Colder Is Not Necessarily Better: CAPITAL CHILL Results

By Jamie Diamond, MD, MPH, Medical Correspondent on

Key Points:

  • Survivors of out of hospital cardiac arrest (OHCA) suffer very poor outcomes including high rates of mortality and neurologic dysfunction, with optimal protocol for therapeutic hypothermia (TH) unknown
  • CAPITAL CHILL is the first double-blinded, randomized trial to evaluate outcomes of moderate hypothermia (31 degrees Celsius) as opposed to mild hypothermia (34 degrees Celsius) after suffering OHCA
  • Trial results showed that patients who were cooled below 31 degrees Celsius for 24 hours (moderate TH) showed no difference in terms of death or poor neurological outcomes at six months compared with patients receiving guideline-recommended cooling of 34 degrees Celsius (mild TH)

Out-of-hospital cardiac arrest (OHCA) is a significant healthcare issue, with an estimated rate of 36–128/100,000 patients worldwide. These patients are extremely high risk for neurological dysfunction and have a mortality rate between 65% and 95%. In the United States alone, 424,000 people utilize emergency medical service assessment for OHCA annually. Therapeutic hypothermia (TH) has previously been shown to decrease mortality and improve neurological outcomes amongst OHCA who achieve return of spontaneous circulation (ROSC). The optimal therapeutic hypothermia (TH) protocol for comatose survivors of out-of-hospital cardiac arrest (OHCA) was previously unknown, though current guidelines recommend targeted temperature management (TTM) of ranging from 32 to 36 degrees Celsius for 24 hours. Prior studies have suggested that moderate TH (28 degrees Celsius and 32 degrees Celsius) could improve clinical outcomes.

In order to further study this issue, the University of Ottawa Heart Institute (UOHI) in Ontario, Canada initiated the Mild Versus Moderate Therapeutic Hypothermia in Out-of-hospital Cardiac Arrest Patients (CAPITALCHILL, NCT02011568) study to determine the effect of various TH protocols on neurologic outcomes at six months ( This trial, led by Michel Le May MD, Director of the University of Ottawa Heart Institute Regional STEMI Program, an interventional cardiologist and Professor of Medicine, is a single-center, randomized, double-blind investigator-initiated prospective clinical trial which recruited comatose OHCA survivors.

The 360 participants were randomized between August 2013 and March 2020 to receive mild hypothermia (TH at 34 degrees Celsius) or moderate hypothermia (TH at 31 degrees Celsius). Patients were included if they had sustained OHCA, were 18 years of age or older, were unconscious at time of presentation (Glasgow Coma score of ≤8) and had presumed cardiac cause for the arrest.  Upon randomization, patients were stratified based on the initial rhythm at the time of the OHCA, i.e. shockable (ventricular fibrillation or ventricular tachycardia) vs non-shockable (asystole or pulseless electrical activity). All patients underwent cooling with an endovascular device with a double-blind protocol so that only nurses were aware of cooling protocol assignments, and there was separate charting of temperatures. Early coronary angiography was also performed. Once the target temperature was achieved, the temperature was maintained for 24 hours before the patient was warmed to normal body temperature at a rate of 0.25 degrees per hour.

For the primary outcome of the proportion of patients experiencing death or a poor neurologic outcome at six months after OHCA, neurologic outcome was assessed according to the Disability Rating Scale (DRS), an ordinal scale evaluating functional dependence ranging from 0 (no disability) to 29 (vegetative), with >5 considered poor neurologic outcome. Secondary outcomes include death, stroke, stent thrombosis, and bleeding. A blinded central committee was responsible for adjudicating on the primary and secondary outcomes.

Study results, announced at the American College of Cardiology meeting Late Breaking Clinical Trial session on May 17th, showed that there was no significant difference at baseline between the mild and moderate hypothermia groups. Furthermore, outcomes of coronary angiography including need for PCI, stenting, intra-aortic balloon pump, and amount of contrast dye used were similar between the two groups. At 180 days, there was no significant difference in the composite of death or poor neurologic outcome between the mild TH (34 degrees Celsius) and moderate TH (31 degrees Celsius) groups (45.4% vs. 48.4%, respectively) (RR 1.07, 95% CI: 0.86-1.33, P=0.56). When breaking down the primary composite into its components, again no significant differences were seen (RR 1.06; 95% CI, 0.83-135; p=0.63 for mortality, RR 1.12; 95% CI, 0.44-2.84; p=0.81 for poor neurologic outcome).

Dr. Le May noted in his presentation that the majority of adverse events occurred early. The researchers performed subgroup analyses to evaluate outcomes by sex, age and other variables but found no significant differences in any subgroups analyzed. The key secondary outcomes of pneumonia, renal replacement therapy, seizure and stroke were also not significantly different between both groups, though there was a trend towards worse outcomes in all parameters for the moderate hypothermia as opposed to mild hypothermia group. Length of stay was noted to be longer in the moderate hypothermia group which was attributed to longer cooling and re-warming times needed from 31 degrees Celsius to normothermia.

A key limitation to the study is the single-center nature of the study, which may limit its generalizability but was helpful in ensuring consistent treatment protocols. Next steps include evaluating other factors such as bystander cardiopulmonary resuscitation, paramedic arrival times, and hospital-level factors in improvement in outcomes following OHCA. Better tools to study brain activity after OHCA would also help determine which patients benefit most from certain TTM protocols and allow physicians to tailor therapies to particular situations.

Leave a Reply