The Critical Care Reviews meeting in Belfast this month previewed the REMAP-CAP study findings on the use of corticosteroids and community acquired pneumonia. Unexpectedly, this adaptive trial showed a signal of harm for steroids (hydrocortisone 50mg for 7 days) in hospitalized severe pneumonia. Community acquired pneumonia is exactly what it sounds like: lung infection in patients who come in from the community. We commonly diagnose CAP in my emergency department, and if they are sick, septic, hypoxic, or unstable, they are admitted to the hospital. (Hospital-acquired pneumonia is a different and worse thing, usually, for evolutionary reasons that we described here.) Most admitted community acquired pneumonia in adults is caused by a bacterial infection and most are prescribed antibiotics, which generally show benefit. The perennial medical debate surrounds the decision to prescribe corticosteroids. For hypoxic COVID pneumonia, the Recovery trial provided evidence that the corticosteroid dexamethasone could be life saving. The Cape Cod trial, recently published in the New England Journal of Medicine, showed a survival benefit to severe pneumonia patients with CAP who received steroids. As far as I can tell, the REMAP-CAP steroid results are yet unpublished, but I, for one, am not surprised that steroids in that trial were ineffective. Here is why:
The question of when and whether to give steroids has evolutionary implications. We should be cautious in prescribing powerful drugs that hinder our immune defenses, because that is the strategy of our enemies. Blocking immune defenses is what successful pathogens do. We see evidence of this in real time with each new COVID variant that makes the news. On the other hand, hosts evolve ways of making immune defenses difficult to bypass and evade. This ongoing co-evolution is ancient – we evolved alongside many pathogens such as Streptococcus pneumoniae that cause CAP. The fact that every one of your ancestors managed to survive lung infections in childhood and young adulthood to produce offspring highlights the general effectiveness of your immune system. Therefore, when physicians restrain the immune system in our patients we’d better have a damn good reason. Moreover, we’d better have an evolutionary rationale. That rationale is most often missing from the debate. Alix Masters and I did a deep dive into exactly this.topic in Cytokine Storms, Evolution and COVID 19, published in Evolution Medicine and Public Health. The bottom line: overreaction by the immune system is expected to be rare, especially when the immune system is dealing with a co-evolved pathogen. If that prediction is true, then we would not expect the routine use of steroids for any infection to be helpful. Despite the results of RECOVERY, this expectation is borne out even in Covid 19. From a meta-analysis in NEJM evidence last year: In conclusion, our meta-analysis shows that administering glucocorticoids in hospitalized patients with Covid-19 not receiving oxygen is likely associated with worse clinical outcomes. Because the vast majority of Covid-19 cases are in the less severe category, the notion that corticosteroids would be harmful overall in the set of patients with symptomatic Covid-19 is probably true.
With all of the above, why are my colleagues still prescribing steroids in massive numbers for respiratory infections in patients well enough to go home? The numbers are concerning. A study in 2019 suggested that 6.8% of patients presenting with respiratory infection received corticosteroids. A 2020 review reported a range from 11-70%. One in five adults received a short term steroid prescription over three years in a sample of US adults with health insurance over three years. In that study, a steroid prescription was associated with a 5 fold increase in the risk of sepsis, a nearly two-fold increase in fractures, and a three fold increase in venous thromboembolism.
To recap (sorry!), I am not surprised a study like REMAP-CAP showed a signal of harm from steroids. It would in fact be more surprising if steroid trials in pneumonia uniformly showed a survival benefit. The clinical rationale for steroids is that the host immune response in pneumonia is maladaptive. Possible? Yes, but evolution teaches us that maladaptive responses should be the exception, not the norm. Steroids should be beneficial only in a few narrowly defined subsets of patients for whom the immune system does more harm than good. If true, REMAP-CAP results provide more clarity on the question of steroids for pneumonia, not more confusion.
Categories: Uncategorized
Joe Alcock
Emergency Physician, Educator, Researcher, interested in the microbiome, evolution, and medicine
EvolutionMedicine 
Agree with all the above. Invariably, the docs & APP’s rx steroids for cough + sniffles in outpatient setting where I’m at (FL). Ignoring the fact that those z-packs are being provided for a likely limited, viral illness—I often sadly wonder about the population-level harm steroids alone must be causing. It’s unfortunate bc it’s such the norm in my community, patients expect it. I imagine this, coupled with reimbursement tied to pt satisfaction, makes it even more difficult for clinicians to change their ways. Their “bad habits” are reinforced by patient expectations and the system. I’m often on the receiving end of it.
Signed,
Your friendly community ER Doc
I think it is pretty universal at this point. Like you I have to wonder about the effect on population health. This is a factor in bad patient care and moral injury for us healthcare workers, when a confluence of forces keep us from doing what we think is right for our patients.