CENSER: Early use of Norepinephrine in Septic Shock Resuscitation

Permpikul C, AJRCCM, 2019. Published online February 1. doi:10.1164/rccm.201806-1034OC

Clinical Question

  • In adult patients presenting to the emergency department with septic shock, does early low-dose norepinephrine compared with standard care increase shock control at six hours?


  • The optimal timing for introduction of vasopressor therapy in patients with septic shock is not known. Traditionally, patients with sepsis and hypotension are given fluids first and the introduction of vasopressors follows in an ad hoc fashion depending on response to the fluid therapy
  • The Surviving Sepsis Campaign Bundle, updated in 2018, suggests rapid initial resuscitation with the aim to achieve the following in the first hour
    • measure lactate (repeat if initial elevated >2mmol/L)
    • obtain blood cultures before antibiotics
    • administer broad-spectrum antibiotics
    • begin rapid administration of 30ml/kg of crystalloid if hypotensive or lactate >4mmol/L
    • Apply vasopressors if hypotensive during or after fluid resuscitation to maintain MAP >65mmHg
  • Overall, there are elements of this bundle that are controversial and based on weak levels of evidence
  • There has not been any large prospective studies in humans to define the optimal timing of introduction of vasopressor support, ie before fluid resuscitation vs during fluid resuscitation vs following a trial of fluid resuscitation
  • A pilot trial examining a regimen of restricted fluids and early vasopressor compared to usual care for initial resuscitation of hypotension due to suspected sepsis has been completed and published (REFRESH trial) https://www.ncbi.nlm.nih.gov/pubmed/30382308


  • Single centre, double-blind, RCT
  • 1:1 assignment to groups
  • Allocation concealment achieved by a computer-generated randomisation table
  • Blinding; the study drug, noradrenaline, and the placebo were packaged in identical containers
  • The outcome assessors were unaware of the group allocation
  • A sample size of 300 would provide 80% power to detect a predicted rate of shock resolution of 80% in the norepinephrine group vs 60% in the placebo group
  • Analysis was by intention to treat


  • The trial was conducted at Siriraj Hospital, a large university-based, tertiary referral hospital in Thailand
  • Recruitment occurred between October 2013 and March 2017


  • Inclusion: patients >18 years, presenting to the emergency department with MAP <65mmHg with infection as the suspected cause
  • Exclusion: septic shock >1 hour, acute CVA, ACS, APO, status asthmatics, active cardiac arrhythmia, acute GIH, pregnancy, burns, advanced cancer, refusing medical therapies
  • 456 patients with MAP<65mmHg were screened. 320 of these met the inclusion criteria. Seven patients in the intervention arm and 3 in the control arm withdrew consent to participate. This left 310 patients, allocated to 2 groups with 155 patients per group
  • Baseline demographics were well matched, including age, source of sepsis, co-morbidities and APACHE II scores


  • 4mg of norepinephrine was added to 250ml of 5% Dextrose and administered through a peripheral line or central line at 0.05 ug/kg/min (for a 70kg person this equals 13ml/hr, or 3.5 mcg/min)
    • The drug was infused for 24 hours without titration


  • 250ml of 5% dextrose run peripherally or centrally at an identical rate without titration

Management common to both groups

  • All other treatments were as per the SSC guideline of 2012, that is, crystalloid bolus, antibiotics, source control and organ support
  • If MAP>65mmHg not achieved after 30ml/hr of crystalloid then open label vasopressors were permitted
  • Patients were managed in the ICU OR the ward, with a central line OR peripheral line, with an arterial line OR with 15 minutely BP checks
  • Patients were admitted to the ICU if they needed mechanical ventilation, renal replacement therapy or invasive haemodynamic monitoring


