Skip Navigation LinksHome > June 2009 - Volume 41 - Issue 3 > The Use of Hypothermia as a Treatment for Traumatic Brain In...
Journal of Neuroscience Nursing:
doi: 10.1097/JNN.0b013e3181a23ec1

The Use of Hypothermia as a Treatment for Traumatic Brain Injury

Rupich, Kristin

Free Access
Article Outline
Collapse Box

Author Information

Questions or comments about this article may be directed to Kristin Rupich, MSN, at She is a CRNP at the Hospital of the University of Pennsylvania, Philadelphia, PA.

Collapse Box


Hypothermia has been shown to have neuroprotective effects and may have benefit in the treatment of head injuries. However, it is a controversial treatment in traumatic brain injury, and to date, there are no specific recommendations for its use. This article examines six research studies investigating the use of hypothermia as a treatment in patients with traumatic brain injury. All studies were prospective trials and compared a controlled normothermia group with a hypothermia group. Studies were compared by sample population, methods of hypothermia, outcomes, and conclusions. The leading variable in each study was hypothermia. However, each study used a different method of cooling, goal temperature, and duration of cooling. Through the comparison of these studies, a recommendation for change in practice cannot be made. Nevertheless, there may be benefits to hypothermia in traumatic brain injury, and suggestions for future research are identified.

Each year, in the United States, 1.4 million people sustain a traumatic brain injury. Of this number, 235,000 individuals are hospitalized and survive, with approximately 80,000 to 90,000 experiencing long-term disability. Traumatic brain injury is responsible for 50,000 deaths per year (Centers for Disease Control and Prevention, 2004). Traumatic brain injury encompasses primary and secondary injuries. The damage incurred at impact is referred to as primary injury. Secondary injury is a biochemical and cellular response to the primary injury that begins within minutes to hours of the initial damage. Secondary injury is global and harms tissue that was not involved in the primary injury. It is the amount, severity, and frequency of secondary injury that leads to widespread tissue damage and poor neurological outcomes (Dutton & McCunn, 2003). In the first 24 hr after injury, hypocapnia, hypotension, hyperglycemia, hypoxia, acidosis, intracranial hypertension, and seizures may all occur (Jeremitsky, Omert, Dunham, Protetch, & Rodriguez, 2003). These secondary injuries cause significant neurological damage in part by affecting brain oxygenation. Accordingly, it is vital to manage these secondary injuries immediately to prevent permanent damage (Dutton & McCunn, 2003).

The international medical community has established guidelines for the treatment of traumatic brain injury. These guidelines depict the current standard of care for the treatment of these patients (Brain Trauma Foundation, 2007). Although these guidelines are followed as the first line of treatment, many other experimental strategies are employed to decrease secondary injury, including hypothermia.

Back to Top | Article Outline

The Effect of Hypothermia

The use of hypothermia as a treatment of traumatic brain injury remains controversial. The definition of induced mild hypothermia varies between research studies, but is generally considered cooling to a temperature of 32°C-35°C. The Brain Trauma Foundation classified hypothermia as a class III level of evidence. Although it appears to provide benefits, there is not enough research available to make formal recommendations on its use (Brain Trauma Foundation, 2007). It is thought that hypothermia has neuroprotective effects and will prevent or minimize the effects of secondary brain injury. Hypothermia protects the brain by inhibiting massive depolarization in the brain and the release of glutamate and aspartate. Hypothermia, therefore, stabilizes the blood-brain barrier and prevents cell death. It also decreases cerebral metabolic rate, with a resultant decrease in carbon dioxide and lactate buildup. Moreover, hypothermia may prevent uncoupling of the metabolic supply demand regulation and prevent loss of cerebral autoregulation (Wright, 2005). Thus, although still considered a controversial treatment, many institutions employ hypothermia as treatment for the prevention of secondary brain injury (Brain Trauma Foundation, 2007).

