Some General Considerations in Neuroanaesthesia
John M. Turner
Induction of Anaesthesia 174
Conduct of a Craniotomy 175
The anaesthetist in the operating theatre must endeavour to provide good intracranial operating conditions. This means ensuring that the intracranial pressure (ICP) does not rise and that, when the skull is open, brain bulk is not increased. The neurosurgeon must be able to retract the brain easily; too high a pressure on the retractors may result in neuronal damage. Understanding the mechanisms underlying the alterations in ICP in disease, the actions of drugs on ICP and the way they interrelate are important requirements for safe neuroanaesthesia. In intensive care, the manipulation and control of ICP is an important part of care of the neurosurgical patient.
Neurosurgical operations are relatively high-risk procedures. The patient with intracranial space occupation is at risk from the development of high ICP and as the space-occupying lesion (SOL) grows, the danger increases. The patient who has suffered a subarachnoid haemorrhage (SAH) is at further risk from repeated bleeds at any moment. There may also be the danger of vascular spasm and cerebral infarction. Anaesthesia and surgery in the patient with cerebral arterial insufficiency are complicated by the fact that arterial disease affects many vascular beds and the patient with cerebral arterial disease may also have significant disease in the coronary or renal circulations. A head-injured patient is not only exposed to the initial neuronal damage but also to the extension of the neuronal damage by high ICP, cerebral oedema and cerebral ischaemia. Unconsciousness may also worsen the injury by producing hypoxia and hypercarbia.
The anaesthetist must make a careful assessment of the neurosurgical patient's general condition and the extent of the neurosurgical disease. Although many clinical situations require a degree of urgency, the high-risk nature of neurosurgery means that when time allows, conditions which would affect the safety of the patient are evaluated and corrected before surgery.
A preoperative evaluation of pulmonary function is essential. Intraoperative hypoxia and hypercarbia will cause brain swelling, making surgery difficult or impossible, so any potential cause needs to be identified and treated preoperatively. Active pulmonary infection requires appropriate treatment before anaesthesia and surgery start. Any tendency towards asthma needs to be noted together with any treatment the patient needs. Chest X-rays only need to be ordered if there are signs or symptoms in the respiratory system but as a high proportion (at least 20%) of intracranial tumours are metastases1 and frequently from bronchogenic carcinoma, a chest X-ray is indicated to detect a primary tumour.
The prevalence of cardiovascular disease in the population at large means that for neurosurgery, as for any major surgery, the careful assessment of cardiovascular function is essential. Pre-existing untreated arterial hypertension can predispose to cerebrovascular complications and during the perioperative period, the patient is at risk from fluctuations in blood pressure. The limits for autoregulatory control of cerebral blood flow (CBF) are reset during hypertension2 and if inadvertent hypotension is caused, there is considerable danger of cerebral and myocardial ischaemia. In the untreated hypertensive patient, a period of high blood pressure can lead to cerebral swelling, as the autoregulatory limit is exceeded, or cause intracranial haemorrhage. If time allows, it is better to establish a previously unrecognized hypertensive patient on treatment. Treated hypertensive patients should be maintained on their treatment and the anaesthetic technique modified to take into account the drugs used. Hypertension may be a response to high ICP, especially following head injury or subarachnoid haemorrhage.
Ischaemic heart disease is also common and needs to be evaluated by a careful history of the patient's exercise tolerance and by electrocardiograph (ECG). The ECG will also show evidence of previous myocardial infarction, whether old or recent. ECG abnormalities are frequently present in patients who have had a SAH or who have raised ICP.
The clinical neurological assessment needs to elucidate not only the diagnosis but also complicating factors that affect the anaesthetic. The level of consciousness, neurological deficits and occurrence of seizures need to be noted. The existence of intracranial space occupation and raised ICP (persistent headache, vomiting, papilloedema) must be evaluated.
Posterior fossa tumours may cause bulbar palsy and the lower cranial nerves should be examined for impairment of swallowing or laryngeal palsy. A history of repeated aspiration of stomach contents, perhaps with nocturnal bronchospasm, reveals laryngeal incompetence.
A skull X-ray and CT scan or MRI scan may reveal not only the presumptive diagnosis but also the extent of any secondary features such as cerebral oedema, brain shift, hydrocephalus or compression and distortion of the ventricular system. Space-occupying lesions producing more than 10 mm shift of the midline structures suggest significant compromise of the intracranial dynamics.3
A reduced level of response presents many problems. It may be associated with hypoxia and hypercarbia, which may in turn reduce still further the level of response by increasing ICP. Reduced ventilation will allow the development of basal pulmonary collapse and consolidation, leading to pneumonia. Dehydration will develop rapidly as the patient is unable to eat and drink and the increased blood viscosity will predispose to venous thrombosis.
Patients with epilepsy need to have their drugs continued up to the time of operation. Sudden withdrawal of antiepileptic drugs is likely to lead to a worsening of the seizure activity. Cerebral oedema complicating intracranial tumours is frequently successfully treated with steroids such as dexamethasone 8 mg tds. Such a regime often produces a significant improvement, with confused patients becoming rational and the level of response improving.
The prospect of intracranial neurosurgery will be daunting for many patients and it is wise to limit the natural anxiety they feel by using a form of sedative or anxiolytic premedication. This is of value particularly when hypertension poses a danger for the patient, as in aneurysm surgery. Temazepam (10-30 mg) or diazepam (10-20 mg) have proved useful and safe for some years.
Spinal procedures are usefully premedicated with a non-steroidal anti-inflammatory analgesic such as diclofenac, assuming there is no contraindication to the NSAID. We have used one of the modified-release preparations of diclofenac for some years together with one of the benzodiazepines. A patient may be in quite severe pain from a prolapsed intervertebral disc and an analgesic premedication may be indicated.
Induction of Anaesthesia
The detailed management of anaesthesia will be discussed in the specific chapters; here the outlines only are discussed.
The anaesthetist needs to aim for a smooth induction of anaesthesia, avoiding coughing, straining or the production of undue hypo- or hypertension. In the adult an intravenous induction of anaesthesia is normal; either thiopentone or propofol may be used and will produce a fall in ICP by lowering the cerebral metabolic rate (CMRO2) and the CBF.45 Propofol has been demonstrated to produce a 32% fall in cerebrospinal fluid (CSF) pressure 2 min after the induction of anaesthesia with 1.5 mg/kg. The fall in CBF and ICP may be quite short; with propofol, the CSF pressure returned to normal values within 3 min.
Nevertheless, such a fall in ICP is most valuable because many other factors are operating during the induction sequence, which tend to raise ICP. Laryngoscopy, for example, may cause a rapid rise in blood pressure and also be associated with jugular venous obstruction as the laryngoscope is pulled forward to visualize the vocal folds. This action has been implicated in kinking the jugular veins, thus causing an obstruction to cerebral venous drainage.
