British Journal of Neurosurgery
ISSN: 0268-8697 (Print) 1360-046X (Online) Journal homepage: http://www.tandfonline.com/loi/ibjn20
Interstitial white matter brain oedema does not alter the electroencephalogram Ian R. Whittle, Margery Clarke, Alberto Gregori, Ian R. Piper & J. Douglas Miller To cite this article: Ian R. Whittle, Margery Clarke, Alberto Gregori, Ian R. Piper & J. Douglas Miller (1992) Interstitial white matter brain oedema does not alter the electroencephalogram, British Journal of Neurosurgery, 6:5, 433-437, DOI: 10.3109/02688699208995032 To link to this article: http://dx.doi.org/10.3109/02688699208995032
Published online: 06 Jul 2009.
Submit your article to this journal
Article views: 1
View related articles
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=ibjn20 Download by: [RMIT University Library]
Date: 16 April 2016, At: 12:47
British Journal of Neurosurgery (1 992) 6, 433-437
ORIGINAL ARTICLE
Interstitial white matter brain oedema does not alter the electroencephalogram IAN R. WHITTLE, MARGERY CLARKE, ALBERT0 GREGORI, IAN R. PIPER Downloaded by [RMIT University Library] at 12:47 16 April 2016
& J. DOUGLAS MILLER
Department of Clinical Neurosciences, Western General Hospital, Edinburgh EH4 2XU, Scotland
Abstract An experimental study was performed to determine the effects of interstitial white matter oedema on the electroencephalogram (EEG). Using both rodent and feline infusion models of focal brain oedema no difference was found between the EEG waveforms recorded epidurally from the infused and control hemispheres. It is concluded that where focal slow-wave EEG abnormalities overlie oedematous brain the EEG abnormalities are not primarily related to the brain oedema but arise from either local biomechanical or other pathophysiological mechanisms.
Key words: Brain oedema, electroencephalogram.
Introduction
A localized increase in the proportion of slow component (delta and theta waveform) of the EEG is frequently recorded from the area immediately surrounding brain tumours. The contribution of peritumoural oedema to this slow-wave EEG activity is uncertain. A number of authors have reported a correlation between EEG abnormalities and the extent of peritumoural edema,'-^ whilst others consider it Experimental studies evaluating the effects of interstitial white matter brain oedema on the EEG suggest that cerebral damage, cerebral ischaemia and secondary brain herniations are more important than perilesional oedema in the production of focal slow-wave abnormalities.s-'2 This study utilized the feline13 and rodent14 infusion brain oedema models to evaluate the effects of interstitial white mater brain oedema on the EEG. The particular advantage of these
models is that the experimental brain oedema produced is quantitatively and qualitatively very similar to that seen with human vasogenic brain oedema but free from confounding variables such as raised intracranial pressure (ICP), cerebral ischaemia, brain herniation, primary brain tissue destruction or a mass lesion.
Methods
Focal extracellular white matter brain oedema was generated in anaesthetized rats and cats using the infusion oedema model^.'^.'^ Both procedures were performed under licence according to the provisions of the UK Animals (Scientific Procedures) Act 1986. In brief, rodents were anaesthetized with intraperitoneal urethane (1.2 mg/kg), paralysed and mechanically ventilated. A 25-gauge roundtipped needle was stereotactically placed ( 1
433
Downloaded by [RMIT University Library] at 12:47 16 April 2016
434
Zan R. Whittle et al.
