Early hyperthermia after traumatic brain injury in children: risk factors, influence on length of stay, and effect on short-term neurologic status.

OBJECTIVES: a) To determine the risk factors for early hyperthermia after traumatic brain injury in children; b) to identify the contribution of early hyperthermia to neurologic status at pediatric intensive care unit (PICU) discharge and to PICU length of stay in head-injured children. STUDY DESIGN: Observational cohort study. SETTING: PICU at a tertiary care, university medical center. PATIENTS: Children (n = 117) admitted to a PICU from July 1995 to May 1997 with traumatic brain injury. These children had a median age of 5.4 yrs (3 wks to 15.2 yrs old), and 33.4% were girls. MEASUREMENTS AND MAIN RESULTS: Early hyperthermia (temperature >38.5 degrees C within the first 24 hrs of admission) occurred in 29.9% of patients admitted to the PICU with traumatic brain injury. Risk factors predicting early hyperthermia included Glasgow Coma Scale score in the emergency department < or =8, pediatric trauma score < or =8, cerebral edema or diffuse axonal injury on initial head computed tomography scan, admission blood glucose >150 mg/dL (8.2 mmol/L), admission white cell count >14,300 cells/mm3 (14.3 x 10(9) cells/L), and systolic hypotension. The presence of early hyperthermia significantly increased the risk for Glasgow Coma Scale score <13 at PICU discharge (odds ratio [OR] 9.7, 95% confidence interval [CI] 2.8, 24.4) and PICU stay > or =3 days (OR 13.8, CI 5.1, 37.5). When we used multiple logistic regression models including injury severity and hypotension, early hyperthermia remained an independent predictor of lower Glasgow Coma Scale score at PICU discharge (OR 4.7, CI 1.4, 15.6) and longer PICU length of stay (OR 8.5, CI 2.8, 25.6). CONCLUSIONS: Early hyperthermia is independently associated with a measure of early neurologic status and resource utilization in children with traumatic brain injury serious enough to require PICU admission. These results support the prevention of hyperthermia in the management of traumatic brain injury in children. Further research is required to understand the mechanisms of this response and to identify appropriate preventive or therapeutic interventions.

Effect of arrest time and cerebral perfusion pressure during cardiopulmonary resuscitation on cerebral blood flow, metabolism, adenosine triphosphate recovery, and pH in dogs.

OBJECTIVES: To test the hypothesis that greater cerebral perfusion pressure (CPP) is required to restore cerebral blood flow (CBF), oxygen metabolism, adenosine triphosphate (ATP), and intracellular pH (pHi) levels after variable periods of no-flow than to maintain them when cardiopulmonary resuscitation (CPR) is started immediately. DESIGN: Prospective, randomized, comparison of three arrest times and two perfusion pressures during CPR in 24 anesthetized dogs. SETTING: University cerebral resuscitation laboratory. INTERVENTIONS: We used radiolabeled microspheres to determine CBF and magnetic resonance spectroscopy to derive ATP and pHi levels before and during CPR. Ventricular fibrillation was induced, epinephrine administered, and thoracic vest CPR adjusted to provide a CPP of 25 or 35 mm Hg after arrest times of O, 6, or 12 mins. MEASUREMENTS AND MAIN RESULTS: When CPR was started immediately after arrest with a CPP of 25 mm Hg, CBF and ATP were 57 +/- 10% and 64 +/- 14% of prearrest (at 10 mins of CPR). In contrast, CBF and ATP were minimally restored with a CPP at 25 mm Hg after a 6-min arrest time (23 +/- 5%, 16 +/- 5%, respectively). With a CPP of 35 mm Hg, extending the no-flow arrest time from 6 to 12 mins reduced reflow from 71 +/- 11% to 37 +/- 7% of pre-arrest and reduced ATP recovery from 60 +/- 11% to 2 +/- 1% of pre-arrest. After 6- or 12-min arrest times, brainstem blood flow was restored more than supratentorial blood flow, but cerebral pHi was never restored. CONCLUSIONS: A CPP of 25 mm Hg maintains supratentorial blood flow and ATP at 60% to 70% when CPR starts immediately on arrest, but not after a 6-min delay. A higher CPP of 35 mm Hg is required to restore CBF and ATP when CPR is delayed for 6 mins. After a 12-min delay, even the CPP of 35 mm Hg is unable to restore CBF and ATP. Therefore, increasing the arrest time at these perfusion pressures increases the resistance to reflow sufficient to impair restoration of cerebral ATP.

