Subdural haematoma represents the most common traumatic lesion in severe head injury and carries the highest mortality rate. Ischaemic damage is the most common neuropathological finding in the patients who die after acute subdural haematoma . Increased intracranial pressure (ICP) due to the mass effect of haematoma decreases cerebral perfusion pressure (CPP), which produces ischaemic injury. However, even with prompt surgical evacuation, only about one-third of patients recover functional independence . Therefore, in addition to the mass effect of the haematoma, other factors appear to be the determinant of neurological outcome. Secondary, potentially treatable mechanisms have been suggested to be responsible for much of the ischaemic damage after subdural haematoma .
Increased glutamate concentration in the cerebrospinal fluid has been reported in severely head-injured patients, suggesting that an excessive release of glutamate may be involved in the process of neuronal damage . Using an acute subdural haematoma model, Bullock and his colleagues  reported a significant increase (three- to eightfold) of excitatory amino acids (glutamate and aspartate) in the ischaemic area beneath the haematoma and in the hippocampus. They also found that the volume of the ischaemic area correlated well with the level of both glutamate and aspartate . Ischaemic damage after subdural haematoma can be lessened by glutamate (N-methyl-D-aspartate: NMDA) receptor antagonists such as dizocilpine and CGS 19755; even though these drugs were given 20–30 min after insult . These findings suggest that excessive release of excitatory amino acids may produce the neural damage after subdural haematoma and that NMDA receptor antagonists may become valuable therapeutic drugs .
Ketamine is an NMDA receptor antagonist and is used as an intravenous (i.v.) induction agent and as an analgesic. Although early studies demonstrated that ketamine increases cerebral blood flow (CBF)  and ICP , recent reports showed that ketamine may not be contraindicated in patients at risk from intracranial hypertension because ketamine helped to decrease ICP in traumatic brain injury patients during propofol sedation . In the present study, using an acute subdural haematoma model in rats which consistently produces ischaemic damage beneath the haematoma, we investigated the effect of ketamine and dizocilpine on ICP and histopathological changes; the treatment starting 30 min after the induction of an acute subdural haematoma.
The protocol of this study was approved by the Ethics Committee for Animal Experimentation at Yamaguchi University School of Medicine.
Anaesthesia and surgical preparation
Thirty-six adult male Wistar rats weighing 350–450 g with free access to food and water were used. Anaesthesia was induced using 4% isoflurane and nitrous oxide (60%) in oxygen in a Plexiglas chamber. After tracheal intubation, anaesthesia was maintained with 1.2% isoflurane and nitrous oxide (60%) in oxygen throughout the surgical procedure. PaCO2 was maintained at 4.1 ± 4.9 kPa throughout the experimental period. The right femoral artery and vein were cannulated with polyethylene catheters for blood pressure monitoring, blood sampling, and drug infusion. Pancuronium 0.1 mg kg−1 was injected i.v. followed by 0.05 mg kg−1 every 2 h. The animals were fixed in a Kopf stereotactic frame. A 20-mm mid-line scalp incision was made and a burr hole 2 mm in diameter was drilled 2 mm to the left side of the sagittal suture and 2 mm caudal to the coronal suture. Using an operating microscope, the dura was incised and a blunt J-shaped 23-gauge needle was carefully inserted into the subdural space and fixed with rapid-curing cyanoacrylate glue, which was used later for the subdural injection of blood or silicone. In order to measure ICP, another 3-mm burr hole was drilled at the same co-ordinate in the right cranium. After incision of the dura, a zero-calibrated fibreoptic device (Camino Laboratories, San Diego, CA, USA) was inserted 1 mm into the right parietal cortex. After surgery, nitrogen (60%) was substituted for nitrous oxide and anaesthesia was maintained with 1.2% isoflurane throughout the experimental period. At least 30 min was allowed to elapse before starting the subdural injection of blood or silicone (see below). Mean arterial pressure (MAP) and ICP were continuously monitored. CPP was calculated by subtracting ICP from MAP. Electroencephalograms (EEGs) were monitored on both hemispheres using bipolar front-occipital needle electrodes. A thermistor probe was inserted into the right temporal muscle and the temperature of the animals was maintained between 36.8 and 37.2°C with a heating lamp.
