Cerebral ischemia may have a role in the pathogenesis of traumatic brain injury in humans [1-5], although its full significance is yet to be determined. Over the past 20 yr, histologic studies of human brains have provided circumstantial evidence that cerebral ischemia occurred after traumatic brain injury [6,7]; however, studies where cerebral blood flow (CBF) was measured have not always confirmed the histologic findings. [For a review of the subject, see Bouma and Muizelaar ]. The reasons that ischemia may have been difficult to detect with blood flow measurements are probably threefold: First, in most studies, CBF was not measured in the early hours after injury (12-15 h), a time when the frequency of ischemic episodes may be the greatest [3-5]. Second, the occurrence of ischemia appears to be dependent on the severity of injury. The ischemic episodes were predominantly found in the more severely injured patients, classified as persistently vegetative or those who died, whereas patients who suffered less severe injury, including those with severe deficits, showed infrequent episodes of ischemia . Unless the clinical studies focused on the more severely injured patients, ischemia would not have been as likely found. Third, the methods used to measure CBF often lacked the spatial resolution necessary to detect focal ischemia when present . Thus, on at least three accounts, ischemia associated with traumatic brain injury may have eluded researchers.
Animal models have been developed for the study of traumatic brain injury with different models emphasizing different aspects of injury as it relates to the human condition . The most commonly used model, fluid percussion, deforms the cortex by rapidly injecting fluid into the closed cranium [8-10]. Injury produced by fluid percussion is dependent on the severity and can include hemorrhage, vascular disruption, contusion, diffuse axonal injury, and neurologic and cognitive deficits [8-13]. With fluid percussion there are some differences depending on whether the primary injury site is along the midline (central) or lateral to the midline. Central fluid percussion injury is characterized by greater brainstem injury, whereas lateral fluid percussion injury is characterized by selective damage to the ipsilateral hemisphere and greater hippocampal damage .
Although a global reduction in CBF of up to 50% of normal has been reported in fluid percussion injury [14-18], ischemia is not a common feature unless other insults are added to the trauma . Other animal models of traumatic brain injury either do not produce ischemia  or have not been characterized in terms of CBF.
Recently, an injury model termed controlled cortical impact injury (CCII), involving direct cortical deformation using a rigid cylinder, was described for ferrets and later for rats [21,22]. Briefly, the model consists of impacting the exposed dura and underlying cortex after a craniectomy with a cylinder driven by a pneumatic impactor. Direct cortical impact produces a relatively consistent biomechanical deformation of the brain and can induce many pathologic features of human brain injury, including contusion, intraparenchymal hemorrhage, subdural hematoma, subarachnoid hemorrhage, diffuse axonal injury, and neurologic and cognitive deficits [21-23]. The model allows the severity of the injury to be controlled by independently manipulating the cylinder velocity and/or the amount of brain deformation. The pathologic and physiologic changes that occur after the cortical impact correlate with the severity of impact [21-25].
We have used the CCII model to study regional cerebral blood flow (rCBF) in rats. We hypothesized that focal ischemia surrounding the site of impact accompanies CCII. We used an autoradiographic method for measuring rCBF which provides an image of flow and thus high spatial resolution. The high resolution is important in order to determine focal areas of ischemia and to determine the ischemic boundaries if ischemia is present. Although autoradiographic methods have high spatial resolution, temporal resolution is poor. We, therefore, chose to include cortical perfusion measurements using laser-Doppler flowmetry. The laser-Doppler technique provides good temporal resolution but poor spatial resolution. Our studies support the above hypothesis by demonstrating that focal ischemia is present 30 min after injury and persisted for at least 4 h.
