Water Cooler Breakdown: The ProCESS Trial

Chan, D. "we will die of sepsis" https://www.flickr.com/photos/denn/4311117490/in/photostream/ Creative Commons License ShareAlike 2.0
Chan, D. "we will die of sepsis" https://www.flickr.com/photos/denn/4311117490/in/photostream/ Creative Commons License ShareAlike 2.0

In March of 2014, Derek Angus and colleagues published the ProCESS trail in the NEJM (1)(N Engl J Med 2014;370:1683-93. DOI: 10.1056/NEJMoa1401602). In ProCESS, they explore the time-honored theory in EM-resuscitation that EGDT as described by Rivers (NEJM 2001) is the dominant strategy to improve survival in severe sepsis and septic shock. Despite the marked reduction in mortality that is reported in Rivers’ study, the study itself has not been successfully reproduced in a multicenter trial.

It is important to note that the influence of the Rivers trial was so great that it is entirely possible that a true RCT of EGDT vs “usual” therapy can never really happen again. The days of “2 liters, some antibiotics and wait for an ICU bed” are long gone and most EM clinical “trialists” wouldn’t participate in a “placebo” controlled sepsis trial at this point—there are simply too much compelling data driving our bedside decisions today to allow us to revert to giving lower volumes of fluids and using greater doses of pressors than we currently do. Would you let your loved ones be enrolled in a trial like that?

So along comes the ProCESS trial.  1341 patients with severe sepsis or septic shock from 31 academic hospitals (without mandatory pre-existing sepsis protocols and which di not routinely use ScVO2 monitors) were enrolled in a (1:1:1) randomization into three arms of study.

  • InclusionCriteria:
    • > 18 yo
    • 2/4 SIRS
    • lactate > 4, or
    • SBP < 90 after 30 mL/kg or 1000 mL (protocol change mid trial)
  • Outcome Measures:
    • Primary: 60 day mortality
    • Secondary:
      • 90 day and 1 year mortality
      • duration of pressor needs
      • Acute Renal Failure
      • Acute Respiratory Failure

Trial Arms

EGDT” (Rivers style)

STANDARD”  Protocolized Sepsis Care for trial

USUALCare(what a prudent EP does)

Description of Therapies

  • CVP 8 - 12
  • ScVO2 > 70
  • MAP >65 but <90
  • 500 mL Fluid boluses as needed to achieve targets
  • Sufficient peripheral IV access
  • SBP >100
  • Shock index < 0.8
  • Clinical eval of hypoperfusion and fluid status hourly
  • 500 mL boluses to chase shock index and SBP
  • HGB > 7.5
  • Bedside care per discretion of treating team, NOT study team.
  • No mandated study intervention protocol


There was no difference in the primary outcome measure of mortality between groups at 60 or 90 days or 1 year. 

Secondary outcomes were interesting but not statistically significant, with the exception that a greater proportion of patients in the EGDT arm were admitted to an ICU, but the overall rate of ICU admission was > 85% for all patients in the trial. There was no difference in the incidence or duration of mechanical ventilation, dialysis or pressor usage between groups.

Buried within the supplemental materials is more comparative post-hoc data that shows some minor but interesting differences between groups:

  • EGDT patients had a greater than 2-fold increase in placement of central lines (may suggest they aren’t always needed as some have argued)
  • More dobutamine was used in EGDT group (not surprisingly)
  • More pRBC given in EGDT group, also not surprising, but statistically significant with no clear benefit in terms of tissue oxygenation.

Also contained within the supplemental materials are the data relating to the effects of the resuscitative process of care and the differences in elements seen in the first 6 hrs. The key elements to note are that the protocol group received more fluids than the EGDT group (3.3 vs 2.8 L) and the “usual” care group only received an average of 2.3L demonstrating once again that providers often limit resuscitative fluids compared to protocolized endpoint driven fluid strategies. Also notable is that there were no differences in serious adverse advents between groups, despite the increased intensity of the EGDT protocol.

Bottom Line(s)

  • All therapies worked in this trial, as long as you had clinicians paying attention to the patients
  • EGDT is aggressive and invasive, but didn’t appear to be associated with more harm than the other strategies. You should certainly feel justified, but perhaps not compelled, to follow EGDT as we do
  • Fluid is critical to survival in sepsis; find a relevant endpoint and target it
  • The SHOCK INDEX (SI) is a wonderful addition to the arsenal to help us identify cryptogenic decompensated sepsis (Heart Rate ÷ Systolic BP); SI > 0.7 is NOT normal; > 1.2 is legitimately something to fear. This can easily be incorporated into an EPIC vital sign-derived trigger in triage
  • There is no longer wild-type severe sepsis and septic shock in the ED; we will never be able to really perform another RCT in this patient population because no one will ever be willing to randomize to limited fluids in sepsis when we know that survival usually requires > 30 mL/kg of resuscitative fluids


  • Yealy, D. et al (The ProCESS Investigators) (2014) New England Journal of Medicine. A Randomized Trial of Protocol-Based Care for Early Septic Shock. 370. 1683-1693. DOI: 10.1056/NEJMoa1401602
  • Chan, D. "we will die of sepsis". https://www.flickr.com/photos/denn/4311117490/in/photostream/.  Accessed 5/30/14.  Creative Commons ShareAlike license 2.0.