The rats were randomly assigned into four groups (nine in each); control, ketamine, dizocilpine and silicone group. The control, ketamine and dizocilpine groups had a subdural haematoma produced by an injection of 150 µL of freshly drawn non-heparinized autologous venous blood into the subdural space over 7.5 min. Thirty minutes after the SDH, the ketamine group received a ketamine infusion of 60 mg kg−1 h−1 for 90 min followed by 20 mg kg−1 h−1 for 6 h (total dose of 210 mg kg−1), and the dizocilpine group received a 0.12-mg kg−1 bolus injection of dizocilpine followed by 0.12 mg kg−1 h−1 for 7.5 h (total dose of 1.0 mg kg−1). The doses of each drug were chosen based on previous reports showing neuroprotective effects in various models of brain injury [10–15]. The silicone group had a subdural injection of 150 µL of silicone gel (a 1:1 mixture of silicone grease and mineral oil). In a preliminary study, this mixture was found to produce a subdural gel of similar size of mass and distribution to subdural blood.
Eight hours after induction of haematoma or silicone subdural injection, the brain was fixed via a perfusion through a temporary cannula placed into the aorta via the left ventricle, using 100 mL of heparinized saline followed by 100 mL of 10% formalin in phosphate buffer. Seven days after fixation, the brain was cut, dehydrated and embedded in wax. Histological sections 7 µm thick were cut and stained with haematoxylin and eosin. Sections corresponding to the predetermined stereotactic planes distributed throughout the forebrain were identified and the extent of ischaemic damage was measured with an image-analysing computer, and the volume of ischaemic damage was calculated as described by Osborne and his colleagues . The volume of ischaemic damage was expressed in cubic millimetres of the left hemisphere. The cortices, hippocampus, striatum and thalamus were also evaluated microscopically by an investigator who was unaware of the experimental groups.
The data are expressed as mean ± SD. For physiological variables, statistical differences among groups were examined by two-way ANOVA for repeated measures. If the F statistic of analysis of variance was significant, Scheffé's method was applied for multiple comparisons. For volumetric histopathology, the Kruskal–Wallis test was used. When appropriate, the Mann–Whitney U-test using the Bonferroni correction was applied for multiple comparisons. Statistical significance was assumed when P < 0.05.
All rats survived for 8 h (the intended experimental time period) after induction of the subdural haematoma and silicone. Changes in temperature, blood glucose, PaO2 and PaCO2 are summarized in Table 1. There were no significant differences in these variables amongst the four groups. Figure 1 shows the cerebral haemodynamic changes during the study. The ICP increased during the induction of haematoma or silicone and then recovered to preinduction levels over 2–3 h. There were no significant differences in ICP amongst the four groups throughout the study. The MAP in the ketamine and dizocilpine groups was not different from that of the control group. The MAP in the silicone group was significantly lower than those of haematoma groups after induction of haematoma. The CPP was decreased during the induction of haematoma or silicone but recovered in 5 min. There were no significant differences in CPP amongst the four groups during subdural injection. After subdural injection, the CPP in the silicone group was significantly lower than those in the ketamine and dizocilpine groups. However, throughout the study, the CPP was maintained above the lower limit of autoregulation, except for a short period during induction of haematoma or silicone.
Representative EEGs are shown in Figure 2. In the control group, the EEG showed a burst (sharp or spike wave, 50–100 µV, 0.5 s) and suppression (0.5–1 s) pattern before induction of subdural haematoma. Immediately after induction of subdural haematoma, in four of the nine rats in the control group, EEGs showed a flat pattern. In other rats, some burst activities were observed. This suppressive pattern was maintained for 5–10 min and gradually (1–2 h) returned to the preinduction pattern. In the ketamine and dizocilpine groups, burst and suppression patterns similar to those in the control group were observed for up to 4–5 h after SDH (not shown). Six to 8 h after SDH, the suppression pattern faded, and continuous activities appeared in the ketamine and dizocilpine groups, with a slight intergroup differences in frequency (12–15Hz vs. 8–12 Hz). In the silicone group, the EEG pattern was similar to that of the control group for up to 3–4 h after injection of silicone. Thereafter, continuous activities (15–17Hz) appeared. In all groups, the EEGs at the contralateral side exhibited a similar pattern to that of the ipsilateral side, but with slightly greater amplitude.