The protocol for this study was approved by the Animal Protocol Review Committee of Baylor College of Medicine. Experiments were performed on 17 male Long Evans rats weighing between 297 and 426 g. Anesthesia was induced with 3.5% isoflurane. After endotracheal intubation with a 16-gauge tube, each rat was artificially ventilated with 2% isoflurane and 30% oxygen in a balance of nitrogen. Catheters were inserted into the jugular vein, the left femoral vein, the left femoral artery, and the tail artery. The tip of the catheter in the jugular vein was advanced to the right atrium. The catheters in the jugular vein, femoral vein, and femoral artery were used in the CBF measurement (see below). The catheter in the tail artery was used to monitor blood pressure and sample arterial blood for determination of blood gases and pH. Rectal temperature was monitored and maintained at 37 degrees C using a temperature monitor that controlled a heating lamp. Arterial blood was periodically sampled and analyzed for PO2, PCO2, and pH. The ventilator was adjusted to maintain PaCO2 near 35 mm Hg.
The head of each rat was fixed in a stereotaxic frame and a craniectomy was performed in the right parietal bone in preparation for the CCII. In 10 rats a 3F microsensor transducer (Codman and Shurtleff, Randolph, MA) was inserted into the parietal cortex through a twist hole in the left parietal bone to monitor intracranial pressure (ICP). Since the ICP probe altered rCBF in the cortical area where it was inserted, ICP was not measured in seven rats in order to obtain autoradiographic images of blood flow (see below) without interference from the probe. A second drill hole was made in the left frontal bone for measuring cortical perfusion using laser-Doppler flowmetry. The Periflux PF3 with a PF318 flow probe having a diameter of 0.6 mm (Perimed Inc., Piscataway, NJ) continuously monitored changes in cortical perfusion.
The CCII was produced by a metal cylinder (6 mm diameter) impacting the exposed dura overlying the right parietal cortex [22,25]. A pneumatic impactor propelled the cylinder at a velocity of 5.2 m/s and deformed the brain 2 mm. The time the metal cylinder was in contact with the brain was 125 ms .
CBF was measured using14 C-isopropyliodoamphetamine (IPIA) and quantitative autoradiography . IPIA has advantages as a flow tracer since it is sequestered by a cytoplasmic protein which retards its movement in the tissue . Consequently, areas such as border zones surrounding ischemic areas are much sharper allowing the area of an ischemic zone to be better defined. Furthermore, if there are areas of hyperperfusion surrounding the ischemic area [27-32], the likelihood of it being seen is high since diffusion of the tracer to brain areas having a lower concentration of the tracer is diminished.
IPIA (0.05 mu Ci/g body weight) was injected into the femoral vein. The IPIA (29 mu Ci/mmole) was custom synthesized by Du Pont New England Nuclear (Boston, MA). Blood was withdrawn from the femoral artery at a rate of 0.4 ml/min beginning 5 s before infusing the IPIA and continuing until the rat was killed. Pentobarbital (300 mg pentobarbital in 1 mL physiologic saline) was infused into the right atrium through the catheter inserted into the jugular vein 15 s after the bolus infusion of IPIA . After death, the brain was rapidly removed, frozen in isopentane chilled to -40 to -50 degrees C, and stored at -50 degrees C until it was sectioned. Each brain was cut into 20-micro meter thick sections using a cryostat (-18 degrees C). Representative sections were mounted on glass slides and placed in contact with radiograph film in light-tight cassettes. After a 15-day exposure the film was developed producing autoradiographic images. Concentrations of the tracer (radioactivity/g tissue) were determined by comparing the optical densities of various brain regions to the optical densities produced by calibrated standards packed with the tissue sections in the cassettes. The total radioactivity in the blood withdrawn was calculated according to the following equation: Equation 1 Blood for individual regions was calculated according to the following equation: Equation 2 where blood flow has the units of milliliters per 100 grams per minute and 0.4 is the rate of blood withdrawal in milliliters per minute from the femoral artery. rCBF was graphically displayed and calculated using an image analysis system (MCID M1; Imaging Research, Inc., St. Catharines, Ontario, Canada). Identification of brain regions for the rCBF measurements was according to the atlas of Palkovits and Brownstein . For the purposes of this study we defined ischemia as blood flow less than 20 mL centered dot 100 g-1 centered dot min-1.
Three groups of rats were included in this study: (a) a control group where rCBF was measured 30 min after sham injury, (b) a group where rCBF was measured 30 min after CCII, and (c) a group where rCBF was measured 4 h after CCII.