Effect of the no-flow interval and hypothermia on cerebral blood flow and metabolism during cardiopulmonary resuscitation in dogs.

BACKGROUND AND PURPOSE: We sought (1) to determine the effect of brief periods of no flow on the subsequent forebrain blood flow during cardiopulmonary resuscitation (CPR) and (2) to test the hypothesis that hypothermia prevents the impact of the no-flow duration on cerebral blood flow (CBF) during CPR. METHODS: No-flow intervals of 1.5, 3, and 6 minutes before CPR at brain temperatures of 28 degreesC and 38 degreesC were compared in 6 groups of anesthetized dogs. Microsphere-determined CBF and metabolism were measured before and during vest CPR adjusted to maintain cerebral perfusion pressure at 25 mm Hg. RESULTS: Increasing the no-flow interval from 1.5 to 6 minutes at 38 degreesC decreased the CBF (18. 6+/-3.6 to 6.1+/-1.7 mL/100 g per minute) and the cerebral metabolic rate (2.1+/-0.3 to 0.7+/-0.2 mL/100 g per minute) during CPR. Cooling to 28 degreesC before and during the arrest eliminated the detrimental effects of increasing the no-flow interval on CBF (16. 8+/-1.0 to 14.8+/-1.9 mL/100 g per minute) and cerebral metabolic rate (1.1+/-0.1 to 1.3+/-0.1 mL/100 g per minute). Unlike the forebrain, 6 minutes of preceding cardiac arrest did not affect brain stem blood flow during CPR. CONCLUSIONS: Increasing the no-flow interval to 6 minutes in normothermic animals decreases the supratentorial blood flow and cerebral metabolic rate during CPR at a cerebral perfusion pressure of 25 mm Hg. Cooling to 28 degreesC eliminates the detrimental impact of the 6-minute no-flow interval on the reflow produced during CPR. The brain-protective effects of hypothermia include improving reflow during CPR after cardiac arrest. The effect of hypothermia and the impact of short durations of no flow on reperfusion indicate that increasing viscosity and reflex vasoconstriction are unlikely causes of the "no-reflow" phenomenon.

Detection of intravascular injection of regional anaesthetics in children.

PURPOSE: Detection of intravascular injection of local anaesthetic during placement of regional blocks in children by using epinephrine-induced tachycardia or hypertension may produce false positive and false negative findings. This study evaluates ECG changes as markers of intravascular injection of local anaesthetics with epinephrine, during placement of epidural blocks in children. METHODS: Observational study in a teaching hospital of all epidural anaesthetics administered to paediatric patients during one year. General anaesthesia, where used, was not controlled. An ECG rhythm strip was recorded during test dose injection and analyzed for changes in rate, rhythm, and T-wave configuration. RESULTS: During the study period, 742 paediatric epidural blocks were administered. There were 644 caudal (284 without catheters), 97 lumbar, and one thoracic epidural anaesthetics. Satisfactory placement was achieved in 97.7% of patients. Intravascular injection was detected in 42 (5.6%) of epidural anaesthetics (3.8% and 6.7% of straight needle and catheter injections, respectively). Detection was by immediate aspiration of blood in six patients, and by heart rate increases > 10 bpm in 30. Five had heart rate decreases suggesting a baroreceptor response. Five had heart rate increases < 10 bpm that were possible responses to noxious stimuli. Of 30 patients with known intravascular injection and for whom ECG strips were available, 25 (83%) had T-wave amplitude increases > 25%, and 29 (97%) had ECG changes in T-wave or rhythm in response to the epinephrine injection. There were no false positives. CONCLUSION: In order to reduce risks associated with epidural anaesthesia in children, epinephrine should be added to the local anaesthetic test dose, the ECG should be monitored continuously for changes in heart rate, rhythm, and T-wave amplitude. Epidural injections should be given in small increments.

Acidemia and brain pH during prolonged cardiopulmonary resuscitation in dogs.

