Sepsis 2.5

By Erica Remer, MD, FACEP, CCDS, ACPA-C

In the old days, you walked into a patient’s room, visually surveyed them, and exclaimed, “(Spicy expletive deleted!) This patient is septic!” and then you rushed out to start aggressive treatment and transfer to the intensive care unit. The reason the pundits tried to specify what constituted sepsis in the 1990s was to avoid missing cases and incurring deaths.

The problem that we have with sepsis right now is that all the attempts to codify and operationalize it left out the most important characteristic. The patient is sick. I don’t mean sick like with a cold or a stomachache; I mean SICK, in all-caps.

I believe sepsis is part of the progression of non-self-limited untreated infections. It doesn’t happen with an adenoviral upper respiratory infection. It also isn’t present in every elderly patient with cystitis.

Sepsis is the penultimate stop prior to dying from an infection. When I used to present at mortality conferences, if I saw a patient who died from an infection, but sepsis was never recognized or explicitly documented, I referred the case to quality for education of the clinical personnel.

But how to recognize it? The general criteria of the systemic inflammatory response syndrome (SIRS) of fever/hypothermia, tachycardia, tachypnea, and abnormal white blood cell count was a start. These were a cue that something was amiss, but they were too nonspecific. Zillions of conditions could result in abnormal vital signs or an abnormal white blood cell count.

“Life-threatening organ dysfunction caused by dysregulated host response to infection” was closer. In writing this, I realized the problem is in the construction of this phrase. The organ dysfunction doesn’t really have to be immediately life-threatening. The condition of sepsis is what is life-threatening; we recognize it is present by the organ dysfunction it has caused.

Howard Rodenberg et al. published an article called, “Sepsis-2.5: Resolving Conflicts Between Payers and Providers,” on behalf of the Society of Critical Care Medicine. Putting aside the fact that I wish they had invited me to be a co-author, because I have been preaching what they wrote for a long time now, I really like their no-nonsense conceptual definition of sepsis:

Sepsis is present when a patient with infection exhibits evidence of organ dysfunction at a site external to the seat of infection, or more than what is routinely expected from a localized infection.

Septic shock is present when a patient with sepsis exhibits persistent hypotension following initial fluid resuscitation.

I used to say that sepsis is when a patient is sicker than the average patient with that underlying infection, heralded by organ dysfunction. A colleague disagreed with the sicker-than-average characterization – he pointed out that if a patient is in the ICU, an intensivist might not have the same definition of “sicker than average,” because the intensivist’s patients are all sick. I mean sicker than all others with that same underlying infection of pneumonia or cellulitis or UTI, not only in the cohort of patients for whom you are caring personally.

They propose that sepsis is identified by “an ill-appearing patient, documented or suspected infection, and evidence of organ dysfunction.” I like it!

I am not sure where some of the specific guidelines stem from (e.g., hyperlactatemia being > 3.0 mmol/L); had I been a co-author, I would not have set strict criteria. Hypoxia may be sufficient; does it have to meet the threshold of acute hypoxic respiratory failure? I would not endorse a specific change from the baseline. If someone normally has platelets of 120,000 and they have thrombocytopenia of 70,000 in the proper setting, I don’t think it is reasonable to reject that as clinically significant thrombocytopenia because the decrease from baseline is less than 50 percent.

I don’t want providers to have to tick off checkboxes and grab their calculator to see if a patient qualifies. I know this was crafted with payor input, but not all payers play by the same rules. I always think that a competent provider using sound clinical judgment should be able to make a diagnosis without being pigeonholed into a specific level or change from a baseline.

Their conceptual definition does explicitly resolve the issue of organ dysfunction involving the infected system. It says “or more than what is routinely expected from a localized infection.” Acute hypoxic respiratory failure is not routinely expected in pneumonia. Acute kidney injury is not typical for urinary tract infection.

The other thing that doesn’t sit well with me is the name. I don’t like “Sepsis-2” or “Sepsis-3” or “Sepsis-2.5.” That implies that the condition of sepsis has been changing and evolving. It has not! Sepsis has always been and will continue to be a real medical condition. The only thing changing is the words we use to describe and characterize the condition.

How about “life-threatening progression of infection identified by causing organ dysfunction?” I propose we don’t give clinicians more hoops to jump through making sure the patient meets specific thresholds and rigid criteria, but let’s allow providers to make the diagnosis and treat the patient, aggressively and with alacrity.

Is this Sepsis-3.1?

Should it be?