Data is reported for all groups as mean +/- SEM. Comparison of physiologic data and rCBF between groups was performed using a one-way analysis of variance and Fisher's test for multiple comparisons. Comparison of rCBF in the left and right hemispheres for a single group was performed using a paired t-test. The level of statistical significance was considered to be P < 0.05.
The physiologic measurements in the control and two postinjury groups are given in Table 1. Mean arterial blood pressure and arterial blood pH showed tendencies to decrease after injury in both the 30-min and 4-h postinjury groups; however, the decreases were not statistically different from the control. ICP and arterial blood gases were not different from the control after injury.
Changes in cortical perfusion recorded from the surface of the left frontal cortex, contralateral to the injury, are shown in Figure 1. Although cortical perfusion fluctuated over time in all three groups, the changes were not significant from the initial perfusion value prior to sham-injury or injury.
rCBF for the three groups are shown in Table 2. In the control group, there were no significant differences in rCBF contralateral and ipsilateral to the craniectomy indicating that the craniectomy alone did not affect blood flow.
Thirty minutes after injury, brain regions contralateral to the injury showed no significant changes in rCBF compared to the control group. However, ipsilateral to the injury, rCBF was significantly reduced in the motor areas of the frontal and parietal cortices, the somatosensory area of the frontal cortex, cingulate gyrus, occipital cortex, caudate-putamen, hippocampus, and lateral geniculate. Many, but not all, of these regions showed significant differences between ipsilateral and contralateral sides.
Four hours after injury, blood flow to brain regions contralateral to the injury were similar to the corresponding regions in the control group. Ipsilateral to the injury, the reduction in rCBF showed patterns of reduced flow similar to the 30-min postinjury group. Additionally, the somatosensory area of the parietal cortex and the anterior nucleus of the thalamus showed statistically significant reductions in rCBF.
Computerized images of rCBF from individual rats, generated from autoradiographs, for the control group and two postinjury groups are shown in Figure 2. Note the ischemic areas (CBF < 20 mL centered dot 100 g-1 centered dot min-1) on the right side in the two postinjury groups. In the 30-min postinjury group, severe ischemia was seen in the motor area of the parietal cortex, the region of direct impact, and areas removed from the site of impact including the motor area of the frontal cortex and in some rats the occipital cortex Table 2. Islands of flow in the middle of the ischemic area occurred in two rats. Severe ischemia was only seen in the cortical regions 30 min after injury. CBF in deeper structures varied 30 min after injury. In some rats, rCBF in parts of the caudate-putamen (three rats), hippocampus (two rats), or thalamus (one rat) was increased compared to the contralateral side. In other rats, flow in these regions was not changed or reduced but flow was never below 20 mL centered dot 100 g-1 centered dot min-1 in the subcortical structures. In the 4-h postinjury group, the ischemic areas in three rats also included parts of the caudate-putamen, hippocampus, and thalamus in addition to the cortex.
(Figure 3) shows the volumes of ischemia, defined as CBF < 20 mL centered dot 100 g-1 centered dot min-1, in the two postinjury groups. Both individual and group data are given in the figure. The mean ischemic area was 82 +/- 12 (n = 6) and 169 +/- 53 mm3 (n = 5) for the 30-min and 4-h postinjury groups, respectively. Although the ischemic area was twofold greater after 4 h, it was not statistically different from the area in the 30-min postinjury group (P = 0.11). The two rats in the 4-h postinjury group showing the greatest area of ischemia had portions of the caudate-putamen, hippocampus, and thalamus with rates of blood flow that met our definition of ischemia.
Regions of apparent hyperemia were often seen in areas surrounding the ischemic zone in the 30-min postinjury group. The magnitude, size, and exact location in relation to the ischemic zone varied among rats. Although hyperemia could be seen in individual animals in subcortical structures, it was most prevalent in the cortical regions directly below the ischemic zone Figure 4. In some rats the hyperemic regions in the cortex occurred rostral to and above the ischemic zone. The areas of hyperemia seen 30 min after injury were not present in the 4-h postinjury group. Table 3 summarizes the rCBF below the ischemic area in the frontal and parietal cortices in the two postinjury groups. There was more than a twofold increase in rCBF in this hyperemic region of the 30-min postinjury group compared to the corresponding area of the contralateral side.