BACKGROUND AND PURPOSE: Cardiopulmonary resuscitation (CPR) generating low perfusion pressures and beginning immediately after cardiac arrest maintains cerebral ATP but not cerebral pH or arterial pH. We tested the hypothesis that preventing severe arterial acidemia prevents cerebral acidosis, whereas augmenting arterial acidemia augments cerebral acidosis. METHODS: In dogs anesthetized with pentobarbital and fentanyl, cerebral pH and ATP were measured with 31P MR spectroscopy and blood flow was measured with radiolabeled microspheres. A pneumatically controlled vest was placed around the thorax, and chest compressions were begun immediately after electrically induced cardiac arrest. Cerebral perfusion pressure was maintained with epinephrine at 30 mm Hg for 90 minutes. The arterial acidemia observed during CPR was untreated in a control group, corrected to a pH of 7.3 with the use of sodium bicarbonate, or maintained below pH 6.5 with intravenous lactic acid after 14 minutes of CPR. RESULTS: At 10 minutes of CPR, cerebral ATP (99 +/- 1.5%, control), blood flow (35 +/- 3 mL/min per 100 g), O2 consumption (4.0 +/- 0.2 mL/min per 100 g), and cerebral pH (7.05 +/- .03) were unchanged from prearrest values (mean +/- SEM). After 10 minutes of CPR in the control group, cerebral pH progressively fell (6.43 +/- 0.10 at 90 minutes) in parallel with cerebral venous pH. In the bicarbonate group cerebral pH was maintained higher (6.91 +/- 0.08). Cerebral blood flow, O2 consumption, and ATP were sustained near prearrest values in both groups. In the lactate group, however, the rate of decrease of cerebral pH was augmented (6.47 +/- 0.06 by 30 minutes), and cerebral blood flow and metabolism were significantly reduced. CONCLUSIONS: Cerebral pH decreased in parallel with blood pH when resuscitation was started immediately upon arrest even when cerebral O2 consumption and blood flow were near normal. Although cerebral metabolism was near normal during the first hour of CPR, systemic bicarbonate administration ameliorated the cerebral acidosis. This finding indicates that the blood-brain pH gradient is important at the subnormal cerebral perfusion pressures seen in CPR.

Somatosensory evoked potential and brain water content in post-asphyxic immature piglets.

Depression of somatosensory evoked potentials (SEP) after a single episode of complete asphyxia with near cardiac arrest was evaluated to determine whether persistent SEP depression is related to postresuscitation edema in cortical gray matter or subcortical white matter. Piglets (< 7 d of age) were anesthetized with sodium pentobarbital and fentanyl. Asphyxia was produced by occlusion of the endotracheal tube for 7 min. Arterial O2 saturation fell to 5%. Resuscitation was achieved in < 2 min with ventilation, epinephrine, and chest compressions. Regional brain water content was determined from the difference between wet and dry weight. Two control groups were also analyzed; one immediately after (n = 5) and one 6 h after induction (n = 7) of anesthesia. SEP amplitude became isoelectric during asphyxia and recovered to 50 +/- 13% (n = 7) of baseline 6 h after resuscitation. In the 6-h control group, SEP amplitude remained above baseline. The percent water content (mean +/- SEM) among the three groups (asphyxia versus time control versus brief anesthesia control) was not different in the cortical gray matter (83.0 +/- 0.7% versus 82.4 +/- 0.4% versus 83.2 +/- 0.3%) or subcortical white matter (75.6 +/- 0.8% versus 74.8 +/- 0.9% versus 75.6 +/- 0.5%). In seven other piglets, cerebral blood flow and O2 consumption recovered to baseline by 1 h after asphyxia. Therefore, we found that the sustained depression of SEP amplitude, after 7 min of asphyxia in immature piglets, is not related to brain edema or persistent decreases in global cerebral O2 consumption.

Effect of vest cardiopulmonary resuscitation on cerebral and coronary perfusion in an infant porcine model.

