The end of early-goal directed therapy?

a b s t r a c t

Emergency medicine practitioners around the world have been confronted with the increasing challenge of managing patients in severe sepsis and septic shock. Introduction of early goal-directed therapy revolu- tionized sepsis care and was adopted worldwide. Since then, multiple randomized controlled trials have been published questioning the superiority of EGDT. The purpose of this article is to review and provide clinical com- mentary on the ProCESS, ARISE, and ProMISE trials, which address whether invasive, expensive interventions are needed to achieve mortality reduction goals in septic patients. This article discusses that EGDT bundled care is not necessary to achieve mortality reduction goals.

(C) 2015

  1. Introduction

Severe sepsis and septic shock have been a growing challenge for emergency medicine practitioners across the world and in Canada, with a significant increase in incidence (17.2% from 2004 to 2009), emergency department (ED) admissions (79%), hospital length of stay (12 days), in- tensive care unit length of stay (11 more days vs nonseptic patients), and a high overall mortality rate (30.5% of all septic patients, 45.2% with severe sepsis) [1]. Introduction of early goal-directed therapy (EGDT) by Rivers et al [2] revolutionized sepsis care and was adopted rapidly worldwide with considerable success [3]. Practice guidelines for sepsis resuscitation in Canada and globally have been anchored on EGDT [7-11].

Barriers to uptake of EGDT in smaller ED settings have included time constraints, resource limitations, lack of critical care training, and need for invasive monitors [9]. Furthermore, it was unclear whether or not invasive, expensive interventions were needed to achieve mortality reduction goals. This article reviews the recently published ProCESS, ARISE, and ProMISE trials, which were designed to test this hypothesis. Furthermore, it provides clinical commentary and addresses the contro- versies regarding the necessity of EGDT.

  1. Methods

All 3 studies were partially blinded, randomized, controlled trials, and their purpose was to test whether protocol-based resuscitation with central hemodynamic monitoring to guide use of fluids, vasopressors,

? Financial support: This research received no specific grant from any funding agency, commercial, or not-for-profit sectors.

?? Conflict of interest: None.

* Corresponding author. Division of Emergency Medicine, Department of Medicine, McMaster University, 237 Barton St East, Hamilton, ON, Canada L8L 2X2.

E-mail address: [email protected] (S. Sharif).

blood transfusions, and dobutamine was superior to usual care [4-6]. All author groups harmonized their end points prior to trial initiation, so that subsequent data pooling for meta-analysis would be seamless [12].

All 3 trials consisted of an EGDT group where strict protocolized care (based on the Rivers study) was conducted for 6 hours and a control group (usual care) where the bedside physician directed all care for 6 hours. In addition, the ProCESS trial had a third group, which was a protocol-based Standard therapy group, where relaxed protocolized care (ie, did not require placement of Central lines, inotropes, or blood transfusions) was conducted for 6 hours with a dedicated doctor, nurse, and research assistant that provided prompts.

Inclusion and exclusion criteria were similar across all studies (Tables 1 and 2). The primary outcome of ARISE and ProMISE was 90-day mortality, whereas that of ProCESS was 60-day mortality. Selected secondary outcomes are outlined in Table 3. Patients in all trials had received antibiotics and roughly 2L of fluids prior to randomi- zation to establish whether they had refractory shock [4-6].

  1. Results

All 3 studies showed that EGDT was not superior to usual care. Pro- CESS was conducted in large, busy, Tertiary care centers in the United States [4]. Conversely, ARISE was a multinational trial that was completed in small and large hospitals [5]. Furthermore, ProMISE examined health- related quality of life between usual care and EGDT, as well as resource use and costs [6]. Both outcomes were not significantly different among the 2 groups [6].

Interestingly, 90-day mortality was lower in ARISE (Table 3) [6]. The lower Acute Physiology and Chronic Health Evaluation scores of patients in ARISE in comparison to the other trials may explain this incongruence by indicating that patients in ARISE were not as sick (Table 3) [6].