We tested the hypothesis that CCII in the rat produces focal ischemia. For the purposes of the present study, we have defined ischemia as rates of blood flow less than 20 mL centered dot 100 g-1 centered dot min-1. The results of this study support our hypothesis by showing that ischemia was present as early as 30 min after CCII and persisted for at least 4 h. The ischemia was most prevalent in the cerebral cortex; however, 4 h after injury, deeper structures in some rats were also ischemic. Although brain temperature was not measured in the present study, we did measure it in a previous study using the identical model . Brain temperature, which averaged 34.8 degrees C compared to 37.1 degrees C for rectal temperature, was not significantly different in injured rats compared to shams and did not change over time.
CCII most affected cortical regions ipsilateral to the impact. The cortical regions included not only the motor area of the parietal cortex, the region directly impacted, but also other cortical regions adjacent to and remote from the injury site Table 2. These areas include the motor area of the frontal cortex, the occipital cortex, and the somatosensory area of the frontal and parietal cortices in the 4-h postinjury group. The somatosensory area of the motor cortex in the 30-min postinjury group and the cingulate gyrus in both injured groups also showed significant reductions in rCBF, but the reductions were not as severe as other cortical regions described above. Subcortical structures (caudate-putamen, hippocampus, and thalamus) in the 30-min postinjury group had rates of rCBF that were significantly decreased, although flow was not less than 20 mL centered dot 100 g-1 centered dot min-1 in any of the rats. Deeper structures in the 4-h postinjury group also had significantly reduced rates of flow with three of five rats in the group having areas that were ischemic. With the exception of the occipital cortex, the effects of CCII on rCBF seemed to be diminished as a function of distance from the impact site.
Continuous monitoring of cortical perfusion in the frontal cortex contralateral to the craniectomy showed no significant change in perfusion in the sham group or either of the injured groups Figure 1. Furthermore, we have previously measured cortical perfusion using the laser-Doppler technique for up to 8 h in sham rats without a significant change in perfusion . The finding that cortical perfusion is not altered due to the CCII is consistent with the quantitative measurement of blood flow using the radioactive tracer. In fact, rCBF was not significantly changed in any region contralateral to the injury at either 30 min or 4 h postinjury Table 2.
The finding that flow was unchanged in all regions contralateral to the injury and decreased in only selected ipsilateral regions differs from previous studies using other models of traumatic brain injury in the rat. After fluid percussion and acceleration injury in the rat, CBF decreased globally after an initial transient increase [14-18,20]. The reduction in CBF was as much as 50% from the preinjury rate and persisted for 30-60 min before gradually returning to baseline. We believe that the absence of a global CBF reduction in our study was a function of the severity of injury induced. Cherian et al.  demonstrated that CCII produced changes in cortical perfusion contralateral to the injury that were dependent on the severity of injury. Using the same level of injury as in our study (5.2 m/s and 2 mm deformation), cortical perfusion briefly increased during the impact but quickly returned to baseline where it remained for at least 8 h. At more severe levels of injury, produced by increasing the deformation of the brain from 2 to 2.5 and 3 mm, cortical perfusion decreased to a similar extent as seen in other injury models in the rat [14-18,20]. Although Cherian et al.  only monitored perfusion at a single region, the results suggest that perfusion was reduced globally since the region monitored was contralateral to the injured hemisphere. While CCII in our study was severe enough to produce focal ischemia, it was not severe enough to produce global reductions in rCBF.