OBJECTIVES: To determine cerebral and myocardial blood flow rates during vest cardiopulmonary resuscitation (CPR) without direct cardiac compression in an infant porcine model. Also, to determine if circumferential chest compression without the chest deformity ordinarily associated with precordial compression maintains cerebral and myocardial blood flow rates during prolonged CPR. Finally, to establish the effect of compression rate and duty cycle on cerebral and myocardial blood flow rates during vest CPR in this model. DESIGN: Prospective, randomized comparison of two compression rates and two duty cycles in four groups during prolonged CPR. SETTING: University cerebral resuscitation laboratory. SUBJECTS: Thirty-two infant domestic swine. INTERVENTIONS: Microsphere-determined cerebral and myocardial blood flow rates, perfusion pressures, and chest dimensions, were measured before and during prolonged vest CPR. Immediately after ventricular fibrillation, epinephrine administration was started and thoracic vest CPR was performed using a single combination of compression rates of 100 or 150/min and duty cycles of 30% or 60%. Measurements were made before and at 5, 10, 20, 35, and 50 mins of CPR. MEASUREMENTS AND MAIN RESULTS: Five minutes into CPR, between-group comparisons showed that cerebral blood flow was 16 to 20 mL/min/100 g and myocardial blood flow was 34 to 45 mL/min/100 g (48% to 62% and 25% to 33% of prearrest values). When CPR was prolonged, cerebral blood flow deteriorated similarly in all groups. Myocardial blood flow decreased over time but was better maintained in the groups with a 30% duty cycle (24 vs. 4 mL/min/100 g; p < .006). There were no differences between the two compression rates. Chest deformity after cessation of 50 mins of compression was < 3%. CONCLUSIONS: Cerebral and myocardial blood flow rates produced by vest CPR are comparable with rates reported using other types of CPR in this model. Deterioration in blood flow during prolonged CPR occurs despite a lack of chest deformation. The deterioration in myocardial blood flow during prolonged CPR is greater when a long duty cycle is used in this model.

Reduced blood-brain barrier permeability after cardiac arrest by conjugated superoxide dismutase and catalase in piglets.

BACKGROUND AND PURPOSE: Cardiac arrest and resuscitation in immature piglets result in a delayed increase in blood-brain barrier permeability. We tested the hypothesis that pretreatment with oxygen radical scavengers reduces postischemic permeability. METHODS: Permeability was assessed by measuring the plasma-to-brain transfer coefficient of the small amino acid, alpha-aminoisobutyric acid, in 2- to 3-week-old anesthetized piglets. Three groups were studied: (1) a nonischemic time control group (n = 5), (2) an ischemia group (n = 8) pretreated with 5 mL of polyethylene glycol vehicle, and (3) an ischemia group (n = 8) pretreated with polyethylene glycol conjugated to superoxide dismutase (10,000 U/kg) and to catalase (20,000 U/kg). The ischemia protocol consisted of 8 minutes of ventricular fibrillation, 6 minutes of cardiopulmonary resuscitation, defibrillation, and 4 hours of spontaneous circulation. RESULTS: The mean +/- SEM of the transfer coefficient of alpha-aminoisobutyric acid in cerebrum was (in microL/g per minute): 1.54 +/- 0.37 in the nonischemic group, 2.04 +/- 0.26 in the ischemia group treated with vehicle, and 1.29 +/- 0.25 in the ischemia group treated with oxygen radical scavengers. Postischemic values with scavenger treatment were significantly lower than those with vehicle treatment in cerebrum, cerebellum, medulla and cervical spinal cord. CONCLUSIONS: Pretreatment with oxygen radical scavengers reduces postischemic blood-brain barrier permeability by a small amino acid. These data are consistent with oxygen radical-mediated dysfunction of cerebral endothelium in a pediatric model of cardiopulmonary resuscitation.

Effect of cerebral blood flow generated during cardiopulmonary resuscitation in dogs on maintenance versus recovery of ATP and pH.