The total amount of fluids administered in 6 hours in all 3 trials ranged from 4.0 to 5.5 L, which is similar to the Rivers trial where

0735-6757/(C) 2015

S. Sharif et al. / American Journal of Emergency Medicine 34 (2016) 292294 293

Table 1

Brief outline of the protocol and inclusion criteria of the ProCESS, ARISE, and ProMISE trials

Table 3

Primary outcomes, select secondary outcomes, and select baseline characteristics of patients in the ProCESS, ARISE, and ProMISE trials


Countries US Australasia, Finland, UK Hong Kong, Ireland








60-d mortality








90-d mortality








APACHE II scores








refractory hypotension








Lactate >= 4








Average lactate levels








Admission to intensive








Intervention ProCESS ARISE ProMISE

No. of patients 1341 1588 1251



20-30 mL/kg (changed during study)

N 1000 mL N 1000 mL

Intervention EGDT EGDT EGDT Protocol-Based Usual Usual Usual

Patient enrollment

care unit

Age >= 18 y

  • ? ? Central lines









  • ? ? IVF received









prerandomization (L)

>= 2 SIRS criteria








Refractory hypotension or

  • ? ? Vasopressors








lactate >= 4

Dobutamine use








b 6 h from ED arrival

? ? ? PRBC transfusions








b 12 h from ED arrival








b 2 h from shock criteria ? ? ? Abbreviations: APACHE II, Acute Physiology and Chronic Health Evaluation II; IVF, in vitro

patients received 5.0 L [2,4-6]. Fewer patients received vasopressors, dobutamine, and blood transfusions in the usual-care group in all 3 trials. Although insertion of central lines was only mandated in the EGDT group, more than 50% of patients in the usual-care group across all 3 trials had a central venous catheter inserted [4-6].

  1. Discussion
    1. Methodological issues

The methodology of all 3 trials was rigorous, with the exception of inability to blind treating physicians and patients during initial resusci- tation. However, data outcomes were locked until trial completion and blindly analyzed. Furthermore, near-perfect follow-up was achieved. Intention-to-treat analyses were used in all trials.

Another key methodology consideration is that all 3 trials were designed to test the superiority of EGDT vs usual care. Equally important, however, is the notion that one treatment protocol should not be significantly inferior to the other. The noninferiority of treatment strategies is determined by the magnitude of the difference in risk of out- comes between the 2 interventions, as compared with a previously

Table 2

Summary of the exclusion criteria of ProCESS, ARISE, and ProMISE




Age b 18 y




Known or suspected pregnancy




Hemodynamic instabilitya




Requirement for immediate surgery




Known Hx of AIDS




advanced directives restricting protocol




Contraindication to central line




Contraindication to blood transfusionb




Resuscitation deemed futile




Transfer from in-hospital setting




Protocol not followedc




Ongoing participation in another interventional study




Underlying disease process with Life expectancy b 90 d




Another primary diagnosisd




a Hemodynamic instability secondary to active bleeding or active gastrointestinal hemorrhage.

fertilization; PRBC, packed red blood cell.

a Adverse events include pneumothorax, hemopneumothorax, arrhythmia, bleeding, thrombosis, pulmonary emboli, vascular catheter infection/thrombosis, pulmonary edema, blood transfusion reaction, myocardial ischemia, and peripheral ischemia.

defined noninferiority margin for Risk difference [13]. The risk differences for the contemporary trials are listed in Table 4.

The Rivers trial showed a large risk difference in favor of EGDT (14%), which may have ranged from 2% to 26%. The contemporary trials show much smaller risk differences, likely reflecting significant improvements in sepsis “usual care” since 2001. The risk difference ranges also suggest true equivalence and noninferiority of usual-care to EGDT, because the 95% confidence interval ranges fall within 10% of the measured difference, a noninferiority limit that would likely be considered reasonable in the EGDT era of ED sepsis care.

