Systemic hypotension in neurosurgery

Controlled hypotension and blood loss during frontoorbital advancement

OBJECT

Controlled hypotension is routinely used during open repair of craniosynostosis to decrease blood loss, although this benefit is unproven. In this study the authors analyzed the longitudinal relationships between intraoperative mean arterial pressure (MAP) and calculated blood loss (CBL) during frontoorbital advancement (FOA) for craniosynostosis.

METHODS

The authors reviewed the records of infants with craniosynostosis who had undergone primary FOA between 1997 and 2009. Anesthesia records provided preoperative and serial intraoperative MAP. Interval measures of CBL had been determined during the course of the operation. The longitudinal relationships between MAP(mean), MAP(change), and CBL(change) were assessed over the same time interval and compared between adjacent time intervals to determine the directionality of associations.

RESULTS

Ninety infants (44 males and 46 females) underwent FOA at a mean age and weight of 10.7 ± 12.9 months and 9.0 ± 7.0 kg, respectively. The average intraoperative MAP was 56.1 ± 4.8 mm Hg, 22.6 ± 12.1% lower than preoperative baseline. A negative correlation was found between CBL(change) and MAP(mean) over the same interval (r = -0.31, p < 0.05), and an inverse relationship was noted between CBL(change) of the previous interval and MAP(change) of the next interval (r = -0.07, p < 0.05). Finally, there was no significant association between MAP(change) of the previous interval and CBL(change) of the next interval.

CONCLUSIONS

Calculated blood loss demonstrated a negative correlation with MAP during FOA. Directionality testing indicated that MAP did not affect intraoperative blood loss; instead, blood loss drove changes in MAP. Overall, these findings challenge the benefit of controlled hypotension during open craniofacial repair.

Basilar tip aneurysm - adenosine induced asystole for the treatment of a basilar tip aneurysm following failure of temporary clipping

We report on a giant basilar tip aneurysm in a 48-year-old woman that could not be clipped despite temporary occlusion of the basilar trunk. Adenosine induced cardiac asystole reduced the aneurysm's wall tension and coupled with increased exposure resulting from brain relaxation allowed for the aneurysm to be satisfactorily occluded via an extended right pterional craniotomy. In difficult vascular cases, adenosine induced cardiac standstill is a useful technique which may allow for technical success when other methods such as temporary clipping fail. It acts to both reduce aneurysm wall tension and increase operative exposure by reducing vascular, and hence brain volume.

Inhibition of nicotinic receptor-mediated catecholamine secretion by Dryobalanops aromatica in bovine adrenal chromaffin cells

Effect of the aqueous extract from a medicinal plant Dryobalanops aromatica(Dipterocarpaceae) on catecholamine secretion was investigated in bovine adrenal chromaffin cells. The aqueous extract inhibited [(3)H]norepinephrine ([(3)H]NE) secretion induced by 1,1-dimethyl-4-phenylpiperazinium iodide (DMPP), a nicotinic acetylcholine receptor (nAChR) agonist, with a half-maximal inhibitory concentration (IC(50)) of 8.4 +/- 1.7 microgml(-1). Increases in cytosolic calcium ([Ca(2+)](i)) and sodium ([Na(+)](i)) induced by DMPP were also inhibited by the extract. However, the binding of [(3)H]nicotine to nAChRs was not affected by the addition of the extract in receptor binding competition analysis, suggesting that active components in the extract and nicotine do not share the binding site in the nAChR. On the other hand, [Ca(2+)](i)increases induced by high K(+), ionomycin, bradykinin, angiotensin II, and thapsigargin were not inhibited by the extract. The data suggest that the extract from D. aromatica specifically inhibits catecholamine secretion by blocking nAChR in a noncompetitive manner.

The use of hypothermia: a role in the treatment of neonatal asphyxia?

Perinatal asphyxia remains one of the most devastating neurologic processes. Although the understanding of the pathophysiology after perinatal asphyxia is extensive, there are few therapeutic interventions available to prevent or even mitigate the devastating process that unfolds after injury. The search for a safe and efficacious therapy has prompted scientists and clinicians to consider various promising therapies. One such therapy is therapeutic hypothermia. On the basis of adult, pediatric, and animal research, there is increasing evidence to suggest that therapeutic hypothermia may be an effective intervention to lessen the secondary neuronal injury that ensues after a hypoxic-ischemic insult. In this article the historic and modern-day uses of therapeutic hypothermia are first reviewed. The pathophysiology of neonatal asphyxia is examined next, with emphasis on the changes that occur when therapeutic hypothermia is implemented. Potential side-effects of the therapy in the neonate and the debate over systemic vs selective hypothermia are discussed. Lastly, although hypothermia as a potential treatment modality for neonates with hypoxic-ischemic encephalopathy is supported by numerous studies, the need for well-designed multicenter trials with detailed patient entry criteria and therapeutic conditions is emphasized.