BACKGROUND AND PURPOSE: Cardiopulmonary resuscitation with external chest compression generates low perfusion pressures that may be inadequate for restoring cerebral metabolism and may worsen intracellular pH. We tested the hypothesis that cerebral reperfusion with a low perfusion pressure after arrest restores brain adenosine triphosphate (ATP) and pH to levels attained at the same perfusion pressure without preceding complete ischemia. METHODS: Brain ATP and intracellular pH were measured by magnetic resonance spectroscopy, and cerebral blood flow was measured with microspheres in anesthetized dogs. External chest compressions were begun in group A (n = 6) immediately after the onset of arrest (ie, arrest time zero) and in group B (n = 10) after 6 minutes of arrest (ie, arrest time 6 minutes). In both groups, mean cerebral perfusion pressure was regulated at 30 mm Hg for 70 minutes by adjustment of inflation pressure of a pneumatic thoracic vest. RESULTS: At 12 minutes of resuscitation, cerebral blood flow was 27 +/- 4 mL/min per 100 g in group A and 21 +/- 4 mL/min per 100 g in group B, but ATP in group B (58 +/- 10% of prearrest) was less than in group A (105 +/- 6%). With prolonged resuscitation, ATP deteriorated to near zero levels in dogs in group B, with blood flow less than 15 mL/min per 100 g. Dogs with greater blood flow never achieved complete metabolic recovery. In group B, intracellular pH was unchanged from the 6.3 value at the start of resuscitation, even in those dogs with extremely low blood flows. CONCLUSIONS: Levels of cerebral perfusion pressure sufficient to maintain cerebral oxidative metabolism without complete ischemia during cardiopulmonary resuscitation are not sufficient to restore metabolism after complete ischemia during cardiopulmonary resuscitation. However, low "trickle" blood flow did not worsen intracellular acidosis.

Brain bioenergetics during cardiopulmonary resuscitation in dogs.

Cardiac arrest causes a rapid loss of cerebral adenosine triphosphate [corrected] (ATP) and a decrease in cerebral intracellular pH (pHi). Depending on the efficacy of cardiopulmonary resuscitation (CPR), cerebral blood flow levels (CBF) ranging from near zero to near normal have been reported experimentally. Using 31P magnetic resonance spectroscopy, the authors tested whether experimental CPR with normal levels of cerebral blood flow can rapidly restore cerebral ATP and pHi despite the progressive systemic acidemia associated with CPR. After 6 min of ventricular fibrillation in six dogs anesthetized with fentanyl and pentobarbital, ATP was reduced to undetectable concentrations and pHi decreased from 7.11 +/- 0.02 to 6.28 +/- 0.09 (+/- SE) as measured by 31P magnetic resonance spectroscopy. Application of cyclic chest compression by an inflatable vest placed around the thorax and infusion of epinephrine (40 micrograms/kg bolus plus 8 micrograms/kg/min, intravenously) maintained cerebral perfusion pressure greater than 70 mmHg for 50 min with the dog remaining in the magnet. Prearrest cerebral blood flows were generated. Cerebral pHi recovered to 7.03 +/- 0.03 by 35 min of CPR, whereas arterial pH decreased from 7.41 +/- 0.4 to 7.08 +/- 0.04 and cerebral venous pH decreased from 7.29 +/- 0.03 to 7.01 +/- 0.04. Cerebral ATP levels recovered to 86 +/- 7% (+/- SE) of prearrest concentration by 6 min of CPR. There was no further recovery of ATP, which remained significantly less than control. Therefore, in contrast to hyperemic reperfusion with spontaneous circulation and full ATP recovery, experimental CPR may not be able to restore ATP completely after 6 min of global ischemia despite restoration of CBF and brain pHi to prearrest levels.

Effect of adrenergic drugs on cerebral blood flow, metabolism, and evoked potentials after delayed cardiopulmonary resuscitation in dogs.

BACKGROUND AND PURPOSE: Epinephrine administration during cardiopulmonary resuscitation increases cerebral blood flow by increasing arterial pressure. We tested whether potential beta-adrenergic effects of epinephrine directly influence cerebral blood flow and oxygen consumption independently of raising perfusion pressure. METHODS: Four groups of seven anesthetized dogs were subjected to 8 minutes of fibrillatory arrest followed by 6 minutes of chest compression, ventricular defibrillation, and 4 hours of spontaneous circulation. Cerebral perfusion pressure was increased to approximately equivalent ranges during resuscitation by either 1) epinephrine infusion, 2) epinephrine infusion after pretreatment with the lipophilic beta-adrenergic antagonist pindolol, 3) infusion of the alpha-adrenergic agonist phenylephrine, or 4) descending aortic balloon inflation without pressor agents. RESULTS: We found no difference in cerebral blood flow, oxygen extraction, or oxygen consumption during chest compression among groups. After ventricular defibrillation, depressed levels of cerebral blood flow, cerebral oxygen consumption, and somatosensory evoked potential amplitude were not different among groups. CONCLUSIONS: We detected no evidence that after 8 minutes of complete ischemia, epinephrine administration during resuscitation substantially influences cerebral blood flow or cerebral oxygen consumption independent of its action of raising arterial pressure or or that epinephrine has a negative impact on immediate metabolic or electrophysiological recovery attributable to its beta-adrenergic activity.

