Secondary Logo

Journal Logo

InFocus: Interventions for Sepsis The ProCESS Trial

Roberts, James R. MD

doi: 10.1097/01.EEM.0000450843.19165.15

Dr. Robertsis the chairman of emergency medicine and the director of the division of toxicology at Mercy Catholic Medical Center, and a professor of emergency medicine and toxicology at the Drexel University College of Medicine, both in Philadelphia. Read The Procedural Pause, the blog by Dr. Roberts and his daughter, Martha Roberts, ACNP, CEN, at, and all of Dr. Roberts' past columns at





Treatment for sepsis has been continually scrutinized, and remarkable advances have been made ever since the study by Rivers et al. was published more than 12 years ago. That single-center trial reported that patients with severe sepsis and septic shock had lower mortality when a six-hour protocol of early goal-directed therapy (EGDT) was initiated in the ED. (New Engl J Med 2001;345[19]:1368.)

The mortality rate in the study was 30.5 percent for EGDT and 46.5 percent for non-protocol therapy, giving rise to a medical gospel that outlined critical actions for the early treatment of sepsis. The original EGDT protocol used a central venous catheter (CVC) and monitored central venous oxygen saturation (ScvO2) and central venous pressure (CVP) to guide intravenous fluids, vasopressors, antibiotics, and packed red blood cells. Certain trigger points required interventions. Fluids were infused, for example, if the CVP was less than 8 mm Hg. Transfusion of packed red cells to targeted hematocrits was mandated if the ScvO2 was less than 70%. Dobutamine or other inotropic agents were initiated if the hematocrit was already 30%.

EGDT rapidly became the mantra of sepsis investigators, and every hospital rushed to provide similar interventions to be current. All of the interventions were suggested as being critical because no one was certain which had the most beneficial effects.

Much has been learned since the initial publication of EGDT for sepsis and severe septic shock. Most EDs have dramatically improved early detection and have instituted early aggressive sepsis monitoring and therapies. Aggressive timely assessment and therapy were considered lacking before this. It was later determined, however, that dobutamine did not improve microvascular profusion in septic shock, that transfusions to a hematocrit of 30% appeared harmful in some critically ill patients, and that lactate measurements soon became an adequate substitute for invasively obtained parameters. The ProCESS trial, a long overdue multicenter randomized trial, was done to assess the efficacy of the elements of the EGDT protocol.

Back to Top | Article Outline

A Randomized Trial of Protocol-Based Care for Early Septic Shock: The ProCESS Investigators

N Engl J Med

March 18, 2014 (PMID: 24635773)

The authors note that more than 750,000 cases of severe sepsis and septic shock occur in the United States each year, with most patients receiving their initial care in the ED. The short-term mortality rate is 20 percent or more, but mortality rates declined in 2001 after Rivers et al. reported success when sepsis was treated according to the six-hour early goal-directed therapy protocol. The EGDT concept assumed that the then-standard therapy was lax and failed to intervene with aggressive and timely assessment and treatment. The authors concluded that central monitoring to review oxygen saturation and venous pressure measurements to guide the use of IV fluids, vasopressors, pack cells, and dobutamine were all beneficial. The question addressed in the ProCESS study was whether all of the elements are still necessary.

It made sense to assess whether the protocol-driven intervention was superior to standard care, though current garden-variety regimens reflected already-instituted clinical advancements. Particularly of concern were invasive monitoring techniques that are time-consuming and associated with complications when performed in the ED. These authors' randomized trial, the Protocolized Care for Early Septic Shock (ProCESS), was conducted at 31 U.S. hospitals, and was originated and published by the University of Pittsburgh.

Importantly, all of the participating sites were academic EDs with voluminous resources and intense continuing education. The treating clinicians entered and studied patients with criteria for the systemic inflammatory response syndrome and those who had refractory hypotension or a serum lactate 4 mmol per liter or higher. Refractory hypotension was defined as less than 90 mm Hg or requiring vasopressors to maintain blood pressure above that level after an IV fluid challenge of 20 mls/kg or more.

Patients were randomly assigned to one of three treatment groups in an equal ratio: a formal protocol-based EGDT, a protocol based on current albeit improved standard therapy, or so-called non-protocol standard care. Patients assigned to the EGDT group had a central venous catheter to monitor CVP pressures and ScvO2 promptly placed, and clinicians initiated the EGDT protocol to determine the amount and timing of IV fluid administration, and they followed the criteria for vasopressors, dobutamine, or packed red cells.

