If your you grind your teeth when you have resuscitated a patient post cardiac arrest in fear of having to initiate a cumbersome protocol with hypothermia, the TTM trial may be just what you were looking for. Up to this point only 2 randomized trials have been performed comparing “usual care” to hypothermia, totaling about 350 patients. These 2 studies have changed the way we approach post-cardiac arrest resuscitation, but many questions were left unanswered.
Why is this topic important?
All emergency physicians will face caring for patients after cardiac arrest, and few interventions have been shown to improve neurologic survival. Therapeutic hypothermia has demonstrated in two RCT to improve neurologically intact survival with a number needed to treat of six. However, despite the dramatic effects seen with treatment many unanswered questions remained. How cold, how fast, how long and how to rewarm these patients? The TTM attempted to answer one small piece of this puzzle.
What does this study attempt to show?
The study hypothesis was that targeting “relative” normothermia at 36 degrees and avoiding febrile episodes would be non-inferior to hypothermia at 33 degrees Celsius. The authors believed most of the benefit from trials comparing hypothermia to control was derived from avoiding hyperthermia and not necessarily from the neuroprotective effects of hypothermia.
Patients aged 18 years and older who had a GCS < 8, with out of hospital cardiac arrest (OOHCA) irrespective of initial rhythm and had maintained pulses > 20 consecutive minutes were eligible.
The primary outcome of the study was all-cause mortality at the end of the trial, with a predefined secondary outcome of poor neurologic function or death defined as a CPC >2 (basically unable to function independently). I think this is a great primary end-point mostly because it is binary; as compared with CPC scores which we know from the stroke studies that the mRS can be highly subjective.
What are the essential findings?
Between January 2010 and January 2013 950 patients were enrolled, with 476 assigned to the 33 degrees Celsius and 473 to the 36 degrees Celsius. The groups had similar pre-randomization characteristics, although some skeptics have argued that fewer patients in the normothermia group had AMI and ischemic heart disease, although there were not more interventions completed for this previously.
Of the patients randomized to T33C 253/473 (50%) had died by the end of the trial as compared to 255/466 (48%) in the T36C cohort (OR 1.06 95% CI 0.98-1.28, p = 0.51). Minimal neurologic deficit (CPC 1 and 2) occurred in 47% in both arms and mRS </= 2 occurred in 45 and 44% in the 33 degree and 36 degree groups respectively.
There was shorter duration of mechanical ventilation in the T36C group: T33C = 0.83 versus T33C = 0.76 median days receiving mechanical ventilation/days in ICU (P=0.006)
How is patient care impacted?
Everyone will likely have a different opinion on this. The fact of the matter is survival after cardiac arrest has improved since the original publication of the HACA and Bernard hypothermia trials. New interventions like hands only CPR, early defibrillation, and cardiac catheterization likely have a part in some of the improvements in mortality seen, but certainly a proportion of this improvement must be attributed to the hypothermia protocol.
That being said, in the largest trial performed to date on optimal temperature management there was no difference in the primary outcome of mortality. The HACA and Bernard papers were flawed in that they only included patients with VF/VT arrest and had NO regulation of temperature in the control arm, which resulted in many patients becoming febrile.
Hypothermia induces diuresis, electrolyte disturbances, shivering, and often requires heavy sedation and sometimes paralytics. These nuances make targeted temperature control a more attractive option, with the best available evidence suggesting at least equivalent outcomes.
As a clinician in both the emergency department and ICU I am always looking for the simplest answer for a complex problem. Thus in my view TTM to 36 degrees seems, at least for the time being, to be a reasonable alternative. The MAJOR drawback of TTM is for clinicians to get relaxed about temperature management and patients may develop fever which is has been demonstrated, albeit in different patient populations, to result in worse neurologic outcome. THEREFORE, if you adopt a 36-degree protocol you should think of the temperature like you would blood pressure, and aggressively treat fluctuations.
Major Limitations of the Study
The study does not reflect the subset of patients seen in most urban county hospitals, with prolonged downtime and most initial rhythms in this study being shockable. That being said, the Bernard and HACA trials excluded all patients with non-shockable rhythyms and these studies still dramatically changed our practice.
Also the study was powered to detect an 11% changed in mortality, which is rather large. Excluding a smaller improvement in mortality is not possible, however there was no trend or even suggestion based on this study that this would be the case.
Lastly the improved survival in all of these studies could reflect increase attentiveness to critically ill patients may result in improved outcomes regardless of the actual intervention employed.
TAKE HOME POINTS:
-No difference in mortality between T36C and T33C by the end of the study.
-T36C is less cumbersome and associated with less physiologic perturbations and is simpler.
-Ignoring temperature management all together SHOULD BE DISCOURAGED, and temperature management should be considered a critical component of resuscitation until proven otherwise.
-Dave