So here's the paper, and also the supplementary appendix as without it the paper lacks too much background to allow any reasonable appraisal.
In a nutshell three groups were compaired. The first had their therapy guided by the protocol of Rivers. The third were managed as per usual, whatever that usual may be - so my first question is does that count as a control group? Will there be elements of the Rivers 'bundle' in this standard care? Is the 'standard care' described better, worse, or different to local practice?
The second group received 'protocolised standard care' (described in figure S2 of the appendix). Does this management reflect our practice? You will note they relied on two variables to guide decision making, SBP and the shock index. Are there any problems with using SBP (as oppose to CO, MAP etc.) and how valid is the shock index? If one of the hypotheses is that guiding therapy by CVP, MVO2 is not appropriate, is it wise to compare with the shock index? Would someone please find and post a reference looking at the validity or use of the shock index in sepsis?
Does the study sample match our patient group? You will note that <20% had a lactate >2 at randomisation. Only 50ish% required vasopressors. Are these patient less sick than our patient population, or are they appropriately and agressively managed early such that they simply 'get better'?
Was this a trial compairing process, treatment, investigation or monitoring? Does this matter?
The response of the surviving sepsis group is available here. Do you think they are right to not change their guidance as a result of this paper? Clearly Rivers has left a legacy, and whilst criticism of some aspects is worthwhile, advances made in the recognition of sepsis as a medical emergency are not to be underestimated.
Apologies about the lengthy post this time - I could go on as I think this is a fascinating paper. I look forward to reading your thoughts........