Fluid percussion injury, at present the most widely used injury model in rodents, does not produce ischemia, although flow may be significantly reduced [14-18]. Although both fluid percussion and CCII produce injury by deforming the brain, we believe that the biomechanical loading of the brain is different and this difference accounts for the absence or presence of ischemia. With fluid percussion, the brain is briefly deformed by a pressure pulse applied to the dural surface . The pressure pulse is delivered via a fluid column through a port placed in the cranium. Dixon et al.  reported that the fluid injected into the cranial cavity during injury moves not only perpendicularly to the surface of the brain, but also laterally between the skull and dura. This lateral movement reduces the biomechanical loading at the site of the injection and, thus, spreads the load over a larger area. On the other hand, the biomechanical loading of the brain during CCII is more localized, with all the energy directed to a defined area determined by the diameter of the cylinder. Thus, loading with CCII is more localized, resulting in greater focal damage including cerebrovascular injury which results in ischemia. This explanation accounts only for the tissue directly impacted; it does not account for distant ischemic areas, such as the occipital cortex.
Reductions of CBF in regions outside the ischemic area could be secondary to the ischemia and not directly related to the injury itself. Destruction of cortical regions could effectively produce deafferentation in subcortical and cortical target regions resulting in a reduced energy requirement and metabolic rate. If flow and metabolism are tightly coupled after CCII, reduced flow in some of these regions would follow the reduced metabolic rate. The direct effects of the injury and the secondary effects will need to be determined in future studies.
Cellular functions are critically dependent on the rate of CBF. In rats, when CBF decreases to less than 55 mL centered dot 100 g-1 centered dot min (-1), protein synthesis is compromised ; at less than approximately 30 mL centered dot 100 g-1 centered dot min-1, extra- and intracellular cellular pH decreases [38,39]; at less than approximately 20 mL centered dot 100 g-1 centered dot min-1, brain electrical activity ceases, phosphocreatine decreases, and inorganic phosphates increase ; at less than approximately 15 mL centered dot 100 g-1 centered dot min-1, adenosine triphosphate decreases and ion homeostasis is disrupted [37-39]. In our study, cellular function in several brain regions with very low flow, primarily cortical regions, should have been affected to a large extent after CCII. Other areas such as the caudate-putamen (45 mL centered dot 100 g-1 centered dot min-1 at 4 h) and hippocampus (41 mL centered dot 100 g-1 centered dot min-1 at 4 h) should have had fewer disruptions of cellular processes. However, the rates of CBF for these critical thresholds, and thus tissue viability, are not solely dependent on the rate of blood flow but also on the duration of the reduced flow. The longer CBF is compromised, the greater the likelihood of permanent tissue damage [37,40]. Most regions affected by CCII showed reduced rates of flow for at least 3.5 h (between 30 min and 4 h). Other regions such as the somatosensory area of the parietal cortex, which showed a deterioration of CBF between 30 min and 4 h, were ischemic for an unknown, but shorter, period of time during the course of the study.
Although we do not know whether CBF would have later recovered in any of the regions where it was severely reduced, no signs of recovery occurred during our study. In fact, several of our findings suggest that the ischemic core may have been spread to bordering areas. First, several areas showed deteriorating rates of CBF between 30 min and 4 h Table 2. Second, in the 30-min postinjury group, tissues with blood flows less than 20 mL centered dot 100 g-1 centered dot min-1 were confined solely to the cortical region; in the 4-h postinjury group three of five rats had ischemia in subcortical structures. Third, the area of ischemia was increased, although not statistically significant (P = 0.11), 4 h after injury (169 mm3) compared to 30 min after injury (82 mm3) Figure 3.
Other studies from our laboratory  appear to be inconsistent with the outcomes suggested by the present blood flow studies. Goodman et al.  reported that the contusion volume, measured 14 days post-CCII using an identical injury level, was only 5 mm3. In the present study, the volume of tissue with rates of blood flow less than 20 mL centered dot 100 g-1 centered dot min-1 was 82 mm3 at 30 min and 169 mm3 at 4 h. From the blood flow studies, we would have predicted contusion volumes much greater than those found by Goodman et al. . One difference in the two studies was the duration of the anesthesia. In the study by Goodman et al. , the rats were removed immediately from anesthesia after injury, whereas in the present study the anesthesia was continued throughout the entire experiment (30 min and 4 h depending on the group). It is possible that the continued presence of the anesthetic interfered with CBF regulation. Alternatively, the presence of the anesthetic may have had no effect on the long-term outcome of the tissue. In this case, CBF would have been restored in much of the ischemic tissue.