If non-depolarizing relaxants are used for tracheal intubation it is important to wait for complete relaxation to occur, because any attempt to intubate an incompletely relaxed patient will cause straining and a consequent elevation in blood pressure and intrathoracic pressure.6 Adequate ventilation must be maintained as the action of the relaxant is developing so that hypoxia and hypercarbia are avoided. The delay implied means that the disturbance produced by laryngoscopy and intubation may well occur after the return to normal of the CBF and ICP following the initial dose of induction agent. Laryngoscopy and intubation provoke a transient but marked arterial hypertension and tachycardia. This is a considerable danger for many neurosurgical patients. The rapidity of the rise in blood pressure is such that it is likely to outstrip the ability of autoregulation to control CBF, so that CBF and ICP will rise with the blood pressure. If there is an area of brain in which autoregulation is impaired or absent, then any rise in blood pressure will lead to a rise in ICP. The patient with a cerebral aneurysm is in danger of aneurysmal rupture, especially if the blood pressure rise is rapid.
Control of the pressor response is important and many ways have been suggested. The administration of a second bolus of the intravenous induction agent before laryngoscopy is one and one study7 has suggested that propofol (0.5-1.0 mg/kg) is rather more effective at preventing the development of hypertension than thiopentone. The use of opioids provides only a partial protection against the pressor response89 and some authors suggest delaying the opioid until after the muscle relaxant is given to avoid the danger of increased chest wall stiffness.10 Lignocaine (1.5 mg/kg)11 given 90 s before intubation has been recommended to blunt the blood pressure and ICP response to laryngoscopy. p-Blockers have also been used. Airway
Neurosurgical operations may take several hours and the head may be inaccessible under the sterile drapes. It is important to make sure that the airway and pulmonary ventilation are secure for the duration of the operation. The use of an armoured tracheal tube -that is, a tube with a metal or nylon spiral reinforcing the wall - is necessary. The neurosurgeon will put the head and neck into the best position for the planned operation and this may on occasion mean that the head and neck are flexed, so that the end of the tracheal tube is advanced farther into the bronchial tree. The tracheal tube therefore must be placed with care, especially ensuring that the end of the tube is clear of the carina so that flexion of the neck does not advance the tube into the right main bronchus.
Fixation of the tracheal tube must also be done with care. Tapes should never be tied around the neck to fasten the tube in place, because there is a considerable danger of causing cerebral venous obstruction. The tube may be effectively fixed to the face by adhesive strapping, using hypoallergenic materials if required. Any method of fixation needs to take into account the length of the surgical procedure, the fact that secretions or surgical skin preparation fluids may loosen the strapping and the position required for the operation.
Neurosurgical patients may be positioned supine, in a lateral position, prone or sitting. In the prone or sitting position, there is a great danger that the tracheal tube may be dragged out. These are the positions in which it would be most difficult to reintubate the trachea, especially if the head is fixed in pins. It is clearly important, then, to fix the tube especially securely in these patients and to ensure that the weight of the anaesthetic tubing does not drag on the tracheal tube. A practice that was once commonplace is to support the tracheal tube with an oropharyngeal pack. The pack should not be placed rigidly, because it may obstruct the internal jugular veins, but firmly enough to support the tracheal tube and to keep it stationary. Even the reinforced tubes may kink in an extreme position; placing a pack supports their kink-resisting properties. Once a pack is placed, all members of the anaesthetic team must be informed that there is a pack to be removed before extubation at the end of the procedure.
During intracranial surgery the head is often fixed using a frame with pins, which are applied to the skull. The fixation produces painful stimuli and a hypertensive response may occur, which obviously has the same dangers as the hypertensive response to tracheal intubation. The eyes must be carefully protected.
The start of a craniotomy is also painful, particularly the skin incision and reflecting the galea. The process of cutting the bone flap may also be painful and the physical pressure on the skull involved in drilling a burr hole may cause an increase in ICP.6 Surgery inside the skull is not markedly painful, unless the surgeon stretches the dura, which is most likely to happen near points of dural attachment. Once the bone flap is cut, the surgeon needs to retract the brain to obtain access to deeper structures.
During this period it is important that anaesthesia is arranged to produce good operating conditions, reducing ICP and brain bulk wherever possible. When the skull is open and the ICP is low, the dura can be seen to be slack but moves in response to the cardiac impulse and the pressure changes produced by the lung ventilator. A high ICP will be seen to be distending the dura and dural pulsation in response to the cardiac cycle and the ventilator will not be visible. In such a circumstance, the ICP must be reduced before the dura is incised because a high ICP will extrude the brain through the dural incision, physically damaging the neurones, and the edge of the dura will cut off the blood supply, causing infarction.
The choice of a volatile anaesthetic agent or propofol infusion is controversial. All the volatile agents (including nitrous oxide) have the potential to increase CBF and therefore make the brain difficult to retract. The use of hyperventilation may mask the effects of some of the volatile agents on CBF but this mechanism cannot be relied on, especially when extensive space occupation means that the mechanisms compensating for space occupation are near exhaustion.12
The discussion was developed by Todd et al,13 who prospectively studied three anaesthetic techniques for patients undergoing craniotomy for supratentorial tumour. The patients were assessed carefully, ICP was measured and the extent of the brain swelling noted. Patients were assigned to one of three groups. Group 1 received propofol induction and maintenance of anaesthesia with fentanyl 10 |g/kg loading dose followed by
2-3 |g/kg/h. The propofol infusion was set between 50 and 300 |g/kg/min. In group 2, anaesthesia was induced with thiopentone and maintained with nitrous oxide, oxygen and isoflurane; fentanyl up to 2 |g/kg was given after bone flap replacement. Group 3 also received thiopentone induction of anaesthesia but with fentanyl, as in group 1, and a lower dose of isoflurane. The ICP before craniotomy in group 1 was 12 ± 7 mmHg, in group 2 15 ± 12 mmHg and in group 3 11 ± 8 mmHg. Two patients in groups 1 and 3 but nine patients in group 2 had ICP greater than 24 mmHg. The authors suggest that all three anaesthetic regimes were acceptable. The group also noted speed of emergence and total stay in hospital and hospital costs.
Propofol infusion avoids the need for vasodilating anaesthetic agents and not only Todd et al but also Ravussin et al14 reported on the successful use of propofol infusion. Ravussin et al considered that propofol gave better control of responses to painful stimuli and faster recovery than thiopentone-isoflurane. Fox et all15 used 2.0-2.5 mg/kg of propofol for induction of anaesthesia and 12 mg/kg/h for 10 min followed by 9 mg/kg/h for another 10 min and then 3-6 mg/kg/h for the rest of the study. They showed that the responsiveness of the cerebral circulation was well maintained.
