Traumatic Brain Injury: Treatment with Mild Prolonged Hypothermia

Introduction: Hypothermia has been used as a method of brain protection in patients with traumatic brain injury (TBI) for many years. The protective effects of hypothermia are related to the inhibition of the excitatory amino acids (EAA) release including glutamate. The hypothermic decline of the cerebral metabolic rate of oxygen is also another mechanism of brain protection because it maintains the aerobic metabolism of the brain. Aim of the work: To study the effect of mild hypothermia on the outcome and complications in severe head trauma. Patients and methods: Ninety four patients (16-60 years old) with severe head trauma (Glasgow coma scale ≤ 8) were classified into hypothermic group (n=47) and normothermic group (n=47). The jugular venous bulb oxygen saturation, the jugular venous lactate, the Glasgow outcome scale at hospital discharge (GOS-HD), the length of ICU stay and other variables were measured and recorded. Results: Mild hypothermia improved the jugular venous bulb oxygen saturation, decreased the jugular venous lactate, with no effect on (GOS-HD). Conclusion: Prophylactic mild hypothermia lacks any effect on the outcome in patients with TBI, although there is improvement in the cerebral oxygenation and decrease the oxygen extraction and prevention of the anaerobic metabolism by decreasing the level of serum lactate. A subgroup of patients with intracranial hematoma may benefit from this type of treatment.


Introduction
Hypothermia has been widely recognized as an effective cytoprotectant. It has been studied to improve outcome from myocardial infarction, organ transplantation, cardiopulmonary bypass, spinal cord injury, intestinal ischemia, and neonatal hypoxia-ischemia [1]. Traumatic brain injury (TBI) has been shown to cause marked elevation in glutamate concentrations, which is responsible through excitotoxicity for neuronal injury or death, making secondary injury as potentially lethal as primary injury [2].
Hypothermia's protective effect has long been thought to be due to preservation of metabolic stores, as cooling the brain leads to approximately 5% reduction in oxygen and glucose consumption per degree centigrade [3].
The aim of this work is to study the effect of mild hypothermia on the outcome and complications in severe head trauma.

Patients and Methods
This study was carried out in the trauma ICU at Assiut University Hospital. The research protocol was approved by the Clinical Research Ethics Committee of our university. In this randomized controlled clinical trial, we compared the effect of mild prolonged hypothermia versus normothermia in 94 adult patients with severe TBI with a score ≤ 8 on Glasgow coma scale (GCS).
We excluded patients with major systemic injuries, prolonged hypoxemia, sustained hypotension, previous C.N.S lesion, pregnancy and patients with hepatic, renal or diabetic diseases.
Patients met our entry criteria were assigned randomly into one of two groups: HT group (mild hypothermia group) or NT group (normothermia group) using computer generated random number table. An informed consent was obtained from patients guardians.
Patients of both groups were admitted within 6 hours to the trauma ICU. Arterial line, jugular venous bulb (JVB) and central venous catheters were inserted for all patients. Management of the patients were according to our institutional protocol for head injured patients including intubation and mechanical ventilation with sedation and relaxation for at least 48 hours, dehydrating measures, arterial CO 2 level (PaCO 2 ) was kept around 35 mmHg, and serum glucose was controlled below 200 mg/dl. Induction of cooling in HT group performed with both invasive and non-invasive techniques, invasive techniques included injection of intravascular cold saline 8 º C at the rate of 30 ml/kg/hour and nasogastric lavage with iced saline. Non-invasive techniques included skin exposure, fans, application of ice packs, refrigerated cooling pads. Once the nasogastric temperature reached 34ºC, invasive maneuvers stopped and non-invasive were continued to maintain temperature between 34 º C and 36 º C for the next 72 hours. The invasive maneuvers may be used again if non-invasive failed to maintain the target temperature. After 72 hours of cooling, patients were passively rewarmed over a period of 6 hours to a temperature not more than 37.5°C by discontinuation of the above mentioned measures.
For patients in NT group, temperature was kept between 37°C and 37.5°C for the entire 3 days after trauma. Any drop in temperature below desired level in both groups was treated using warming blankets.
For measurement of JVB blood oxygenation, the internal jugular vein was cannulated in the cephalad direction, and the position of the catheter tip was confirmed by either a lateral or an antero-posterior (AP) neck radiograph. On the lateral film, the catheter tip must be above the disc of C1/C2 and as close to the skull base as possible. On AP view, a correctly placed tip should lie cranial to a line extending from the atlanto-occipital joint space and caudal to the lower margin of the orbit. The JVB samples were withdrawn at a rate of 0.5 ml/min to avoid extracranial contamination [4]. For estimation of the arterialjugular oxygen content difference (AJDO 2 ) we used the following equations: AJDO 2 (ml.dl -1 ) = arterial oxygen content -jugular venous oxygen content AJDO 2 = (Hb × 1.39 × SaO 2 /100 +0.003 × PaO 2 ) -(Hb × 1.39 × SjO 2 +0.003× PjO 2 ).

