Authors: Mark Winstanley, Charlotte Davies, David McCreary, Andy Neill, Chris Connolly, Becky Maxwell, Nikki Abela, Dan Lumsden / Codes: / Published: 05/07/2021

Clinical Question

But no really, does hypothermia improve outcomes in post ROSC OOHCA patients at 6 months when compared to just avoiding fever?

Title of Paper

Hypothermia versus Normothermia after Out-of-Hospital Cardiac Arrest (The TTM2 Trial)

Journal and Year

NEJM. 2021.

Lead Author



– Therapeutic hypothermia post cardiac arrest has been around for a while (depending on which centre you work in)
– Some early studies in dogs showed promising outcomes which lead to preliminary human trials ([such as this one from Bernard]( published almost 20 years ago
– They showed some significant improvement in good outcomes
– This was further tested in the [TTM1 trial]( (33 degrees vs 36 degrees) showing no difference in outcomes
– This finally permitted the research group to ask the question the wanted to all along – how does hypothermia compare to just avoiding fever?

Study Design

– International, multicentred, open-label, outcome assessor-blinded, parallel group, randomised superiority trial

Patients Studied

– Adults post out-of-hospital cardiac arrest
– Presumed cardiac or unknown cause, irrespective of initial rhythm
– Unconscious – not able to obey verbal commands and no verbal response to pain
– ROSC for >20 minutes


– ROSC to screening >3 hours
– Unwitnessed arrest with asystole initial rhythm
– Limitations of care


– Cooled to 33 degrees for 28 hours
– Then rewarmed to 37 degrees by 1/3 degree per hour
– 95% of patients received cooling


– Maintenance of <37.5oC with active cooling measures if ≥37.8
– 46% of patients needed a device for cooling


Primary: any cause mortality at 6 months
Secondary: poor functional outcome at 6 months (MRS 4-6) per structured questionnaire
– Or binary “Good” (MRS 0-3) vs “Poor” (MRS 4-6) based on available data if structured questionnaire couldn’t be performed (thanks, COVID)

Summary of Results

– Powered to detect 15% relative mortality reduction (ARR 7.5%) – 1862 pts
– 1900 enrolled → ITT population 1861

Primary Outcome (death at 6 months): No difference

– Hypothermia: 465/925 (50%)
– Normothermia: 446/925 (48%)
– RR w/ hypothermia 1.04 [0.94-1.14]

Secondary Outcome (functional status): No difference

– Hypothermia: 488/881 (55%) had MRS 4-6 | 54% had “poor”
– Normothermia: 479/866 (55%) had MRS 4-6 | 54% had “poor”
– RR 1.0 [0.92-1.09] for MRS 4-6 | RR 1.0 [0.91-1.09] for “poor”

Adverse Events

– Arrhythmia more common in hypothermia: 24% vs 17% (p<0.001)

Authors Conclusion

In patients with coma after out-of-hospital cardiac arrest, targeted hypothermia did not lead to a lower incidence of death by 6 months than targeted normothermia.

Clinical Bottom Line

I’m keeping it simple here…it looks like there is currently no benefit in cooling patients post cardiac arrest. Remember though, you still don’t want them getting fevers – so just because you’ll hear people calling “cooling isn’t a thing any more” – doesn’t mean you can take your eye off the ball. With bed pressures as they are everywhere, post arrest patients can be in your ED for some time, certainly long enough for the pyrexia to hit so keep an eye on it. Remember 46% of the normothermic patients needed active cooling measures.

There are oh so many more clever opinions than mine out there (see below for a few) with people discussing the nitty gritties of the data available in this trial and I’m sure the authors will be publishing lots of secondary analyses from all that lovely data.

Steve Bernard (off of the original study) has said himself there is now no evidence to support the use of therapeutic hypothermia for post cardiac arrest patients outside of clinical trials. But he does think that they still need to be looking at the benefits of earlier cooling as time to cooling was (I still reckon a pretty impressive) around 3 hours to a temp of 34, for example.

