Cite this post as:
Scott Weingart, MD FCCM. EMCrit Wee – The Targeted Temperature Trial Changes Everything. EMCrit Blog. Published on November 18, 2013. Accessed on January 20th 2025. Available at [https://emcrit.org/emcrit/emcrit-wee-targeted-temperature-trial-changes-everything/ ].
Financial Disclosures:
The course director, Dr. Scott D. Weingart MD FCCM, reports no relevant financial relationships with ineligible companies. This episode’s speaker(s) report no relevant financial relationships with ineligible companies unless listed above.
CME Review
Original Release: November 18, 2013
Date of Most Recent Review: Jul 1, 2024
Termination Date: Jul 1, 2027
You finished the 'cast,
Now Join EMCrit!
As a member, you can...
- Get CME hours
- Get the On Deeper Reflection Podcast
- Support the show
- Write it off on your taxes or get reimbursed by your department
.
Get the EMCrit Newsletter
If you enjoyed this post, you will almost certainly enjoy our others. Subscribe to our email list to keep informed on all of the Resuscitation and Critical Care goodness.
This Post was by the EMCrit Crew, published 11 years ago. We never spam; we hate spammers! Spammers probably work for the Joint Commission.
NOTE: These are more questions for my own learning (I’m new to this whole EBM thing). Take any of these with a pound of salt. Why are the p-values in the NEJM study so high for most of their findings? The studies were both well done with regards to methodology. The NEJM study had a very good bystander CPR rate (median 1 min for both groups). What was the rate of bystander CPR in the original studies? Another question – what were the vent settings used/how was oxygenation managed in these patients? What were the PaO2 values at each time-point… Read more »
There is an excellent power analysis on the stemlyns blog. This study was powered to find an 11% difference between the 33, and 36 degree group. For a stat. sig. 2% difference (NNT 50 (i.e. ASA in MI, TSA in trauma)) sample size would have needed to be 20,000.
http://stemlynsblog.org/whats-target-temperature-oohca-cooling-st-emlyns/
This is holding trials such as this up to a much different standard than any of the other clinical trials out there that compare a new therapy to standard care. To say a 1000 patient hypothermia study doesn’t do much is probably to set one’s expectations way too high for future research.
Scott, I’m not disagreeing with you that an adequately powered study would be difficult to do here but I don’t think that takes away from Simon’s statistical analysis. It’s pretty easy to let studies like this lead us in directions that the data doesn’t support. We have prior data that did find significant difference between TH and nothing. Now we have a study that didn’t find significant difference between 33 and 36C, but was only powered to find a pretty huge difference. So we can trust this study to say there is <11% mortality difference between these 2 groups, but… Read more »
Agree with every word of that
Also, it is interesting to note how St. Emlyn’s mentions CRASH 2. This trial did manage to enroll a cohort of ~20 000 pts.
Running a RCT of that scale for TTM at 33 vs 36 would be possible and has been done for RCT’s of other interventions – one of the the only problems being that the logistics would be (VERY) nasty.
Any future RCTs will have this problems as advancement in medical care kicks in. As mortality drops, we need bigger sample size. The solution is either change the primary outcome (ie MOPETT trial) or bundling it in al together (ie EGDT). However even if we were unable to find 20 000 pt to validate the 2% differences, it is too small and the benefit and conveniences of 36 C is too great to pass. For us who has not even started TTM extensively due to practicality/cost/logistic issues, this study is a gift from heaven.
Roslan
Emergency Physician
Sarawak, Malaysia
Will this change our practice much? Probably not. Here’s what we will most likely be taking out of it, based on all that’s come through from the AHA meeting and other publications: We will continue to not warm in the ED, and will still be inserting a CVC cooling catheter in these patients once we get them into ICU. We may well remove the “2 litres of 4 degree saline IV” section from our protocol in the ED also; as there is now prehospital evidence of harm/no benefit which might be transferable. Jury is out. The target temp will alter… Read more »
John,
Will this change practice?
If you set the goal on your cooling caths to 35 or 36 now, that is an ENORMOUS change.
Hi Scott, I suspect it will change our practice to targeting 34 – 36 degrees; though we are still thrashing it out looking at the overall balance of evidence The catheters we use are very, very good – when you dial in a temp, that’s exactly what you get. Our real term audited figures are bang-on what we set the machine for. As such, I’d be happy aiming for 34-36 degrees and not expect to have any unanticipated overshoots to hyperthermia. Indeed, we currently use them for the controlled rewarming phase in our patients at the moment, and never have… Read more »
John, Look at the Day 3 SOFA-C scores in the table I added above to see why that dial change on the machine may have big implications.
Thanks Scott,
SOFA C is just a surrogate marker of inotrope use though.
We already know that hypothermia reduces inotrope requirements in cardiogenic shock, so would you not just chalk up the SOFA C variance as an observational vagrancy of the intervention itself?
The inotrope requirement did not bear out a mortality difference, so is it not little more than a “number on the bedside pump?”
-John
John,
I misread the table that same way the first time (b/c I expected hypothermia group to need less pressors). It was exactly the opposite–36 got off pressors more frequently than 33
Oops – my bad!
A perfect example of Confirmation Bias in action for the pair of us!!
What’s the chat in your place Scott – is a protocol change likely?
-John
yes. I think we will convert all v-fib and PEA to 36
Hi Scott,
The original studies looked at shockable rhythms, this study at all rhythms. How does this study provide us evidence to change our hypothermia protocols for the shockable rhythms?
Kind regards,
René
this study looked at shockable + PEA, very few PEA. I would look at it as directly comparable to the pt pop. of the previous trials.
hey EMCrit family would like to comment on the target population of this study. The study excluded patients who are still in arrest at the time of arrival in the ED. Furthermore, a large percentage of the patients in the study (80%) had shockable rhythms, and 73% had bystander CPR, with the median time to initiation of BLS being 1minute. Though median time to ROSC was 25min, it strikes me that the patients included in the study may have substantially high degrees of brain perfusion during their arrest events as supported by moderate acidemia (pH 7.2 +/- 0.2) and moderate… Read more »
i’m interested in the overlap between therapeutic hypothermia for post-arrest vs. traumatic TBI and it’s interesting to note this paper from 2009: http://www.ncbi.nlm.nih.gov/pubmed/19131820?dopt=Abstract
Great podcast Scott, and key to point out as you did that 36 isn’t the old “standard of care.” I don’t think we’ve yet seen the last of hypothermia, and maybe really deep is better… Have you ever seen the vids of Peter Safar’s re-animated dogs?
I ramble a little more about it here:
http://thinkingcriticalcare.com/2013/11/29/is-it-still-cool-to-cool-in-cardiac-arrest-the-new-ttm-hypothermia-trial-foamed-foamcc/
thanks!
Phil
Yep! Some of my buds are talking about a multi-center RCT of 36 vs 31.
Hi all – I’m presenting this paper at a meeting today so I’m enjoying reading everyone’s insightful comments! Phil – I think the Safar paper that you mention is interesting but it shows only that hypothermia can prevent primary neurological injury as the dogs got cooled 2 minutes into their arrests. Hypothermia to prevent anticipated primary neurological injury is well established, e.g. in cardiac surgery. In an unexpected arrest that would require intra-arrest cooling. The cooling that Nielsen et. al performed was some time post arrest (median 25 minutes to ROSC, cooling started < 4hrs post ROSC) and can only… Read more »