Advanced practice nurses, bedside nurses, and physicians are responsible for creating treatment plans for this population of patients and therefore must have knowledge of the latest research. There has been a marked increase in the research surrounding hypothermia as a treatment option due to its potential to improve patient outcomes. The advanced practice nurse on the neurosurgery team is often responsible for the acute management of traumatic brain injury. Advanced practice nurses are vital in exploring the latest research and potential treatments for these patients (Freeborn, 2004).

To identify research articles, a search was performed using Medline, CINHAL, Embase, and PubMed. Key words used alone and in combination were hypothermia, traumatic brain injury, and brain oxygenation. Forty-three primary research articles from 2000 to 2006 written in English were reviewed. Four research studies were identified based on these criteria for selection: (a) inclusion of controlled trials using therapeutic hypothermia for at least 24 hr versus normothermia in adults with traumatic brain injury and (b) outcomes measured by the Glasgow Outcome Scale (GOS). Variables measured were intracranial pressure (ICP), cerebral perfusion pressure (CPP), arterial blood pressure, continuous temperature, and management appropriate for traumatic brain injury. A fifth study was identified by the other researchers as a landmark hypothermia study and was therefore included in the analysis. The last study included did not have a normothermia control group; rather, it compares long-term mild hypothermia to short-term mild hypothermia.

Several meta-analyses have explored the possibility of hypothermia as a beneficial treatment in traumatic brain injury. The most recent meta-analysis examined 14 clinical trials of comparable groups and suggested that hypothermia reduces mortality and results in favorable neurological outcomes when maintained for greater than 48 hr. In addition, the greatest improvements in outcome were found when patients were evaluated 1 to 2 years after injury rather than at 6 months. It also appears that the patients who respond best to hypothermia are those who respond well to standard measures of ICP control, excluding barbiturate therapy (Peterson, Carson, & Carney, 2008). Although the meta-analysis was not able to make a recommendation for the use of hypothermia, its results recognized the potential of the therapy. This article will further explore the topic and compare and analyze the mentioned research studies and discuss the implications of hypothermia as a treatment on nursing practice.

Back to Top | Article Outline

Analysis of Research

Six studies were compared to explore the effects of hypothermia on neurological outcome after traumatic brain injury. Clifton et al. (2001); Gal, Cundrle, Zimova, and Smrcka (2002); Marion et al. (1997), Polderman, Tjong Tjin Joe, Peerdeman, Vandertop, and Girbes (2002); and Jiang et al. (2006) evaluated the effects of hypothermia on patient outcome in the traumatic brain injury population with intracranial hypertension. Shiozaki et al. (2001) looked at the effect of hypothermia on neurological outcome in patients with normal ICP. Although it examines a slightly different aspect of brain injury, it is comparable with the other studies and provides a contrast to high ICP. The study of Shiozaki et al. used the same methods of treating traumatic brain injury; however, they chose to include patients who did not have high ICP in the hypothermia group. This was notable because the study of Shiozaki et al. demonstrated the effect of hypothermia on patients with normal ICP.

Back to Top | Article Outline
Selection of Patients

In each study reviewed, patients were chosen on admission to the hospital. Inclusion criteria between the studies were comparable, with minimal difference between the selection processes. Patients were accepted if they had a closed head injury and Glasgow Coma Scale (GCS) of 3 to 8. All studies accepted patients aged 16 to 65 years, except the study by Marion et al. (1997), which accepted patients up to 75 years old, and by Shiozaki et al. (2001), which included one patient younger than 16 years.

Exclusion criteria in all six studies included clinical brain death, prolonged hypoxia or hypotension, a gunshot wound, pregnancy, an undetermined time of injury, inability to begin cooling within 6 hr, organ failure of another system, or normal findings on a computerized tomography (CT) scan. The sample populations were comparable across studies. The study by Shiozaki et al. (2001) was the exception and also excluded patients who were unable to maintain their ICP below 25mm Hg on admission despite conventional therapies. The selection of patients was similar between the groups and did not affect the results.