Hyperventilation has been a part of neurosurgical anaesthesia for many years, notwithstanding the concern that metabolism and flow were being uncoupled. CBF changes 4% for each mmHg change in PaCO2; thus at high values (10.6 kPa, 80 mmHg) the cerebral vasculature is maximally dilated and CBF is approximately doubled. Maximum vasoconstriction occurs below 2.6 kPa (20 mmHg), at which level CBF is reduced by 40%.16 Harp and Wollman17 studied safety of hyperventilation and found no evidence of brain hypoxia, even during marked hypocapnia. The vasoconstriction and the reduction in cerebral blood volume (CBV) reduce the ICP and although in the normal brain when hyperventilation is instituted, CBF returns to normal values in about 3 h, the ICP returns to the starting value more slowly. The difference is due to the reduction in brain extracellular fluid volume. Cerebral vasoconstriction reduces the intravascular hydrostatic pressure in the capillary, so extracellular water returns to the circulation under the influence of the plasma oncotic pressure. The result is that brain extracellular fluid volume is reduced, at least in the short term.
The introduction of jugular venous oxygen content measurement18 has allowed the study of hyperventilation and its effects. The study quoted comments that the use of hyperventilation during neurosurgical procedures can result in cerebral venous oxygen desaturation, which the authors define as jugular venous bulb oxygen saturations less than 50%, in up to 40% of patients. The slowing of the cerebral circulation allows increased oxygen extraction by the brain and the authors comment that the desaturation is suggestive of a decreased margin of safety and possibly indicative of a limited oxygen supply with impending tissue hypoxia. In a subsequent study, they investigated the effect of high oxygen tensions (PaO2 = 100-200, 201-300, 301-400 and >400 mmHg) on jugular venous oxygen content at two levels of hyperventilation (PaCO2 = 3.3 kPa and PaCO2 = 4 kPa). They suggested that hyperoxia during acute hyperventilation should be considered for those patients in whom hyperventilation is contemplated and cerebral ischaemia considered a risk. Hyperventilation should also be considered if volatile anaesthetic agents are to be used.19
A profound degree of muscle relaxation is important in the production of good operating conditions. Incomplete muscle relaxation is associated with an increased mean intrathoracic pressure and therefore central venous pressure (CVP); the increased CVP is transmitted to the cerebral veins. Good muscle relaxation is also advisable because of the obvious danger of an incompletely paralysed patient coughing during an intracranial operation.
An intraoperative rise in ICP will impede surgery because if the brain is bulky, it is more difficult to retract. Increased force will be needed on the brain retractors, producing local damage. If, during neurosurgery, the ICP is noted to be high, the likely cause must be identified and corrected. There are also specific methods for lowering ICP.
Checklist for Causes of High ICP During Surgery
• Position Head up
Clear cerebral venous drainage
No abdominal compression
• Adequate ventilation PaCO2 low
Long expiratory pause No PEEP
• Good muscle relaxation
• Avoid cerebral vasodilating drugs
Treatment of High ICP
Mannitol, which is perhaps the agent most frequently used, has many systemic and cerebral effects. Given in doses of 0.5-1.0 g/kg, it raises serum osmotic pressure so that water is drawn into the vascular system from the tissues. The increased oncotic pressure draws water from the brain and reduces brain bulk. As the action develops, the circulating blood volume rises and the haematocrit falls.20 The blood volume remains elevated for 15-30 min and during this time, the blood pressure and CVP may also be elevated. The diuresis limits the rise in blood volume. The decreased haematocrit allows a greater CBF and in patients with intact autoregulation, cerebral vasoconstriction occurs, keeping oxygen supply in balance with demand. The result is that the CBV is reduced and therefore so is the ICP. If autoregulation is impaired then the increased CBF persists, though ICP still falls, though to a small extent.21
Frusemide has been studied the most, though many diuretics have useful intracranial effects. Frusemide22 1.0 mg/kg produces a fall in ICP similar to that produced by 1 g/kg mannitol. It acts by inhibition of sodium and chloride reabsorption in the ascending limb of the loop of Henle and has a separate action in reducing CSF production by suppressing sodium transport. It lowers ICP by mobilizing normal brain extracellular fluid and cerebral oedema. The diuresis reduces blood volume and therefore the low cerebral venous pressure allows resorption of CSF. Frusemide appears not to affect the volume/pressure response, as does mannitol.
The effect of steroids in reducing the oedema related to tumours preoperatively has been mentioned earlier. They are particularly effective in patients with focal lesions and are ineffective when there is widespread brain injury. They have little place in the control of intraoperative high ICP but may be given intraoperatively to reduce postoperative swelling. They reduce the extrachoroidal production of CSF.23
If high ICP occurs, then it is important to check the PaCO2with an arterial sample to avoid any inadvertent hypercapnia. Metabolic Suppression
Shapiro24 described the use of barbiturates intravenously during periods of intraoperative high ICP. The aim is to reduce cerebral metabolic activity, while reducing brain bulk by producing a cerebral vasoconstriction. Propofol and etomidate have also been used to lower ICP and have the advantage over the barbiturates that they are metabolized quickly.
Lignocaine (1.5 mg/kg) may also be used to lower ICP, especially in the patient with cardiovascular instability. 1.5 mg/kg lignocaine is said to be as effective as 3 mg/kg thiopentone in lowering ICP.25
The end of the neurosurgical anaesthetic has been studied less than the start. Leech et al26 measured ICP at the end of surgery, before and after the reversal of the muscle relaxants and after removal of the tracheal tube. After surgery but before reversal of the relaxants, the mean ICP was 11 mmHg but after reversal, with the patient breathing spontaneously, the mean ICP had risen to 21 mmHg. Just before return to the ward, the ICP was still high at 19 mmHg. These are surprisingly high values considering that craniotomy had been performed and CSF drainage had taken place.
The possible role of nitrous oxide in causing an increase in ICP by diffusing into the air space left in the skull after craniotomy was studied by Domino et al.29 They found that all patients in their series had intracranial air but that nitrous oxide was not associated with an increase in ICP as the dura and skull were closed. Indeed, the patients who were maintained on nitrous oxide demonstrated a significant fall in ICP postoperatively. The air disappears only slowly; in one series27 all patients were shown to have intracranial air immediately after craniotomy or craniectomy. In the second postoperative week, 11.8% still had an intracranial air collection large enough, in the opinion of the authors, to put them at risk if nitrous oxide was used for another anaesthetic.
The disturbance of pharyngeal suction and extubation may cause a rise in MAP and an increase in intrathoracic pressure. Hypertension may provoke bleeding from arteries that have been cut during the surgical procedure and the increase in intrathoracic pressure may provoke bleeding from venous channels opened during surgery. The anaesthetic should be continued so that there is no tendency for the patient to cough and strain during the application of head bandages, which often results in head movement. Muscle relaxants should not be reversed until after the bandages have been applied.
