Welcome back to journal club, this month we are going to continue our theme of celebrating those who dare to question received wisdom….
Back at the turn of the millennium a hugely influential paper was published Rivers et al (http://www.nejm.org/doi/full/10.1056/NEJMoa010307) that built on earlier work that suggested targeting “supranormal” physiological values improved outcomes in the critically ill……
The subsequent years have eroded clinical confidence in this mantra as each individual component of the “EGDT” algorithm has been shown to either be detrimental or ineffective (ie aggressive fluid resuscitation, multiple transfusions, dobutamine, targeting CVP’s….).
Finally equipoise was reached once more to allow several trials to retest the hypothesis that EGDT improved outcomes. The ARISE trial(http://www.nejm.org/doi/full/10.1056/NEJMoa1404380) did just that…..
Please find Dr Rick Leggett’s excellent summary below. Have a read of the papers and his summary and please feel free to add your comments, I’ll be chipping in later with my thoughts too!
“Goal-Directed Resuscitation for Patients with Early Septic Shock – the ARISE study
We all know the background, sepsis is bad, sepsis kills, the golden hour, sepsis six, bufalo etc… Traditional teaching in sepsis has been fluids, fluids, fluids, antibiotics and more fluids if they are still responsive. This comes partially from a chap called Rivers, who in 2001, published his study in the New England Journal of Medicine suggesting patients managed with early ‘goal-directed therapy’ (EGDT) had better outcomes. By goal directed therapy, they mean the EGDT group had mixed venous saturations continuously measured (hint: how often do we see this on the unit?!), alongside MAP, UO and CVP. Fluid to maintain CVP of 8-12mmHg were the mainstay of treatment before starting vasoactive agents.
Not all of the outcomes reached statistical significance, but non-the-less, the world celebrated and a new era EGDT was born. The surviving sepsis campaign released guidelines advocating a CVP of greater than 8 and maintaining mixed venous sats greater than 70% as part of their 6 hour care bundle. Problem solved…
So what happened?
People started to become critical of EGDT, they questioned the validity of the original trial design (it was a single-centre trial with potential for significant author bias), felt that patients were over-transfused, excess resources were being used and that it is difficult to make a ‘one-size fits all’ approach. The Australasian Resuscitation in Sepsis Evaluation ‘ARISE’ trial was born.
What did they do?
They recruited 51 centres to study EGDT, mainly in New Zealand and Australia, but also in Hong Kong, Finland and the Republic of Ireland. 1600 patients were deemed necessary to power (so, the primary did teach me something relevant!) the study appropriately to detect any differences between EGDT and usual therapy.
They then took patients who fulfilled the criteria of SIRS with refractory hypotension (MAP <65, SBP <90) or hypoperfusion (lactate >4) after treatment with a litre of crystalloid and randomised them into EGDT or usual care at a ratio of 1:1. They remained in this group for the initial 6 hours and then got the same care. The only requirement was that all patients got abx before starting the trial. The EGDT group was managed by the trial team but interestingly the group assigned to conventional therapy had an ‘anything-goes’ approach; they could be managed on ICU or in the ED, with lines and treatments given according to the treating clinician.
The primary outcome was 90-day mortality but they had all the usual secondary outcomes – duration of ventilation, ICU stay, vasopressors etc. They also accounted for variabilities in age, APACHE II scores etc in their statistical analysis.
There was no difference in the primary outcome of 90-day mortality between the two groups (EGDT 18.6% and usual therapy 18.8%). The only significant difference found between the two was an increased use of vasopressors in the EGDT group (76.3% vs 65.8%) – however this did not affect the outcome. All other parameters were statistically the same.
Are these results valid though?
Patients randomly recruited into each arm were of similar average age, similar percentage of each sex, illness severity, had similar presenting vital signs and received similar volumes of pre-trial fluids; thus adding validity to any results found.
Adherence to the EGDT was good, with approximately 90% of goals reached in all areas.
They removed analytical bias by planning the amount of patients needed and the type of statistical analysis they would use before obtaining the results.
On a negative note, due to the nature of the study blinding was not possible, but selection was randomised to prevent bias. The authors admit that some of the clinicians in the study may have been influenced by their practice in the EGDT group when treating the usual therapy group.
Two other trials of a similar nature (ProCESS and ProMISe) were conducted at the same time as ARISE, but in the USA – they also found no improvement with EGDT.
The authors did not find EGDT beneficial compared with usual treatment. There is therefore no significant evidence for CVP and mixed venous sats monitoring in the treatment of sepsis, therefore the surviving sepsis campaign guidelines have been updated accordingly…”