Back to Top | Article Outline
Comparison of Patient Population

The sample population across the six studies was equivalent in terms of age. The mean age of patients in both the normothermia and hypothermia groups was 31 to 42 years and was not a statistically significant variable. Significantly more men participated in the studies than did women, a fact which is reflective of the greater traumatic brain injury population where males are twice as likely as females to sustain an injury (Centers for Disease Control and Prevention, 2004).

All studies used GCS to measure neurological status of patients on admission and for continued assessments throughout the studies. The scale was defined and used in the same manner for each study, which eliminated error in measuring neurological status. With the exception of the study of Gal et al. (2002), the other studies acknowledged differences in the severity of GCS by comparing the hypothermia and normothermia groups as a whole and as subclasses of patients with a GCS of 3 or 4 and 5 to 8. A lower GCS is indicative of more severe injury and thus should be looked at separately from the higher scores. By making this distinction, the studies looked at the effect of hypothermia on degree of injury.

In addition to identifying GCS, the specific head injuries sustained were identified and classified in different ways. Marion et al. (1997), Polderman et al. (2002), and Shiozaki et al. (2001) classified injuries by CT scan using the classification categorized by Marshall et al. (1991). Clifton et al. (2001), Gal et al. (2002), and Jiang et al. (2006) categorized type of injury by location and surgical procedure performed. Each study acknowledged that severity of injury was not always equal but was mathematically adjusted for and thus was not statistically significant. Many patients also experienced secondary injuries, which were acknowledged in three studies. The studies by Marion et al. and Shiozaki et al. specified the additional injuries involved that included abdominal injuries, chest injuries, pelvic or leg fractures, and arm fractures. The number of injuries in the hypothermia and normothermia groups was equivalent. Although Clifton et al. (2001), Polderman et al., and Jiang et al. did not specify other injuries sustained, the studies noted that all additional variables between the groups were equal. The study by Gal et al. did not specify additional injuries, an omission which makes it difficult to compare the severity of other complications.

Back to Top | Article Outline
Methods of Randomization

Methods of randomization into hypothermia and control groups differed by study. The main differences were that Clifton et al. (2001) and Shiozaki et al. (2001) both involved 11 centers. The study of Clifton et al. had the largest sample size of 392 patients and was a randomized prospective study. The study of Jiang et al. (2006) was the only other study that involved multiple centers. Jiang et al. included three centers and 215 patients. Shiozaki et al. and Marion et al. (1997) both conducted their studies at one health center. Their studies, like that of Clifton et al., were randomized prospective studies with a relatively large sample size. Polderman et al. (2002) conducted the study at a single hospital and divided the patients on whether or not the ICP responded to the last step of their protocol, a barbiturate coma. Those who did not respond were placed in the hypothermia group, and those who did respond made up the control group. The study that differed was performed by Gal et al. (2002). This study included only 30 patients and assigned the first 15 to the hypothermia group and the second 15 to the normothermia group. This small sample size may not be reflective of the population.

Back to Top | Article Outline

In the populations studied, hypothermia was the variable affecting outcome. Although each group was treated using hypothermia, there were differences in the applied methodology. These differences included time from injury to ideal cooling, cooling temperature selected, and duration of cooling. In addition to the cooling temperature chosen, the methods of cooling varied. Marion et al. (1997), Shiozaki et al. (2001), and Clifton et al. (2001) used surface cooling mechanisms and nasogastric lavage with iced saline. Gal et al. (2002), Jiang et al. (2006), and Polderman et al. (2002) used water-circulating cooling blankets that were noninvasive and posed no risk of infection.

For the five studies with a control group, the normothermia group's temperatures were maintained between 36.5°C and 38.5°C with cooling blankets. Although the normothermia groups served as the control groups, the amount of time they were maintained at normothermia differed for each study. Jiang et al. (2006) maintained both groups between 33°C and 35°C. Length of cooling ranged from 24 to 96 hr. See Table 1 for comparison of hypothermia characteristics.