The removal of the tracheal tube should be carefully performed when it is certain that the patient is able to breathe effectively. The use of IV lignocaine (1.5 mg/kg) at least 90 s before extubation28 blunts the cardiovascular responses effectively.
The neurosurgical patient requires close observation postoperatively, because many of the preoperative disease processes will still be active, such as cerebral oedema. Haematomas may form and oedema may spread, raising ICP and reducing conscious level. Vascular spasm may worsen, producing local cerebral ischaemia, shown as a neurological deficit. The patient may develop seizures. Monitoring must therefore be close and include neurological observations, observation of the conscious level, such as the Glasgow Coma Scale, observation of the pupil size and reactivity and frequent measurements of pulse rate, blood pressure and respiration.
It is crucial therefore that the patient should be awake and responding at the end of the anaesthetic and choosing the anaesthetic drugs and techniques to achieve not only good operating conditions but also an awake patient without neurological deficit is part of the challenge of neuroanesthesia.
1. Kendall BE. The detection of intracranial tumours. Br J Hosp Med 1980; 23: 116.
2. Strandgaard S, Olesen J, Skinh''y E, Lassen NA. Autoregulation of brain circulation in severe arterial hypertension. BMJ 1973; 1: 507.
3. Bedford RF, Morris L, Jane JA. Intracranial hypertension during surgery for supratentorial tumor: correlation with pre-operative computed tomography scans. Anesth Analg 1982; 61: 430-433.
4. Michenfelder JD. The interdependency of cerebral function and metabolic effects following maximum doses of thiopental in the dog. Anesthesiology 1974; 41: 231.
5. Ravussin P, Guinard JP, Ralley F, Thorin D. Effect of propofol on cerebrospinal fluid pressure and cerebral perfusion pressure in patients undergoing craniotomy. Anaesthesia 1988; 43(suppl): 37-41.
6. Shapiro HM, Wyte SR, Harris AB, Galindo A. Acute intraoperative intracranial hypertension in neurosurgical patients: mechanical and pharmacological factors. Anesthesiology 1972; 37: 399-405.
7. Harris CE, Murray AM, Anderson JM, Grounds RM, Morgan M, Effects of thiopentone, etomidate and propofol on the haemodynamic response to tracheal intubation. Anaesthesia 1988; 43(suppl): 32.
8. Martin DE, Rosenberg H, Aukburg SJ et al. Low dose fentanyl blunts circulatory responses to tracheal intubation. Anesth Analg 1982; 61: 680-684.
9. Kautto UM. Attenuation of the circulatory response to laryngoscopy and intubation by fentanyl. Acta Anaesth Scand 1982; 26: 217-221.
10. Spiekermann BF, Stone DJ, Bogdonoff DL, Yemen TA. Airway management in neuroanesthesia. Can J Anaesth 1996; 43: 820834.
11. Bedford RF, Persing JA, Pobereskin L, Butler A. Lidocaine or thiopental for rapid control of intracranial hypertension? Anesth Analg 1980; 59: 435.
12. Grosslight K, Foster R, Colohan AR, Bedford RF. Isoflurane for neuroanesthesia: risk factors for increases in intracranial pressure. Anesthesiology 1985; 63: 533.
13. Todd MM, Warner DS, Sokoll MD et al. A prospective comparative trial of three anesthetics for elective supratentorial craniotomy. Propofol/fentanyl, isoflurane/nitrous oxide and fentanyl/nitrous oxide. Anethesiology 1993; 78: 1005-1020.
14. Ravussin P, Tempelhoff R, Modica PA, Bayer-Merger MM. Propofol vs thiopental-isoflurane for neurosurgical anaesthesia: comparison of hemodynamics, CSF pressure and recovery. J Neurosurg Anesthesiol 1991; 3: 85.
15. Fox J, Gelb AW, Enns J, Murkin JM, Farrar JK, Manninen PH. The responsiveness of cerebral blood flow to changes in arterial carbon dioxide tension is maintained during propofol-nitrous oxide anesthesia in humans. Anesthesiology 1992; 77: 453.
16. Harper AM, Glass HI. Effect of alterations in arterial carbon dioxide tension on the blood flow through the cerebral cortex at low and normal arterial blood pressures. J Neurol Neurosurg Psychiat 1965; 28: 449.
17. Harp JR, Wollman H. Cerebral metabolic effects of hyperventilation and deliberate hypotension. Br J Anaesth 1973; 45: 256.
18. Matta BF, Lam AM, Mayberg TS, Shapiro Y, Winn HR. A critique of the intraoperative use of jugular venous bulb catheters during neurosurgical procedures. Anesth Analg 1994; 79: 745-750.
19. Jung R, Reisel R, Marx W, Galicich J, Bedford RF Isoflurane and nitrous oxide: comparative impact on cerebrospinal fluid pressure in patients with brain tumours. Anesth Analg 1992; 75: 724-728.
20. Muizelaar JP, Wei EP, Kontos HA, Becker DP. Mannitol causes compensatory cerebral vasoconstriction and vasodilatation in response to blood viscosity changes. J Neurosurg 1983; 59: 822.
21. Muizelaar JP, Lutz HA, Becker DP. Effect of mannitol on ICP and CBF and correlation with pressure autoregulation in severely head injured patients. J Neurosurg 1984; 61: 700.
22. Cottrell JE, Robustelli A, Post K, Turndorf H. Furosemide and mannitol induced changes in ICP and serum osmolality and electrolytes. Anesthesiology 1977; 47: 28.
23. Martins AM, Ramirez A, Soloman LS, Weise GM. The effect of dexamethasone on the rate of formation of cerebrospinal fluid in the monkey. J Neurosurg 1974; 41: 550.
24. Shapiro HM. Intracranial hypertension: therapeutic and anesthetic considerations. Anesthesiology 1975; 43: 445.
25. Bedford RF, Persing JA, Poberskin L, Butler A. Lidocaine or thiopental for rapid control of intracranial hypertension? Anesth Analg 1980; 59: 435-437.
26. Leech PJ, Barker J, Fitch W. Changes in intracranial pressure during the termination of anaesthesia. Br J Anaesth 1974; 46: 315.
27. Reasoner DK, Todd MM, Scamman FL, Warner DS. The incidence of pneumocephalus after supratentorial craniotomy. Observations on the disappearance of intracranial air. Anesthesiology 1994; 80: 1008-1012.
28. Bidwai AV, Bidwai VA, Rogers CR, Stanley TH. Blood-pressure and pulse-rate responses to endotracheal extubation with and without prior injection of lidocaine. Anesthesiology 1979; 51: 171-173.
29. Domino KB, Hemstad JR, Lam AM. Effect of nitrous oxide on intracranial pressure after cranial dural closure in patients undergoing craniotomy. Anesthesiology 1992; 77: 421-425.