Selective cooling of brain using profound hemodilution in dogs

A new method of selective cooling of the brain was studied under profound hemodilution in 17 dogs. The carotid and vertebral arteries were bilaterally exposed, and the right vertebral artery was destroyed to provide an infusion route for cold solution for brain cooling. After the other three cerebral arteries were clamped simultaneously in the neck under low-dose heparinization, cold Ringer's lactate solution was immediately perfused into the right vertebral artery. Brain temperatures fell gradually in two dogs, and the experiments were terminated. In 10 dogs, the brain temperature fell to 28 degrees C within 4.4 +/- 1.5 minutes and was maintained at 27.0 +/- 1.0 degrees C for 60 minutes. During this interval, the body temperature was 33.9 +/- 1.6 degrees C, the stump pressure of the vertebral artery was 58 +/- 15 mm Hg, and the hematocrit value of cerebral venous blood was 7.2 +/- 4.2%. Inspection of the brain during infusion revealed paleness of the cortical vessels and no evidence of swelling. All animals survived in good condition until the time of death at 10 weeks. Histological examination of the brain revealed no evidence of ischemic injury. In a control study of five dogs, Ringer's solution at 38 degrees C was infused in the same manner as the cold solution. None of these dogs recovered from anesthesia. It is concluded that selective cooling of the brain under profound hemodilution has a protective effect on cerebral ischemia and provides a relatively bloodless operative field.

Carbon dioxide reactivity and local cerebral blood flow during prostaglandin E1- or nitroglycerin-induced hypotension

The aims of this randomized study were to determine the effect of prostaglandin-(PGE1) or nitroglycerin-(TNG) induced hypotension on local cerebral blood flow (LCBF) and carbon dioxide reactivity during isoflurane anaesthesia in 20 patients after subarachnoid haemorrhage (SAH) scheduled for aneurysm clip ligation. Mean arterial blood pressure decreased immediately, after giving either PGE1 or TNG. The LCBF, measured using a thermal gradient blood flowmeter, was unchanged after PGE1, while the LCBF increased after TNG infusion (control; 47.6 + 10.0, 60 min after infusion; 55.1 +/- 6.5 (P < 0.05), before clipping; 55.5 +/- 7.8 (P < 0.05)) but returned to control values after its discontinuation. Carbon dioxide reactivity, calculated from % delta LCBF/delta PaCO2 was unchanged during PGE1- or TNG-induced hypotension (PGE1; 2.13 +/- 0.9, 2.48 +/- 0.68 and 2.31 +/- 0.79%/mmHg for before, during and after hypotension respectively) (TNG; 2.08 +/- 0.68, 2.17 +/- 0.64 and 2.02 +/- 0.69%/mmHg for before, during and after hypotension respectively). Carbon dioxide reactivity correlated with presurgical neurological status (rs = -0.7, -0.648 and -0.458 for before, during and after hypotension respectively) and the initial LCBF (rs = -0.605). These results suggest that both PGE1 and TNG are useful drugs for induced hypotension for cerebral aneurysm surgery, because neither decreased LCBF.

Carbon dioxide reactivity during prostaglandin E1 induced hypotension for cerebral aneurysm surgery

The cerebral vasomotor reactivity to carbon dioxide was studied, using a thermal gradient blood flow meter in 43 patients with intracranial cerebral aneurysm under deliberate hypotension induced by prostaglandin E1 (PGE1) infusion. The patients were divided into three groups according to the neurological status. Patients in Groups A and B had subarachnoid haemorrhage due to ruptured cerebral aneurysms. Group A consisted of 23 patients with a neurological grade of I-II and Group B consisted of 11 patients with a grade of III-V. Nine patients with non-ruptured cerebral aneurysm served as controls (Group C). After the dura was opened, local cerebral blood flow (LCBF) was measured. The PGE1 was started with an initial dose of 0.1 microgram.-kg-1.min-1 and the dose was adjusted to maintain MAP at about 70 mmHg. The LCBF and carbon dioxide (CO2) reactivity were estimated during and after PGE1 administration. The LCBF did not change among groups throughout the study period. Carbon dioxide reactivity was estimated as follows: absolute; delta LCBF/delta PaCO2, and relative; % delta LCBF/delta PaCO2 after changing PaCO2 by increasing minute ventilation.(ABSTRACT TRUNCATED AT 250 WORDS)