Clinical commentary

The Surviving Sepsis Campaign (SSC) recommends protocolized resuscitation of patients with sepsis, but achieving all aspects of the 6- hour bundle is difficult for most EDs in Canada [8,15]. ProCESS, ARISE, and ProMISE challenge the belief that the 6-hour Sepsis Bundle devel- oped by Rivers et al is necessary to improve mortality from severe sepsis and septic shock.

Usual care in 2015 is the result of more than a decade of early work by Rivers et al and the SSC. As a result, early broad-spectrum antibiotics, source control, and the provision of adequate amounts of intravenous

(IV) fluid are aspects of usual care that transcend protocols.

The ProCESS, ARISE, and ProMISE trials demonstrated a significant in- crease in the use of vasopressors, dobutamine, and blood transfusions, without a reduction in mortality with EGDT. Because of the robust findings in all 3 trials, the SSC has revised the 6-hour bundle. In the event of persistent arterial hypotension (mean arterial pressure b 65 mm Hg) despite volume resuscitation or initial lactate of 4 mmol/L or greater, measurement of central venous pressure and CVO2 saturation are op- tions, along with bedside ultrasound, passive leg raise, and physical ex- amination to reassess a patient’s volume status and tissue perfusion.

Table 4

Risk differences estimates for sepsis trials

b Contraindication to blood transfusion or high likelihood of refusal (eg, Jehovah’s Witness).



Risk difference

Rivers (2001)

60-d mortality

14% (2% to 26%)

ProCESS (2014)

90-d mortality (EGDT vs usual care)

1% (5% to 7%)

ARISE (2014)

90-d mortality

-0.3% (-4% to 3.6%)

ProMISE (2015)

90-d mortality

-0.3% (-5.4% to 4.7%)

c Not able to commence Resuscitation protocol within 1 hour of randomization or complete 6 h of protocol treatment from commencement.

d Diagnoses include acute cerebral vascular event, acute coronary syndrome, acute

pulmonary edema, Status asthmaticus, major cardiac arrhythmia, seizure, drug overdose, injury from burn or trauma, requirement for immediate surgery, and known CD4 count less than 50/mm2.

294 S. Sharif et al. / American Journal of Emergency Medicine 34 (2016) 292294

Repeating lactate measurements remains a principle of ongoing care in the 6-hour bundle, and it should be noted that following serial lactate, levels have been shown to provide equivalent information as measure- ment of CVO2 saturation [16]. Likewise, the recently published Transfu- sion Requirements in Septic Shock trial demonstrated no difference in 90-day mortality in patients with septic shock who received a transfu- sion trigger of 7 g/dL vs 9 g/dL [17]. Of note, peripheral vasopressor infu- sions in proximal IV sites may be safe for short-term use (b 6 hours) [18]. A critical understanding from these trial results, however, is that bundled sepsis care is not to be abandoned completely. All 3 author groups strongly endorse the use of sepsis care paths/algorithms for the ED care of severe sepsis/septic shock patients. These results just confirm that such bundles do not need to be anchored on aggressively invasive, expensive interventions to meet the original EGDT end points. Rather, such bundles should focus on early recognition, lactate screening, broad-spectrum antibiotics, and liberal fluid resuscitation therapy with IV crystalloids. This is critically relevant evidence for all levels of ED practice for septic patients and should be implemented

into clinical practice immediately.

After completion of these trials, the SSC revised its recommenda- tions for Initial resuscitation improvement bundles to exclude invasive components of EGDT and have focused now on noninvasive monitoring measures for ED care of septic shock patients [14].

  1. Conclusion

These 3 trials confirm that an invasive, expensive EGDT-based sepsis care bundle in the ED is not necessary to achieve mortality reduction goals, provided that early recognition, liberal crystalloid resuscitation, broad-spectrum antibiotics and transfer to definitive care are achieved.


The authors would like to thank McMaster University.