Other aspects of our previous study  are consistent with the present study. Previously, we found that the depth and volume of cortical contusion and hippocampal neuronal loss produced by CCII increased with the increasing severity of injury. The area of impact showed contusions with cavitation, macrophage infiltration, and marked gliosis in the surrounding parenchyma . This area of damage at the impact site corresponds to the primary zone of ischemia.
Our data show that areas of hyperemia were found at the rim of the ischemic area in the 30-min postinjury group but not in the 4-h postinjury group. In some of the rats in the 30-min postinjury group, hyperemia was apparent in the hippocampus, caudate-putamen, or thalamus. All rats in the 30-min postinjury group showed hyperemia in the cortex below the ischemic area Figure 4, Table 3. Additionally, some rats showed hyperemia above (or medial) and rostral to the ischemic area. Hyperemia is commonly reported in tissue surrounding focal ischemia [27-32]; however, there appears to be some differences in our study compared to other studies. Whereas we found the hyperemia predominantly in the cortex, other studies found the hyperemia predominantly in deeper structures [27-32]. Cortical hyperperfusion was only reported in cortical regions after 18 h [31,32], a time after which the hyperemia had disappeared in our study. The differences in our results and the results of others may be due to the nature of the ischemia. Ischemia was produced by impacting the brain in our study and by occluding the middle cerebral artery in other studies.
Little is known about the mechanism and significance of the peri-ischemic hyperemia. Nedergaard et al.  reported that hyperemic regions surrounding the ischemic zone produced by middle cerebral artery occlusion in the rat were associated with increased metabolic rates in some hyperemic regions and decreased metabolic rates in others. According to one hypothesis, referred to as "luxury perfusion," the increase in blood flow is due to vasodilation produced by acid metabolites from the ischemic zone . Furthermore, the fate of the hyperemic areas is unknown. One study found that hyperperfusion was correlated with pathologic changes while another found no relationship [31,32]. More studies will be needed to determine the mechanisms of the hyperemia and the fate of the hyperemic tissue after traumatic brain injury.
The nature and type of head trauma, in both humans and animals, are very complex. Care must be taken by the reader not to extrapolate our findings, that ischemia occurs during a specific type of injury (controlled cortical impact), to other animal models and to head injury in general. A likely scenario is that ischemia in humans and animals is specific to certain types of head injury when the injury is severe enough. For those types of injuries that resemble CCII (focal contusion injury where the energy of the injury is localized), it is interesting to speculate that focal ischemia may occur. Bouma et al.  demonstrated that focal ischemia occurred in the first few hours after injury in two patients with focal contusions. Although the majority of the patients with focal contusions did not show regional ischemia, the authors did note that there was great enough variability that further subdividing these patients may have been warranted. Furthermore, Adams and Graham  reported that histopathologic damage, consistent with ischemia, was associated with focal contusion due to impact. We speculate that the patients with focal contusion and showing regional ischemia are a subset of head-injured patients having injury resembling the CCII in rats. Further studies are necessary to demonstrate the validity of this hypothesis. Knowledge of the type of traumatic brain injury that is directly associated with ischemia will be important in its prevention. However, hypoxia/ischemia secondary to hypotension, hypoxia or other systemic factors affecting the cerebral circulation is likely to be a problem involved with head trauma in general especially when multiple injuries occur. Separating the direct causes of ischemia from the secondary causes will also be a challenge facing the intensivist when treating head-injured patients.
In summary, we report that CCII produced ischemia on the injured side of the brain. The ischemia was predominantly confined to the cortex and was present not only in the cortex directly impacted but also cortical areas distant from the impact site. Four hours after injury, deeper structures sometimes became ischemic. Border zones surrounding the ischemic areas may be at risk after traumatic brain injury as they are after stroke . Accordingly, appropriate pharmacologic interventions may be important therapeutically in reducing the infarct volume after brain injury.