In all SDH groups, the brain showed a variable degree of mid-line shift to the right and narrowing of the left lateral ventricle due to the brain swelling. These changes were less in the silicone group. In all animals, the ischaemic damage was confined to the cortex underlying the subdural haematoma or silicone. A schematic illustration and corresponding light microphotographs of a rat in the control group are shown in Figure 3. The ischaemic damage is sharply demarcated from normal cortex and is most extensive where the haematoma (or silicone) was thickest. Histological damage included necrosis beneath the haematoma, a spongy state of neuropil in the marginal area of ischaemic damage and pyknosis of the neuronal cells. This damage was not found in the contralateral cortex, bilateral hippocampi, striatum or thalamus.
The volume of ischaemic damage is summarized in Figure 4. The control group had an ischaemic damage volume of 11.9 ± 3.8 mm3. Ischaemic damage in the ketamine group (7.8 ± 5.0 mm3) was smaller but was not significantly different from the control group. In contrast, the volume of ischaemic damage in the dizocilpine group (6.1 ± 3.8 mm3) was significantly smaller than that in the control group (P < 0.05). The volume of ischaemic damage in the silicone group (1.3 ± 1.2 mm3) was significantly smaller than that in the control group (P < 0.001).
This study demonstrates in rats that (a) a subdural silicone injection produces significantly less ischaemic damage than that produced by haematoma, (b) ketamine does not increase ICP but fails to decrease significantly the volume of ischaemic damage after subdural haematoma and (c) dizocilpine significantly decreases the volume of ischaemic damage after subdural haematoma.
Several studies have shown that subdural haematoma increases ICP and causes ischaemic brain injury by decreasing CPP, in addition to the hemispheric swelling. However, many patients with poor prognosis often exhibit a ‘lucid interval’ at the early stage, and thus the secondary events other than decreased CPP may possibly be involved. The significantly greater ischaemic damage in the control group (subdural haematoma) compared with the silicone group in this study suggests that the damage is increased by soluble factors released from the clotted blood. There has been a clinical study suggesting an excitotoxicity as a mechanism for the neuronal damage after head injury . In the subdural haematoma model in animals, the concentrations of excitatory amino acids (glutamate and aspartate) have been shown to increase in the ischaemic/peri-ischaemic cortex beneath the haematoma and in both hippocampi . In addition, hypermetabolism was observed in these areas 2 h after induction of SDH but not after silicone injection . Treatment with NMDA antagonist (D-CPP-ene) ameliorated glucose hypermetabolism . Treatment with NMDA antagonists (dizocilpine, CGS19755, D-CPP-ene) produced a significant reduction in the volume of ischaemic damage when evaluated 2 and 4 h after SDH [5,19]. In this study, the histological damage was confined to the cortex, and not the hippocampi up to 8 h after subdural haematoma. The sparing of the hippocampi may be due to the smaller volume of blood injected in our study compared with the other studies [5,19]. This is similar to Miller and colleagues' study  in which cortical damage was observed with no hippocampal damage. On balance, it is possible that hypermetabolism in the peri-ischaemic cortex beneath the haematoma may be caused by the excessive release of excitatory amino acids, which disturbs energy balance and thus amplifies the damage associated with SDH.
Ketamine and dizocilpine are non-competitive NMDA receptor antagonists and their effects have been examined in various models of brain ischaemia and trauma. In previous studies, the effect of ketamine in brain ischaemia was inconclusive with some showing neuroprotective effects [10–12] and some not [21,22]. In contrast, in head injury models, two studies have shown the neuroprotective effects of ketamine [13,14]. To date, there have been no studies that have compared the neuroprotective effects of ketamine and dizocilpine in a given head injury model. In our study, dizocilpine significantly reduced the ischaemic damage beneath the SDH, while ketamine failed to reduce the damage. The dose of ketamine (210 mg kg−1) used in this study was larger than or equal to those reported in the previous studies [10–14]. The dose of dizocilpine (1.0 mg kg−1) in our study was based on a study by Gill and his colleagues , who showed that this dose reduced the volume of ischaemic damage maximally in a focal ischaemia model without causing hypotension. In a comparative study of the potency of NMDA antagonism using isolated chick embryo retina, the concentration of antagonist required to totally prevent the toxic action of 200 µmol NMDA is 0.1 and 5.0 µmol for dizocilpine and ketamine respectively . Wong and his colleagues  also reported that dizocilpine was a hundred times more potent as an antagonist of NMDA responses than ketamine in a rat cortical-slice preparation. The dose of ketamine used in this study appears larger than that of dizocilpine as a potency of NMDA antagonism. Therefore, the failure of ketamine to reduce ischaemic damage in this study is not explained by an inadequate dose.
