Fellow's quick read · Critical Care / Infectious Disease
Corticosteroids in Sepsis, ARDS, and CAP
SCCM 2024 Focused Update · reviewed 2026-08-24
Personal study digest for a new ID fellow. Recommendations are from the SCCM 2024 Focused Update (DOI: 10.1097/CCM.0000000000006172); the "What's changed since 2024" section is reviewer synthesis of newer evidence, each claim cited. Not a substitute for the full guideline.
Sources digested: SCCM 2024 Focused Update on Corticosteroids in Sepsis, ARDS, and Community-Acquired Pneumonia (Chaudhuri et al., Crit Care Med 2024).1
In one line
Low-dose corticosteroids reduce short-term mortality in septic shock (RR 0.93), ARDS (RR 0.82), and severe CAP (RR 0.62) — the sicker the patient, the larger the benefit. High-dose/short-duration regimens are harmful in septic shock and explicitly recommended against.
What does the guideline say?
| Disease state | Recommendation | Strength / certainty |
|---|---|---|
| Septic shock | Suggest administering corticosteroids | Conditional, low certainty |
| Septic shock — high dose | Recommend against high-dose/short-duration (>400 mg/d HC equiv for <3 d) | Strong, moderate certainty |
| ARDS (adult) | Suggest administering corticosteroids | Conditional, moderate certainty |
| Severe CAP | Recommend administering corticosteroids | Strong, moderate certainty |
| Less-severe CAP | No recommendation | — |
No pediatric recommendations were made for any disease state due to limited data.
Key effect sizes
- Septic shock, short-term mortality: RR 0.93 (95% CI 0.88–0.98; moderate certainty; 46 RCTs)
- ARDS, 28-day mortality: RR 0.82 (95% CI 0.72–0.95; moderate certainty; 18 RCTs)
- Severe CAP, hospital mortality: RR 0.62 (95% CI 0.45–0.85; moderate certainty; 10 RCTs)
- Less-severe CAP, mortality: RR 1.08 (95% CI 0.83–1.42; low certainty) — no benefit
- Shock reversal (sepsis): RR 1.24 (95% CI 1.11–1.38; high certainty)
How do I dose steroids for each syndrome?
| Syndrome | Regimen | Duration |
|---|---|---|
| Septic shock | Hydrocortisone 200 mg/d IV (continuous infusion or 50 mg q6h) ± fludrocortisone 50 µg PO daily | 7 d or until ICU discharge |
| Early ARDS (≤24 h) | Dexamethasone 20 mg IV daily × 5 d, then 10 mg IV daily × 5 d (until extubation) | 10 d |
| Early ARDS (≤72 h) | Methylprednisolone 1 mg/kg IV bolus, then 1 mg/kg/d × 14 d → taper over 14 d (if extubated d 1–15, advance to d 15 regimen) | 28 d |
| Unresolving ARDS (7–21 d) | Methylprednisolone 2 mg/kg IV bolus, then 2 mg/kg/d divided q6h × 14 d → prolonged taper over 18 d | 32 d |
| Severe CAP (CAPE COD regimen) | Hydrocortisone 200 mg IV × 1, then 10 mg/hr IV infusion | 7 d (median actual 5 d, guided by clinical improvement) |
| Severe CAP (methylprednisolone) | Methylprednisolone 40 mg IV bolus, then 40 mg/d × 7 d → taper to 4 mg/d over 20 d total (continuous infusion in ICU, then divided bid IV/PO) | 20 d |
Key decisions a fellow owns
- Start steroids early in septic shock. The 2024 update removes the 2017 qualifier that steroids are only for shock "not responsive to moderate- to high-dose vasopressors" — benefit applies to any patient with septic shock requiring vasopressors.
- Add fludrocortisone? An IPD meta-analysis (Pirracchio, NEJM Evidence 20232) suggests HC + fludrocortisone (RR 0.86) outperforms HC alone (RR 0.96). The guideline notes this but stops short of a formal recommendation — know the data and discuss with your attending.
- Never high-dose/short-course for septic shock. >400 mg/d HC equivalent for <3 d is strongly recommended against (moderate certainty).
- Distinguish severe from non-severe CAP. The mortality benefit is confined to severe CAP (PSI class IV–V, CURB-65 ≥3, or ICU admission). No recommendation exists for less-severe CAP.
- Pick the molecule by syndrome. Dexamethasone or methylprednisolone for ARDS; hydrocortisone for septic shock and severe CAP. No subgroup analysis showed a credible difference by molecule within a syndrome, but the RCT evidence tracks these pairings.
- Monitor for harm. Hyperglycemia (RR 1.11–1.13 across syndromes, moderate certainty) and possible neuromuscular weakness (RR 1.21 in sepsis, low certainty) are the main adverse effects. Screen glucose and manage pharmacologically.
Special populations
- Sepsis without shock: No recommendation for corticosteroids. However, if the patient has sepsis with severe CAP or sepsis with ARDS, treat per those recommendations.
- Pediatrics: No recommendation for any syndrome — limited data.
- Cirrhosis + septic shock: A 2025 RCT (Meersseman et al., Liver Int 20257) found no mortality benefit and was stopped early for futility — use with caution in this subgroup.
- COVID-19 ARDS: Six of 18 ARDS RCTs enrolled COVID patients; no credible subgroup interaction by COVID status. Dexamethasone 6 mg/d (RECOVERY) is the standard in COVID ARDS.
