(EBM) ProCESS study





TL;DR:

Routine use of CVL monitoring (CVP and SvO2) does NOT improve survival in the context of recents changes of the management of sepsis, especially with...

  • Earlier recognition of sepsis
  • Earlier administration of antibiotics / fluid resuscitation

A. Background

  • ProCESS = Protocolised Care for Early Septic Shock

  • Published in NEJM 2014; 370:1683-93

Context:

In 2001, Rivers et all reported significantly lower mortality among patients with severe sepsis or septic shock treated according to a 6 hour protocol of early goal-directly therapy (EGDT)

EGDT used CVL to monitor:

  • CVP
  • SvO2 (central venous oxygen saturation)

... which are used to guide therapy

Purpose of study

To find out if all elements of the EGDT are still necessary, given the changes in the management of sepsis

  1. Is protocol-based resuscitation (EGDT or standard therapy) superior to usual care?
  2. Is a protocol that uses CVL monitor to guide Rx superior to a simpler protocol that does not?
    • i.e. Is EGDT approach superior?

B. Method

  • Prospective randomised controlled trial
    • In USA (31 emergency departments)

Study groups / arms

3 groups:

  1. Protocol-based EGDT group

    • EGDT = early goal-direct therapy
    • Follows protocol similar to that used by Rivers et al

  2. Protocol-based standard therapy group

    • Less aggressive
      • CVL only inserted if peripheral access was insufficient
      • Transfusion only if Hb < 75 g/L

  3. Usual-care group

    • The bedside providers directed all care

Inclusion criteria

Must meet all of the following:

  • At least 18 years old
  • Two or more criteria for systemic inflammatory response syndrome (SIRS)
  • Either:
    • Refractory hypotension (SBP < 90 mmHg despite fluid challenge)
    • Serum lactate ≥ 4 mmol/L

Outcome / end-points

Primary end-point = 60-day in-hospital mortality

Secondary outcomes include:

  • Longer-term mortality
    • 90-day mortality
    • 1-year mortality
  • Need for organ support

C. Findings

Numbers

  • Study ran from March 2008 through May 2013
  • 1351 underwent randomisation
    • 10 withdrew consent

Thus, 1341 patients enrolled

  • Protocol-based EGDT = 439
  • Protocol-based standard therapy = 446
  • Usual care = 456

Resuscitation

CVL placed:

  • 93.2% of EGDT group
    • Placement also occurred earlier
  • 56.5% of standard therapy group
  • 57.9% of usual-care group

Volume of IV fluid differs significantly:

  • 2.8L in EGDT group
  • 3.3L in standard therapy group
  • 2.3L in usual-care group

Vasopressors:

  • More patients in the two protocol-based groups than usual-care group received vasopressors (p = 0.003)
    • 54.9% in EGDT group
    • 52.2% in standard therapy group
    • 44.1 in usual-care group

Dobutamine:

  • More patients in EGDT group received dobutamine (p < 0.001)
    • 8.0% in EGDT group
    • 1.1% in standard therapy group
    • 0.9% in usual-care group

PRBC transfusion:

  • More patients in EGDT group received packed RBC transfusion (p = 0.001)
    • 14.4% in EGDT group
    • 8.3% in standard therapy group
    • 7.5% in usual-care group

No significant difference in use of antibiotics, glucocorticoids, and activated protein C.

Outcome:

No significant difference in primary and secondary outcomes:

  • 60-day in-hospital mortality
    • 92 in EGDT group
    • 81 in standard-therapy group
    • 86 in usual-care group
    • No difference among all groups
    • No difference between protocol-based groups and usual care group
  • 90-day mortality
  • 1-year mortality
  • Need for organ support

NB:

  • Acute renal failure requiring renal-replacement therapy is higher in standard therapy group
  • ICU admission rate is higher in EGDT group

D. Other notes

Limitation

  • ProCESS study did NOT address the issue of patients with severe sepsis without septic shock

Comparison with Rivers et all study in 2001

Results of this study differs from Rivers

Possible explanations:

  • Cohort in the Rivers study was different

    • Slightly older
    • Higher rates of pre-existing heart and liver disease
    • Higher initial serum lactate level

  • Adherence in this study is good, but not as good as it was in Rivers study

  • Other aspects of management have changed since Rivers study, e.g.

    • Earlier recognition of sepsis
      • Earlier administration of antibiotics
        ← Possibility the most important factor
      • Earlier volume resuscitation
    • Lower Hb as a threshold for transfusion
      • i.e. Transfusion less likely given
    • Tighter control of BSL
    • Lung-protection strategies
      • i.e. low tidal-volume ventilation

NB:

  • In the ProCESS study,

    • More than 2 L of fluid resuscitation was given to all groups
    • More than 75% of patients received antibiotics prior to randomisation

  • Mortality rate ≈ 20%

    • which is less than predicted (30 to 46%) before study commenced.

Related studies

  • Jones et al, 2010

    • EGDT protocol based on serial measurement of serum lactate was not inferior to an EGDT protocol that used SvO2 monitoring

  • Two related studies that are still ongoing

    • ARISE (Australasian Resuscitation in Sepsis Evaluation) in Australia
    • ProMISe (Protocolised Management in Sepsis) in USA