Table 1
Table 1
Image Tools

All studies measured similar variables to determine efficacy of hypothermia. The values measured were arterial blood pressure, ICP, CPP, jugular bulb oximetry, and rectal or bladder temperature. These variables were managed and maintained within the standard treatment guidelines for each institution. Additional factors measured were brain oxygenation and temperature, carbon dioxide, and pH of the blood through brain oxygenation monitors and frequent laboratory tests. Electrolytes and clotting factor values were evaluated at least daily from the hypothermia and normothermia groups. All studies identified the difference between oral, rectal, or bladder temperature and found that they were interchangeable.

Back to Top | Article Outline

Each study compared the values recorded in a slightly different way. All studies evaluated nonparametric data using t tests, chi-square tests, Wilcoxon rank-sum tests, or Fisher's exact tests. These statistical tests were used for comparison of baseline characteristics, complications, and outcomes. Tests were chosen by the researcher or statistician based on sample and data size. All values were presented as means ± standard deviation. Statistical relevance for each study was a p value of <.05 as significant.

Back to Top | Article Outline

Outcomes of Studies

Intracranial Pressure, CPP, and GOS

All studies measured and evaluated GOS, ICP, and CPP. A finding in all studies was that the ICP was decreased and CPP was increased within the hypothermia groups. Of all factors recorded, hypothermia had the most effect on ICP and CPP, which may in turn result in a more favorable outcome. Decreased ICP and increased CPP may improve outcomes by increasing cerebral blood flow and oxygenation to the brain (Wright, 2005). Although Shiozaki et al. (2001) examined hypothermia in traumatic brain injury with low ICPs, the study utilized hypothermia in patients with uncontrollable intracranial hypertension. This demonstrates that hypothermia is effective at deceasing ICP in patients experiencing intracranial hypertension that is unresponsive to conventional methods.

In all studies, outcome was measured by the GOS. According to the GOS, good recovery is defined as functional independence with minor disability, moderate disability is defined as functional independence with more substantial disability, and severe disability is defined as functional dependence (Jennet & Bond, 1975). In each of the studies, favorable outcome included good recovery and moderate disability, whereas poor outcomes were considered severe disability and death. Outcomes were measured by the GOS at various times from 3 to 12 months. GOS was not recorded at the same intervals for each study.

The examined research revealed divergent outcomes on the benefits of hypothermia in traumatic brain injury. Clifton et al. (2001) found that 57% of the patients in both groups had poor outcomes. Of all patients, 53 (28%) in the hypothermia and 48 (27%) in the normothermia group died. Although the study reported the mortality rates, they did not specify the differences in unfavorable and favorable outcomes between hypothermia and normothermia groups. Despite a higher mortality rate in the hypothermia group, it is possible that there was still an improvement in favorable outcome in those who did survive. It, therefore, would be useful to know the percentage of surviving patients who had favorable versus unfavorable outcomes. Shiozaki et al. (2001) found no benefits to using hypothermia in brain injury patients with low ICP. Mild hypothermia, however, was used to successfully control ICP in 16 of 73 patients who experienced high ICP during the study. Of these patients with intractable intracranial hypertension who were treated with hypothermia as a last resort, 6 of those 16 patients (37%) had a favorable outcome.