The second group, the protocol-based standard study arm, used similar but less aggressive criteria than the formal EGDT. It also utilized adequate peripheral venous access instead of a central line and gauged the administration of fluids and vasoactive agents to reach goals for systemic blood pressure. In contrast to the formal EGDT protocol, red cell transfusions were only given if the hemoglobin was less than 7.5 g/dL. The amount of fluid administered was decided by the team to be enough to replace the loss and provide adequate volume resuscitation.

Bedside providers directed all care on their own clinical estimates of needs and requirements for the third usual-care group. The individuals monitoring the care could not suggest various interventions or therapies.

The primary outcome of the study was the rate of in-hospital death from any cause in 60 days. Further outcomes included duration of acute cardiovascular failure, the need for vasopressors, acute respiratory failure, acute renal failure, and duration of hospital stay. A total of 1341 patients were enrolled and analyzed over five years. All three groups were well matched at baseline. The leading diagnoses were pneumonia, urinary tract infection, intra-abdominal infection, and skin soft tissue infection. One of the main differences in therapy was the volume of intravenous fluids administered over six hours. During the initial treatment, the protocol-based EGDT group received 2.8 liters, the protocol-based standard therapy group received the greatest fluid load of 3.3 liters, and the usual-care group received the least fluid, 2.3 liters. Crystalloid was the predominant fluid in all groups. Patients in the two protocol groups received vasopressors more often, while the standard group received vasopressors significantly less frequently. Central catheters were eventually placed in about half of the non-EGDT patients. The use of antibiotics, corticosteroids, and activated protein C was similar across all groups. All patients received low-tidal volume ventilation and moderate glycemic control.

The number of patients dying in each group was statistically the same, around 20 percent, after 30 days. No difference in 90-day and one-year mortality was seen. No significant differences were seen among any of the groups in length of ICU stay, duration of cardiovascular failure or respiratory failure, or in length of hospital stay. Measuring serial lactates was not inferior to the prior goal-directed ScvO2 monitoring.



These authors concluded that this multicenter randomized trial resulted in similar antibiotic use and other resuscitation measures. Patients in the two protocol-driven arms had greater CVP use as well as more intravenous fluids, vasoactive agents, and blood transfusions. Mortality and morbidity rates were good without the use of routine initial intravenous catheterizations or central hemodynamic monitoring. The authors found no benefit of protocol-based resuscitation guidelines over usual bedside care provided according to the treating physician's clinical judgment for patients identified as being in septic shock in the ED. No specific benefit of routine central venous catheterization could be found.

Comment: Just when one has mastered the art of early goal-directed therapy, the actual value of adhering to the exact protocol becomes less obvious. Lactate measurements have largely replaced venous pressure monitoring in the ED, and educating physicians about early use of adequate fluid resuscitation and broad-spectrum antibiotics is probably the most important in providing a better outcome. It's a relief when a central line and complicated testing are not needed to obtain the same outcome, and it seems that good old physician judgment, albeit newly informed and educated, appears to fit the bill.

Keep in mind that all patients were treated in academic medical centers where state-of-the-art care should be extant. The middle group in this study was actually a mild hybrid of EGDT. The bar has been raised regarding so-called standard care, especially for sepsis and septic shock, based largely on the original EGDT. But it is less certain whether all EGDT is feasible or if it has filtered through to smaller community or rural hospital EDs. The usual-care group actually received the smallest volume resuscitations, but all groups in reality received aggressive care compared with 10 years ago. Care was clearly stellar in these academic centers. A total of 97 percent of patients received antibiotics and 76 percent had antimicrobials within six hours of arrival, even before they were randomized.

We still have not figured out how to get patients out of the ED rapidly, and transfer of care to the ICU team is still problematic. This is still probably the biggest deficit to be tackled. Patients in septic shock are pounced on and aggressively treated by most ED staffs with broad-spectrum antibiotics and adequate fluids. Even the patient who seems to be stable or the one about to decompensate can be readily prognosticated by an elevated lactate.

It is rather interesting that all three treatment methods, though using similar approaches, demonstrated the same survival rate at numerous medical centers. Much of this can be credited to the alacrity of recognition and the aggressive initial care by the ED staff. The so-called usual standard care, not always present at every hospital, performed as well as sophisticated and intense protocols. Early sepsis is still frustrating to identify, predict, diagnose, and treat adequately. Unidentified patients can decompensate in a very short time while the ED staff lollygags around other seemingly more important patients.

