This is Andy Shorr, with the Medscape Pulmonary and Critical Care literature update. Today I want to highlight an article by Nielsen and colleagues in the December 5 issue of the New England Journal of Medicine. This article focused on temperature management following out-of-hospital cardiac arrest.
Current guidelines recommend cooling with a goal temperature of 33°C, particularly in patients who have had a ventricular fibrillation (v-fib)-associated cardiac arrest. Many of the early trials that focused on this topic were limited because they were small, or because they did not address the lack of blinding when assessing neurologic outcome at the follow-up period. In addition to being small, the studies were also very selective in terms of whom they enrolled.
Despite that and on the basis of meta-analyses, international guidelines recommend cooling, particularly in out-of-hospital v-fib cardiac arrest in patients who have return of spontaneous circulation. Cooling is recommended predominantly because of data showing mortality benefit.
Nielsen and colleagues tried to look at this question more seriously. They randomly assigned 939 unconscious adults to cooling with a goal temperature of 33°C vs maintenance of normothermia with a temperature of 36°C. Their effort was intended to address cooling and also to examine whether the issue is cooling or prevention of fever, because we know that fever in these patients is associated with poor neurologic outcome and poor prognosis.
Patients in this trial included those with a v-fib arrest, but about 20% had an event other than that. Mean age was in the mid-60s. They excluded patients who did not have return of spontaneous circulation and patients who had a dismal prognosis initially. Sedation management for the first several hours or the first day and a half of the trial was also part of the protocol.
The investigators found no difference in mortality or neurologic outcome at 180 days between the group that had a goal temperature of 33°C vs those with a goal temperature of 36°C.
In this study, compliance with the cooling protocol was very high, and the patients' temperature curves in the 2 groups separated very early and remained separated for the duration of the intervention period. Thus, they achieved their goal of getting people to 33°C or maintaining normothermia.
Again, at the end, they saw no difference in neurocognitive outcomes or in mortality.
Compelling Outcomes
This study has a number of important strengths that compel us to look at it seriously. First, it helps us to sort out the question of whether good outcomes are related to cooling or to preventing fever. This study clearly confirms that prevention of fever is a good thing. The mortality and outcomes rates in the 36°C arm are very good when compared with historical controls.
Second, this is the largest trial of its kind. Third, these authors also included a protocol for addressing conversations with families about withdrawal of care. Withdrawal of care in both groups occurred at about the same rate; that is important, because imbalances or differential rates of withdrawal of care might affect outcomes.
As one of the editorialists comments, this study also tells us that overall, about 50% of patients who survive an out-of-hospital arrest will survive at 180 days. Many of them will have a neurocognitive impairment, but that is a huge improvement compared with historical controls.
What does this mean for the practicing clinician? First, for the non-v-fib, ventricular tachycardia population, which was about 20% of this cohort, there is certainly no reason to consider cooling. Second, preventing fever is important in this population, and people should have protocols for that.
Third, we have to question the true value of cooling below 36°C. This is important because cooling is not without side effects. We do not cool just the brain, we cool the entire body; rewarming has consequences from coagulation; cooling requires more sedation at times; and sometimes cooling actually prolongs suffering, because there are patients who will do poorly no matter what, and by cooling we just delay the inevitable.
Regardless, this is a very important study that will influence the evolution of guidelines. I urge you to read it.