  • Primary outcome: shock control by 6 hours. This was defined as sustained MAP>65mmHg (>15 minutes) AND 2 consecutive hours of urine output >0.5ml/kg/hr OR decrease in serum lactate >10% from the initial lactate level.
    • Early norepinephrine group vs the control group: 76.1% vs 48.4% achieved shock resolution at 6 hours (OR 3.4, 95% CI 2.09-5.53, p<0.001)
  • Secondary outcome: Early norepinephrine group vs control group
    • Mortality at Day 28: 15.5% vs 21.9% (p=0.15)
    • Shock resolution in hours (median): 4:45hrs vs 6:02hrs (p=<0.001)
    • Intravenous fluid before open label norepinephrine (median): 2080ml vs 1900ml (p=0.32)
    • Need for open label norepinephrine: 67.7% vs 80% (p=0.014)
    • Time from diagnosis to any norepinephrine (median hours): 1:10 vs 2:47 (p<0.001)
    • Admission to ICU: 54.8% vs 51.6% (p=0.57)
    • Mechanical ventilation: 37.4% vs 38.1% (p=0.91)
    • Renal Replacement Therapy: 12.3% vs 14.8% (p = 0.51)
    • Pulmonary Oedema: 14.4% vs 27.7% (p=0.004)
    • New-onset arrhythmia: 11% vs 20% (p=0.03)
    • Skin necrosis: 0.6% vs 0.6%

Authors’ Conclusions

  • Early norepinephrine was associated with increased shock control at 6 hours


  • This phase II clinical trial proposes an important clinical question that will add to the literature regarding optimal timing of vasopressor use in septic shock
  • It is the first prospective trial examining this specific component of the surviving sepsis bundle.
  • Allocation concealment, blinding, intention to treat analysis, complete follow-up achieved
  • Lactate clearance has previously been found to be a useful surrogate for mortality prediction in patients with septic shock


  • Single centre trial in a developing nation with limited ICU resources – reduces the external validity of the trial
  • Clinicians may be able to guess the allocation of the patient group as they may see an increment in the blood pressure, thus unblinding the trial. The authors point out that around 20% of the control group also had an increment in blood pressure
  • The primary outcome was not clinically meaningful.  Secondary outcomes were clinically meaningful but the trial was not powered to detect differences in these and thus these are hypothesis generating outcomes which may guide design of future trials in this area
  • Of interest, patients received similar median volumes of resuscitative fluids. Early use of vasopressors in septic shock has been promoted as a way to reduce fluid volumes administered but this was not seen here

The Bottom Line

  • Overall the results seem encouraging and would certainly provide a basis to conduct a larger, multi-centre trial to explore the important question of timing of initiation of vasopressor therapies in septic shock
  • Low-dose, dilute norepinephrine was given safely through a peripheral line in more than half the patients,  without adverse effect. This is reassuring that this practice is safe in time-poor or resource-poor environments
  • This is a well-conducted trial that really adds to the literature base for management of septic shock

External Links


Summary author: Celia Bradford
Summary date: February 14th 2019
Peer-review editor: Fraser Magee


  • Rob

    Thanks for a great summary of an interesting article. There are a few points that I think should be drawn out further.
    1) recruitment 2013 – 2017. Why 4 years? Sepsis with shock is far far more common than this perhaps this was due to:
    2) exclusion if septic shock > 1hour. This is odd. How was shock defined, why 1 hour, and how often do we see the first hour of shock anyway? This was a developing nation study – transport times must have been long. Not a real-world selection I would suggest.
    3) non-titrated norad infusion is a useful research tool but again not real-world. Most guidelines would suggest achieve MAP>65, and keep it there for as long as the patient requires. Then it can be weaned as the shock resolves in the conventional sense.
    3) As mentioned in the appraisal, the primary outcome of shock control is also not real-world or clinically meaningful. Too early and potentially transitory. How many patients had a MAP or urine output drop after the outcome had been met?
    So – Interesting hypothesis-generating study, which will inform further work, such as ARISE-FLUIDS, shortly to begin recruitment in Australasia (early vasopressor and reduced fluids vs standard care in septic shock patients presenting to ED).
    Thanks for the opportunity to comment.


    What i dont get in this study is the definition of early NE. Does it mean they started NE even when the MAP was > 65 mmHg? I am just thinking fluids are given early and if after about 2000 ml or so of fluid and the shock is still refractory we start vasopressors to target MAP > 65. But early NE was given when? to be defined as early.

Leave a Reply

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

This site uses Akismet to reduce spam. Learn how your comment data is processed.