Today we have another Pulm PEEPs Pearls episode about a core critical care topic. Furf and Monty will be giving a high level overview of the use of steroids in sepsis including a review of the relevant literature and recent guidelines, and pragmatic bedside points.
Contributors
This episode was prepared with research by Pulm PEEPs Associate Editor George Doumat.
Dustin Latimer, another Pulm PEEPs Associate Editor, assisted with audio and video editing.
Key Learning Points
Why Steroids in Sepsis?
Steroids do not treat the infection — antimicrobials are always first and remain the cornerstone. The goal is addressing critical illness–related corticosteroid insufficiency (CIRCI), where cortisol production cannot keep up with the overwhelming inflammatory demand of septic shock.
Hydrocortisone helps in two main ways:
- Blunts the dysregulated inflammatory response — tempers the excessive vasodilation and febrile response that drive harm beyond the infection itself.
- Restores vascular sensitivity to catecholamines — sepsis downregulates adrenergic receptors; steroids turn that responsiveness back on.
Clinical takeaway: The first thing you notice is vasopressor weaning (or a bend in the escalation curve) — not a rapid improvement in fever or white count.
Caveat: These trials predate modern sepsis phenotyping. None distinguish hyperinflammatory vs. hypoinflammatory responders — they treat all comers.
The Evidence: Four Landmark Trials
Every IM resident and critical care fellow will eventually journal-club these four. The most consistent signal across all of them is faster shock reversal and reduced vasopressor use; the mortality question remains unsettled.
| Trial (Year) | N | Regimen | Key Finding |
| Annane (2002) | ~300 | Hydrocortisone + fludrocortisone | Mortality benefit in ACTH non-responders; criticized methodology and messy cortisol-response testing; not cleanly replicated. |
| CORTICUS (2008) | ~500 | Hydrocortisone alone | Faster shock reversal but no mortality benefit, regardless of cortisol responsiveness. Raised (later allayed) superinfection concern. Cornerstone for abandoning routine cort-stim testing. |
| ADRENAL (2018) | ~3,800 | Hydrocortisone alone | Faster vasopressor weaning; no 90-day mortality benefit. |
| APROCCHSS (2018) | ~1,200 | Hydrocortisone + fludrocortisone | Mortality benefit at 90 days. |
Bottom line: Faster shock reversal is consistent. Mortality benefit appears in 2 of 4 trials (both used fludrocortisone) but not the others. A 2026 meta-analysis showed benefit for hydrocortisone + fludrocortisone vs. placebo, but
not for hydrocortisone + fludrocortisone vs. hydrocortisone alone — suggesting hydrocortisone drives the main effect.
Who Gets Steroids, and When?
- 2021 Surviving Sepsis: Consider steroids for norepinephrine or epinephrine ≥ 0.25 mcg/kg/min for ≥ 4 hours despite adequate resuscitation — a reasonable bedside trigger.
- Early 2026 update: Moved away from a specific numeric trigger — consider steroids when a septic patient is not responding well to vasopressors or has escalating requirements. Make a clinical decision. (Quality of evidence: low to moderate.)
- Go faster than the threshold when: Known/suspected adrenal insufficiency or home steroids, or florid pressor-requiring shock on arrival.
A practical escalation sequence: escalating norepinephrine → add vasopressin (per VASST) → then add steroids if requirements keep climbing.
Do NOT wait for an ACTH stimulation test. It does not reliably predict who responds and only delays treatment. Sepsis is an elevated-cortisol state but can dissociate ACTH and cortisol, and cortisol-binding globulin is depleted — the test is too messy to guide care.
What to Give: The Regimen
- Standard dose: Hydrocortisone 200 mg/day, typically 50 mg IV Q6H. (Original trials often used continuous infusions, rarely used in the U.S.) Some start with a 100 mg bolus to gain control.
- Higher dose: If chronically on steroids / adrenally insufficient, consider ~300 mg/day (e.g., 100 mg Q8H).