mm anterior to bregma, 3 mm lateral to the midline and at a depth of 2.8 mm) and 100 pl of saline infused over 60 min. EEG was recorded from Ag/AgCl epidural electrodes, positioned lateral to and 1 mm behind the plane of infusion and from a similar position on the contralateral hemisphere. A reference electrode was positioned at the nasion. EEG recordings were taken every 20 rnin during the infusion and thereafter each 30 min for 2 h. The cats were anaesthetized with intraperitoneal pentobarbital (45 mg/kg) intubated, paralysed and mechanically ventilated (66% N 2 0:0,). A 23-gauge needle was positioned stereotactically in the forebrain white matter (+ 19 mm, 9 mm lateral, 9 mm depth) and 600 pl of autologous serum protein (ranging from 50 to 66% concentration) infused over 3 h. EEG was recorded from Ag/AgCl epidural electrodes positioned over each ectoslvian gyrus with reference the nasion. EEG records were taken every 30 min during the infusion and then hourly for 3 h. A DISA 26 10-channel EEG recording system was used with high-pass filter at 0.3 s and low-pass filter at 100 Hz. The traces were visually assessed for symmetry of frequency and amplitude.
Results Five rats and three cats were studied. In all experiments mean arterial pressure (rodents 90 SD f 10 mmHg; cats 120 SD t 12 mmHg), mean p 0 2 ( > l o 0 mmHg) and pCOz (34 SD Ifr 2 mmHg) remained stable. These models of infusion brain oedema cause: (1) significant increases in white matter specific gravity and an increased white matter tissue water content of between 10.4 and 12.5 ml/ 100 g tissue; (2) the classical histological features of extracellular white matter oedema; (3) small rises in mean ICP (rodents 2 mmHg, cats 10 mmHg) but no cerebral herniation or ischaemia (mean rat rCBF 35 m1/100 g/min; mean cat rCBF 45 m1/100 g/min); and (4) no change in the latency or amplitude of either cortical SEPs elicited by forepaw stimulation or cortically generated spinal M E P s . ' ~ J ~ The baseline rodent EEGs were all symmet-
rical and showed background slow waves (mainly in the delta range) with intermittent alpha and theta activity. Occasional large amplitude sharp components (100 p V ) were seen but general background amplitude was 30-40 pV. The pattern of bilateral slow-wave activity persisted throughout the experiment but at no time did the infused hemisphere show asymmetrical slow-wave activity (Fig. 1). The baseline feline EEGs showed symmetrical theta (6-7 Hz) waveform (30-50 p V ) with faster components to the beta range and intermittent sharp wave transients of between 100 and 150 pV in both hemispheres. During the course of the experiment waveforms remained symmetrical. After 4 h there was a symmetrical increase in the amount of delta activity with decreased sharp waves (Fig. 2). In none of the experiments did the infused hemisphere show focal EEG abnormalities. Overall the feline and rodent EEG waveforms were felt appropriate for the level of anaesthesia. Discussion This study adds additional support to the contention that interstitial brain oedema, uncomplicated by cerebral ischaemia, brain herniations or significant rises in ICP, does not cause significant focal abnormalities in the EEG. Although this model of vasogenic brain oedema can cause a shift to anaerobic metab o l i ~ r n 'and ~ expansion of the white matter extracellular space, the infused fluid has no inherent neuro- or glial toxicity and does not impair rCBF or alter cortical SEPs or cortically generated spinal MEPs.'~-'' The paucity of changes in either SEP17or EEG3!6,7waveforms in many patients with brain tumours correlates well with P E T studies that show only minor perturbations of rCBF and cerebral metabolism in peritumoural Experimental brain tumours in cats that cause profound peritumoural oedema also do not produce either abnormalities in cortical SEPs or EEG waveforms whether examined visually or using Fourier a n a l y ~ i s . ~ The combined mass effect of a brain
Interstitial white matter oedema
435
Downloaded by [RMIT University Library] at 12:47 16 April 2016
R
L
FIG. 1. (a) Rodent EEG from right (R) and leff (L) hemispheres before intracerebral infusion. There is some symmetrical beta activity with occasional high amplitude sharp potentials and irregular slow-wave components (MAP 95 mmHg, p C 0 , 33 mmHg). (b) Sixty minutes after 100 pl saline infusion into the left hemisphere the EEG waveforms remain symmetrical with mainly slow-wave activity but some beta components (MAP 86 mmHg; pCOz 35 mmHg).
tumour and surrounding oedematous brain will eventually cause secondary brain herniation and cerebral ischaemia. It is most likely that these biomechanical effects combined with focal brain parenchymal destruction by the
tumour are responsible for focal and diffuse EEG changes in patients with brain ~ u ~ o u ~ s . Resolution ~ ~ ~ J ~of, focal ~ ~ JEEG ~ abnormalities with ‘anti-oedema’ therapy in some patients with brain turn our^'**^^ is most
Zan R. Whittle et al.
