CAPITAL CHILL… on the hypothermia

We recently released an Emergency Medicine Cases Journal Jam episode covering all the available literature on ‘therapeutic’ hypothermia. Unfortunately, JAMA didn’t think to inform us that they had another big RCT in the pipeline, so the “CAPITAL CHILL” trial got left out of our summary. CAPITAL CHILL is somewhat unique, looking to compare two truly hypothermic populations, to see if moderate hypothermia (31°C) is better than mild hypothermia (34°C). Although it isn’t likely to change our conclusions, or anyone’s management, let’s do a quick summary here.

The paper

Le May M, Osborne C, Russo J, et al. Effect of Moderate vs Mild Therapeutic Hypothermia on Mortality and Neurologic Outcomes in Comatose Survivors of Out-of-Hospital Cardiac Arrest: The CAPITAL CHILL Randomized Clinical Trial. JAMA. 2021 Oct 19;326(15):1494-1503. doi: 10.1001/jama.2021.15703. PMID: 34665203

The Methods

This is a single-center, double-blind, randomized trial. 

Patients

Adults with out of hospital cardiac arrest who were alive but comatose at the time of admission to hospital.

Exclusions: known inability to perform activities of daily living, cardiac arrest caused by intracranial bleed, severe coagulopathy with clinical evidence of bleeding, coma not attributable to cardiac arrest, and suspected survival of less than 1 year due to reasons other than the cardiac arrest. 

Intervention

Target temperature of 31°C.

Comparison

Target temperature of 34°C.

Shared Treatment

Other than the target temperature, both groups were treated with the same protocol. Hypothermia was with an endovascular cooling catheter. All patients received immediate cardiac catheterization. Sedation, analgesia, and neuro-muscular blockase with all done by standard protocol. Patients were maintained at their assigned temperature for 24 hours, and then rewarmed by 0.25°C/hour, so that everyone had 48 hours of control temperature. 

Outcome

The primary outcome was a composite of all-cause mortality or poor neurologic outcome at 180 days after randomization.

The Results

They included 381 patients. The mean age was 61 and about 80% were male. Cardiac arrest was witnessed in 85% of patients, and 70% received bystander CPR.  85% of patients had an initial shockable rhythm.

Targeted temperature management was successful, with a clear separation between the groups. 

For the primary outcome of all cause mortality and poor neurologic outcomes, there was no difference between the groups. It occurred in 48.4% of the 31°C group and 45.4% of the 34 °C group (absolute difference 3%, 95% CI -7.2% to 13.2%, p=0.56). All cause mortality was 43.5% vs 41.0%.

The only significant difference among secondary outcomes was a longer ICU length of stay in the 31°C group (10 days versus 7 days, p=0.004). 

My thoughts

This is a well done clinical trial, and it is pretty clear that colder is not better. Seeing as almost everyone is moving away from using even mild hypothermia in favour of controlled normothermia, this information probably doesn’t matter very much.

There was 1 previous trial – the FROST-I trial – that looked at 3 different temperatures for hypothermia (32°C, 33°C, and 34°C), and also found no difference, but it was only a pilot trial with 50 patients in each group. (Lopez-de-Sa 2018)

The outcomes actually look worse across the board in the 31°C group. Nothing is even close to statistically significant, but there is a 3% absolute difference in mortality between the groups, which would clearly be clinically significant if real. I doubt that this trial will be repeated, as we move away from hypothermia altogether. However, I find the difference interesting from a dose-response perspective. There is certainly no indication that more is better when it comes to hypothermia.

All trials have limitations. This is probably the highest quality trial of hypothermia to date, considering that they tried to blind almost everyone (treating physician, patient, and family members.) However, the bedside nurse had to be aware of the target temperature, and that certainly raises the possibility of unblinding, especially if the nurses are not well versed in the reasons for or importance of blinding. (Unlike the other trials, where blinding would have been practically impossible, it seems like it might have been possible to alter the machines so that the display temperature was off by 3 degrees in one of the groups, leaving even the nurses blinded.)

The big problem with this trial is, once again, the lack of a true control group. Comparing 2 unproven therapies against each other doesn’t make a lot of sense. Although I understand that hypothermia became standard practice after the 2 early (but far from definitive) RCTs, this trial may simply be a comparison of placebo against placebo. Before testing different doses, or different protocols, we really should have solid evidence that the treatment is better than placebo. To spend the time and money required for a large scale RCT, and ultimately compare 2 interventions that could both be worse than nothing at all is wasteful (and is especially problematic when you are using invasive methods of temperature control). Basically, any future hypothermia RCTs clearly need to have a true comparison group with no temperature control.

Bottom line

Colder is not better after cardiac arrest. This RCT showed no benefit of moderate over mild hypothermia after cardiac arrest. At this point, it is fairly clear that hypothermia is not beneficial. The real question is whether temperature control matters at all, or whether we have once again overzealously and prematurely spent millions of dollars chasing an unproven therapy based on inadequate evidence.

Other FOAMed

Journal Jam 19 Therapeutic Hypothermia After Cardiac Arrest – Mixed Evidence

TTM2: The big chill on ‘therapeutic’ hypothermia

Hypothermia for cardiac arrest

References

Le May M, Osborne C, Russo J, et al. Effect of Moderate vs Mild Therapeutic Hypothermia on Mortality and Neurologic Outcomes in Comatose Survivors of Out-of-Hospital Cardiac Arrest: The CAPITAL CHILL Randomized Clinical Trial. JAMA. 2021 Oct 19;326(15):1494-1503. doi: 10.1001/jama.2021.15703. PMID: 34665203

Lopez-de-Sa E, Juarez M, Armada E, et al. A multicentre randomized pilot trial on the effectiveness of different levels of cooling in comatose survivors of out-of-hospital cardiac arrest: the FROST-I trial. Intensive Care Med. 2018 Nov;44(11):1807-1815. doi: 10.1007/s00134-018-5256-z. PMID: 30343315