Data collection
Patients' characteristics as age, sex, GCS on admission were recorded. Mean arterial blood pressure, heart rate, arterial pH, arterial oxygen tension and temperature were recorded every day for the first 3 days. JVB samples for ABG and JVB lactate were taken before hypothermia and after 24, 48, and 72 hours, arterial and JVB blood samples were used for estimating AJDO 2 .
Hemoglobin level, white blood cell count, prothrombin time, serum potassium, serum creatinine were recorded at the end of the third day.
At discharge from ICU: duration of mechanical ventilation, length of ICU stay and incidence of complications were recorded.
At hospital discharge: length of hospital stay and Glasgow outcome score [5] was recorded (Table 1).

Death
Severe injury or death without recovery of consciousness 2. Persistent vegetative state Severe damage with prolonged state of unresponsiveness and a lack of higher mental functions 3. Severe disability Severe injury with permanent need for help with daily living 4. Moderate disability No need for assistance in everyday life, employment is possible but may require special equipment.

Low disability
Light damage with minor neurological and psychological deficits.

Statistical analysis
The Statistical Package for the Social Sciences software version 19 (SPSS) was used for statistical analysis; Comparisons of nominal data were performed using chi-square test. Continuous data are presented showing the mean ± standard deviation.
Their comparison was performed using the Student's t-test and Mann-Whitney test as appropriate. Statistical significance was accepted at a probability level of less than 0.05

Patients' characteristics
Over a period of one year, we included 94 patients who met the eligibility criteria. They were randomly assigned to hypothermia and normothermic therapy. Baseline characteristics of the patients who were enrolled in the study are presented in Table 2.
There were no significant differences observed between both groups as regard age, sex, BMI, GCS on admission, and trauma severity score. Data were represented as (mean ± SD) unless otherwise indicated. P-value<0.05 is considered significant

Laboratory data
We only observed a statistically significant prolongation in the prothrombin time in the HT group compared to NT group (

Jugular venous bulb data
JVB O 2 saturation showed significant increase (p<0.001) at all studied times in both groups compared with baseline readings. JVB O 2 saturation was higher in HT group compared to NT group which is statistically significant (Figure 1). AJDO 2 difference (arterio-jugular oxygen difference): AJDO 2 difference also decreased significantly (p<0.001) at all studied times when comparing each group with the baseline readings. AJDO 2 difference was lower in HT group compared to NT group which is statistically significant (Figure 2). JVB lactate: Serum lactate level decreased significantly (p<0.001) at all studied times in HT group NT group when compared with baseline readings JVB lactate was lower in HT group compared to NT group which is statistically significant (Figure 3).

Study Outcomes
On the basis of GOS-HD we classified patients into those with favorable outcome and unfavorable outcome. Patients with favorable outcome include those with good recovery and moderate disability on GOS-HD. Patients with unfavorable outcome include patients with severe disability, vegetative state and death. Data for primary outcomes were available for all patients. Twenty three patients out of 47 patients (49%) in the HT group and 19 out of 47 patients (40%) in the NT group had a favorable outcome at hospital discharge with no significant difference between both groups(p-value=0.214). In a subgroup analysis, in patients with diffuse brain injury we found that 13 out of 29 (45%) of patients with diffuse brain injury in the HT group had a favorable outcome compared to 11 out of 27 (41%) in NT group with no significant difference between both groups (p-value=0.793). On the other hand for patients with intracranial hematoma 12 of 18 (67%) of patients in HT group had a favorable outcome compared to 8 out of 20 (40%) of patients in NT group, although it still statistically insignificant (p-value= 0.119) there is a strong trend toward favorable outcome in HT group in this type of injury and perhaps with larger sample size a significant level may be achieved.
In a subgroup analysis of the patients on the basis of GCS at time of admission, in patient with GCS (between 3 and 5), 46% of patients in HT group had favorable outcome compared to 35% in NT group. In patients with GCS on admission between 6 and 8, 62% on HT group had favorable outcome compared to 48% in NT group. No statistically significant difference observed between both groups (Tables 4 and 5).   As regard duration of mechanical ventilation, length of ICU stay and length of hospital stay there were no significant differences between both groups.