Downloaded by [RMIT University Library] at 12:47 16 April 2016
436
FIG.2. (a) Baseline feline EEG waveforms from right (R) and left (L) ectosylvian regions. Predominant rhythm is theta (6-7 Hz) with occasional faster components. Sharp wave transients are seen on both sides with no overall asymmetry (MAP 112 mmHg;pCO, 33 mmHg; ICP 5 mmHg). (b) Two hours after 600 p1 autologous serum protein (50%) infusion into the right forebrain white matter the EEG is dominated by symmetrical slow-wave activity but some beta activity is still seen. There are fewer sharp wave transients (MAP 121 mmHg;pC02 35 mmHg; ICP mmHg). (c) Even at this time induced hypercapnia ( p C 0 , 5 9 mmHg) produces no asymmetry of EEG waveform. Although the activity is generally of higher amplitude, slow components are more prominent and remain symmetrical.
Interstitial white matter oedema likely attributable to alterations in cerebral perfusionla and biomechanical changes in the brain19 since significant reductions in local brain tissue water content and C T imaged brain oedema occur only after several days of steroid therapy.*O
Acknowledgement
Downloaded by [RMIT University Library] at 12:47 16 April 2016
This work was supported by a Project Grant from the MRC (8500319).
Address for correspondence: Mr I. R. Whittle, Department of Neurosurgery, Western General Hospital, Edinburgh EH4 2XU, Scotland.
References 1 Aguglia U, Gambardella A, Oliveri RL, Lavano A, Camerlingo R, Quattrone A. Triphasic waves and cerebral tumors. Eur Neurol 1990; 3O:l-5. 2 Matsuoka S, Arakaki Y, Numaguchi K, Ueno S. The effect of dexamethasone on electroencephalograms of patients with brain tumours. J Neurosurg 1978; 48~601-8. 3 Nagata K, Gross LE, Kindt GW et al. Topographic electroencephalographic study with power ratio index mapping in patients with malignant brain tumours. Neurosurgery 1985; 17:613-19. 4 Soejima T, Okuda S, Ishikawa T, Matsuoka S. [The effect of glycerol on EEG-abnormality (polymorphic delta wave) in patients with brain tumors.] No-ToShinkel 1987; 39553-60. 5 Steudel WI, Beck V, Becker H et al. EEG in patients with twnours of the cerebral hemispheres. In: Lanksch W, Kazner E, eds. Cranial computerized tomography. Berlin: Springer, 1976: 188-200. 6 Gastaut JL, Michel B, Sabet Hassen S, Corda M, Gastaut H. Electroencephalography in brain oedema (127 cases of brain tumour investigated by computerized tomography). Electroencephalogr Clin Neurophysiol 1979; 46231-55.
