Politics and Hypothermia—What Might They Have in Common? Editorial Comment on Silasi and Colbourne,2011

The late United States Speaker of the House of Representatives, Thomas Philip “Tip” O'Neil is credited with popularizing the phrase “All politics is local.” And as politics has evolved, most feel that this statement has and will continue to stand the test of time. Similarly, as therapeutic hypothermia has evolved both in clinical use and in randomized controlled trials (RCTs), for conditions such as traumatic brain injury (TBI) and stroke, the ability to cool locally rather than systemically has taken on renewed interest. Specifically, the ability to cool locally might be critical to the successful application of this powerful therapy—one that might otherwise be precluded from consideration because of unwanted systemic side effects. In conditions such as severe stroke and severe TBI, rewarming upon completion of a course of mild or moderate hypothermia can precipitate hypotension and rebound intracranial hypertension, in the presence of persistent brain swelling, as was observed in the recent Hypothermia trial for children with severe TBI from the Canadian Critical Care Trials Group (Hutchison et al., 2008). Similarly, hypotension during its application was seen in other RCTs of moderate hypothermia in adult TBI, as reported by Clifton et al. (2001). Mild systemic hypothermia is also well known to have inhibitory effects on cytochrome P450-mediated drug metabolism that may predispose patients to toxicity and complications (Tortorici et al., 2006, 2007, 2009; Hostler et al., 2010), and inhibit leukocyte migration, which may increase the risk of infections such as pneumonia (Whalen et al., 1997; Ishikawa et al., 1999). Obviating these problems, particularly in the setting of TBI and stroke might be critical to its successful use—where RCTs have yet to be designed such that whole body cooling can demonstrate benefit. For example, a recent study in experimental TBI in rats by Yao et al. (2011) using a local cooling collar has shown neuroprotective benefit. And local cooling for stroke, TBI, and other conditions has also been subject to clinical exploration (Abou-Chebl et al., 2011). Lest we forget that in the seminal work by Busto et al. (1987), isolated brain cooling in incomplete forebrain ischemia in rats had profound neuroprotection. Indeed, within 2 years of publication of that paper, it became next to impossible to publish work successfully on brain ischemia without controlling brain temperature (Busto et al., 1989).

Recently, in Therapeutic Hypothermia and Temperature Management, Silasi and Colbourne (2011) reported on a new strategy to induce local cooling of the rat brain using a metal cooling tube secured to the skull—to cool the ipsilateral hemisphere focally such that the hippocampus beneath the cooling site can be readily cooled to 32°C in an awake animal. Body temperature and contralateral hippocampus were kept normothermic, and the cooling system was coupled to the metal cooling tube in such a way that the rat was allowed to move freely. Using this approach, local cooling for 2 days beginning at 1 hour after a 10 minute period of incomplete forebrain ischemia was highly protective to CA1 hippocampus. Of particular interest, contralateral neuroprotection was not seen. This certainly represents one of the cleverer approaches to selective brain cooling in an experimental paradigm and, with the lack of need for sedation and use of a freely moving system, could facilitate studies where the confounding effects of sedatives can be eliminated—and yet the environment is one where stress appears to have been minimized.

Although this approach could provide valuable information on issues such as the mechanism of hypothermic protection in the absence of anesthetics, it is important to realize that unlike TBI or stroke, the clinical condition that most commonly produces selective neuronal death in CA1 and other selectively vulnerable brain regions in global cerebral ischemia is cardiac arrest. Cardiac arrest is a condition where total body hypothermia has been shown to be efficacious, and although neurological outcome appears to be the key target of protection in cardiac arrest victims treated by hypothermia, we should be mindful to recognize that potential beneficial effects of mild hypothermia on the myocardium and possibly other organs may play a role. For example, Nozari et al. (2004) demonstrated that early (but not delayed) application of hypothermia after prolonged VF cardiac arrest in dogs protected against neuronal injury, but also dramatically reduced the development of multiple organ failure. And recently, Ye et al. (2012) demonstrated marked benefit of an early short application of mild hypothermia on myocardial function and microcirculation after VF cardiac arrest in rats—further highlighting this possibility. However, a study on the use of systemic versus brain cooling in neonatal brain injury did not show a difference in the prevalence of multi-organ failure (Sarkar et al., 2009). Indeed, preference of systemic versus local cooling in perinatal brain injury remains unclear (Higgins et al., 2011). Even in the setting of cardiac arrest in adults, there are some patients in whom whole body hypothermia is contraindicated, such as in the setting of coagulopathy and possibly infection, among others. And the merits of the trade-off of potential extracerebral benefit from whole body cooling after cardiac arrest versus the potential merits of using local cooling with a reduced need for sedation and neuromuscular blockade—potentially allowing earlier rehabilitation and shorter ICU stay—remain undefined. In contrast, unlike cardiac arrest, local cooling in TBI and stroke may be fundamentally essential to successful application—if it indeed is possible to show robust benefit from hypothermia in these two conditions. In addition, one might also speculate that the best approach to cooling may importantly differ for each type of CNS insult and that, for example, some combination of local and systemic cooling might be optimal in some conditions. There is much to be learned about this complex therapy.

In 1989, Busto et al. in a brief review in the journal Stroke stated:

Can selective brain cooling be accomplished feasibly and safely in humans? Studies on this point are lacking. Nonetheless, the experimental evidence is now compelling that moderate degrees of selective brain cooling…may protect against ischemic neuronal injury…As concerned investigators and clinicians, we should begin promptly to design and implement the studies needed to resolve these important issues.

Now, two decades later, considerable progress has been made with regard to the use of mild hypothermia in clinical neuroprotection, including studies across the age spectrum and in a number of clinical conditions, and investigations of both its efficacy and its physiological and mechanistic actions (Clifton et al., 2011; Shankaran et al., 2005, 2012; Tiainen et al., 2003; Dietrich 2012; Marion et al., 1997; Rutherford et al., 2010; Bayır et al., 2009; Abend et al., 2009, 2012; Fink et al., 2012; Smith et al., 2011; Clifton 2011; Patel et al., 2011). However, the original question raised on selective or local brain cooling is still unanswered and, for TBI and stroke, deserves to be addressed. Hopefully potential solutions to these questions won't get mired in politics.