The studies that recorded hypothermia as a potentially useful treatment were those of Gal et al. (2002), Marion et al. (1997), Polderman et al. (2002), and Jiang et al. (2006). Gal et al. found that the difference in GOS between the hypothermia and normothermia groups of patients after 6 months was not statistically significant. The study, however, found that favorable neurological outcome was reached in 13 of 15 patients in the hypothermia group (87%) as opposed to 7 of 15 patients in the normothermia group (47%). This showed a 40%increase in favorable neurological outcome 6 months after injury. The lack of statistical significance in the study of Gal et al. may be due to the small sample size and reveals promise for further study. Marion et al. had positive results; however, they found that hypothermia did not decrease ICP or improve outcomes for patients with an initial GCS of 3 or 4. Of patients with a GCS of 5 to 7, 16 (73%) in the hypothermia group and 9 (35%) in the normothermia group had good outcomes at 6 months. Within the group of patients with a GCS of 5 to 7, the hypothermia group had a slightly better class of CT scan, meaning their injury was less severe. The researchers therefore adjusted for this, and this variable should not have changed the statistical significance of the groups. Like Marion et al., Polderman et al. found that patients with a GCS of 5 or 6 had significantly improved outcomes. The number of patients with favorable neurological outcome was 15.7% in the hypothermia group versus 9.7% in the normothermia group. Of the subgroup of patients with GCS of 5 or 6 at admission, 29% have favorable neurological outcome versus 8% in the normothermia group. Jiang et al. examined a slightly different aspect and compared short-term with long-term hypothermia. They found that 43.5% of patients in the long-term hypothermia group had favorable outcomes and 56.5% had unfavorable outcomes. In comparison, only 29% of patients in the short-term hypothermia had favorable outcomes and 71% had unfavorable outcomes in the short-term hypothermia group. See Table 2 for comparison of outcomes.

Table 2
Table 2
Image Tools

When examining outcomes, it should be noted that the study by Clifton et al. (2001) was stopped before the goal of 500 patients was reached. This was the result of an interim analytic review by the patient safety and monitoring board, which concluded that hypothermia was not effective at improving outcomes. Shiozaki et al. (2001) also discontinued their study due to the high rate of complications encountered and the inability to show improved outcome. The discontinuation of these two studies is significant because they not only found no benefits to hypothermia, but they also found that it was detrimental to the patients.

Back to Top | Article Outline
Complications Associated With Hypothermia

It has been suggested through past research that a higher incidence of infection and complications is found in hypothermia (Odette, Colford, Good, & Matz, 2002). The prevalence of complications differed between the hypothermia and normothermia groups for each study. Significant complications related to hypothermia are pneumonia, infection, hypotension, bradycardia, and changes in laboratory values (Odette et al., 2002). Shiozaki et al. (2001) found that 30 of 43 patients in the hypothermia group experienced infection as opposed to 12 of 40 patients in the normothermia group. The incidence of pneumonia was approximately three times higher in the hypothermia group than in the normothermia group. Clifton et al. (2001) found that 10% of the patients in the hypothermia group and 3% of those in the normothermia group had hypotension (mean arterial pressure <70 mm Hg) for two or more consecutive hours. Finally, 16% of the patients in the hypothermia group and 4% of the patients in the normothermia group experienced bradycardia (Clifton et al., 2001). In terms of laboratory values, Shiozaki et al. found a higher incidence of electrolyte abnormalities in the hypothermia group and Clifton et al. found that prothrombin and partial thromboplastin times were slightly higher. Although these complications were noted and may appear significant, Marion et al. (1997) and Gal et al. (2002) showed no increase in complications between the two groups. Polderman et al. (2002) acknowledged the potential for side effects; however, they stated that with aggressive treatment these potential complications can be prevented. In their study, the hypothermia group received more vasopressors, antiarrhythmic medications, and electrolyte replacement than did the normothermia group; however, this prevented further complications and did not appear to have negative effects on the patients. Although Jiang et al. (2006) mentioned pneumonia, electrolyte imbalances, arrhythmias, stress ulcers, and seizures as potential complications, there was not a statistically significant difference between long-term and short-term hypothermia groups.

The rates of complications are difficult to compare because the periods when complications occurred during a patient's treatment are not specified. Laboratory values compared in the articles were measured during the time of hypothermia. The results that were not within normal limits may have corrected without negative effects once hypothermia was discontinued. There is no documentation of when the infection rates were recorded. If pneumonia and other infections were acquired during the period of hypothermia, these results may be significant. Consequently, if they began after the period of hypothermia, there may not be a correlation between the infection rate and hypothermia. As a result, it is important to identify the relationship between hypothermia and abnormalities before drawing a conclusion.