Anaesthesia for Surgery of Supratentorial Space-Occupying Lesions
John M. Turner
Intracranial Tumours 183
General Anaesthesia for Craniotomy 185
The challenge of any anaesthetic for neurosurgery is to provide good intracranial operating conditions, with a slack brain and low intracranial pressure (ICP). When a patient has an intracranial space-occupying lesion (SOL), the achievement of a low ICP during surgery demands a careful choice of the most appropriate anaesthetic and an attention to detail.
Patients present for craniotomy for a supratentorial SOL most often because of a tumour but space occupation may also be caused by subdural, extradural or intracerebral haematomas or an intracranial abscess. Even when a tumour is histologically benign, the processes set in train by intracranial space occupation can be fatal if the tumour is not treated. A badly administered or inappropriate anaesthetic may add to the intracranial problems generated by the space occupation, increasing ICP.
These arise from the arachnoid cap cell which is synthetically active but has a slow rate of cell division. They account for 15% of all intracranial neoplasms and 90% are supratentorial. Tentorial meningiomas account for 2-3% but, because of the high rate of brainstem compression, tend to present earlier, whereas frontal lobe meningiomas arising in a 'silent' area of the brain present much later and may become very large before diagnosis (see Fig. 4.3). In general, meningiomas tend to present in the fifth and sixth decades of life. Agents implicated in causation are trauma, hormonal influence and viral infection and there is a definite link with radiotherapy, when lesions arise in younger patients. They are classically benign, causing compression rather than invasion, but have a high rate of local recurrence, particularly associated with radiotherapy, and can be frankly malignant. They are generally highly vascular, deriving large 'feeding' vessels from local blood vessels, both intracranial and extracranial. They are classified using the Helsinki system which grades increasing malignancy from grade I (benign) to grade IV (sarcomatous). Seizures are common and occur in up to 60% of patients presenting for surgery. Mental deterioration is usually slow in onset but eventually focal deficit will occur with a rise in ICP. At surgery, extension outwards along the vault or skull floor may make complete removal impossible.
These are the commonest types of intracranial neoplasms, arising from brain cellular components and ranging from relatively slow-growing astrocytomas to malignant glioblastomas. The variable natural history and unpredictability of malignant transformation require a histological diagnosis. Positron emission tomography (PET) scanning may show up 'hot spots' for metabolism which are highly suggestive of malignant transformation1. Total excision may not be possible and surgery may need to be followed by radiotherapy.
Astrocytomas tend to present with a long history of epilepsy, with late-onset focal deficit. Glioblastomas comprise 25% of primary intracranial tumours and are always supratentorial. They have a 10% 2-year survival rate. They present with a short clinical history, rapid onset and progression of focal deficit and signs of raised intracranial pressure. Epilepsy is less common. Surgery may be required to improve the neurological deficit prior to radiotherapy. Oligodendrogliomas are uncommon and slow growing. They commonly cause epilepsy and may even calcify. Eventually localizing signs will develop.
Whatever the histology, these tumours tend to present with obstructive hydrocephalus due to obstruction of the third ventricle outflow and aqueduct of Sylvius. Papillomas can occur in the choroid plexus. Colloid cysts arise in the third ventricle and are related to ependymomas, which usually arise in the fourth ventricle from cells lining the ventricular system. Pinealomas do not all arise directly from the pineal gland but from the area around it. They cause similar symptoms and can produce midbrain compression. With all these tumours, the presentation is usually related to the onset of hydrocephalus, the patient complaining of severe intermittent headache. Intermittent blockage of the CSF circulation may produce high ICP and occasionally this can precipitate sudden loss of consciousness. Surgical removal may be difficult because of potential damage to the ventricular system or hypothalamus. An endoscopic approach may be possible to intraventricular tumours such as colloid cysts.
Other Primary Tumours
Dermoid and epidermoid neoplasms can arise although these are rare outside the posterior fossa. Metastatic Disease
Secondary lesions from primary cancers elsewhere are the commonest intracranial neoplasms; 20-40% of all cancer patients develop brain metastases.2 Despite their frequency, care should be taken to ensure that the patient does have a metastasis and not another primary tumour.
Removal of a solitary cerebral metastasis is usually worthwhile because most patients do not die of their brain disease and improvement in neurological function after surgery may greatly improve their quality of life. Clearly, these patients may present with a range of systemic problems which may need attention preoperatively.
The space occupation produced by acute extradural and subdural haematomas develops rapidly and allows less time for the brain to accommodate to the SOL. Figure 4.2 shows the classic appearance of an extradural haematoma, with its biconvex shape. Chronic subdural haematomas, usually affecting the elderly, may present weeks after the initial injury. The patient may become confused and drowsy and the state of consciousness may fluctuate, with a mild hemiparesis. A history of trauma may be difficult to elicit and ICP may be normal or only slightly raised. Burr hole aspiration may be all that is required but occasionally open operation may be necessary if the haematoma recollects. Occasionally the insertion of a conduit from the haematoma cavity to the peritoneum may be required to provide for continuous decompression of the haematoma.
The first successful removal of an intracranial abscess was in 1752, when most abscesses were secondary to gunshot wounds or other trauma. Now supratentorial abscesses usually result from local spread of a source of infection from the frontal sinus or middle ear; occasionally they arise as a result of blood borne infection.
Surgery is the mainstay of treatment despite the huge improvement in mortality which occurred following the introduction of antibiotics to neurosurgery in 1943. The key to further improvement in treatment has come with the greater ability to localize cerebral lesions. Introduction of CT scanning led to much earlier diagnosis and early, accurate drainage. The patient presents with a picture of meningitis and raised ICP occurring over a few days or weeks. Aspiration should be carried out via a burr hole and antibiotics instilled into the cavity. This may need to be repeated until the abscess cavity contracts and becomes sterile. The residual cyst may need to be removed. Some surgeons prefer craniotomy and primary excision. The mortality rate remains related to the neurological status of the patient at initial presentation.
The diagnosis of a space-occupying lesion implies a degree of urgency for surgery because as the SOL develops, the intracranial compensating mechanisms for space occupation are being exhausted. A general history should be taken, with particular reference to cardiac and respiratory disease and to neurological deficit. An examination should be carried out in the usual way, again with special attention paid to neurological deficit which may compromise the airway, hinder safe emergence from anaesthesia or be due to raised intracranial pressure. It is particularly important to identify raised ICP and to assess the degree of intracranial space occupation preoperatively as extra care must be taken to ensure the patient's condition is not worsened by manoeuvres preoperatively or at induction. A history of headache together with papilloedema suggests that high ICP is present.
The patient's full blood count, urea and electrolytes and clotting should be checked and blood crossmatched as appropriate. The amount of blood loss depends on the type of tumour and its vascularity; a large meningioma can be associated with heavy blood loss but for most craniotomies 2-4 units of blood are sufficient. Chest X-ray and ECG should be ordered if indicated. It should be remembered that neurological disease can cause a variety of ECG changes including ischaemic changes and arrhythmias. Some patients will have been prescribed dexamethasone for several days preoperatively to reduce reactive oedema.
