adrenal Insufficiency, GC-Induced — board-level quick read

Personal study digest for IM boards preparation. Recommendations are from the cited guideline; the “What’s changed” section is reviewer synthesis of newer evidence, each claim cited. Not a substitute for the full guideline.

Source digested: Bornstein SR et al. Diagnosis and Management of Glucocorticoid-Induced Adrenal Insufficiency. ESE/Endocrine Society 2024. PMID 38724043 · DOI

In one line

Any patient on supraphysiologic glucocorticoids ≥3–4 weeks—by any route—is at risk. Taper slowly, check an 8–9 AM cortisol to assess HPA recovery, stress-dose during illness or surgery, and never delay hydrocortisone 100 mg IV for suspected adrenal crisis.

When to suspect / diagnose

At-risk population: ≥1% of the general population is on chronic GC. Biochemical AI develops in ~49% on oral GC and ~52% after intra-articular injection, though only ~10% become clinically symptomatic.
Risk factors (Table 3): higher GC potency, systemic route, higher dose, longer duration (>3–4 wk supraphysiologic), obesity (higher BMI), and older age.
Non-oral routes matter: high-dose inhaled (21.5%), intra-articular (52%), topical (4.7%), intranasal (4.2%) can all suppress the HPA axis.
CYP3A4 trap: ritonavir + inhaled fluticasone → fluticasone AUC ↑350× → iatrogenic Cushing syndrome + secondary AI. Same risk with budesonide, triamcinolone.
Cushing features (R 2.10): if present, assume fully suppressed HPA axis—do not test, just plan a taper with stress coverage.
GC withdrawal syndrome ≠ AI: fatigue, myalgia, arthralgia during taper can occur with a normal HPA axis. Distinguish by cortisol testing.

Workup

Morning serum cortisol (R 2.7): draw 8–9 AM, hold GC for 24 h before the test (hold long-acting agents longer).
- >300 nmol/L (10 µg/dL) → HPA recovered. Stop replacement; no stress coverage needed.
- 150–300 nmol/L (5–10 µg/dL) → continue physiologic replacement + stress coverage; recheck in weeks to months.
- <150 nmol/L (5 µg/dL) → continue replacement + stress coverage; recheck in months.
No routine cosyntropin stimulation test (R 2.8): morning cortisol is sufficient for screening. Consider dynamic testing only if the AM cortisol is persistently indeterminate.
No fludrocortisone (R 2.12): mineralocorticoid axis is preserved in GC-induced AI (unlike primary AI).

Taper protocol

When the underlying disease permits, taper toward physiologic replacement (Table 4). Strong recommendation.

Current prednisone dose	Decrease by	Interval
>40 mg/day	5–10 mg	per week
20–40 mg/day	5 mg	per week
10–20 mg/day	2.5 mg	every 1–4 weeks
5–10 mg/day	1 mg	every 1–4 weeks
<5 mg/day	1 mg every 4 wk, or switch to HC 20 mg and taper	every 4 weeks

Switch long-acting → short-acting (R 2.5): convert dexamethasone or betamethasone to hydrocortisone or prednisone once the disease no longer requires a long-acting agent.
Physiologic equivalents: HC 15–25 mg = prednisone 4–6 mg = methylprednisolone 3–5 mg = dexamethasone 0.25–0.5 mg per day.

Stress dosing

Any patient still on supraphysiologic GC or with an unrecovered HPA axis needs stress coverage (Table 8). Strong recommendation.

Stress level	Regimen	Duration
Minor illness / minor surgery	HC 40 mg/day (e.g. 20 + 10 + 10 mg), or prednisone 10 mg/day, or dexamethasone 1 mg/day	Duration of illness; return to baseline when resolved
Major surgery / critical illness	HC 100 mg IV bolus at induction, then 200 mg/24 h continuous infusion (or 50 mg IV q6h)	Taper by 50% per day back to baseline over 2–3 days

Adrenal crisis — don’t delay for labs

Clinical diagnosis (R 3.2): hypotension + nausea/vomiting/fatigue/fever/altered mental status in a patient on or recently withdrawn from GC.

Treat immediately: hydrocortisone 100 mg IV/IM + IV normal saline. Do not wait for cortisol results. HC at this dose provides adequate mineralocorticoid effect—no fludrocortisone needed acutely.

Key decisions

When to check AM cortisol: once the patient reaches a physiologic replacement dose (pred ≤5 mg or HC ≤20 mg), check 8–9 AM cortisol after holding the GC for 24 h.
When to stop replacement: AM cortisol >300 nmol/L (10 µg/dL) on two occasions → recovered. Discontinue and educate about adrenal crisis symptoms for 12 months.
When to refer to Endocrine (R 2.11): no HPA recovery after 1 year of monitoring, or any episode of adrenal crisis.
Patient education: every at-risk patient should carry a steroid emergency card and know to double their GC dose for febrile illness and seek emergency care if unable to take oral medication.