437
7 Passerini P, Ferini Strambi L, Sbacci M et al. EEG pattern in tumours with and without cerebral oedema. Electromyogr Clin Neurophysiol 1983; 23: 117-22. 8 Dempsey RJ, Combs DJ, Olson JW, Maley M. Brain ornithine decarboxylase activity following transient cerebral ischaemia: relationship to cerebral oedema development. Neurol Res 1988; 10:175-8. 9 Hossmann K-A, Szymas J, Seo K et al. Experimental transplantation glioma in the adult cat; Part 2. Pathophysiology and magnetic resonance imaging. Acta Neurochir (Wien) 1989; 98:189-200. 10 Schaul N, Ball K, Gloor P, Pappius H. The EEG in cerebral oedema. In: Pappius H, Feindel M, eds. Dynamics of brain oedema. Berlin: Springer, 1976; 144-9. 11 Sutton LN, Bruce DA, Welsh FA, Jaggi JL. Metabolic and electrophysiological consequences of vasogenic brain oedema. Adv Neurol 1980; 28:241-54. 12 Tranmer BI, Iacobacci RI, Kindt GW. Effects of crystalloid and colloid infusions on intracranial pressure and computerized electroencephalographic data in dogs with vasogenic brain oedema. Neurosurgery 1989; 25:173-8. 13 Whittle IR, Piper IR, Miller JD. The contribution of secondary mediators to the etiology and pathophysiology of brain oedema. Acta Neurochir (Wein) 1990; SUPPI 51~71-3. 14 Whittle IR, Miller JD. A rodent model of infusion brain oedema: methodology and pathophysiological effects of saline and protein infusions. Acta Neurochir (Wien) 1990; 105:158-68. 15 Sutton LN, Barranco D, Greenberg J et al. Cerebral blood flow and glucose metabolism in experimental brain oedema. J Neurosurg 1989; 71:868-74. 16 Hino A, Imahori Y, Tenjin H et al. Metabolic and hemodynamic aspects of peritumoural low density areas in human brain tumour. Neurosurgery 1990; 26:615-21. 17 Penn RD. Cerebral oedema and neurological function in human beings. Neurosurgery 1980, 6249-54. 18 Brooks DJ, Beaney RP, Thomas DGT. The role of positron emission tomography in the study of cerebral tumours. Semin Oncol 1986; 13233-93. 19 Miller JD, Sakalas R, Ward JD. Methylprednisolone treatment of brain tumours. Neurosurgery 1977; 1:114-19. 20 Bell BA, Smith MA, Kean D M et al. Brain water measured by resonance imaging. Correlation with direct estimation and changes after mannitol and dexamethasone. Lancet 1987; i:66-9.
ISSN: 0268-8697 (Print) 1360-046X (Online) Journal homepage: http://www.tandfonline.com/loi/ibjn20
Interstitial white matter brain oedema does not alter the electroencephalogram Ian R. Whittle, Margery Clarke, Alberto Gregori, Ian R. Piper & J. Douglas Miller To cite this article: Ian R. Whittle, Margery Clarke, Alberto Gregori, Ian R. Piper & J. Douglas Miller (1992) Interstitial white matter brain oedema does not alter the electroencephalogram, British Journal of Neurosurgery, 6:5, 433-437, DOI: 10.3109/02688699208995032 To link to this article: http://dx.doi.org/10.3109/02688699208995032
Published online: 06 Jul 2009.
Submit your article to this journal
Article views: 1
View related articles
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=ibjn20 Download by: [RMIT University Library]
Date: 16 April 2016, At: 12:47
British Journal of Neurosurgery (1 992) 6, 433-437
ORIGINAL ARTICLE
Interstitial white matter brain oedema does not alter the electroencephalogram IAN R. WHITTLE, MARGERY CLARKE, ALBERT0 GREGORI, IAN R. PIPER Downloaded by [RMIT University Library] at 12:47 16 April 2016
& J. DOUGLAS MILLER
Department of Clinical Neurosciences, Western General Hospital, Edinburgh EH4 2XU, Scotland
Abstract An experimental study was performed to determine the effects of interstitial white matter oedema on the electroencephalogram (EEG). Using both rodent and feline infusion models of focal brain oedema no difference was found between the EEG waveforms recorded epidurally from the infused and control hemispheres. It is concluded that where focal slow-wave EEG abnormalities overlie oedematous brain the EEG abnormalities are not primarily related to the brain oedema but arise from either local biomechanical or other pathophysiological mechanisms.
Key words: Brain oedema, electroencephalogram.