GOS-HD
As regard complications, there are minimal complications in both groups which can be just attributed to the hypothermia technique (

Discussion
Hypothermia slows down metabolic processes and thereby inhibits deleterious biochemical events such as free radical production [6] and excitatory amino acid release including glutamate [7].
In the present study 94 patients with severe head trauma (GCS ≤ 8) were enrolled. We used the mild type of hypothermia (34-36°C) within 6 hours of patients' admission and lasting for 72 hours as a neuroprotective technique.
The results of our study showed that mild hypothermia was effective in decreasing the JVB lactate level, increasing the JVB oxygen saturation and decrease the AJDO 2 difference. No significant differences were found as regard the duration of mechanical ventilation, length of ICU and hospital stay. As regard neurological outcome, despite non-significant differences between both groups, there is a trend toward good outcome in HT group.
The increased intracranial pressure associated with rewarming may explain the bad neurological outcome. This conflict about the outcome may be due to several factors as the scale used to measure the outcome, the timing for measurement that outcome, the definition of good and bad outcome, the protocol for hypothermia may differ according to the time of initiation of hypothermia, the degree of hypothermia used, the duration of application of hypothermia, the time to reach the coldest temperature and the rate of rewarming. Patient factors include the type and severity of the trauma, and the associated injuries.
In our study we applied mild degree of hypothermia for 72 hours; better outcomes were obtained in many studies where hypothermia was applied for more than two days [15][16][17].
In our study a subgroup of patients with intracranial hematoma may benefit from mild prolonged hypothermia. This correlation between brain pathology and neurological outcome was documented in many studies with the favor of hematoma type over the diffuse brain injury [18,19]. The reperfusion injury occurred after decrease in the intracranial pressure is the pathophysiologic mechanism responsible for secondary brain injury in this type of patients [20] which is similar to the pathophysiologic mechanism of brain injury in cardiac arrest patients who has been successfully treated with hypothermia [21].
Adequacy of brain perfusion after TBI is one of factors that prevent secondary brain injury. AJDO2 inversely correlated to the ratio of cerebral blood flow and Oxygen metabolic rate [22]. In traumatic brain injury AJDO 2 was measured for targeted therapies as for PaCO 2 level [23] or mannitol therapy [24]. The profound episodes of desaturation were associated with poor neurological outcome [25].
Both groups in our study showed comparable rates of complications. Pneumonia was the most frequent complication occurred in our patients followed by cardiovascular complications as arrhythmias and hypotension.
Nearly all body organs are affected to a various degree by hypothermia and minimal organ affection may be hazardous in critically ill patients. Cardiovascular system will suffer from increased incidence of arrhythmia [26], increased venous pressure and systemic vascular resistance [27,28] and increased plasma level of catecholamines [29,30]. The diuretic effect of cooling may results in low electrolytes level. Magnesium ion depletion may play a role in increasing the brain injury following TBI [31,32].
A limitation of the study is the lack of blinding to the intervention, which is problematic in all trials of therapeutic hypothermia. However, because cooling to normothermia was permitted in the standard-care group, it is possible that there was masking of the intervention to patients and relatives in some cases. Outcome scoring was blinded.

Conclusion
Prophylactic mild Hypothermia lacks any effect on the outcome in patients with TBI although there is improving in the cerebral oxygenation and decrease the oxygen extraction and preventing the anaerobic metabolism by decreasing the level of serum lactate. A subgroup of patients with intracranial hematoma may benefit from this type of treatment.