Normal medication, especially anticonvulsants and antihypertensive drugs, should be continued until just before surgery. Sedative premedication may be desirable in order to allay anxiety. Respiratory depression with subsequent hypercarbia should be avoided, especially in the presence of raised ICP. For this reason we avoid opiates and usually prescribe 10-20 mg of temazepam or 10-15 mg diazepam, with 10 mg metoclopramide orally 90 min preoperatively.
General anaesthesia for surgery in patients with an acute mass effect includes head elevation, osmotherapy with mannitol, artificial ventilation to prevent hypoxia and hypercarbia, with controllable hypocarbia, administration of cerebral protective agents such as barbiturates and maintenance of cerebral perfusion pressure with fluids and inotropes if necessary. Emergency craniotomy for decompression may be necessary.
The assessment of supratentorial lesions is made considerably easier by the improvement in imaging techniques, CT scanning and MRI. These allow early, precise location of lesions and give some idea of the probable histological diagnosis. The scans should be examined to give information on:
• ventricular distortion or CSF obstruction;
• degree of contrast enhancement;
• proximity to venous sinus.
The size of the mass depends partly on whether the tumour is developing in a silent or an eloquent area of the brain. Tumours in a silent area may grow so large before they present that they cause a very considerable compromise of intracranial dynamics. Assessment of the degree of intracranial space occupation is important; if there is more than 10 mm shift of the midline structures, for example, volatile agents should be used with care.345 The amount of oedema may turn a relatively small lesion into a more serious problem. The degree of enhancement with intravenous radiographic contrast shows the degree of abnormal or damaged blood-brain barrier (BBB) in the lesion and it is through this damaged BBB that the contrast penetrated to the stroma of the tumour. A vascular tumour may have a low vascular resistance and frequently in angiography cerebral veins draining the tumour fill early, during the arterial or capillary phases of the angiogram, reflecting the fast flow. Such a tumour, especially if it is near one of the venous sinuses, has the potential for causing major blood loss as resection is undertaken.
The increasing availability of metabolic imaging such as PET, MR spectroscopy and single photon emission tomography (using thallium-201 which is specifically taken up by tumour cells but not by necrotic areas) will offer more precise information on the size and location of the tumour.
Patients suspected of having an astrocytoma should undergo stereotactic biopsy prior to craniotomy to confirm the histological diagnosis.
The dangers of intraoperative high ICP in the presence of a SOL mean that the anaesthetist must be especially careful to choose anaesthetic agents and techniques which lower ICP. In particular, it is important to avoid a rise in cerebral venous pressure, cerebral vasodilation, hypercapnia and hypertension. All these circumstances can be provoked during the induction and maintenance of anaesthesia.
The effects of anaesthetic agents and techniques on cerebral blood flow (CBF) and ICP have been discussed in earlier chapters. All volatile agents have the ability to raise CBF and ICP, though their action can be modified by hyperventilation and by the use of opioids which reduce the MAC of the agents. Adams6 demonstrated that isoflurane combined with hypocapnia (PaCO2 about 25 mmHg) was not associated with a rise in ICP and Jung7 makes a similar point.
Muzzi has shown that 1 MAC desflurane causes a gradual but significant rise in ICP in patients with a supratentorial SOL, whereas isoflurane does not.8 Enflurane has the potential to produce high-amplitude spike and wave EEG complexes and whilst in one study,9 the postoperative seizure rate was no higher in the patients receiving enflurane than in those receiving isoflurane, it is probably best avoided. Sevoflurane seems to have similar effects to isoflurane.1011
The potential problems of the volatile agents can be avoided by using intravenous maintenance of anaesthesia with propofol. Barbiturates and propofol reduce CMRO2 and therefore CBF. The reduction in CBF and cerebral blood volume (CBV) is helpful in maintaining as low an ICP as possible. The cerebral effects of propofol, together with its rapid recovery, make possible the use of an infusion technique for maintenance of anaesthesia throughout a craniotomy. Infusion rates have been published12 and one group13 has used a maintenance infusion rate of 100 ^g/kg/min for elective neurosurgery after a bolus dose for induction of 1.5 mg/kg. The infusion should be started at the same time as the bolus dose, to achieve an effective plasma concentration rapidly.
The opioids are a valuable adjunct to neuroanesthesia, not least because of the cardiovascular stability they produce. They also reduce the MAC of volatile agents and therefore allow agents such as isoflurane to be used safely in lower concentrations so that the effect on CBF is lessened. Many opioids can be used as long as they do not induce a fall in MAP.141516
Induction of anaesthesia can be with either thiopentone or propofol. Neuromuscular blockade can be achieved with atracurium or vecuronium, although newer agents such as mivacurium or rocuronium17 may also be used. Intubation should not be performed until enough time has elapsed for the muscle relaxation to be complete. The pressor response to laryngoscopy and intubation will cause an increase in the size of vascular tumours in particular, because the tumour blood supply will not be under autoregulatory control. At the same time, laryngoscopy tends to kink the jugular veins, so that cerebral venous outflow is impaired. The result is a great tendency for a dangerous rise in ICP. The pressor response should, therefore, be controlled. A second dose of the induction agent or the use of IV lignocaine 1.5 mg/kg 90 s before intubation is well established, as is the use of agents such as esmolol.
The airway should be secured with an armoured endotracheal tube, taped on the contralateral side to operation. It may be desirable to secure the tube further using a throat pack. Careful taping is required to prevent both extubation and venous congestion. Pressure ventilation of the lungs is arranged so that there is a slow rate, with a long expiratory pause; PEEP is normally not applied. A nasogastric tube should be considered for long operations.
Operations may last several hours so the patient should be carefully positioned with bony points padded, eyes protected with tape or gel and care taken to ensure tubing is not pressed against the patient's skin. Patients may be positioned supine or slightly rotated, with one shoulder raised, or they may be placed in a lateral position. A head-up tilt of the table of about 15° is essential to aid cerebral venous drainage. Pinealomas may be approached with the patient either prone or sitting.
A urinary catheter is necessary to assess fluid status. The patient's head position on the operating table is fixed either on a horseshoe-type head rest or by the insertion of skull pins. The application of the pins represents another stimulus causing an increase in blood pressure response18 and great care should be taken to ensure that the patient is adequately anaesthetized before the pins are applied.
The choice of volatile or intravenous agents for maintenance of anaesthesia has already been described. Todd et al,19 in a complex and thoughtful study, effectively demonstrated that propofol/fentanyl, nitrous oxide/high-dose isoflurane or fentanyl/low-dose isoflurane techniques of anaesthesia can be used. It is interesting to note that although there were no significant intergroup differences in mean ICP, a relatively high number of patients in the nitrous oxide/highdose isoflurane group (9/40) showed ICP greater than 23 mmHg. In the propofol/fentanyl group, there were only two patients out of 40 and in the fentanyl/lowdose isoflurane only two out of 41 patients with high ICP (greater than 23 mmHg).
