A large multinational study may cool enthusiasm for using hypothermia to treat comatose patients recovering from an out-of-hospital cardiac arrest.

The test of 1,850 adults found cooling the body to 91.4 degrees F (33 degrees C) and keeping it there for 28 hours before gradual warming had no effect on the risk of death from any cause at six months. Nor did it impact functional outcomes half a year after the cardiac arrest.

The findings appear in the New England Journal of Medicine.

International guidelines currently recommend hypothermia as a way to protect the brain once coma has descended on a cardiac-arrest patient.

But the quality of the evidence supporting that strategy has been low and there is a lot of variability among countries, regions and hospital systems in the degree to which they practice hypothermia, senior author Dr. Niklas Nielsen of Lund University in Sweden told Reuters Health by email.

The TTM2 trial was an attempt to show that a lower body temperature was better. It enrolled five times the combined enrollment of all earlier tests and was conducted at 61 hospitals worldwide.

At the six-month mark, the death rate was 50 percent in the hypothermia group versus 48 percent in the group where doctors tried to keep the body at a normal temperature (P=0.37).

The rates of severe disability or worse were identical in each group, at 55 percent.

Dr. Nielsen and his colleagues did find that arrhythmia resulting in hemodynamic compromise was more likely among hypothermia recipients (24 percent vs. 17 percent, P<0.001), but other adverse events were just as likely in the two groups.

Additional analyses failed to find any hypothermia benefit for subgroups, such as those based on age, initial rhythm or time until a return of spontaneous circulation.

An earlier, smaller study from 2013, known as TTM1 had produced similar results.

Because of that and the size of this trial, “it will be difficult to ignore the quite clear-cut results that lowering the body temperature does not confer a benefit for adult cardiac arrest patients,” said Dr. Nielsen, an associate professor of anesthesiology and intensive care at Lund.

Based on the new results, instead of focusing on aggressive cooling, “focus can be on other aspects of care including hemodynamic stabilization, cardiac interventions if indicated, radiological imaging if indicated, etc. Then the patients should be meticulously treated in the critical-care unit according to general intensive-care standards,” said Dr. Nielsen. “It is important to have well-established protocols for prognostication of neurological recovery and not prematurely withdraw care.”

The researchers note that the findings were consistent across all regions and sites, and were based on contemporary standards of care. Earlier studies supporting hypothermia were typically done in the 90s.

To avoid bias, the team also conducted the majority of neurological function outcome assessments face to face with the patient or relatives using assessors who were unaware of the treatment group.

The researchers also “blinded the analysis and writing processes, so in the end we had two manuscripts with the groups interchanged and all authors had to agree on both versions before breaking the randomization code,” said Dr. Nielsen. “With this method we avoided analysis and writing bias which is a real problem in scientific production.”

The authors note that “since we did not include a control group without temperature management, this trial leaves a knowledge gap regarding whether any temperature management is better than no temperature management.”

“There will be many trying to find ways around these results, for example by advocating individualized therapies etc.,” said Dr. Nielsen. “This is all very compelling but (as of) today not supported by any robust evidence. On the contrary, our findings from both this trial and our previous are that the results were consistent in subgroups. So, in summary, I think this trial will change practice.”