- Fludrocortisone: Unsettled. The two mortality-benefit trials added it (50 mcg PO/NG/OG daily), but hydrocortisone already has mineralocorticoid activity and meta-analyses don’t show added benefit over hydrocortisone alone. Most clinicians omit it — adding it is reasonable and safe, just be honest about the uncertainty.
Duration & Tapering
- Typical course: ~7 days is most common. Trial practices varied (ADRENAL ~7 days; VANISH used a taper after 6 days; some continue until pressors are off).
- No taper needed. You do not need to taper for adrenal insufficiency after a short course — just stop. If pressors dramatically rebound, you can restart, but most patients have gained the benefit they’ll get by day 7.
Pitfalls & Safety
- Hyperglycemia: Expected and must be managed (monitor closely; insulin drip if needed). No signal for major DKA / severe complications in the trials.
- Superinfection / fungal infection: The most-quoted concern, but the overall literature does not show a convincing, statistically significant increase. Be disciplined about stopping on schedule.
- Muscle weakness: Steroids can worsen critical illness myopathy; a short 7-day course likely has limited effect, but be aware.
- Other: GI bleeding (follow general PPI prophylaxis guidance) and sodium disturbances (watch for hyper-/hyponatremia).
Two things we know: (1) steroids shorten duration of vasopressor support, and (2) they are relatively safe in sepsis. Whether they improve mortality — and in whom — remains open.
The Five Pulm PEEPs Pearls
- Mechanism: Steroids restore catecholamine vascular sensitivity and blunt dysregulated inflammation. The clinical target is vasopressor weaning, not infection treatment.
- Evidence: Faster shock reversal is the most consistent finding. Mortality benefit is seen in 2 of 4 trials but not the others — still controversial. Some patients likely benefit; we don’t yet know who.
- Trigger: A practical 2021 threshold is levo/epi ≥ 0.25 mcg/kg/min for ≥ 4 hours. Newer guidance drops the strict number — make a clinical decision based on poor pressor response or escalation.
- Dose: Hydrocortisone 200 mg/day (e.g., 50 mg Q6H). Adding fludrocortisone mirrors two trials, but meta-analyses find no benefit over hydrocortisone alone.
- Safety: Steroids appear safe in sepsis. Monitor and treat hyperglycemia; no marked increase in superinfection.
References and Further Reading
Annane, Djillali et al. “Effect of treatment with low doses of hydrocortisone and fludrocortisone on mortality in patients with septic shock.” JAMA vol. 288,7 (2002): 862-71. doi:10.1001/jama.288.7.862
Sprung, Charles L et al. “Hydrocortisone therapy for patients with septic shock.” The New England journal of medicine vol. 358,2 (2008): 111-24. doi:10.1056/NEJMoa071366
Venkatesh, Balasubramanian et al. “Adjunctive Glucocorticoid Therapy in Patients with Septic Shock.” The New England journal of medicine vol. 378,9 (2018): 797-808. doi:10.1056/NEJMoa1705835
Annane, Djillali et al. “Hydrocortisone plus Fludrocortisone for Adults with Septic Shock.” The New England journal of medicine vol. 378,9 (2018): 809-818. doi:10.1056/NEJMoa1705716
Sun, Alin et al. “Correction: Hydrocortisone combined with fludrocortisone for treatment of adults with septic shock: an updated meta-analysis and systematic review.” Frontiers in medicine vol. 13 1811616. 2 Mar. 2026, doi:10.3389/fmed.2026.1811616
Prescott, Hallie C et al. “Executive Summary: Surviving Sepsis Campaign: International Guidelines for Management of Sepsis and Septic Shock 2026.” Critical care medicine vol. 54,4 (2026): 715-724. doi:10.1097/CCM.0000000000007089
Podcast: Play in new window | Download
Subscribe: Apple Podcasts | Spotify | Amazon Music | Android | iHeartRadio | Podcast Index | RSS





