The mechanism of action for dexamethasone would be to counter cytokine storms (an autoimmune-like process). Cytokine storms (vs. the virus itself) are likely the biggest contributors to respiratory failure and other effects such as blood clots. It is routinely used for meningitis patients as they begin to recover. So, while dexamethasone does not have antiviral efficacy, it may in fact save many lives for the sickest patients.

The NEJM authors did a good job at matching the samples. What was interesting is that the C-Reactive protein (CRP), steroid level, and IL-6 inhibitor level of the HCL sample was significantly higher than the control sample. Also, the 02 saturation levels were significantly lower in the HCL sample than the control. The white counts of both samples were comparable. This implies the poor outcomes of the HCL sample were skewed toward Cytokene Storms versus the control. For a Cytokene Storm (an autoimmune phenomenon), steroids and IL-6 inhibitors are the preferred treatment options - hence the elevated levels in the HCL cohort. CRP is a non-specific indicator of inflammation. It is elevated in both autoimmune diseases and infections. HCL probably has some antiviral efficacy. It is proven to have autoimmune efficacy, but it can't cut through a CRP of 140 (normal range is single digits). Having has a life-threatening autoimmune lung inflammation years ago, heavy steroid use saved me. I think doctors have been hesitant to blast these very sick patients with steroids (e.g., solu-medrol) because steroids can suppress the immune system, but that is changing. I would also like to see a controlled HCL study with less sick patients to see its efficacy, or example, as a COVID prophylactic.