If the recorded rates of complications are accurate, then the prevalence may be related to the degree cooled and the length of time maintained at that temperature. It is possible that a cooler temperature increases the incidence of infection (Tokutomi et al. 2003). The method of cooling may also affect infection rates. Patients cooled by nasogastric lavage have the potential for a higher incidence of infection and pneumonia due to an invasive tube (Gal et al., 2002). For the hypothermia groups with high occurrences of pneumonia, the number of patients who aspirated at the time of injury may factor into that number. Because the studies did not record complications in the same manner, it is difficult to state if hypothermia does or does not cause an increase in infection rates and abnormalities of laboratory values.

Back to Top | Article Outline

Discussion of Studies

Three studies make strong recommendations for the use of hypothermia in traumatic brain injury patients. Specifically, Marion et al. (1997) recommended treatment with moderate hypothermia to a temperature of 32°C or 33°C for 24 hr, initiated soon after severe brain injury for patients with an initial GCS of 5 to 7. Polderman et al. (2002) stated that artificial cooling can significantly improve neurological outcome in patients with severe brain injury when used in a protocol with great attention to side effects. They recommended maintaining hypothermia at 32°C for 24 hr with ICP less than 20 mm Hg and then slowly rewarming. If ICP rises during rewarming, they recommended reinstating hypothermia. Jiang et al. (2006) recommended long-term mild hypothermia for 5 days followed by rewarming within 24 hr. Gal et al. (2002) recognized that their sample size was small and therefore may not be reflective of the greater population but found promising data that hypothermia is beneficial and encouraged further research before a change in practice is made. In contrast, Shiozaki et al. (2001) found negative outcomes in the population studied and recommended hypothermia not be used in patients with low ICP. The study of Shiozaki et al., however, conceded that hypothermia may be used with discretion to treat severe head injury patients in whom ICP cannot be controlled.

Clifton et al. (2001) concluded that hypothermia should never be used. A potential problem in this study may have been the large trial size with multiple participating centers. It is possible that the study may have been influenced by differences in treatment protocols and intensive care unit procedures. Clifton et al. also noted observations that may affect its and other studies' outcomes. Specifically, through retrospective chart review, the study found that rewarming patients who were hypothermic on admission may be detrimental to their outcome. They also noticed that patients who were hypothermic on admission tended to have more severe injuries. Clifton et al. did not rewarm hypothermic patients assigned to the normothermia group in their study but mentioned that Marion et al. (1997) rewarmed patients in their study. They also suggested that this could have caused more poor outcomes in the normothermia group of the study of Marion et al., thus falsely elevating the number of good outcomes in the hypothermia group. Clifton et al. also mentioned that their study had a high number of participants with hypothermia on admission, which may have affected their results due to the increased severity of their injury. Hypothermia is a sign of a more severe injury and is associated with an increased mortality rate (Jeremitsky et al., 2003). If these patients had been excluded from the data analysis, the study results may have been more positive. Although these factors are not proven, they are important to note as they may have influenced patient outcomes.

Other factors to consider while using hypothermia is the time from injury to goal temperature,as well as the length of time hypothermia is maintained. Of the studies reviewed, Polderman et al. (2002) had some of the most convincing results. They initiated with a mean time of 5.2 hr to hypothermia and maintained hypothermia for a mean of 4.8 days. Jiang et al. (2006) also cooled patients rapidly, with a mean time of 3 hr, and maintained hypothermia for 5 days. It has been shown that the earlier hypothermia is initiated, the better the results (McIntyre, Ferguson Hebert, Moher, & Hutchison, 2003).