Introduction
A localized increase in the proportion of slow component (delta and theta waveform) of the EEG is frequently recorded from the area immediately surrounding brain tumours. The contribution of peritumoural oedema to this slow-wave EEG activity is uncertain. A number of authors have reported a correlation between EEG abnormalities and the extent of peritumoural edema,'-^ whilst others consider it Experimental studies evaluating the effects of interstitial white matter brain oedema on the EEG suggest that cerebral damage, cerebral ischaemia and secondary brain herniations are more important than perilesional oedema in the production of focal slow-wave abnormalities.s-'2 This study utilized the feline13 and rodent14 infusion brain oedema models to evaluate the effects of interstitial white mater brain oedema on the EEG. The particular advantage of these
models is that the experimental brain oedema produced is quantitatively and qualitatively very similar to that seen with human vasogenic brain oedema but free from confounding variables such as raised intracranial pressure (ICP), cerebral ischaemia, brain herniation, primary brain tissue destruction or a mass lesion.
Methods
Focal extracellular white matter brain oedema was generated in anaesthetized rats and cats using the infusion oedema model^.'^.'^ Both procedures were performed under licence according to the provisions of the UK Animals (Scientific Procedures) Act 1986. In brief, rodents were anaesthetized with intraperitoneal urethane (1.2 mg/kg), paralysed and mechanically ventilated. A 25-gauge roundtipped needle was stereotactically placed ( 1
433
Downloaded by [RMIT University Library] at 12:47 16 April 2016
434
Zan R. Whittle et al.
mm anterior to bregma, 3 mm lateral to the midline and at a depth of 2.8 mm) and 100 pl of saline infused over 60 min. EEG was recorded from Ag/AgCl epidural electrodes, positioned lateral to and 1 mm behind the plane of infusion and from a similar position on the contralateral hemisphere. A reference electrode was positioned at the nasion. EEG recordings were taken every 20 rnin during the infusion and thereafter each 30 min for 2 h. The cats were anaesthetized with intraperitoneal pentobarbital (45 mg/kg) intubated, paralysed and mechanically ventilated (66% N 2 0:0,). A 23-gauge needle was positioned stereotactically in the forebrain white matter (+ 19 mm, 9 mm lateral, 9 mm depth) and 600 pl of autologous serum protein (ranging from 50 to 66% concentration) infused over 3 h. EEG was recorded from Ag/AgCl epidural electrodes positioned over each ectoslvian gyrus with reference the nasion. EEG records were taken every 30 min during the infusion and then hourly for 3 h. A DISA 26 10-channel EEG recording system was used with high-pass filter at 0.3 s and low-pass filter at 100 Hz. The traces were visually assessed for symmetry of frequency and amplitude.
Results Five rats and three cats were studied. In all experiments mean arterial pressure (rodents 90 SD f 10 mmHg; cats 120 SD t 12 mmHg), mean p 0 2 ( > l o 0 mmHg) and pCOz (34 SD Ifr 2 mmHg) remained stable. These models of infusion brain oedema cause: (1) significant increases in white matter specific gravity and an increased white matter tissue water content of between 10.4 and 12.5 ml/ 100 g tissue; (2) the classical histological features of extracellular white matter oedema; (3) small rises in mean ICP (rodents 2 mmHg, cats 10 mmHg) but no cerebral herniation or ischaemia (mean rat rCBF 35 m1/100 g/min; mean cat rCBF 45 m1/100 g/min); and (4) no change in the latency or amplitude of either cortical SEPs elicited by forepaw stimulation or cortically generated spinal M E P s . ' ~ J ~ The baseline rodent EEGs were all symmet-