The authors thank C. Edward Dixon, PhD, and Ronald Hayes, PhD, Department of Neurosurgery, University of Texas Health Science Center, Houston, Texas, for their valuable advice and aiding in the injury model.
1. Robertson CS, Contant CF, Gokaslan ZL, et al. Cerebral blood flow, arteriovenous oxygen difference, and outcome in head injured patients. J Neurol Neurosurg Psychiatry 1992;55:594-603.
2. Bouma GJ, Muizelaar JP, Choi SC, et al. Cerebral circulation and metabolism after severe traumatic brain injury: the elusive role of ischemia. J Neurosurg 1991;75:685-93.
3. Bouma GJ, Muizelaar JP. Cerebral blood flow, cerebral blood volume, and cerebrovascular reactivity after severe head injury. J Neurotrauma 1992;9(1 Suppl):S333-S348.
4. Bouma GJ, Muizelaar JP, Stringer WA, et al. Ultra-early evaluation of regional cerebral blood flow in severely head-injured patients using xenon-enhanced computerized tomography. J Neurosurg 1992;77:360-8.
5. Overgaard J, Mosdal C, Tweed WA. Cerebral circulation after head injury. Part 3: Does reduced regional cerebral blood flow determine recovery of brain function after blunt head injury? J Neurosurg 1981;55:63-74.
6. Graham DI, Adams JH. Ischemic brain damage in fatal head injuries. Lancet 1971;1:265-6.
7. Adams H, Graham DI. The pathology of blunt head injuries. In: Critchley M, Jennett B, O'Leary J, eds. Scientific foundations of neurology. Philadelphia: FA Davis, 1972:478-91.
8. Gennarelli TA. Animate models of human head injury. J Neurotrauma 1994;11:357-68.
9. McIntosh TK, Vink R, Noble L, et al. Traumatic brain injury in the rat: characterization of a lateral fluid-percussion model. Neuroscience 1989;28:233-44.
10. Dixon CE, Lyeth BG, Povlishock JT, et al. A fluid percussion model of experimental brain injury in the rat. J Neurosurg 1987;67:110-9.
11. Smith DH, Okiyama K, Thomas MJ, et al. Evaluation of memory dysfunction following experimental brain injury using the Morris water maze. J Neurotrauma 1991;8:259-69.
12. Hicks RR, Smith DH, Lowenstein DH, et al. Mild experimental brain injury in the rat induces cognitive deficits associated with regional neuronal loss in the hippocampus. J Neurotrauma 1993;10:405-14.
13. Lyeth BG, Jenkins LW, Hamm RJ, et al. Prolonged memory impairment in the absence of hippocampal cell death following traumatic brain injury in the rat. Brain Res 1990;526:249-58.
14. Yamakami I, McIntosh TK. Alterations in regional cerebral blood flow following brain injury in the rat. J Cereb Blood Flow Metab 1991;11:655-60.
15. Muir JK, Boerschel M, Ellis EF. Continuous monitoring of post-traumatic cerebral blood flow using laser-Doppler flowmetry. J Neurotrauma 1992;9:355-62.
16. Yamakami I, McIntosh TK. Effects of traumatic brain injury on regional cerebral blood flow in rats as measured with radiolabeled microspheres. J Cereb Blood Flow Metab 1989;9:117-24.
17. Yuan XQ, Prough DS, Smith TL, DeWitt DS. The effects of traumatic brain injury on regional cerebral blood flow in rats. J Neurotrauma 1988;5:289-301.
18. Okiyama K, Rosenkrantz TS, Smith DH, et al. (S)-Emopamil attenuates acute reduction in regional cerebral blood flow following experimental brain injury. J Neurotrauma 1994;11:83-95.
19. Ishige N, Pitts LH, Berry I, et al. The effect of hypoxia on traumatic head injury in rats: alterations in neurological function, brain edema, and cerebral blood flow. J Cereb Blood Flow Metab 1987;7:759-67.
20. Nilsson B, Nordstrom C-H. Experimental head injury in the rat. Part 3. Cerebral blood flow and oxygen consumption after concussive impact acceleration. J Neurosurg 1977;47:262-73.
