52 Articles: Early Goal Directed Therapy in Sepsis

Main Points:

  • In comparison to standard care, early goal-directed therapy (EGDT) was demonstrated to significantly reduce in-hospital mortality for patients presenting with severe sepsis or septic shock.
  • Patients in the EGDT group also demonstrated higher mean SCVO2, a lower lactate concentration, a lower base deficit, and a higher pH when compared to the standard therapy group, suggesting aggressive optimization of hemodynamic parameters and targeting of predefined resuscitation endpoints contributed to their better outcomes. 
  • New multicenter studies have now called into question the value of EGDT as compared to current standard care.   


Severe sepsis and septic shock are common and have been recognized as a substantial cause of morbidity and mortality in the United States.

Goal-directed therapy, i.e. guiding resuscitation by predefined endpoints based upon laboratory data and invasive hemodynamic monitoring, has long been used and validated in the ICU setting. In this landmark 2001 trial, Dr. Rivers, a physician triple-boarded in emergency medicine, internal medicine and critical care, sought to examine the utility of implementing goal-directed therapy in the emergency department setting for the treatment of severe sepsis and septic shock. Hence, this was “early” goal-directed therapy as opposed to waiting to initiate such treatment in the ICU. More broadly, Rivers’ thought process mirrored the rising trend of “bringing upstairs care downstairs” and the continuing evolution of critical care in emergency medicine.

For Rivers, sepsis was a particularly salient target of early goal-directed therapy, as it was a disease state that could be altered through timely intervention, with pathology that was largely driven by changes in hemodynamics. According to this school of thought, sepsis spans a pathophysiologic continuum that begins with the systemic inflammatory response syndrome and, if left untreated, progresses to end organ failure and death. The circulatory abnormalities that result from sepsis, namely intravascular volume depletion, peripheral vasodilation, myocardial depression and increased metabolism, create an imbalance between oxygen delivery and oxygen demand throughout the body. It is this imbalance that chiefly drives and is responsible for global tissue hypoxia, shock, and death. Rivers thought that progression along this continuum of disease could be halted by early recognition and correction of the systemic imbalance between oxygen delivery and demand. To that end, promptly optimizing hemodynamics by meeting specific resuscitation end points or “goals,” could reduce the morbidity and mortality of sepsis.


This was a prospective, randomized study that enrolled 263 patients who presented to a single urban emergency department over a three-year period with severe sepsis or septic shock. Criteria for inclusion were 2/4 SIRS criteria along with an SBP ≤ 90mm Hg after a 20-30cc/kg bolus of IV crystalloid OR a blood lactate concentration of ≥ 4mmol/L. Exclusion criteria were numerous and included: age < 18 years old, pregnancy, acute CVA, ACS, acute pulmonary edema, status asthmaticus, cardiac dysrhythmias, GI bleeding, seizure, drug overdose, burn injury, trauma, cancer on chemotherapy, immunosuppression, contraindications to central venous catheterization or DNR status. Patients were randomized by computer assignment to either early goal-directed therapy or to standard therapy. Of note, the treating clinicians in the ED were not and practically speaking could not be blinded to these assignments.

Standard therapy was described in this study as being “at the discretion of the clinician,” and consisted broadly of central venous and arterial catheterization with a goal of maintaining a CVP of 8-12mm Hg, MAP > 65mm Hg, and urine output > 0.5mL/hr. Cultures were obtained and antibiotics administered. Critical care consultation was also obtained and patients were admitted from the ED as soon as possible.  

Conversely, patients in the early goal-directed therapy group were treated by a specific protocol for at least six hours in the ED and then transferred to inpatient beds. The critical care clinicians who subsequently assumed care of these patients were not aware of their study group assignments. According to the EGDT protocol, 500cc IVF crystalloid boluses were given every 30 minutes to achieve CVP of 8-12 mm Hg. If MAP was < 65, vasopressors were given. If MAP was > 90, vasodilators were given. If central venous oxygen saturation was < 70%, pRBCs were transfused to achieve a hematocrit of at least 30%. If the CVP, MAP and hematocrit were optimized and SCVO2 was still < 70%, dobutamine was started at 2.5mcg/kg/min and increased until SCVO2 was 70% or higher or the maximum dose of dobutamine was given. If the patient was still not hemodynamically optimized after these interventions, they would be mechanically ventilated and sedated if they were not already.

The primary efficacy end point was in-hospital mortality. Secondary end points for resuscitation such as temperature, heart rate, blood pressure, urine output, and CVP were also measured out to 72 hours. APACHE II, SAPS II and MODS scores were further calculated as was consumption of health care resources.

Overall, the study found a significant reduction of in-hospital mortality in the EGDT group as compared to the standard therapy group (30.5% vs. 46.5%, relative risk of 0.58 (0.38-0.87), P = 0.009). With regard to other secondary resuscitation end points, the study authors found that between 7-72 hours, those in the EGDT group had a significantly higher mean central venous oxygen saturation (70.4±10.7 percent vs. 65.3±11.4 percent), a lower lactate concentration (3.0±4.4 vs. 3.9±4.4 mmol/L), a lower base deficit (2.0±6.6 vs. 5.1±6.7 mmol/L), and a higher pH (7.40±0.12 vs. 7.36±0.12) than the patients assigned to standard therapy (P < 0.02 for all comparisons). They also found APACHE II, SAPS II and MODS scores to be significantly higher in patients assigned to standard therapy as opposed to those assigned to EGDT (P= < 0.001), suggesting a greater degree of end organ dysfunction in the standard therapy group as compared to the EGDT group.

There were interesting differences noted in the amount and timing of resuscitation between the two groups. Perhaps unsurprisingly, in the first 6 hours the EGDT group received more IVF (5 vs. 3.5L, p < 0.001), pRBC transfusions (p < 0.001), and inotropic support (p < 0.001). Between 7-72 hours however, the standard care group received more pRBC transfusions (p < 0.001), more vasopressors (p=0.03), more mechanical ventilation (p<0.001) and more PA catheterization (p=0.04). This suggests that the standard therapy group was initially under-resuscitated in the first 6 hours, contributing substantially to longer term morbidity and mortality. 

Where to go from here:

Although fundamentally altering the standard of care for sepsis, Rivers’ study has since come under increasing scrutiny, particularly in light of the recent ARISE and ProCESS trials, two multicenter studies that demonstrated no difference for in-hospital mortality between EGDT and standard therapy (see relevant articles below). These new studies, however, have been questioned given the possibility of significant overlap now between “standard” care and EGDT as a result of Rivers’ work having altered the standard of care itself. While the utility of certain aspects of EGDT, such as following SCVO2 to guide resuscitation, have been questioned, Rivers’ work set the standard for the aggressive treatment of sepsis that we see today. This in turn has resulted in a significant decline in the morbidity and mortality of sepsis over the past several decades.

Level of evidence:

Based on the ACEP therapeutic grading scheme this article is Level I evidence.

Resident Reviewer: Dr. Anatoly Kazakin
Faculty Reviewer: Dr. Matt Siket

Relevant articles:

Peake, S. et al. “Goal-directed resuscitation for patients with early septic shock.” NEJM 371;16 Oct 16, 2014.

Yealy, D. et al. “A randomized trial of protocol-based care for early septic shock.” NEJM 370;18 May 1, 2014.