Phencyclidine and related drugs (ketamine, dizocilpine) have been shown to cause vacuole formation in the posterior cingulate and retrosplenial cerebrocortical neurones . Although the difference in the potency of ketamine and dizocilpine to induce these pathomorphological changes has not been clearly defined, Olney and his colleagues  estimated that the order of potencies with which these drugs induced this effect (dizocilpine > ketamine) was the same as their order of affinities for binding to the phencyclidine receptor and for their order of potencies in antagonizing either the excitatory or neurotoxic actions of NMDA. However, the dose of ketamine relative to the dose of dizocilpine used in this study is far greater than that used by Olney and his colleagues. Therefore, we can not rule out that this possible adverse effect of ketamine may have contributed to the failure to show a neuroprotective effect. Ketamine has other adverse effects; the increase of cerebral metabolism , a change of monoamine metabolism (release and uptake)  and a proconvulsant effect . These adverse effects may have contributed to the failure to demonstrate any neuroprotective effect of ketamine. The lack of significant neuroprotection with ketamine cannot be attributed to its effect on cerebral haemodynamics as ketamine neither increased ICP nor decreased MAP.
It is possible that the small number of animals may have caused a type II error for the protective effect of ketamine. If means and variance remain largely unchanged, the volume of ischaemic damage in the ketamine group would be significantly smaller than that of the control group at a sample size of about 26 in each of the ketamine and control groups by a power analysis (β = 0.1). Thus, we believe that the neuroprotective effect of ketamine, if any, is clinically not so meaningful.
In conclusion, the ischaemic damage after acute SDH is attributable not only to the mass effect of a blood clot but also to a secondary factor such as an excessive release of excitatory amino acids. The ischaemic damage is reduced by the NMDA receptor antagonist dizocilpine but not convincingly by ketamine despite the latter's lack of effect on cerebral haemodynamics.
1 Seelig JM, Becker DP, Miller JD, Greenberg RP, Ward JD, Choi SC. Traumatic acute subdural haematoma
: major mortality reduction in comatose patients treated within four hours. N Engl J Med
1981; 304: 1511–1518.
2 Stone JL, Rifai MHS, Sugar O, Lang RGR, Oldershaw JB, Moody RA. Subdural
Hematomas. I. Acute subdural
hematoma: Progress in definition, clinical pathology and therapy. Surg Neurol
1983; 19: 216–231.
3 Baker AJ, Moulton RJ, MacMillan VH, Shedden PM. Excitatory aminoacids in cerebrospinal fluid following traumatic brain injury in humans. J Neurosurg
1993; 9: 369–372.
4 Bullock R, Butcher SP, Chen MH, Kendall L, McCulloch J. Correlation of the extracellular glutamate concentration with extent of blood flow reduction after subdural haematoma
in the rat. J Neurosurg
1991; 74: 794–802.
5 Kuroda Y, Fujisawa H, Strebel S, Graham DI, Bullock R. Effect of neuroprotective N-Methyl-D-Aspartate antagonists on increased intracranial pressure: studies in the rat acute subdural haematoma
1994; 35: 106–112.
6 Bullock R, Kuroda Y, Teasdale GM, McCulloch J. Prevention of post-traumatic excitotoxic brain damage with NMDA antagonist drugs: a new strategy for the nineties. Acta Neurochirurgica
, 1992; 55 (Suppl.): 49–55.
7 Takeshita H, Okuda Y, Sari A. The effects of ketamine
on cerebral circulation and metabolism in man. Anesthesiology
1972; 36: 69–75.
8 Shapiro HM, Wyte SR, Harris AB. Ketamine
anaesthesia in patients with intracranial pathology. Br J Anaesth
1972; 44: 1200–1204.