Eschewing the initial central venous catheter to monitor CVP and oxygen levels is good news to most EPs. CVP monitoring, however, is likely still very important for assessing ongoing fluid resuscitation and need for vasopressors in the ICU. This study goes a long way in resolving the longstanding clinical debates about how to manage sepsis patients best within the first few hours of treatment because good results were reported in academic EDs even when doctors did not follow specific protocols. The fancy initial invasive monitoring techniques add little to the final outcome as long as sepsis is recognized promptly and patients are adequately treated with fluids and antibiotics.

Two other sacred cows for treating sepsis seem to have been put out to pasture, though they are not generally ED-related. The first is the routine use of high- or even low-dose steroids. Some intensivists still administer hydrocortisone to patients with a systolic blood pressure less than 90 that persists despite adequate fluid resuscitation and vasopressor administration, but this is still questioned by many. The second previously praised bromide is tight glycemic control in critically ill patients. Both interventions have been found to be ineffective and potentially harmful, but these issues are largely out of the realm of emergency medicine. The once-hallowed ACTH stimulation test may be unreliable in critically ill patients, and the optimal level of cortisol necessary to survive critical illness is unclear. Finally, results have recently demonstrated that etomidate when used as a single dose for intubation in the ED is not harmful, though it may initially suppress the hypopituitary adrenal axis.

Access the linksin EMN by reading this on our website or in our free iPad app, both available Comments?Write to us

Back to Top | Article Outline

Core Strategies Make the Difference

Read an editorial about the ProCESS trial by Alan Chiem, MD, and Luis M. Lovato, MD, on p. 6.



Reader Feedback: Readers are invited to ask specific questions and offer personal experiences, comments, or observations on InFocus topics. Literature references are appreciated. Pertinent responses will be published in a future issue. Please send comments to

Dr. Roberts: I am writing to thank you for the two articles that you wrote in EMN about pyomyositis back in 2005. (EMN 2005;27[2]:23,; 2005;27[1]:20, I recently saw a generally healthy (DM, some EtOH abuse) middle-aged man in our ED with a few hours of fever. We measured his temperature at something like 102.7°F. He had some cough and congestion, but he also complained of two weeks of mid-right thigh pain. His two previous evaluations at an outside clinic for the thigh pain resulted in the diagnosis of some sort of muscle strain, although there was no recalled injury. He was given an NSAID, and he hadn't had a fever on those clinic visits.

When I looked at his right thigh, it appeared completely normal. And when touched it with my hand, it didn't feel unusually warm. Then, for some reason (the mind is a funny thing), a light bulb went on in my head, and I remembered your pyomyositis articles from way back: I could feel a distinct area of swelling, almost the size of my fist, when I pushed firmly against the patient's thigh. His CT scan showed a 3×3×7 cm abscess in his vastus intermedius and vastus medialis muscles. I later stuck a needle in the abscess and aspirated 6 cc of bloody pus. Not surprisingly, it was growing methicillin-sensitive Staphylococcus aureus.

Thanks again for your articles. I probably would have failed to make this interesting and important diagnosis if I hadn't read them. — Charlie Stengel, MD, Newport, RI

Dr. Roberts responds: Amazing, fantastic, great pickup. Thanks for the feedback.

Back to Top | Article Outline

Read InFocus and Earn CME!

Earn CME by completing a quiz about this article. You may read the article here, on our website, or in our iPad app, and then complete the quiz, answering at least 70 percent of the questions correctly to earn CME credit. The cost of the CME exam is $10. The payment covers processing and certificate fees.

Visit for more information about this educational offering and to complete the CME activity. This enduring material is available to physicians in all specialties, nurses, and other allied health professionals. Lippincott Continuing Medical Education Institute, Inc., is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. Lippincott Continuing Medical Education Institute, Inc., designates this enduring material for a maximum of 1 AMA PRA Category 1 Credit™. Physicians should only claim credit commensurate with the extent of their participation in the activity. This activity expires June 30, 2015.

Learning Objectives for This Month's CME Activity: After participating in this CME activity, readers should be better able to prepare a plan for treating sepsis and septic shock patients.

© 2014 by Lippincott Williams & Wilkins