Early Use of Noradrenaline in Septic Shock resuscitation (CENSER): A Randomized Trial

Chairat Permpikul et al. Am J Respir Crit Care Med Vol 199, Iss 9, pp 1097–1105, May 1, 2019.

Background

• Murine models show increased organ perfusion with increased mean arterial blood pressure
• There are retrospective studies in humans that show improvements in mortality

Aim

Does administering low-dose noradrenaline at the beginning of sepsis-induced hypotension resuscitation accelerate shock control compared to fluid resuscitation followed by noradrenaline (usual practise)?

Design

• Randomised, double-blinded, placebo-controlled trial
• Single centre in Bangkok (tertiary teaching hospital)
• October 2013 – March 2017

Eligibility criteria

• All patients >18 years meeting Surviving Sepsis Campaign 2012 diagnostic criteria for sepsis with a MAP <65mmHg attributable to infection
• Exclusion criteria attempted to exclude other causes of hypotension

Methodology

• Randomisation by computer generated randomization table produced by co-author (sequential 1:1 ratio)
• Randomised to NA in 5% dextrose in water at 0.05mcg/kg/min or placebo (5% Dextrose in water) from onset of resuscitation through a peripheral or central line
• Patient & medical staff were blinded to trial infusion
• All subjects got anti-infective agent, source control and organ support as needed
• Open-label vasopressors were permitted for refractory shock after 30ml/kg of intravenous fluid
• Those needing further organ support went to ICU (for arterial-line monitoring), others went to general medical ward (no arterial-line) with nursing ratio 1:3

Outcomes

Primary: Control of shock at 6 hours

• MAP >65mmHg (on 2 consecutive BP measurements 15 mins apart)
  • PLUS, evidence of organ perfusion: either UO >0.5ml/kg/hr for 2 consecutive hours OR >10% reduction in lactate from baseline

Secondary: 28-day mortality, hhospital mortality and rate of resp failure requiring mechanical ventilation, RRT, days requiring organ support up to Day 28

Safety: New onset arrhythmias, organ ischaemia, pulmonary oedema (till hospital d/c or death) and death

Results

• Well matched baseline characteristics
• Primary outcome: Treatment with noradrenaline from the start resulted a greater number of patients achieving MAP target of >65mmHg plus evidence of organ perfusion by 6 hours. 118/155 patients in treatment group compared to 75/155 patients in placebo group, OR 3.4 (2.09-5.53 95%CI), p<0.001.
• Secondary outcomes: there were no significant difference between the groups in need for organ support, ICU admissions, length of ICU stay, length of hospital stay, 28-day mortality or hospital mortality
• Safety outcomes: Treatment with noradrenaline compared to placebo group resulted in a lower incidence of cardiogenic pulmonary oedema (22/155 patients vs 43/155 patients. RR 0.70 (95% CI 0.56-0.87), p=0.004. No significant difference in ARDS, new onset cardiac arrhythmias, hospital acquired infection, upper gastro-intestinal haemorrhage, acute limb/intestinal ischaemia, or skin necrosis.
• Ther was no significant difference in volume of fluid administered or use of open label vasopressors.

Limitations

• Relatively small number in each group (n = 150)
• Although blinded, clinicians may have seen the effects of intervention group which would have added bias
• Half of intervention group went to general medical ward (nursed 1:3!) on NA 
• Fluid resuscitation rate was not controlled (but should have equalled out between groups)
• Single centre trial in Bangkok– and may not be generalisable
• No cardiac output monitoring details
• Cannot make conclusions about secondary outcomes given lack of power to assess these

Discussion

• Expectedly, early use of vasopressors results in faster resolution of shock! Early NA didn’t result in less fluid given?!
• Improves macro and micro circulation but with no significant improvement in organ function seen in this study
• Interesting that they infused low dose noradrenaline through central (43%) or peripheral access (57%)
• One case of skin necrosis in both groups and no difference in evidence of splanchnic hypoperfusion
• Fitting with local practise of starting low dose peripheral metaraminol infusions… metaraminol/phenylephrine vs NA in sepsis debate
• Less cardiogenic shock (= fluid, ? due ↑CO) & fewer arrhythmias in NA group?
• Safety data on peripheral NA – sample size too small to see issues here?
• Unlikely to change practise because of the safety concerns of peripheral noradrenaline

Summary by Dr Jason Van Schoor.