Steroids in severe community-acquired pneumonia: dangerous, worthless, or miracle cure? The roller coaster of clinical trials

Marwan Bouras , Jordi Rello , Antoine Roquilly

Main research Goal: 

To determine the efficacy of steroid administration in cases of severe pneumonia - not influenza pneumonia however 

Why is this important to emergency medicine practice:

Community acquired pneumonia (CAP) is the leading cause of infectious disease and the 2nd leading cause of disability adjusted life years in the world (2019). That being the case, we do see a large amount of patients with pneumonia. I myself have not worked with any attending who routinely gives steroids in these cases, and if it is clinically indicated to improve outcomes then we should be administering this medication. 

One paper showed 5mg of dexamethasone decreased hospital stay, and 50mg/day for 7 days reduced time to clinical stability with no increase in complications, but again failed to improve patient outcomes. 2021 meta-analysis showed there was not an association with decrease in mortality but did show a decrease in disease progression with less people having to be admitted to the ICU or be mechanically ventilated. 

Escape Trial - 586 randomized critically ill patients in 42 american hospitals received 40 mg/day of methylprednisolone or placebo for 20 days with a tapering plan showed no significant difference in 60 day mortality ( 16 vs 18%) or other secondary outcomes 

Cape Cod trial - 31 french hospitals , double blind placebo controlled trial evaluation 200mg/day infusion of hydrocortisone for 8 or 14 days planned taper in patients admitted in ICU with severe non viral CAP, this trial was stopped early due to covid but the authors reported a significant decrease in mortality of the hydrocortisone group of 6.2% and 11.9% of the placebo group dead at 28 days with a p of .005 

Study Design:

Meta analysis mainly of the Cape cod and escape trial. Cape cod was a placebo controlled double blind study with 31 french hospitals and escape trial was a randomized control trial in 42 american hospitals. 

Methods:

• Population

  • Escape Trial - 586 randomized critically ill patients in 42 american hospitals

  • Cape Cod trial - 31 french hospitals , double blind placebo controlled trial 

• Intervention or exposure 

  • Steroids vs placebo 

• Outcome measure 

  • Mortality, length of hospital stay 

Results:

• Key Findings 

  • Cape cod trial hydrocortisone was most effective in inflammatory patients with CRP over 150mg 

  • Both studies showed a decrease in mortality and should be administered early in the disease course and is safe, but should not be used in pneumonia due to influenza 

  • These results should not be extrapolated to hospital acquired pneumonia 

• Were the results statistically significant 

  • Not in the escape trial 

Discussion and conclusion 

• Did the authors discuss strengths and limitation of the study, was there any limitations not addressed 

  • Inclusion criteria of the 2 trials were different. The Cape cod trial did not use viral CAP because steroids were associated with unfavorable outcomes in severe influenza pneumonia , steroids do show a increase in ventilator free days in covid 

  • Timing of steroid administration. In the ESCAPe trial allowed for randomization up tp 72-96 hours after hospital admission while in the CAPE COD trial patients in the hydrocortisone group received hydrocortisone administered continuously within 24 hours after onset of severity criterion 

    • Authors point out in the first few hours of steroids administration can reduce inflammatory cytokine production, neutrophils, and T Cell recruitment and adhesion, while in later days it can have the opposite effects on the inflammatory response by reducing anti inflammatory cytokine secretion or increasing phagocytic capacity 

  • 96% of patients in the usa trial were male compared with 69% in the french study  with a sub group analysis showing possibly better benefits among females 

  • ESCAPe trial did not reach significance likely due to lack of power 

• Were the conclusions drawn by the author supported by the data 

  • Basically, but it is a little worrisome the escape trial did not reach statistical power 

• How do the findings compare to previous research in the same area 

  • 2017 Cochrane systematic review showed a relative risk of death was .66 in CAP patients treated with steroids 

Impact on clinical practice 

• If you have a patient with severe pneumonia, which to me means increased oxygen requirement or shock you can give steroids. I’d likely get a CRP level to correlate current inflammatory response. 

Summary by Dr. Kenneth Appel PGY-2 , 2025