Journal Club - Recent Additions

November - 2013

Previous Next    

Showing Journal 5 of 12


Targeted Temperature Management at 33°C versus 36°C after Cardiac Arrest

Niklas Nielsen, Jorn Wettersley, Tobias Cronberg, et al for the TTM Trial Investigators New Eng J Med, 2013, online first

Comment

This large Swedish therapeutic hypothermia (TH) in OHCA trial challenges our current paradigm of treatment. The authors argue that TH needed another trial despite the 2 previous positive trials, because the findings need to be confirmed or refuted, and the issues around management of...


November



Previous Comments

Very interesting trial, and more for the questions that it raises than those it answers. The biggest problem I see with this trial is that it is significantly underpowered. The authors, based on the 2002 trials, state that they are looking for an absolute risk reduction for death of 11%. Surely all this trial now shows is that there isn't an 11% risk reduction, and that potentially clinically meaningful reductions (such as a 2-3% ARR, like Tranexamic Acid for instance) are possible.
John-25 Nov, 2013 11:56:10 AM

Very interesting trial indeed. Practice changing, may be... Prevention of fever may be the crux to better outcomes. (The 2002 trial had temperatures mostly above 37.5 in the 'control' arm, hence may the worse outcomes in that trial.) Also cooling is not a benign intervention with problems of coagulopathy/ electrolyte imbalance and vasodilation with the subsequent re-warming.
ravibk123-28 Nov, 2013 01:09:29 PM

I've already put this comment on Chris Nickson's excellent blog post (at http://lifeinthefastlane.com/reports-therapeutic-hypothermia-death-greatly-exaggerated/), but the power calculation of a study refers to a theoretical space in which a state of nature, referred to in the calculations, is assumed and the behaviour of the entire potential result set of repeated clinical trials is simulated. It's not something to apply retrospectively to a trial which has been done. Taking this trial alone, the point estimate is for an absolute increase in mortality associated with therapeutic hypothermia, of 2%. Another trial conducted using the same protocol may well have come up with a different answer but we don't have the data for that hypothetical trial. We do have the data for the other large trials of therapeutic hypothermia and it's OK to use those data in making overall decisions. Power is "what might happen if...", results are "what happened when...". Lewis Campbell
LTC-03 Dec, 2013 02:04:26 PM

I'm not entirely sure what you're trying tot say Lewis, but a trial is pointless if it isn't powered to detect a clinically meaningful difference. I suspect that many people thought that the mortality benefit associated with the previous 2 trials was likely to be optimistic. I also believe that many would happily accept a positive mortality benefit of 2-3% as "worth cooling for". So the reality is that this trial doesn't answer the question. Does it?
Anni-09 Dec, 2013 04:15:45 AM