21. Lighthall JW. Controlled cortical impact: a new experimental brain injury model. J Neurotrauma 1988;5:1-15.
22. Dixon CE, Clifton GL, Lighthall JW, et al. A controlled cortical impact model of traumatic brain injury in the rat. J Neurosci Methods 1991;39:253-62.
23. Lighthall JW, Goshgarian HG, Pinderski CR. Characterization of axonal injury produced by controlled cortical impact. J Neurotrauma 1990;7:65-76.
24. Goodman JC, Cherian L, Bryan RM Jr, Robertson CS. Lateral cortical impact injury in rats: pathological effects of varying cortical compression and impact velocity. J Neurotrauma 1994;11:587-98.
25. Cherian L, Robertson CS, Contant CF, Bryan RM Jr. Lateral cortical impact injury in rats: cerebrovascular effects of varying cortical compression. J Neurotrauma 1994;11:573-86.
26. Bryan RM Jr, Myers CL, Page RB. Regional neurohypophysial and hypothalamic blood flow in rats during hypercapnia. Am J Physiol 1988;255:R295-R302.
27. Nedergaard M, Gjedde A, Diemer NH. Focal ischemia of the rat brain; autoradiographic determination of cerebral glucose utilization, glucose content, and blood flow. J Cereb Blood Flow Metab 1986;6:414-24.
28. Tamura A, Graham DI, McCulloch J, Teasdale GM. Focal cerebral ischaemia in the rat. 2. Regional cerebral blood flow determined by [(14
) C]iodoantipyrine autoradiography following middle cerebral artery occlusion. J Cereb Blood Flow Metab 1981;1:61-9.
29. Blair RDG, Waltz AG. Regional cerebral blood flow during acute ischemia. Neurology 1970;20:802-8.
30. Shigeno T, McCulloch J, Graham DI, et al. Pure cortical ischemia versus striatal ischemia. Surg Neurol 1985;24:47-51.
31. Yamaguchi T, Waltz AG, Okazaki H. Hyperemia and ischemia in experimental cerebral infarction: correlation of histopathology and regional blood flow. Neurology 1971;21:565-78.
32. Bolander HG, Persson L, Hillered L, et al. Regional cerebral blood flow and histopathologic changes after middle cerebral artery occlusion in rats. Stroke 1989;20:930-7.
33. Bryan RM Jr. A method for measuring regional cerebral blood flow in freely moving, unstressed rats. J Neurosci Methods 1986;17:311-22.
34. Palkovits M, Brownstein MJ. Maps and guide to microdissection of the rat brain. New York: Elsevier, 1988.
35. Lighthall JW, Dixon CE, Anderson TE. Experimental models of brain injury. J Neurotrauma 1989;6:83-97.
36. Dixon CE, Lighthall JW, Anderson TE. Physiologic, histopathologic, and cineradiographic characterization of a new fluid-percussion model of experimental brain injury in the rat. J Neurotrauma 1988;5:91-104.
37. Mies G, Ishimaru S, Xie Y, et al. Ischemic threshold of cerebral protein synthesis and energy state following middle cerebral artery occlusion in rat. J Cereb Blood Flow Metab 1991;11:753-61.
38. Naritomi H, Sasaki M, Kanashiro M, et al. Flow thresholds for cerebral energy disturbance and Na+
pump failure as studied by in vivo31
Na nuclear magnetic resonance spectroscopy. J Cereb Blood Flow Metab 1988;8:16-23.
39. Harris RJ, Symon L. Extracellular pH, potassium, and calcium activities in progressive ischemia of rat cortex. J Cereb Blood Flow Metab 1984;4:178-86.
40. Siesjo BK. Pathophysiology and treatment of focal cerebral ischemia, Part I. Pathophysiology. J Neurosurg 1992;77:169-84.
41. Hoedt-Rasmussen K, Skinhoj E, Paulson O, et al. Regional cerebral blood flow in acute apoplexy. Arch Neurol 1967;17:271-81.