What’s changed since 2024

Reviewer synthesis — not the guideline. Each item cited.

Still current — tapering reduces AI. A 2026 systematic review/meta-analysis (12 studies, 1170 participants) found pooled AI prevalence of 30% after systemic GC. Tapering significantly reduced observed prevalence (p < 0.01), and testing ≥15 days after GC cessation showed lower rates (15%) than testing at cessation (48%). Validates the guideline’s taper-first approach. Bowley et al., Eur J Intern Med 2026; DOI
Still current — routine cosyntropin testing not justified post-taper. A prospective PMR cohort (n=60) showed GIAI peaked at 25% at month 6 during GC use but recovered in 92% by 12 months. Routine short synacthen testing after planned GC cessation was not justified by the data. Gaster et al., Rheumatology 2026; DOI
Still current — risk quantified by route. A TriNetX network analysis (n=243,430 systemic; 315,237 inhaled) confirmed systemic GC carries HR 6.32 for AI diagnosis and HR 3.52 for AI hospitalization vs NSAIDs. Inhaled GC: HR 1.55 for diagnosis but no significant increase in hospitalization (HR 1.26, p=0.17). Vaduva et al., Eur J Endocrinol 2026; DOI
Practice-evolving — tapering below physiologic dose may promote recovery. A 2026 cohort (n=65 adults) found each 1 mg prednisone reduction was associated with a +48.8 nmol/L rise in morning cortisol, even when starting cortisol was <150 nmol/L. This challenges the guideline’s conservative “continue, recheck in months” approach at low cortisol and suggests ongoing dose reduction itself drives HPA recovery. Mehta et al., Endocrine Connections 2026; 15(1)
Emerging — waking salivary cortisone. A non-invasive home-based salivary cortisone assay achieved AUC 0.95 for AI diagnosis in validation cohorts, and real-world evidence supports its use for monitoring HPA recovery. The 2024 guideline acknowledges it as “promising”; not yet widely available. Debono et al., NEJM Evidence 2023; PMID 38320042
Newly available — Efmody (modified-release hydrocortisone). EU indication expanded April 2026 from congenital adrenal hyperplasia to all adrenal insufficiency. The CHAMPAIN trial showed it restores physiological early-morning cortisol (median 417 vs 6 nmol/L with Plenadren) and improves quality of life. Not yet available in the US.
Open — TOASST trial (NCT03153527): RCT comparing abrupt vs gradual GC discontinuation at pred 7.5 mg. Protocol published 2023 (PMID 37018188). Results not yet published.
Open — RESCUE & REPLACE trials. Two ongoing Danish RCTs in PMR/GCA testing supplemental HC during stress (RESCUE, NCT05435781) and HC replacement for GC withdrawal syndrome (REPLACE, NCT05193396). Results expected 2026–2027. Borresen et al., BMJ Open 2026; DOI. Dreyer et al., BMJ Open 2026; DOI

Anki cards minted this run

AM cortisol thresholds for HPA recovery — core boards knowledge; the 10/5 µg/dL cut-points drive the stop/continue/refer decision.
Stress dosing: minor vs major — actionable dosing for the medicine ward; HC 40 mg/day vs 100 mg IV bolus + 200 mg/24 h.
Adrenal crisis: treat first, test later — must-not-miss; HC 100 mg IV/IM without waiting for labs.
Mehta 2026: tapering below physiologic dose promotes HPA recovery — practice-changing currency finding; each 1 mg pred reduction → +48.8 nmol/L morning cortisol.

Sources

[1] Bornstein SR et al. Diagnosis and Management of Glucocorticoid-Induced Adrenal Insufficiency. J Clin Endocrinol Metab 2024. PMID 38724043. DOI
[2] Bowley J et al. Adrenal insufficiency following systemic glucocorticoid therapy: a systematic review and meta-analysis. Eur J Intern Med 2026;146:106707. DOI
[3] Gaster T et al. Repeated short corticotropin testing in patients with polymyalgia rheumatica. Rheumatology 2026;65(2). DOI
[4] Vaduva P et al. Risk of hospitalization for adrenal insufficiency following long-term systemic and inhaled corticosteroid use. Eur J Endocrinol 2026;194(3):422–430. DOI
[5] Mehta A et al. Physiological dose tapering promotes HPA recovery in glucocorticoid-induced adrenal insufficiency. Endocrine Connections 2026;15(1).
[6] Debono M et al. Waking salivary cortisone as a diagnostic test for adrenal insufficiency. NEJM Evidence 2023. PMID 38320042.
[7] Borresen SW et al. RESCUE study protocol. BMJ Open 2026;16(6):e113110. DOI
[8] Dreyer AF et al. REPLACE study protocol. BMJ Open 2026;16(2):e111334. DOI