Other #FOAMed Resources / References:

The Bottom Line have their usual high-quality summary

Critical Care Reviews really put on a show with the live presentation of the trial results, including an editorial with Steve Bernard and a panel discussion that is worth a watch


Dr Charlotte Davies


B12 deficiency, cobalamin deficiency or pernicious anaemia is also known as Biermer’s anaemia, Addison’s or Addison-Biermer anaemia. It’s one of the megaloblastic anaemias.


Normally, B12 is absorbed through the ileum with the help of intrinsic factor which is secreted by the parietal cells.
Various things can stop this happening – antibiodies to parietal cells, like in atrophic gastritis, antibodies to intrinsic factor like in pernicious anaemia, lack of a terminal ileum, or disease of the ileum – like Crohn’s disease, and reduced b12 intake.
It takes a while for the disease to become apparent as there’s 3-5 years worth of reserves in the liver.
Lack of B12 means that DNA replication is slow, so cells divide less but grow bigger. This gives you your macrocytic or megaloblastic anaemia.
Lack of B12 means the conversion of homocysteine to methionine is reduced, so there are high levels of homocysteine – causing things like atherosclerosis, thromboembolism and osteoporosis.
More importantly, B12 is a cofactor in the conversion of methylmalonic acid into succinyl CoA. Without this, the dorsal and lateral spinal columns are demyelinated, through an unknown mechanism, so you get neurological symptoms.


FBC will show macrocytic anaemia, and maybe neutropaenia and thrombocytopenia.

A blood film will show anisocytosis and poikilocytosis.
Bilirubin may be increased because of haemolysis.
Then you can look for autoantibiodies, and maybe an absorption test like the Schilling test.


Symmetrical, legs > arms
Loss of position sense
Loss of vibration sense

Low grade pyrexia, weight loss, diarrhoea, jaundice due to haemolysis, pallor due to anaemia, premature greying of the hair.


If B12 levels are low, then patients need intramuscular B12 and a haematology referral. If there is neurological involvement, 5 – 6 loading doses of 1000mcg, followed by maintenance of 1000mcg every three months.
If B12 levels are borderline, then oral B12 may have a response and be diagnostic.
Patients often feel better within 24hours of starting treatment.
If patients are asymptomatic, there is debate about their treatment- monitoring seems preferred.

B12 is naturally found in animal products – fish, meet, eggs, milk. It’s not generally present in plant foods, but is in fortified cereals. It’s worth mentioning that pabrinex doesn’t contain B12!



Dr Chris Connolly
Dr Becky Maxwell


Dr Dan Lumsden
Dr Nikki Abela

This was recorded at PEMFest 2018 in Birmingham


Andy Neill
Dave McCreary

Clinical Question

should we be giving high flow O2 to kids with pneunonia?

Paper Title

Randomised controlled trial of oxygen therapy and high‑flow nasal therapy in African children with pneumonia


Maitland, Intensive Care Medicine, 2021


Maitland has been doing great research for years and is willing to quesiton some of the sacred cows of critical illness such as do fluids work for sepsis with FEAST trial. These types of questions aren’t really an issue for us because these therapies are easily available and seem obvious. In a limited resource setting expecting people toprovide such therapies might be a bit of a stretch. If it turned out they weren’t helpful then it might make health care delivery easier and more accessible

– so now we have the COAST trial where they question if oxygen is really that helpful in pneunonia


– kids 1 month to 12 years with pneumonia and hypoxia
– there are 2 streams here
1) if sats <80 then randomised high flow v standard flow
2) if sats 80-91 then could go into high flow, low flow or simply target sats 80-89
– 48 hr and 28 day mortality as the outcome
– orginally powered or 4200


– stopped early as people in uganda complained that it was unethical
– 1800 enrolled at this point
– 400 in the severe hypoxaemia stratum
– in severe hypoxaemia (high flow v low flow) high flow was a little better 9% mortality v 13%
– in the 3 arm, sats 80-90 stratum the highet mortality (2.5% was in the low flow group with permissive hypoxia and high flow doing better
– these are all single figure or low double digit number of kids so understandably no significance from a stat perspective


– interestingly they found that using high flow with room air actually improved hypoxaemia (reinforcing all the non FiO2 gains of high flow)
– the high flow is the most likely to be actually beneficial but would need further testing
– hardly a practice changer but an insight into how hard it is to establish equipoise

Worth listening to

Critical Care Reviews Podcast