From the research by Marion et al. (1997), Gal et al. (2002), Polderman et al. (2002), and Jiang et al. (2006), hypothermia shows promise as a treatment, yet additional research on methodology of cooling and ideal temperature is needed. Although there have been a few studies on the specifics of hypothermia, examining the ideal temperature and length of time to cool patients, these results also vary. One study that examined the ideal temperature to cool patients found that 35°C offered the most benefits with the least amount of complications (Tokutomi et al. 2003). In contrast, Jiang et al. demonstrated that long-term mild (33-35°C) hypothermia maintained for 5 days is more effective than if maintained for 2 days. The reasoning for this was that, when hypothermia was maintained for only 2 days, there were occurrences of rebound intracranial hypertension, which affected outcomes (Jiang et al., 2006). Rebound intracranial hypertension is a consequence of rewarming patients too quickly and can be detrimental to their outcome (Jeremitsky et al., 2003). Although Marion et al. displayed positive outcomes, when the data for ICP and CPP were further explored, it was found that, in the 37 to 60 hr after hypothermia was discontinued, the hypothermia group had a statistically significantly lower CPP. This is noteworthy because a lower CPP leads to decreased cerebral blood flow and decreased oxygenation. Although the study stated that the difference in ICP is not statistically significant, it should be noted that it was higher in the hypothermia group than in the normothermia group after rewarming. This phenomenon may have occurred in the other studies as well and contributed to poor outcomes.

Hypothermia is now used as a standard of care after cardiac arrest to improve neurological function. It has been demonstrated that hypothermia at 32-34°C for 24 hr improves neurological outcome after cardiac arrest. In one study, mortality at 6 months after cardiac arrest was 41% in the hypothermia group compared with 55% in the normothermia group (Hypothermia After Cardiac Arrest Study Group, 2002). Although hypothermia after cardiac arrest was proven to be beneficial and has more standard guidelines, these guidelines cannot be applied to the traumatic brain injury population. The cardiac arrest patients do not experience the same secondary neurological injuries as traumatic brain injury patients and thus do not have the same difficulty with rewarming procedures (McIntyre et al., 2003). Although it is appropriate to rewarm cardiac arrest patients after 24 hr, it may not be beneficial for traumatic brain injury patients who are still experiencing secondary injuries.

Back to Top | Article Outline

Implications for Practice

Hypothermia shows promise as a treatment of traumatic brain injury. However, there are several important points to consider when contemplating its use. Hypothermia as a treatment of traumatic brain injury should be utilized in hospitals with specialized neuroscience units that have continuous resident coverage. In addition, nurses are at the front line of initiating the treatment and must be properly taught to care for these patients. There are many potential complications of hypothermia that nurses must be aware of and trained to aggressively treat. The nursing care involved in caring for a patient with a severe brain injury is complex, and it is crucial that they have the support and appropriate nursing ratios to care for these patients.

Back to Top | Article Outline


From the research reviewed, a recommendation cannot be made for changing practice. However, it appears that hypothermia may have benefits for patients with severe traumatic brain injury, specifically those with a GCS of 5 to 8. It also appears that there is no benefit to hypothermia for those patients with low ICP. From the research reviewed, it can be recommended that hypothermia be initiated as soon as possible after injury and that patients who are cooled for at least 48 hr tend to have better outcomes. If hypothermia is employed as a treatment option, careful attention to side effects is crucial for improved patient outcome. Time, temperature, and methodology are all variables that must be considered if hypothermia is employed for patients with traumatic brain injury.

Traumatic brain injury is a devastating event that can potentially result in a lifelong debilitation. The guidelines that govern care have aided healthcare providers in treating the sequelae of traumatic brain injury. Unfortunately, there are still many people with poor outcomes. Continuing research endeavors for treatments that can improve outcomes for these patients and their families is paramount. Hypothermia as a neuroprotective mechanism is a promising therapy but must be further explored before a suggestion for a change in practice can be made. Advanced practice and bedside nurses working with this population of patients have the opportunity to further explore hypothermia as a treatment of traumatic brain injury.

Back to Top | Article Outline


Brain Trauma Foundation. (2007). Guidelines for the management of severe traumatic brain injury: 3rd Edition. Journal of Neurotrauma, 24(Suppl. 1), S1-S106.

