Protocol-Based Approach to Sepsis Still Necessary?

Andrew F. Shorr, MD, MPH


December 22, 2014

This feature requires the newest version of Flash. You can download it here.

This is Andy Shorr from Washington, DC, with the pulmonary and critical care literature update.

Today I want to highlight the ARISE trial.[1] This is the second of three recently completed randomized controlled trials looking at early goal-directed therapy (EGDT) in the treatment of patients who present to the emergency department (ED) with septic shock.

In this trial, conducted in Australia among 1600 patients, participants were randomly assigned to receive standard care or EGDT. The EGDT protocol required the placement of a catheter to measure the central venous oxygen saturation (SCV02), to clarify how the use of this catheter and this specific protocol alter outcomes in early septic shock.

As with the ProCESS study,[2] the ARISE trial found no difference in 90-day mortality with the use of EGDT vs standard of care in this group of patients who presented to the ED. However, with EGDT, they spent more on the catheter, required more dobutamine, and gave more transfusions. There are some nuances that I believe are important before we conclude that a protocol-based approach to septic shock has no value.

A Few Caveats to the ARISE Findings

Before patients were enrolled in the study—and they were enrolled quickly after presenting to the ED—patients in both arms received, on average, a 30 cc/kg bolus of crystalloid. This is very different from what we have seen historically, at least in the United States, where the use of fluids in the ED has been much more conservative. In the ARISE trial, everyone received 1 or 2 L of fluids up front, before they were enrolled. That has to affect outcomes.

Second, after randomization, patients in the EGDT arm did receive more crystalloid than patients in the standard-of-care arm, but on average that difference was less than 300 cc. Thus, it appears overall that patients received essentially the same amount of crystalloid, regardless of which arm they were assigned to. That suggests that our practice of giving fluids has changed. When you look at how crystalloid was used in Manny Rivers' early study[3] of EGDT and in other observational studies over a decade ago, you see that patients were getting much smaller boluses of fluid in the ED or the critical care unit (CCU), even when they were critically ill. What this study really tells me is that practice has changed and perhaps that explains why mortality rates from severe sepsis and septic shock have been falling globally for over a decade.

Moreover, this study enrolled patients with a very low risk for death. The mean APACHE-2 score of patients in this study was approximately 15; in this group of patients with septic shock, the 90-day mortality rate was only 20%. By contrast, mortality rates in most septic shock studies we have seen historically[3]—other than the most recent ones such as ProCESS—have been in the 30% to 40% range. If you look at the recently completed TRISS trial,[4] which focused on transfusion thresholds, the 90-day mortality was in the 40% range for septic shock in the CCU.

Thus, we are studying different groups of people here and we must understand that severe sepsis at an early time point is very different from established septic shock in the CCU. Similarly, we must understand that not all populations are the same phenotypically.

More Caveats and the Take-Home

In this population, only about half of the patients had a lactate level above 4 mmol/L, so even though these patients presented with hypotension and signs of vasodilation and end-organ dysfunction, they had not developed a lactic acidosis. Given these nuances, it is unclear how having a catheter in these patients would have been valuable in the first place. In one sense, it is as if these investigators created a straw man and knocked it down, because these are patients who, in my hospital, probably would have received 2 L of fluid and never would have been considered for admission to the CCU, but could have gone to our intermediate-care unit or the general medical unit. If you are studying patients with such a low risk for death because they meet the entry criteria, then you have diluted your ability to ascertain where an aggressive fluid resuscitation and protocol-based approach might be valuable.

The study clearly is internally valid. The authors have done an excellent job, and the ANZICS (Australian & New Zealand Intensive Care Society) group has done superlative work in advancing our understanding of critical care through a series of large, randomized controlled trials, whether it is looking at crystalloid and transfusion requirements or things such as this. But I believe that in this case we have to be cautious because, as with ProCESS, it seems that they are studying a group of patients in the ED who will do well with 2 or 3 L of crystalloid and antibiotics, no matter what.

Now, that is good because we can identify those patients, but it also suggests that we should not necessarily generalize this to what we are doing at hour 6, 8, 12, or 24 in the CCU when those patients come to us from the ED.

Again, I believe that the ARISE study by the ANZICS investigators, recently published in the New England Journal of Medicine, is very thought-provoking and something we should all be looking at.

This is Andy Shorr, from Washington, DC.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: