Selective Hypothermia in the Management of Traumatic Brain Injury

Eisha Christian, B.A.; Gabriel Zada, M.D.; Gene Sung, M.D.; Steven L. Giannotta, M.D.


Neurosurg Focus. 2008;25(4):E9 

In This Article


Surface Cooling

Surface cooling remains the simplest method of inducing selective hypothermia. Basic methods of systemic hypothermia via surface cooling have been refined to achieve regional cooling of the cranial contents. These methods have traditionally included ice packs, ice baths, and head fanning, although more complex means such as cooling helmets and caps have been used more recently.

In 2004, Wang et al.[41] reported the results of a RCT on the safety and efficacy of a cooling helmet first developed based on National Aeronautics and Space Administration (NASA) technology. This device has 2 components: a head/neck liner and a conditioning unit. The liner is made from a lightweight, thin, urethane-laminated nylon with 2 layers consisting of a liquid cooling heat exchanger and a pressurizable air bladder. The liner has bilateral frontal openings and an anterior midline neck opening for urgent cranial and tracheal access. The conditioning unit serves as an insulated ice reservoir system that can be operated with either a standard power supply or batteries. The study by Wang and colleagues included patients with severe stroke or TBI and excluded those with systemic diseases, cervical spine injuries, and significant head lacerations. The study protocol was initiated within 24 hours of admission, and patients were randomized to a cooling helmet or no hypothermia intervention. Cerebral and systemic temperatures in both patient groups were monitored for 72 hours after admission. The cooling helmet provided a mean change in the brain-body temperature of -1.6°C, as opposed to a change of +0.22°C in the control group (p < 0.0001). Results showed an average reduction of 1.84°C in the hypothermia group within 1 hour of treatment application, with an average 3.4 hours (range 2-6 hours) to achieve a brain temperature < 34°C. Moreover, an average of 6.7 hours (range 1-12 hours) elapsed prior to any decrease in systemic temperature. No significant complications were reported, although there was 1 episode of asymptomatic bradycardia that resolved after raising the core body temperature by 0.5°C. The main conclusion of their study underlined the practicality of this method in inducing rapid SBC, its relatively low risk profile, and the delayed onset of systemic hypothermia. Note, however, that no outcome benefits were studied in this preliminary trial.

In 2006, Liu et al.[25] described a preliminary RCT in which they compared SBC with mild systemic hypothermia and normothermia. Selective brain cooling to 33-35°C was induced via a cooling helmet and neckband, whereas systemic hypothermia to 33-35°C (rectal temperature) was achieved using a cooling blanket. The cooling helmet consisted of circulating saline at 4°C, and a neckband containing blue ice strips was placed around the neck. Treatment was performed for up to 6 hours/day for the first 3 days following injury. Natural rewarming began on the 3rd day after injury. The mean nadir brain temperatures were significantly lower (33°C) after SBC compared with those following mild systemic hypothermia (35°C). The authors reported lower ICPs in the SBC group compared with those in controls. Furthermore, systemic levels of superoxide dismutase measured on postinjury Days 3 and 7 were higher in the SBC and mild systemic hypothermia groups compared with those in the control group. Finally, survival rates and the proportions of patients with good neurological outcomes (GOS Score 4 or 5) were significantly improved in the SBC group versus the systemic hypothermia and control groups (p < 0.05). There were no significant differences in complication rates (infections, thrombocytopenia, or cardiovascular complications) among the 3 groups.

In another study by Qiu et al.,[33] 90 patients that suffered a TBI were randomized to a selective hypothermia group (a cooling headcap and neckband) or a control group (no intervention). Brain temperatures of 33-35°C were maintained for 3 days after injury. No increased complication rate was noted in the selective hypothermia group. Furthermore, the authors reported statistically significant decreases in ICPs at 24, 48, and 72 hours in the hypothermia group. Improved neurological outcomes were reported in the treatment group at 6 months following injury, with good neurological outcomes (GOS Score 4 or 5) in 68.9% of the hypothermia group and 46.7% of the control group (p < 0.05).

Selective surface cooling has established Class I evidence supporting its use in the management of neonatal encephalopathy.[17] Per initial studies, surface cooling methods applied for several hours a day for the first 3 days after injury appear to be a safe and noninvasive method of implementing regional hypothermia. However, more Class I evidence is required to corroborate these results. The availability of the cooling helmet and circulating hypothermia machine remains an issue of practicality at this time.