rical and showed background slow waves (mainly in the delta range) with intermittent alpha and theta activity. Occasional large amplitude sharp components (100 p V ) were seen but general background amplitude was 30-40 pV. The pattern of bilateral slow-wave activity persisted throughout the experiment but at no time did the infused hemisphere show asymmetrical slow-wave activity (Fig. 1). The baseline feline EEGs showed symmetrical theta (6-7 Hz) waveform (30-50 p V ) with faster components to the beta range and intermittent sharp wave transients of between 100 and 150 pV in both hemispheres. During the course of the experiment waveforms remained symmetrical. After 4 h there was a symmetrical increase in the amount of delta activity with decreased sharp waves (Fig. 2). In none of the experiments did the infused hemisphere show focal EEG abnormalities. Overall the feline and rodent EEG waveforms were felt appropriate for the level of anaesthesia. Discussion This study adds additional support to the contention that interstitial brain oedema, uncomplicated by cerebral ischaemia, brain herniations or significant rises in ICP, does not cause significant focal abnormalities in the EEG. Although this model of vasogenic brain oedema can cause a shift to anaerobic metab o l i ~ r n 'and ~ expansion of the white matter extracellular space, the infused fluid has no inherent neuro- or glial toxicity and does not impair rCBF or alter cortical SEPs or cortically generated spinal MEPs.'~-'' The paucity of changes in either SEP17or EEG3!6,7waveforms in many patients with brain tumours correlates well with P E T studies that show only minor perturbations of rCBF and cerebral metabolism in peritumoural Experimental brain tumours in cats that cause profound peritumoural oedema also do not produce either abnormalities in cortical SEPs or EEG waveforms whether examined visually or using Fourier a n a l y ~ i s . ~ The combined mass effect of a brain
Interstitial white matter oedema
435
Downloaded by [RMIT University Library] at 12:47 16 April 2016
R
L
FIG. 1. (a) Rodent EEG from right (R) and leff (L) hemispheres before intracerebral infusion. There is some symmetrical beta activity with occasional high amplitude sharp potentials and irregular slow-wave components (MAP 95 mmHg, p C 0 , 33 mmHg). (b) Sixty minutes after 100 pl saline infusion into the left hemisphere the EEG waveforms remain symmetrical with mainly slow-wave activity but some beta components (MAP 86 mmHg; pCOz 35 mmHg).
tumour and surrounding oedematous brain will eventually cause secondary brain herniation and cerebral ischaemia. It is most likely that these biomechanical effects combined with focal brain parenchymal destruction by the
tumour are responsible for focal and diffuse EEG changes in patients with brain ~ u ~ o u ~ s . Resolution ~ ~ ~ J ~of, focal ~ ~ JEEG ~ abnormalities with ‘anti-oedema’ therapy in some patients with brain turn our^'**^^ is most
Zan R. Whittle et al.
Downloaded by [RMIT University Library] at 12:47 16 April 2016
436
FIG.2. (a) Baseline feline EEG waveforms from right (R) and left (L) ectosylvian regions. Predominant rhythm is theta (6-7 Hz) with occasional faster components. Sharp wave transients are seen on both sides with no overall asymmetry (MAP 112 mmHg;pCO, 33 mmHg; ICP 5 mmHg). (b) Two hours after 600 p1 autologous serum protein (50%) infusion into the right forebrain white matter the EEG is dominated by symmetrical slow-wave activity but some beta activity is still seen. There are fewer sharp wave transients (MAP 121 mmHg;pC02 35 mmHg; ICP mmHg). (c) Even at this time induced hypercapnia ( p C 0 , 5 9 mmHg) produces no asymmetry of EEG waveform. Although the activity is generally of higher amplitude, slow components are more prominent and remain symmetrical.
Interstitial white matter oedema likely attributable to alterations in cerebral perfusionla and biomechanical changes in the brain19 since significant reductions in local brain tissue water content and C T imaged brain oedema occur only after several days of steroid therapy.*O
Acknowledgement
Downloaded by [RMIT University Library] at 12:47 16 April 2016
This work was supported by a Project Grant from the MRC (8500319).
Address for correspondence: Mr I. R. Whittle, Department of Neurosurgery, Western General Hospital, Edinburgh EH4 2XU, Scotland.