9 Jacques A, Sophile A, Marc R, Laurent T, Bernard A, Claude M. Ketamine
decreases intracranial pressure and electroencephalographic activity in traumatic brain injury patients during propofol sedation. Anesthesiology
1997; 87: 1328–1334.
10 Church J, Zeman S, Lodge D. The neuroprotective action of ketamine
and MK-801 (dizocilpine
) after transient cerebral ischaemia
in rats. Anesthesiology
1988; 69: 702–709.
11 Marcoux FW, Goodrich JE, Dominick MA. Ketamine
prevents ischaemic neuronal injury. Brain Res
1988; 452: 329–335.
12 Hoffman WE, Pelligrino D, Werner C, Kochs E, Albrecht RF, Schulte am Esch J. Ketamine
decreases plasma catecholamines and improves outcome from incomplete cerebral ischaemia
in rats. Anesthesiology
1992; 76: 755–762.
13 Smith DH, Okiyama K, Gennarelli TA, McIntosh TK. Magnesium and ketamine
attenuate cognitive dysfunction following experimental brain injury. Neurosci Lett
1993; 157: 211–214.
14 Shapira Y, Lam AM, Eng CC, Laohaprasit V, Michel M. Therapeutic time window and dose response of the beneficial effects of ketamine
in experimental head injury. Stroke
1994; 25: 1637–1643.
15 Gill R, Brazell C, Woodruff GN, Kemp JA. The neuroprotective action of dizocilpine
(MK-801) in the rat middle cerebral artery occlusion model of focal ischaemia
. Br J Pharmacol
1991; 103: 2030–2036.
16 Osborne KA, Shigeno T, Balarsky AM et al.
Quantitative assessment of early brain damage in a rat model of focal cerebral ischaemia
. J Neurol Neurosurg Psychiatry
1987; 50: 402–410.
17 Kuroda Y, Inglis FM, Miller JD, McCulloch J, Graham DI, Bullock R. Transient glucose hypermetabolism after acute subdural haematoma
in the rat. J Neurosurg
1992; 76: 471–477.
18 Inglis FM, Bullock R, Chen MH, Graham DI, McCulloch J. Glucose hypermetabolism after acute subdural haematoma
is ameliorated by a competitive NMDA antagonist. J Neurotrauma
1992; 9: 75–84.
19 Chen MH, Bullock R, Graham DI, Miller JD, McCulloch J. Ischaemic neuronal damage after acute subdural haematoma
in the rat: effects of pretreatment with a glutamate antagonist. J Neurosurg
1991; 74: 944–950.
20 Miller JD, Bullock R, Graham DI, Chen MH, Teasdale GM. Ischaemic brain damage in a model of acute subdural haematoma
1990; 27: 443–439.
21 Jensen ML, Auer RN. Ketamine
fails to protect against ischaemic neuronal necrosis in the rat. Br J Anaesth
1988; 61: 206–210.
22 Ridenour TR, Warner DS, Todd MM, Baker MT. Effects of ketamine
on outcome from temporary middle cerebral artery occlusion in the spontaneously hypertensive rat. Brain Res
1991; 565: 116–122.
23 Olney JW, Price MT, Salles S, Labruyere J, Frierdich G. MK-801 powerfully protects against N-methyl aspartate neurotoxicity. Eur J Pharmacol
1987; 141: 357–361.
24 Wong EH, Kemp JA, Priestley T, Knight AR, Woodruff GN, Iversen LL. The anticonvulsant MK-801 is a potent N-methyl-D-aspartate antagonist. Proc Natl Acad Sci USA
1986; 83: 7104–7108.
25 Olney JW, Labruyere J, Price MT. Pathological changes induced in cerebrocortical neurons by phencyclidine and related drugs. Science
1989; 244: 1360–1362.
26 Davis DW, Mans AM, Biebuyck JF, Hawkins RA. The influence of ketamine
on regional brain glucose use. Anesthesiology
1988; 69: 199–205.
27 Snell LD, Yi SJ, Johnson KM. Comparison of the effects of MK-801 and phencyclidine on catecholamine uptake and NMDA-induced norepinephrine release. Eur J Pharmacol
1988; 145: 223–226.
28 Lees GJ. Influence of ketamine
on the neuronal death caused by NMDA in the rat hippocampus. Neuropharmacology
1995; 34: 411–417.