Intranasal SBC

Intranasal cooling of the cranial contents appears to be a promising treatment following TBI. Several methods of cooling the brain via the nasopharynx have been attempted in animals; only a few human clinical trials have been completed to date. In a small RCT authors attempted room temperature airflow through the upper respiratory tract of intubated brain-injured patients but demonstrated no benefit in the reduction of brain temperature associated with this method.[1] In a follow-up study, the same authors attempted nasal airflow with supplemental NO as well as bilateral head fanning in a group of 12 patients. In this study, a mild, statistically significant decrease of 0.31°C in brain temperature was observed within 5 minutes in the selective hypothermia group.[19] The authors concluded that heat loss from the upper airways and the skull base can reduce parenchymal brain temperature in brain-injured humans in a rapid, controlled fashion.

Covaciu and associates[13] reported their results for the intranasal placement of balloon catheters filled with chilled, circulating saline. Hypothermia was maintained for 6 hours in a porcine model via this method. No major disturbances in systemic circulation or physiological variables were associated with the intranasal cooling. There was a rapid decrease in cerebral temperature within the first 20 minutes of inducing hypothermia and a mean reduction of 2.8°C in cerebral temperature to a target temperature of 35°C. In comparison, the systemic temperature decreased by a mean of 1.7°C.

More recently, BeneChill, Inc., has developed the RhinoChill System, an intranasal device that sprays an inert evaporative coolant onto the surfaces of the upper airways and skull base via nasal prongs (Fig. 1). The only major contraindication at this time in patients with TBI is the presence of skull base fractures. The liquid coolant is contained in bottles that can be stored at room temperature. It is sprayed into the nasal cavity and disperses widely because of its low surface tension. Once in contact with the nasopharynx, the coolant rapidly evaporates. Heat dissipates through the skull base via conductive mechanisms, and cooled blood along the skull base spreads via the arteries to the remainder of the brain (Fig. 2). The intranasal temperature is ∼ 2°C. In a sheep trial, the RhinoChill device was compared with whole-body surface cooling.[44] In that study, brain cooling was significantly faster with the RhinoChill device than with systemic hypothermia. Furthermore, brain-to-systemic temperature gradients were maintained, and the brain therapeutic hypothermia zone (defined as ≤ 3.5°C below baseline) was reached within 15 minutes. There were no complications associated with this method. In additional studies, the RhinoChill device significantly improved resuscitation rates as well as survival and neurological outcomes in pigs undergoing cardiopulmonary resuscitation following induced ventricular fibrillation.[39]

Image showing the nasal prong system of the RhinoChill device, which is inserted into the nasal cavity and sprays an evaporative perfluorocarbon coolant that cools the skull base. Reprinted with permission from BeneChill, Inc.

Image depicting the method of action of the RhinoChill device. Once the coolant is sprayed into the nasal cavity, it rapidly evaporates. Heat dissipates through the skull base via conductive mechanisms, and cooled blood along the skull base spreads via the arteries to the remainder of the brain. Reprinted with permission from BeneChill, Inc.

Intranasal selective hypothermia appears to be a relatively noninvasive and rapid method of achieving preferential hypothermia with minimal systemic disturbances. Multiple institutions, including our own, are currently enrolling patients into prospective clinical trials to assess the safety and efficacy of the RhinoChill device.

Endovascular Cooling

The infusion of cold saline or blood has been attempted via a transvenous or transarterial route to achieve selective cooling of the brain. The first description of selective hypothermia via the carotid arteries appeared in 1954; Parkins and associates[30] reported the unilateral perfusion of the common carotid artery in dogs (Fig. 3). During a period of complete circulatory occlusion, blood was drawn from the carotid artery, cooled extracorporeally, and reinfused into the carotid artery. The authors were able to achieve cerebral temperatures of 15°C while maintaining systemic temperatures of 32°C. Significant neuronal injury occurred at brain temperatures below 12°C, but neuroprotective effects were noted in the range of 15-20°C. In 1992 Ohta et al.[29] successfully perfused cold lactated Ringer solution via the vertebral artery in dogs while clamping both carotid arteries and reported the maintenance of temperatures of 27-28°C and a mean systemic temperature of 33.9°C for up to 60 minutes. They reported improved neurological outcomes in the hypothermia group compared with those in the normothermic infusion control group, in which no animals survived. No histological evidence of ischemia was observed in the hypothermia group.