References 1 Aguglia U, Gambardella A, Oliveri RL, Lavano A, Camerlingo R, Quattrone A. Triphasic waves and cerebral tumors. Eur Neurol 1990; 3O:l-5. 2 Matsuoka S, Arakaki Y, Numaguchi K, Ueno S. The effect of dexamethasone on electroencephalograms of patients with brain tumours. J Neurosurg 1978; 48~601-8. 3 Nagata K, Gross LE, Kindt GW et al. Topographic electroencephalographic study with power ratio index mapping in patients with malignant brain tumours. Neurosurgery 1985; 17:613-19. 4 Soejima T, Okuda S, Ishikawa T, Matsuoka S. [The effect of glycerol on EEG-abnormality (polymorphic delta wave) in patients with brain tumors.] No-ToShinkel 1987; 39553-60. 5 Steudel WI, Beck V, Becker H et al. EEG in patients with twnours of the cerebral hemispheres. In: Lanksch W, Kazner E, eds. Cranial computerized tomography. Berlin: Springer, 1976: 188-200. 6 Gastaut JL, Michel B, Sabet Hassen S, Corda M, Gastaut H. Electroencephalography in brain oedema (127 cases of brain tumour investigated by computerized tomography). Electroencephalogr Clin Neurophysiol 1979; 46231-55.
437
7 Passerini P, Ferini Strambi L, Sbacci M et al. EEG pattern in tumours with and without cerebral oedema. Electromyogr Clin Neurophysiol 1983; 23: 117-22. 8 Dempsey RJ, Combs DJ, Olson JW, Maley M. Brain ornithine decarboxylase activity following transient cerebral ischaemia: relationship to cerebral oedema development. Neurol Res 1988; 10:175-8. 9 Hossmann K-A, Szymas J, Seo K et al. Experimental transplantation glioma in the adult cat; Part 2. Pathophysiology and magnetic resonance imaging. Acta Neurochir (Wien) 1989; 98:189-200. 10 Schaul N, Ball K, Gloor P, Pappius H. The EEG in cerebral oedema. In: Pappius H, Feindel M, eds. Dynamics of brain oedema. Berlin: Springer, 1976; 144-9. 11 Sutton LN, Bruce DA, Welsh FA, Jaggi JL. Metabolic and electrophysiological consequences of vasogenic brain oedema. Adv Neurol 1980; 28:241-54. 12 Tranmer BI, Iacobacci RI, Kindt GW. Effects of crystalloid and colloid infusions on intracranial pressure and computerized electroencephalographic data in dogs with vasogenic brain oedema. Neurosurgery 1989; 25:173-8. 13 Whittle IR, Piper IR, Miller JD. The contribution of secondary mediators to the etiology and pathophysiology of brain oedema. Acta Neurochir (Wein) 1990; SUPPI 51~71-3. 14 Whittle IR, Miller JD. A rodent model of infusion brain oedema: methodology and pathophysiological effects of saline and protein infusions. Acta Neurochir (Wien) 1990; 105:158-68. 15 Sutton LN, Barranco D, Greenberg J et al. Cerebral blood flow and glucose metabolism in experimental brain oedema. J Neurosurg 1989; 71:868-74. 16 Hino A, Imahori Y, Tenjin H et al. Metabolic and hemodynamic aspects of peritumoural low density areas in human brain tumour. Neurosurgery 1990; 26:615-21. 17 Penn RD. Cerebral oedema and neurological function in human beings. Neurosurgery 1980, 6249-54. 18 Brooks DJ, Beaney RP, Thomas DGT. The role of positron emission tomography in the study of cerebral tumours. Semin Oncol 1986; 13233-93. 19 Miller JD, Sakalas R, Ward JD. Methylprednisolone treatment of brain tumours. Neurosurgery 1977; 1:114-19. 20 Bell BA, Smith MA, Kean D M et al. Brain water measured by resonance imaging. Correlation with direct estimation and changes after mannitol and dexamethasone. Lancet 1987; i:66-9.