Whichever technique is chosen, therefore, there remains a danger of high ICP and it is important for the anaesthetist to check on the state of the brain once the skull has been opened. A tense, bulging dura is dangerous and may prevent surgery proceeding, because if the dura is opened the brain is squeezed through the dural defect and infarcts. Reduction of ICP is necessary, first of all eliminating any faults of technique that have given rise to the high ICP and then using specific methods to lower ICP. A checklist for this has been given in Chapter 12, p. 176.
The use of mannitol at the start of the craniotomy is valuable, because mannitol not only lowers ICP but also appears to reduce the volume/pressure response (VPR).20 This reduces the pressure required on the surgical retractors and therefore reduces the extent of neuronal damage. The dose of mannitol is controversial; suggested doses have ranged from 0.5 to 2.0 g/kg body weight, given as a 20% solution. Smaller doses (0.5-1.0 g/kg) can be given over 15-20 min at the start of an operation, but the larger doses should be infused over 1-1.5h. Mannitol exerts its action in a number of ways. The administration of mannitol raises serum oncotic pressure, so that water is drawn into the vascular system. The withdrawal of brain extracellular fluid reduces brain bulk and therefore ICP. The increased vascular volume is demonstrated by a temporary increase in blood pressure and CVP before the onset of the diuresis. It also produces a reduction in the haematocrit.
Muizelaar et al2122 suggested that the reduction in haematocrit is associated with an increase in cerebral blood flow so that oxygen supply is greater than metabolic demand. When cerebral autoregulation is intact, the balance of oxygen supply and demand is restored by cerebral vasoconstriction which, by producing a fall in CBV, also reduces ICP. In patients with intact autoregulation, there was a 27% fall in ICP but in a group of head-injured patients with impaired autoregulation, mannitol produced a rise in CBF of 17.9%, but a fall in ICP of 4%. Ravussin et al23 infused mannitol rapidly and found that there was a 25% increase in CBV and ICP. The rise in ICP lasted for a short time
(5 min), but the rise in CBV lasted for 15 min. They suggested that the ICP begins to fall after mannitol only when the dehydrating effect has begun to counteract the initial increase in CBV.
Mannitol may theoretically worsen oedema by crossing a damaged BBB.
Frusemide may be used either on its own or in combination with mannitol to reduce intracranial hypertension. Frusemide (1 mg/kg) reduces ICP to the same extent as mannitol (1 g/kg).24 It has the advantage of mobilizing oedema fluid more effectively than mannitol and lowering CVP, so that cerebral venous pressure is low and therefore reabsorption of CSF optimal.
Metabolic suppression with thiopentone,25 propofol and lignocaine26 may also be used to reduce intracranial hypertension.
Careful monitoring is essential and must include oxygen saturation, ECG and end-tidal CO2 analysis. Two large-bore venous cannulae must be placed when a vascular tumour is being excised. Continuous direct measurement of arterial pressure and central venous pressure is essential. A central venous line is also valuable for administration of hypotensive agents when the tumour is vascular. A temperature probe is placed in the oesophagus. Arterial blood gases should be taken early in the procedure to check the validity of the end-tidal reading.
Vascular tumours (notably meningiomas) can be associated with very fast blood loss and if the tumour is on the convexity of a cerebral hemisphere, the blood loss may occur during the cutting of the bone flap, when the surgeon may not be in a position to stop the bleeding.
The anaesthetist thus needs to have everything needed for rapid transfusion ready at the start of a craniotomy for a vascular tumour. Crossmatched blood needs to be in the theatre suite and there must be a large-bore venous cannula in place, together with arterial and CVP measurement. A blood warmer needs to be set up and acid-base estimations available quickly.
Further significant and persistent blood loss may occur during the subsequent resection of a meningioma, so it is important that the initial blood loss is replaced. The cerebral vasoconstriction that occurs with haemorrhage reduces CBF significantly,27 especially in the junctional areas between the major cerebral vessels.
The need for induced hypotension has lessened in neuroanaesthesia following the improvements in surgical techniques. The use of propofol infusion with opioid analgesics and moderate hyperventilation provides a satisfactory surgical field for most neurosurgical operations but occasionally the resection of a large vascular tumour may still require hypotension.
Before hypotension is used, the patient's physical state should be carefully assessed, particularly looking for signs of ischaemia in the cerebral or coronary circulations or a history of hypertension. Hypotension should not normally be applied until the dura is open and the length of hypotension should be kept as short as possible. Hypovolaemia should not be allowed to coexist with induced hypotension; blood replacement must parallel blood loss. CBF is maintained constant as long as the MAP is between 60 and 160 mmHg, so hypotension to a MAP of 60-70 mmHg, which will provide a surgical field with reduced oozing, may be of value.
Monitoring must be extensive and reliable; as well as the monitoring mentioned, jugular venous oxygen content measurements are valuable, as are transcranial Doppler measurements of flow velocity.
Hypotension is most easily induced by the combination of labetalol (10-20 mg) followed by sodium nitroprusside infusion (0.01%). The infusion is best given along the central venous line, so that the time lag between changing the infusion rate and observing the effect on blood pressure is kept to a minimum.
Sodium nitroprusside causes cerebral vasodilatation28 but the effect can be overcome by ensuring that the arterial pressure is lowered to at least 70% of the control value. If the anaesthetic has been so arranged that the patient's cardiovascular system is stable and not responding to painful stimuli, hypotension should be easily achieved with such small doses of propofol that toxicity29 is not invoked. If the patient develops a tachycardia or is resistant to sodium nitroprusside, it is better to supplement the action of sodium nitroprusside by another drug, such as labetalol or propranolol to control heart rate, or to increase the infusion of propofol or the inhaled concentration of isoflurane, rather than using excessive infusion rates of sodium nitroprusside.
Other hypotensive agents are available, such as trimetaphan28 and trinitroglycerine30 31 (GTN). GTN, like sodium nitroprusside, causes an increase in ICP and also a marked tachycardia. Trimetaphan is a less effective drug than either but does not cause a rise in
ICP, unless the patient is suffering from extreme degrees of intracranial space occupation, in which case the ganglionic blockade may produce an increase in CBF by blocking the sympathetic supply to the cerebral vessels.28
When induced hypotension is being used, it is important that the surgeon knows this and that, after the tumour is resected, blood pressure is returned to normal before the dura is closed. This is essential so that the bleeding points can be visualized and sealed.