Diagram showing hypothermia in a dog. Dogs were subjected to a period of complete circulatory occlusion, during which blood was drawn from the carotid artery, cooled extracorporeally, and reinfused to the carotid artery. A.= artery; ENDOTRACH. = endotracheal; INF. = inferior; L. = left; R. = right; SUP. = superior; TEMP. = temperature. Reprinted with permission from Parkins et al: Annals of Surgery 140:284-289, 1954, Lippincott Williams & Wilkins, Inc.

In 1996 Schwartz et al.[36] attempted perfusion of cold autologous blood via the common carotid artery in baboons. Blood was drawn from a femoral access catheter, cooled in an extracorporeal fashion, and pumped back into the common carotid artery at systemic blood pressure. They were able to cool the brain to as low as 18-19°C within a mean time of 26 ±13 minutes, with systemic temperatures stabilizing at an average of 34°C.[36] This same study demonstrated decreases in cerebral blood flow associated with the selective hypothermia treatment and no resulting hemodynamic complications. In 2006, Jiang and colleagues[22] described the perfusion of cold lactated Ringer solution (5°C) versus normothermic lactated Ringer solution (37°C) via the internal carotid artery following cerebral ischemia in Rhesus monkeys. Global cerebral hypothermia was induced in the treatment group to as low as 15-16°C, with minimal changes in systemic temperature. Normal neurological function was reported in the hypothermia group, compared with death in all animals in the normothermia group. Furthermore, no histological evidence of ischemia was noted in the hypothermia group, compared with the finding of medullary necrosis in the normothermia group.

In 2005, Wen et al.[43] performed a feasibility experiment assessing hypothermic retrograde jugular flush as an alternative to carotid artery infusion. The external jugular veins of rats were cannulated and infused with cold saline at 2 temperatures, 4° or 24°C, over a period of 10 minutes. These rats were compared with those from a normal control group. Significant reductions in brain temperatures were achieved in both groups. No outcome measures were studied. The authors suggested further studies of shock models in animals, followed by potential clinical trials in humans.

To date, endovascular methods of selective cerebral hypothermia remain in the experimental phase. The safety of this method as a primary intervention for TBI remains to be determined. The feasibility of emergency carotid artery infusion, especially with hemodilution, remains unlikely at this time in patients with polytrauma. Hypothermic retrograde jugular flush appears to be an effective means of achieving rapid preferential hypothermia, although it is associated with increases in cerebral blood flow.[43] Conflicts regarding ICP dynamics and rates of thrombocytopenia or coagulopathy in conjunction with endovascular methods of selective hypothermia remain to be studied but may pose significant limitations in patients with TBI.

Extraluminal Vascular Cooling

The novel method of extraluminal vascular cooling was recently developed by the Walter Reed Army Institute of Research and involves a cooling cuff containing cold circulating water wrapped around the common carotid arteries. In a model of MCA occlusion in rats, extraluminal vascular cooling of the common carotid arteries was performed starting 30 minutes postinjury and was maintained for 90 minutes.[42] Brain temperatures were reduced by 2-5°C within 20 minutes, whereas the systemic temperature was maintained at 37°C. No cardiovascular complications occurred. The authors were able to demonstrate decreases in infarct size in the MCA territory and diminished rates of periinfarct depolarizations in the treatment group.

Epidural Cooling

Epidural brain cooling is another recently described method of achieving selective hypothermia. In a study by Cheng et al.,[6] cold saline (4°C) was infused into the epidural space of swine via a double-lumen catheter. Brain (deep and surface), epidural, and systemic temperatures were monitored, and cold saline infusion rates were manipulated to achieve target brain temperatures. Epidural pressures were also monitored. Surface brain temperatures were reduced to 17.3-21.6°C within 1 minute, whereas deep brain temperatures were reduced to 27.2-29.7 within 5 minutes. Systemic temperatures were maintained during 6 hours of treatment. No hemodynamic or hematological complications were associated with this treatment, and ICPs remained stable.

In a separate study, the same author group assessed the efficacy of the epidural cooling method used in swine 1 hour after complete MCA occlusion and maintained for a period of 5 hours.[45] Brain temperature was reduced to < 30°C (deep brain) and < 20°C (brain surface), again with no changes in systemic temperature. The study showed a significantly decreased infarction volume (< 50%) with the cooling method as compared with the control group. Epidural cerebral cooling appears to offer a potent neuroprotective benefit, but to date this method has only been attempted in animal studies.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.