Duration and stopping
- Septic shock: 7 days or until ICU discharge, whichever comes first. Taper optional but shorter courses (<72 h total) appear to retain benefit and may be safer.6
- ARDS: Courses >7 days had better survival than ≤7 days (subgroup interaction p = 0.04, moderate credibility). Plan for 7–28 days depending on regimen, with a taper.
- Severe CAP: 5–7 days typical; one protocol extends to 20 days with taper. CAPE COD used clinical improvement criteria to stop at a median of 5 days.
What's changed since 2024
Reviewer synthesis — not the guideline. Each item cited.
- SSC 2026 reaffirms corticosteroids for septic shock. The Surviving Sepsis Campaign's March 2026 comprehensive update (129 statements) maintains the recommendation, citing the same evidence base plus subsequent analyses.3
- Fludrocortisone + HC likely superior to HC alone. A network meta-analysis (Teja et al., AJRCCM 2024) found combined fludrocortisone + hydrocortisone reduced mortality (RR 0.85 vs placebo, 94.2% probability of being superior to HC alone), consistent with the Pirracchio IPD meta-analysis the guideline cites.4
- SONIA trial extends CAP steroids beyond "severe." Lucinde et al. (NEJM 2025; PMID 41159889) randomized 2,180 Kenyan adults with any-severity CAP to oral prednisolone or dexamethasone vs placebo. 30-day mortality HR 0.84 (0.73–0.97). This is the first large RCT showing benefit in ALL CAP, not just severe — but in a sub-Saharan African setting with high HIV prevalence, limiting direct generalizability to US practice.5
- Rapid taper may be better than prolonged courses (sepsis). Denninger et al. (Ann Pharmacother 2025; PMID 41355405) found that tapering steroids in <72 hours was associated with better outcomes than longer tapers, challenging the assumption that slow tapers are universally needed.6
- Cirrhosis + septic shock: no benefit. Meersseman et al. (Liver Int 2025; PMID 40757786) randomized cirrhotics with septic shock to HC and found no mortality benefit; the trial was stopped early for futility. Steroids should not be assumed beneficial in this subgroup.7
- Non-viral ARDS evidence still thin. A 2025 narrative review (Nguyen et al., Pneumonia 2025; PMID 41046258) cautions that recommendations for non-viral ARDS outpace the evidence: most modern ARDS RCTs enrolled COVID patients, and trials in non-viral ARDS are older and smaller.8
- ESCAPe vs CAPE COD: timing and molecule matter for CAP. ESCAPe (methylprednisolone within 36 h of admission, CRP >150) and CAPE COD (hydrocortisone for ICU-admitted severe CAP) both show benefit, but their populations, molecules, and eligibility criteria differ enough that they are not interchangeable. Know which protocol your institution follows.
- Dose-stratified confirmation. A 2025 meta-analysis (BMC Anesthesiol 2025) confirms that low-dose, prolonged corticosteroid regimens drive the mortality benefit in septic shock, while high-dose/short-duration regimens do not — reinforcing the guideline's strong recommendation against high-dose boluses.9
- Still current: The core recommendation framework — conditional for septic shock and ARDS, strong for severe CAP, no recommendation for less-severe CAP — remains unchanged since publication. No newer society guideline has contradicted any of the four recommendations.
Anki cards minted this run
- Septic shock steroid dosing — HC 200 mg/d + the high-dose prohibition
- Severe CAP steroid mortality benefit — RR 0.62, strong recommendation, and the severity gate
- ARDS steroid duration — courses >7 d better than ≤7 d
- Fludrocortisone + HC vs HC alone — the NMA and IPD data on fludrocortisone addition
Sources
- 1. Chaudhuri D, Nei AM, Rochwerg B, et al. 2024 Focused Update: Guidelines on Use of Corticosteroids in Sepsis, Acute Respiratory Distress Syndrome, and Community-Acquired Pneumonia. Crit Care Med. 2024;52(5):e219-e233. doi:10.1097/CCM.0000000000006172
- 2. Pirracchio R, Annane D, Waschka AK, et al. Patient-Level Meta-Analysis of Low-Dose Hydrocortisone in Adults with Septic Shock. NEJM Evid. 2023;2(6). doi:10.1056/EVIDoa2300034
- 3. Surviving Sepsis Campaign 2026 Comprehensive Guidelines Update. Crit Care Med / Intensive Care Med. March 2026.
- 4. Teja B, Guo Y, Engelbrecht C, et al. Fludrocortisone Plus Hydrocortisone vs Hydrocortisone Alone in Septic Shock: A Network Meta-Analysis. Am J Respir Crit Care Med. 2024;209(10):1258-1266. PMID:38271488
- 5. Lucinde RK, et al. Oral Glucocorticoids for Community-Acquired Pneumonia in Kenya (SONIA). N Engl J Med. 2025. PMID:41159889
- 6. Denninger MH, et al. Corticosteroid Tapering Duration in Septic Shock. Ann Pharmacother. 2025. PMID:41355405
- 7. Meersseman P, et al. Hydrocortisone in Cirrhosis with Septic Shock: A Randomized Controlled Trial. Liver Int. 2025. PMID:40757786
- 8. Nguyen CT, et al. Corticosteroids in Non-viral ARDS: A Narrative Review. Pneumonia. 2025. PMID:41046258
- 9. Dose-Stratified Meta-Analysis of Corticosteroids in Septic Shock. BMC Anesthesiol. 2025.