The Sepsis Toolbox

We are about to learn whether giving corticosteroids in sepsis is a good idea or bad. I am on record predicting a negative result.

[Update – The ADRENAL trial, now published in the New England Journal of Medicine showed that corticosteroids did not improve mortality. It did shorten ICU stay by 1 day, and increased blood pressure. In other words, steroid treatment made patients with sepsis appear to be improving, when in reality, they were not. There is much to say about these results. See Lessons from the Adrenal Trial.]

I hope this study prompts a discussion about the bigger picture in sepsis. What exactly is the rationale for intervening?

We use the infected, not the infecting agent, to define and detect sepsis.  Instead of measuring the causative pathogen, we use proxy measures. Fever, low blood pressure, increased respiratory rate, and somnolence alert us that there is a problem. The conventional thinking is that the human body’s response is the problem in sepsis. But is that true?

Some responses to overwhelming infection are no doubt needed. But which ones? Good evidence suggests that fever, mobilization of white blood cells, and production of acute phase response proteins are functional, evolved responses (e.g. adaptations.) This evolutionary insight leaves treating physicians in a bind.  Certainly we need to intervene to some degree. But when?

Can we do better than nature?

Logically, it is unlikely that every host response is optimal in sepsis.

Significant inter-individual variation exists in all traits. In light of variation, an individual may not show the optimal expression in any trait, even adaptive ones like fevers. Measurements in an individual patient, e.g. lactate, TNF-alpha, might be inadequate or excessive.  By extrapolation, this is likely true for any immune regulator and immune effector in sepsis. However, this does not mean that none of them are optimal.

“Sepsis traits”, what we observe in sepsis, are not random. Because of natural selection, many of these traits are more likely to be helpful than harmful. Our ancestors faced infectious challenges since before the first multicellular organism evolved. Organisms with effective host defenses against overwhelming infection were more likely to survive and reproduce, leaving extant organisms with a genetic toolbox of defenses.

Because pathogens evolve too, the toolbox is never perfect. This means that we will never see an organism perfectly adapted to resist lethal infection. Other reasons for non-optimal sepsis traits include aging, prior injury, environmental toxins, energetic or biological constraints, environmental change for which we are not evolved (like the intensive care unit) and immune trade-offs that protect us from one pathogen while leaving an vulnerable to another.

Despite the evolution of host defenses, immune regulators and effectors are probably not optimal. However, several features may still yield fitness-enhancing outcomes. The reasons: immune system redundancy, the ability to correct and compensate, and massively parallel regulation. In sum, these may produce a functional result that is much closer to optimality than for any individual trait.

Back to steroids in sepsis. In order for a immunoregulator (like corticosteroids) to work in our patients with sepsis, we would need to make a variety of assumptions.

1) The sepsis trait must be non-optimal both for an individual patient and for the population.

2) We must know the direction in which the sepsis trait needs to be adjusted.

3) Our therapy must accomplish this better that any endogenous host regulatory mechanism.

3) Changes in other physiologic systems have not adequately compensated for the sepsis trait.

4) No unintended consequences. By “correcting” the sepsis trait, we do not make other traits and compensations suboptimal.

5) The intervention will have a predictable and beneficial effect despite individual trait variation.

6) Pathogens will not benefit from the trait change caused by our therapy.

7) Pathogens will not evolve to compensate for the trait change.

Satisfying these assumptions is a heavy lift.  That’s why no magic-bullet immune drug has worked thus far in sepsis.

Is all hope lost? Maybe not. Applying evolutionary reasoning may lead us to a possible therapy. Evolution can inform when we can expect non-optimality for any trait, including for septic shock:

1) Constraint, e.g. protein manufacture capacity, energy availability.

2) Trade-offs, e.g. efforts against one pathogen gives advantage to another

3) Manipulation by infecting organisms that subvert the immune system and block our defenses.

4) Senescence, selection shadow, and antagonistic pleiotropy.

5) Gene-environment mismatch, e.g. we have not evolved to be in hospitals or to undergo intensive care unit level of care.

The ADRENAL trial is a well designed study that will give us exciting results, no matter the outcome. We will have an opportunity to discuss the complete results after they are presented on January 19th.

We will discuss sepsis and evolution this summer at the 4th annual meeting of ISEMPH – The International Society for Evolution, Medicine and Public Health in beautiful Park City, Utah. Abstract submissions are now open.

Categories: Uncategorized

Joe Alcock

Emergency Physician, Educator, Researcher, interested in the microbiome, evolution, and medicine