Study Calls Into Question Emergency Hypothermia for Children After Cardiac ArrestMay 11, 2015
For adults who have suffered a heart attack, the standard of care is to cool the body from 98 to 91 degrees Fahrenheit, called induced hypothermia. Yet a large-scale, multicenter study co-led by the University of Utah School of Medicine shows that emergency body cooling provides no benefit over actively maintaining normal body temperature in infants or children after cardiac arrest. Dr. J. Michael Dean, professor of pediatrics, describes the research and explains what the surprising results mean and their implications going forward. Learn more here.
Julie: A study calls into question the common medical practice of induced hypothermia. Up next on The Scope. Examining the latest research and telling you about the latest breakthroughs. The Science and Research Show is on The Scope.
I'm talking with Dr. J. Michael Dean, Professor of Pediatrics at the University of Utah. He co-led a study recently published in the New England Journal of Medicine. It calls into question a common medical practice: induced hypothermia. Dr. Dean, first, what is the main takeaway from these findings?
Dr. Dean: The main takeaway from this study is that for children to have a cardiac arrest in the out-of-hospital setting, who arrive in the hospital and have coma, there is no benefit from lowering their body temperature to below normal in comparison to simply preventing fever by making sure that they keep their body temperature normal.
Julie: Explain what induced hypothermia or body cooling is?
Dr. Dean: What we're talking about here is reducing the temperature of the body below normal to help the body recover after an injury such as a cardiac arrest. During a cardiac arrest, the brain doesn't receive as much blood flow as it would normally receive while you and I are awake, for example. And when people give CPR, or cardiopulmonary resuscitation, the blood flow that's generated by that is not normal. So the brain is subjected to low oxygen levels. And after low oxygen levels, a brain can be helped in recovery by making the temperature of the brain lower.
Julie: So you were a co-lead investigator on a large multi-institutional study that studied this procedure in a very specific context, inducing hypothermia in children who have undergone cardiac arrest. Why was it important to do this study?
Dr. Dean: So cardiac arrest in children is not the same as heart attacks in adults, and that's a common misconception. When an adult has a heart attack and they have a cardiac arrest from that, it's because they had a myocardial infarction and that caused the rhythm of their heart to be disturbed suddenly. So adults that have cardiac arrests have a very sudden collapse and they don't have a prolonged period of brain injury.
Children, on the other hand, who have out-of-hospital cardiac arrest, include children who drowned, who got crushed underneath a garage door, or were asphyxiated by putting their head into a cellophane sack by accident. So the child's heart doesn't suddenly stop like an adult would after myocardial infarction. But it takes a while to stop, during which the brain is not receiving adequate oxygenation.
About 15 years ago, a couple of papers were published that indicated that hypothermia, after cardiac arrest in adult patients had a protective effect and improved the survival of that population. Based on those studies, other studies have been done in newborn infants looking at whether hypothermia can improve outcome after having the difficult birth with perinatal ischemia, which is not the same as cardiac arrest, but it is low blood flow. The neonatal studies also showed some potential benefit for this. We decided to study pediatric patients because they are neither adult nor are they newborns, and no one had really looked at hypothermia in this population.
Julie: How was this study performed? Because it was a little bit different than some of the previous studies that have tested induced hypothermia?
Dr. Dean: The important difference between this study and the previous studies is that both groups of children who were in this study had active temperature control with a cooling mattress machine. In our study, we had two groups. One group of children was cooled to a lower body temperature, which was 33 degrees centigrade. And the other group was maintained at an absolutely normal body temperature of just under 37 degrees centigrade. So we did not allow any children to develop fevers, which is in stark contrast with the previous studies that were done in adults.
Julie: And what did you find when you compared these two groups?
Dr. Dean: When we compared these two groups at the end of the study, we found that there was no significant difference between the number of children, or the proportion of children, who had a good neurobehavioral outcome 12 months after the cardiac arrest. Nor was there a difference in mortality between the two groups. So, in summary, lowering the body temperature offered no benefit in comparison with merely maintaining the body temperature at a normal level and preventing fever.
Julie: And it sounds like this might have been a surprising result?
Dr. Dean: It is a surprising result because I think that many people, including physicians and nurses working in critical care, have taken a rather relaxed attitude toward fever. I mean, certainly our grandmothers and great-grandmothers and people hundreds of years ago worried about fevers and thought that fever was the most important thing to treat when somebody was sick. But I think that most of us regard fever, or have regarded fever, as sort of a nuisance symptom for which you tell the patient to, "take some aspirin and call me in the morning."
In the critical care environment, after a cardiac arrest or another brain injury, fever is actually extremely detrimental and we know that this is true. But when we actually go back to intensive care unit records and look at how aggressively was fever prevented in different populations of children in the intensive care unit, it's not very well addressed in the intensive care unit. So if a patient develops a fever and the doctor says to the nurse, "Give the patient some Tylenol," the fever is still going to be there for an hour before the Tylenol takes effect. And in order to really keep the temperature normal here, you have to put these children on a mattress that's got cold water going through it. It's not the same as just taking some aspirin or Tylenol.
Julie: You know, there are differences between kids and adults. There are also differences, as you pointed out, about how they get this cardiac arrest in the first place. Could this explain why induced hypothermia may not be as effective in this population?
Dr. Dean: In the early adult studies, they did not control the temperature of the normal temperature group. And what happened was many of those adult patients became febrile and developed fevers. A recent study that was done in Europe of adult patients, where they also controlled the normal temperature and kept it normal, showed no benefit from hypothermia, which is in contrast with the original adult studies. And that study that came out approximately a year ago had nearly 1,000 patients in it, making it the largest adult study on record concerning hypothermia after cardiac arrest. And our results are the same as the results that they found in that study, i.e. no difference.
Julie: Okay. So your results are consistent with some of these newer studies?
Dr. Dean: That's correct.
Julie: Okay. If you- if you scrutinize some of the numbers in your paper, there are actually some apparent differences. Can you talk about that a little bit?
Dr. Dean: So what we're trying to assess here is are differences between two groups in a study different because of the intervention in the study, or because there are other factors we don't know about that make these children different? The numbers are that 20% of the low temperature group did well, while 12% of the normal temperature group did well. And you can look at those numbers and say, "That seems like a difference of 8%." The p-value, and this is the statistical term that people use for whether something is significant or not, was 0.14. And the p-value that people would usually like to see for saying that something is different is less than .05. So the reader can look at this paper and say, "Well, 20% and 12% seem different." But, the p-value, or the statistical analysis, says to us, "In fact, this is not different because this difference that you are looking at could be by chance."
Julie: Would it be possible to get a different statistical result if you had continued this study and enrolled more patients?
Dr. Dean: It is possible that if you enroll more patients that you are able to see smaller differences between two groups and eventually call them statistically significant. It's also a statistical sin because when you design a study in advance, you are deciding in advance what probability of being wrong do you accept. So when you do a study like this and you fully enroll the subjects to the number of subjects you planned in advance, then you cannot just continue the study because you see a number that is a little bit hopeful.
Julie: That's sort of, yeah, toying with the results a bit.
Dr. Dean: It is toying with the results a lot.
Julie: A lot. Why did you want to bring attention to these results?
Dr. Dean: So it's important for families and parents and clinicians to know that hypothermia is not a magic bullet. And when a hospital says, "We do hypothermia here," that should not indicate to them that that's a higher level of sophistication or a better level of medical care. So I think it's important for people to realize that, based on the only study done to date in this population, there is no benefit from hypothermia.
Julie: Interesting, informative, and all in the name of better health. This is The Scope Health Sciences Radio.