Intravenous fluids should be chosen with care; overtransfusion will lead to a high CVP and therefore predispose to high ICP and use of solutions of glucose in water worsen cerebral oedema. Patients with severe intracranial hypertension may have been drowsy or vomiting preoperatively and therefore may be hypovolaemic. If mannitol has been used, fluids should be given to replace the deficit produced by diuresis. Mannitol may also produce hyponatraemia and hypokalaemia.
Normal saline or Hartmann's solution is indicated for fluid therapy during the procedure and should be given to replace fluid losses and controlled by the CVP to avoid overtransfusion. Colloid solutions, such as modified gelation (Gelofusine), may be given and blood loss over 1 litre should be replaced as appropriate.
The closure of a craniotomy may take a little time. Following surgery for a space-occupying lesion, some postoperative brain swelling is likely and great care should be taken to ensure that the end of the anaesthetic is smooth, without undue hypertension, coughing or straining. During the closure period, the propofol and relaxant infusions can be reduced, the aim being to ensure that the patient is awake at the end of the procedure, but reversal of the muscle relaxant must be left until after any dressings or head bandages have been applied. Much movement of the head may take place then and the patient should not cough or strain.
In order to avoid hypertension on extubation, removal of the endotracheal tube and suctioning of the pharynx may be covered with IV lignocaine 1.5 mg/kg 90 s before extubation. Labetalol can also be given to obtund these responses.32
Patients in whom a tight brain was present during surgery or where there was excessive blood loss or in whom oedema spread is marked should be considered for postoperative pressure ventilation in an intensive care unit. Normally, however, anaesthetic technique should be so judged as to produce an awake, responding patient in the recovery area, so that neurological monitoring to detect the postoperative complications of haematoma formation may be started.
1. De Witte O, Levivier M, Vioon P et al. Prognostic value positron emission tomography with [1 8F] fluoro-2-deoxy-D-glucose in the low-grade glioma. Neurosurgery 1996; 39: 470-476.
2. Kendall BE The detection of intracranial tumours. Br J Hosp Med 1980; 23: 116.
3. Bedford RF, Morris L, Jane JA. Intracranial hypertension during surgery for supratentorial tumor: correlation with pre-operative computed tomography scans. Anesth Analg 1982; 61: 430-433.
4. Campkin TV, Flynn RM. Isoflurane and cerebrospinal fluid pressure - a study in neurosurgical patients undergoing intracranial shunt procedures. Anaesthesia 1989; 44: 50.
5. Grosslight K, Foster R, Colohan AR, Bedford RF. Isoflurane for neuroanesthesia: risk factors for increase in intracranial pressure. Anesthesiology 1985; 63: 533.
6. Adams RW, Cucchiara RF, Gronert GA, Messick JM, Michenfelder JD. Isoflurane and cerebrospinal fluid pressure in neurosurgical patients. Anesthesiology 1981; 54: 97-99.
7. Jung R, Reisel R, Marx W, Galicich J, Bedford RF Isoflurane and nitrous oxide: comparative impact on cerebrospinal fluid pressure in patients with brain tumours. Anesth Analg 1992; 75: 724-728.
8. Muzzi DA, Losasso TJ, Dietz NM, Faust RJ, Cucchiara RF, Milde LN. The effect of desflurane and isoflurane on cerebrospinal fluid pressure in humans with mass lesions. Anesthesiology 1992; 76: 720-724.
9. Christys AR, Moss E, Powell D. Retrospective study of early postoperative convulsions after intracranial surgery with isoflurane of enflurane anaesthesia. Br J Anaesth 1989; 62: 624-627.
10. Scheller MS, Teteishi A, Drummond JC, Zornow MH. The effects of sevoflurane on cerebral blood flow, cerebral metabolic rate for oxygen, intracranial pressure, and the electroencephalogram are similar to those of isoflurane in the rabbit. Anesthesiology 1988; 68: 548-551.
11. Takahashi H, Murata K, Ikeda K. Sevoflurane does not increase intracranial pressure in hyperventilated dogs. Br J Anaesth 1993; 71: 551-555.
12. Roberts FL, Dixon J, Lewis GTR, Tackley RM, PrysRoberts C. Induction and maintenance of propofol anaesthesia. A manual infusion scheme. Anaesthesia 1988; 43(suppl): 14-17.
13. Ravussin P, Guinard JP, Ralley F, Thorin D. Effect of propofol on cerebrospinal fluid pressure and cerebral perfusion pressure in patients undergoing craniotomy. Anaesthesia 1988; 43(suppl): 37-41.
14. Fitch W, Barker J, Jennett WB, McDowall DG. The influence of neuroleptanalgesic drugs on cerebrospinal fluid pressure. Br J Anaesth 1969; 41: 800.
15. Werner C, Kochs E, Bause H, Hoffman WE, Schulte am Esch J. Effects of sufentanil on cerebral hemodynamics and intracranial pressure in patients with brain injury. Anesthesiology 1995; 83: 721-726.
16. Warner DS, Hindman BJ, Todd MM et al. Intracranial pressure and hemodynamic effects of remifentanil versus alfentanil in patients undergoing supratentorial craniotomy. Anesth Analg 1996; 83: 348-353.
17. Schramm WM, Strasser K, Bartunek A, Spiss CK. Effects of rocuronium and vecuronium on intracranial pressure, mean arterial pressure and heart rate in neurosurgical patients. Br J Anaesth 1996; 77: 607-611.
18. Shapiro HM, Wyte SR, Harris AB, Galindo A. Acute intraoperative intracranial hypertension in neurosurgical patients: mechanical and pharmacological factors. Anesthesiology 1972; 37: 399-405.
19. Todd MM, Warner DS, Sokoll MD et al. A prospective comparative trial of three anesthetics for elective supratentorial craniotomy. Propofol/fentanyl, isoflurane/nitrous oxide and fentanyl/nitrous oxide. Anesthesiology, 1993; 78: 1005-1020.
20. Leech P, Miller JD. Intracranial volume/pressure relationships during experimental brain compression in primates. 3. The effect of mannitol and hypocapnia. J Neurol Neurosurg Psychiat 1974; 37: 1105.
■21. Muizelaar JP, Wei EP, Kontos HA, Becker DP. Mannitol causes compensatory cerebral vasoconstriction and vasodilatation in response to blood viscosity changes. J Neurosurg 1983; 59: 822.
■22. Muizelaar JP, Lutz HA, Becker DP. Effect of mannitol on ICP and CBF and correlation with pressure autoregulation in severely head injured patients. J Neurosurg 1984; 61: 700.
■23. Ravussin P, Archer DP, Meyer E, Abou-Madi M, Yamamotu L, Trop D. The effect of rapid infusions of saline and mannitol on cerebral blood volume and intracranial pressure in dogs. Can Anaesth Soc J 1985; 32: 506.
■24. Cottrell JE, Robustelli A, Post K, Turndorf H. Furosemide and mannitol induced changes in ICP and serum osmolality and electrolytes. Anesthesiology 1977; 47: 28.
Was this article helpful?