Centers for Disease Control and Prevention. (2004). Traumatic brain injury (TBI): Topic home. Retrieved January 28, 2006, from

Clifton, G. L., Miller, E. R., Chol, S. C., Levin, H. S., McCauley, S., Smith, K. R., et al. (2001). Lack of effect of induction of hypothermia after acute brain injury. New England Journal of Medicine, 344, 556-563.

Dutton, R. P., & McCunn, M. (2003). Traumatic brain injury. Current Opinion in Critical Care, 9, 503-509.

Freeborn, K. (2004). Neurotrauma-the role of the nurse practitioner in traumatic brain injury. Topics in Emergency Medicine, 26, 225-230.

Gal, R., Cundrle, I., Zimova, I., & Smrcka, M. (2002). Mild hypothermia therapy for patients with severe brain injury. Clinical Neurology and Neurosurgery, 104, 318-321.

Hypothermia After Cardiac Arrest Study Group. (2002). Mild therapeutic hypothermia to improve the neurologic outcome after cardiac arrest. New England Journal of Medicine, 346, 549-556.

Jennet, B., & Bond, M. (1975). Assessment of outcome after severe brain damage. A practical scale. Lancet, 1, 480-484.

Jeremitsky, E., Omert, L., Dunham, M. C., Protetch, J., & Rodriguez, A. (2003). Harbingers of poor outcome the day after severe brain injury: Hypothermia, hypoxia, and hypoperfusion. Journal of Trauma, 54(2), 312-319.

Jiang, J., Xu, W., Li, W., Gao, G., Bao, Y., Liang, Y., et al. (2006). Effect of long-term mild hypothermia or short-term mild hypothermia on outcome of patients with severe traumatic brain injury. Journal of Cerebral Blood Flow and Metabolism, 26, 771-776.

Marion, D. W., Penrod, L. E., Kelsey, S. F., Obrist, W. D., Kochanek, P. M., Palmer, A. M., et al. (1997). Treatment of traumatic brain injury with moderate hypothermia. New England Journal of Medicine, 336, 540-546.

Marshall, L. F., Marshall, S. B., Klauber, M. R., Clark, B. M., Eisenberg, H. M., Jane, J. A., et al. (1991). A new classification of head injury based on computerized tomography. Journal of Neurosurgery, 75S, S14-S20.

McIntyre, L. A., Ferguson, D. A., Hebert, P. C., Moher, D. M., & Hutchinson, J. S. (2003). Prolonged therapeutic hypothermia after traumatic brain injury in adults: A systematic review. JAMA, 289, 2992-2999.

Odette, H. A., Colford, J. M., Good, M. C., & Matz, P. G. (2002). The role of hypothermia in the management of severe brain injury: A meta-analysis. Archives of Neurology, 59, 1077-1083.

Peterson, K., Carson, S., & Carney, N. (2008). Hypothermia treatment for traumatic brain injury: A systematic review and meta-analysis. Journal of Neurotrauma, 25, 62-71.

Polderman, K. H., Tjong Tjin Joe, R., Peerdeman, S. M., Vandertop, W. P., & Girbes, A. R. (2002). Effects of therapeutic hypothermia on intracranial pressure an outcome in patients with severe head injury. Intensive Care Medicine, 28, 1563-1573.

Shiozaki, T., Hayakata, T., Taneda, M., Nakajima, Y., Hashiguchi, N., Fujimi, S., et al. (2001). A multicenter prospective randomized controlled trial of the efficacy of mild hypothermia for severely head injured patients with low intracranial pressure. Journal of Neurosurgery, 94(1), 50-54.

Tokutomi, T., Morimoto, K., Miyagi, T., Yamaguchi, S., Ishikawa, K., & Shigemori, M. (2003). Optimal temperature for the management of severe traumatic brain injury: Effect of hypothermia on intracranial pressure, systemic and intracranial hemodynamics, and metabolism. Neurosurgery, 52(1), 102-112.

Wright, J. E. (2005). Therapeutic hypothermia in traumatic brain injury. Critical Care Nursing Quarterly, 28, 150-161.

© 2009 American Association of Neuroscience Nurses


Article Level Metrics