Much has been made of efforts to detect and treat sepsis as early as possible after presentation, with many post-hoc analyses seeming to demonstrate time-sensitive mortality benefits associated with receiving the various components included in our “quality” measures. However, just like early goal-directed therapy back in the day, it has never truly become clear which element of early sepsis care confers the survival benefit. Absent specific data regarding how to best tailor therapy to the individual patient, we simply bludgeon everyone with the same sepsis bundle.
And, as we see here, that generalization is likely harmful.
This is a small randomized trial from Zambia – not to be confused with Nambia – in which 212 patients presenting with suspected infection, at least 2 systemic inflammatory response syndrome criteria, and hypotension were randomized to an early resuscitation protocol or “usual care”. The early resuscitation protocol sounds, generally speaking, similar to our modern approach to sepsis care – early intravenous fluid boluses with frequent intravascular volume assessment, vasopressors, and transfusions for severe anemia. The usual care cohort received, essentially, the same, only less so.
The two groups enrolled were roughly equivalent – but nothing like our sepsis cohort here in the United States. Approximately 90% of the patients enrolled were positive for the human immunodeficiency virus, half of the positive culture results were tuberculosis, the mean hemoglobin was 7.8 g/dL, and the majority of patients had been bedridden for over a week prior to presentation. Intravascular volume status and response to resuscitation was assessed by evaluating jugular venous pressure, tachypnea, and peripheral oxygenation. Finally, unlike many modern settings, most patients received dopamine as their vasopressor of choice and were admitted to general medical wards rather than intensive care units. In short, these are patients and care settings unlike in most industrial nations.
Mortality, not unexpectedly, was high – but it was much higher in those randomized to the early resuscitation cohort. Mortality in those receiving early resuscitation was 48.1%, compared with 33.0% in those receiving usual care. This mortality different then persisted out to 28 days, with over 60% mortality in the early resuscitation group at that time, compared with just over 40% in usual care.
This trial does not necessarily call into question the general principles of modern sepsis care, but it certainly provides a couple valuable lessons. The first, and most obvious, is the cautionary tale regarding generalizing research findings from one setting to another. Even a reasonably important, beneficial effect size can be transformed into a greater magnitude of harm if applied in another clinical setting. Then, this should clearly make us re-examine our current approach to sepsis care to ensure there is not a subgroup for whom early resuscitation is, in fact, the wrong answer. Our blind pursuit of checking boxes for quality measures, while generating an overall beneficial effect, is probably resulting in waste and harms for a substantial subgroup of those presenting with sepsis and septic shock.
“Effect of an Early Resuscitation Protocol on In-hospital Mortality Among Adults With Sepsis and Hypotension”
http://jamanetwork.com/journals/jama/fullarticle/2654854