  1. Canadian Institute for Health Information. The CIHI Data quality Framework, 2009. Ottawa, Ont.: CIHI; 2009
  2. Rivers E, Nguyen B, Havstad S, Ressler J, Muzzin A, Knoblich B, et al. Early goal- directed therapy in the treatment of severe sepsis and septic shock. NEJM 2001; 345:1368-77.
  3. Jones AE, Brown MD, Trzeciak S, Shapiro NI, Garrett JS, Heffner AC, et al. The effect of a quantitative resuscitation strategy on mortality in patients with sepsis: a meta- analysis. Crit Care Med 2008;36:2734-9.
  4. The ProCESS Investigators. A randomized trial of protocol-based care for early septic shock. NEJM 2014;370:1683-93.
  5. The ARISE Investigators and the ANZICS Clinical Trials Group. Goal-directed resusci- tation for patients with early septic shock. NEJM 2014;371:1496-506. http://dx.doi. org/10.1056/NEJMoa1404380.
  6. Mouncey PR, Osborn TM, Power GS, Harrison DA, Sadique MZ, Grieve RD, et al. Trial of early, goal-directed resuscitation for septic shock. NEJM 2015;372:1301-11.
  7. Green R, Djogovic D, Gray S, Howes D, Brindley PG, Stenstrom R, et al. Canadian Associa- tion of Emergency Physicians Sepsis Guidelines: the optimal management of severe sepsis in Canadian emergency departments. Can J Emerg Med 2008;10(5):443-59.
  8. Dellinger P, Levy MM, Rhodes A, Annane D, Gerlach H, Opal SM, et al. Surviving Sep- sis Campaign: international guidelines for management of severe sepsis and septic shock, 2012. Crit Care Med 2013;41:580-637.
  9. Jones A, Kline J. Use of goal-directed therapy for severe sepsis and septic shock in ac- ademic emergency departments. Crit Care Med 2005;33(8):1888-98.
  10. Francis M, Rich T, Williamson T, Peterson D. Effect of an emergency department sep- sis protocol on time to antibiotics in severe sepsis. Can J Emerg Med 2010;12(4): 303-10.
  11. Sweet D, Jaswal D, Fu W, Bouchard M, Sivapalan P, Rachel J, et al. Effect of an emer- gency department sepsis protocol on the care of septic patients admitted to the intensive care unit. Can J Emerg Med 2010;12(5):414-20.
  12. The ProCESS/ARISE/ProMISE Methodology Writing Committee. Harmonizing inter- national trials of Early goal-directed resuscitation for severe sepsis and septic shock: methodology of ProCESS, ARISE, and ProMISE. Intensive Care Med 2013;39: 1760-75.
  13. Al Deeb M, Azad A, Barbic D. Critically appraising noninferiority randomized con- trolled trials: a primer for emergency physicians. Can J Emerg Med 2015;17(3): 231-6.
  14. Surviving Sepsis Campaign. SSC Six-Hour Bundle Revised. Accessed online July 4, 2015 at

vised.aspx; 2015.

  1. Djogovich D, Green R, Keyes R, Gray S, Stenstrom R, Sweet D, et al. Canadian Associ- ation of Emergency Physicians sepsis treatment checklist: optimizing sepsis care in emergency departments. Can J Emerg Med 2012;14(1):36-9.
  2. Jones AE, Shapiro NI, Trzeciak S, Arnold RC, Claremont HA, Kline JA, et al. Lactate clearance vs Central venous oxygen saturation as goals of early sepsis therapy: a ran- domized clinical trial. JAMA 2010;303(8):739-46.
  3. Holst LB, Haase N, Wetterslev J, Wernerman J, Guttormsen AB, Karlsson S, et al. Lower versus higher hemoglobin threshold for transfusion in septic shock. NEJM 2014;371:1381-91.
  4. Loubani OM, Green RS. A systematic review of extravasation and local tissue injury from administration of vasopressors through peripheral intravenous catheters and central venous catheters. J Crit Care 2015;30(3):653.e9-653.e17.

Leave a Reply

Your email address will not be published. Required fields are marked *