Blood-brain barrier disruption after cardiopulmonary resuscitation in immature swine.

We investigated blood-brain barrier permeability in 2-3-week-old anesthetized pigs during and after cardiopulmonary resuscitation. We assessed permeability by tissue uptake of radiolabeled aminoisobutyric acid, after correcting for plasma counts in tissue with radiolabeled inulin. Among 14 regions examined, the transfer coefficient of aminoisobutyric acid in nonischemic control animals ranged from 0.0018 +/- 0.0001 ml/g/min in diencephalon to 0.0049 +/- 0.0003 ml/g/min in cervical spinal cord. After 8 minutes of cardiac arrest followed by either 10 or 40 minutes of continuous sternal compression, there was no increase in the transfer coefficient. Likewise, during the immediate period after ventricular defibrillation, there was no increase in transfer coefficient despite the brief, transient hypertension. However, after 8 minutes of arrest, 6 minutes of cardiopulmonary resuscitation, and 4 hours of spontaneous circulation, the transfer coefficient was significantly increased by 59-107% in 10 of 11 regions rostral to the pons. Plasma volume in tissue measured by inulin was not elevated, suggesting that the increased transfer coefficient was not due to increased surface area. Thus, after an 8-minute period of complete ischemia, the blood-brain barrier remains intact during and immediately after resuscitation despite large vascular pressure fluctuations. However, in contrast to previous work on adult dogs, immature pigs are prone to a delayed increase in permeability, thereby allowing circulating substances greater access to the brain.

Blood-brain barrier integrity during cardiopulmonary resuscitation in dogs.

Blood-brain barrier integrity during cardiopulmonary resuscitation may be important because of the potential effects of adrenergic agonists administered during arrest on cerebral metabolism and the cerebral vasculature. As an index of blood-brain barrier permeability to small molecules, we measured the brain uptake of [14C]alpha-aminoisobutyric acid during a 10-minute period in 25 anesthetized dogs. To correct for the amount of carbon-14 label in the plasma space, we administered [3H] inulin 2 minutes before death. The mean transfer coefficient in 14 brain regions of five control dogs ranged from 0.002 to 0.007 ml/g/min. After 8 (n = 15) or 15 (n = 5) minutes of cardiac arrest, external chest compression was instituted to maintain aortic blood pressure above 60 mm Hg. The transfer coefficient was not elevated during chest compression (n = 10), immediately following defibrillation (n = 5), or 4 hours after resuscitation (n = 5); in some brain regions the transfer coefficient decreased. However, the decrease in the transfer coefficient was proportional to the decrease in the cerebral plasma volume as measured by the ratio of the [3H]inulin concentration in the tissue to that in the plasma. Thus, it is unlikely that a decrease in capillary surface area masked an increase in blood-brain barrier permeability. Therefore, we found no evidence of blood-brain barrier disruption during or after cardiopulmonary resuscitation in dogs. Despite the large phasic increases in sagittal sinus pressure associated with external chest compression, concurrent increases in cerebrospinal fluid pressure apparently protect the microcirculation from increased transmural pressure.

Blood-brain barrier disruption following CPR in piglets.

We studied blood-brain barrier (BBB) integrity in immature piglets during the following cardiopulmonary resuscitation (CPR). As in our previous work in the dog, there was no disruption during CPR after eight minutes of cardiac arrest, or immediately following resuscitation using a small molecule, alpha-aminoisobutyric acid. However, unlike the dog, where the BBB remained intact, we found delayed disruption of the BBB four hours after resuscitation. Young animals may be more prone to a delayed increase in BBB permeability after cardiac arrest and CPR.