Review Of: Bernard SA, Smith K, Cameron P, et al. Induction of Therapeutic Hypothermia by Paramedics after Resuscitation from Out-of-Hospital Ventricular Fibrillation Cardiac Arrest: A Randomized Controlled Trial. Circulation. 2010;122:679–681.
This is a prospective, randomized control trial that tries to determine if a detectable difference can be found between prehospital initiated therapeutic hypothermia and hospital induced therapeutic hypothermia for out-of-hospital-cardiac-arrest (OHCA) that presented with an initial rhythm of ventricular fibrillation (VF). The study was performed in Melbourne, Australia, over a two-year period. Over that time, 234 patients were enrolled in the study; 118 received paramedic cooling, and 116 received hospital cooling. The study was prematurely halted due to no detectable difference, and it was further determined that continued enrollment would not yield any trends or differences. Out of paramedic-cooled patients, 47.5% had a favorable outcome compared with the 52.6% of hospital-cooled patients who had favorable outcomes.
Dr. Wesley Comments
This is a fascinating study and one that every service that has implemented or is considering implementing return of spontaneous circulation (ROSC) hypothermia should read. The researchers controlled for every possible variable and found that while prehospital infusion of chilled saline can and does lower the body temperature and begin the process of hypothermia, it doesn't result in any significant neurological improvement compared to simply waiting until arrival at the hospital to begin the therapy.
This reminds me of the how aspirin came to be so automatic in EMS as a treatment for ST-elevation myocardial infarction (STEMI) patients. Clearly, there is evidence that aspirin benefits the acute coronary syndrome. However, there is no evidence that giving it in the field is superior to giving in the emergency department (ED).
Inducing hypothermia does improve neurological outcomes for cardiac arrest, but is it worth the cost of implementation in EMS? And, do we really have the resources to perform yet another procedure during the care of these most critical of patients?
If your system has hospitals that are already committed to starting hypothermia in the ED, then I don't believe it's worth doing in the field. On the other hand, if you have hospitals that are hit and miss, then starting in the field may encourage them to do it. Of course, a better solution to this situation is simply not to deliver ROSC patients to their facility.
We have many things to do once we get a patient back: Stabilize their pressures, get a 12-lead ECG and ensure they don't arrest again. Perhaps evidence found in the future will prove that starting hypothermia during resuscitation will help, but for now, at least from this paper, it appears there is no harm in waiting until you get to the hospital.
I agree with the doc—this is a fascinating study and one that should definitely be read by every service currently or looking to perform prehospital therapeutic cooling. However, I think before we say, "Hospital cooling is just as effective as prehospital cooling," we need to take heed to what the investigators suggest—initiating therapeutic hypothermia immediately (during CPR) instead of waiting until a patient has been resuscitated. This way we would know whether prehospital cooling really doesn't have any effect on outcomes.
I'm not disagreeing with the study completely, but I believe the limitations of the study need to be examined. The first issue is how temperature was measured in the prehospital setting—by tympanic readings. These can be problematic because they can produce readings that are artificially lower due to improper placement of the device during the reading. So, it's conceivable that the data reported for prehospital temperature is inaccurate.
The second issue is time. The investigators report various time intervals, but I'd like to know when the ice-cold fluid bolus was initiated. How long after ROSC was obtained did they start the infusion? The third limitation of this study, in my opinion, is the fact that it can't be generalized. This study took place in Melbourne, Australia, with a population of approximately 4 million. Comparing EMS in the U.S. to EMS in Australia isn't fair; we operate two completely different systems—not to mention the Melbourne system is a major urban/suburban one, which isn't very applicable to rural EMS.
With any research, there are always going to be limitations. Although the number of them and the impact they ultimately have are for readers to decide. In my opinion, this study is an example of what not to do in future studies of this nature.