Cerebrovascular and metabolic changes during the delayed vasospasm following experimental subarachnoid hemorrhage in baboons, and treatment with a calcium antagonist

A model has been designed in baboons for simulating the clinical situation during the late phase of vasospasm in patients with subarachnoid hemorrhage (SAH). A total amount of 14-33 ml autologous blood was injected into the cisternal system on 3 occasions in the course of 4 days. Neurological symptoms were seen, and the mortality rate was 29%. Angiography 3 days after the last injection showed arterial vasoconstriction amounting to 23% in the vertebro-basilar system, and 11% (right) and 18% (left) in the carotid system. Cerebral blood flow (CBF) measured by the intra-arterial 133Xe technique and the cerebral metabolic rate of oxygen (CMRO2) were reduced by 18% and 11%, respectively. The hypercapnic CBF response was significantly impaired, from a mean of 3.90 ml/100 g/min to 1.72 ml/100 g/min of flow increase for each mm Hg elevation of paCO2. Autoregulation, tested by administration of angiotensin II, was also significantly affected as evidenced by a pressure-dependent increment of CBF during hypertension in 5 out of 7 animals tested. The impaired autoregulation was reflected in the autoregulatory index, which in the whole group increased from 0.06 ml/100 g/min for each mm Hg increase in MABP in the pre-SAH animals to 0.29 ml/100 g/min per mm Hg post-SAH. Treatment with the calcium antagonist, nimodipine (0.5 microgram/kg/min i.v. during 45 min), enhanced CBF significantly by 17% before experimental SAH, whereas after SAH the effect was slight and did not reach statistical significance; CMRO2 was not significantly affected in either group. Intravenous nimodipine combined with hypertension resulted in a marked increase in the autoregulatory index to 1.58 ml/100 g/min per mm Hg in pre-SAH animals and a less pronounced increment to 0.58 ml/100 g/min per mm Hg following experimental SAH. The beneficial effect of nimodipine reported in SAH patients is therefore, in view of our findings, more likely due primarily to a protective mechanism at the cellular level than to an influence on the vascular bed.

Current management of cerebral aneurysms: is it based on facts or myths?

The overall morbidity and mortality from aneurysmal subarachnoid hemorrhage have not changed significantly over the past 30 years in spite of great progress in surgical technology and diagnostic capabilities. Most series report major morbidity and mortality figures of approximately 60%. The reason for the currently poor prognosis is due, in part, to the sometimes devastating effect of the initial hemorrhage over which we have no control. Another reason for our inability to improve these dismal statistics may be that the currently accepted treatment regimens have proven to be ineffective. Eighteen misconceptions or "myths" surrounding the treatment of aneurysmal subarachnoid hemorrhage have been identified as factors that have largely determined the currently accepted, ineffective treatment protocols, and these are discussed.

Cerebrovascular effects of hypocapnia during adenosine-induced arterial hypotension

Profound arterial hypotension is a commonly used adjunct in surgery for aneurysms and arteriovenous malformations. Hyperventilation with hypocapnia is also used in these patients to increase brain slackness. Both measures reduce cerebral blood flow (CBF). Of concern is whether CBF is reduced below ischemic thresholds when both techniques are employed together. To determine this, 12 mongrel dogs were anesthetized with morphine, nitrous oxide, and oxygen, and then paralyzed with pancuronium and hyperventilated. Arterial pCO2 was controlled by adding CO2 to the inspired gas mixture. Cerebral blood flow was measured at arterial pCO2 levels of 40 and 20 mm Hg both before and after mean arterial pressure was lowered to 40 mm Hg with adenosine enhanced by dipyridamole. In animals where PaCO2 was reduced to 20 mm Hg and mean arterial pressure was reduced to 40 mm Hg, cardiac index decreased 42% from control and total brain blood flow decreased 45% from control while the cerebral metabolic rate of oxygen was unchanged. Hypocapnia with hypotension resulted in small but statistically significant reductions in all regional blood flows, most notably in the brain stem. The reported effects of hypocapnia on CBF during arterial hypotension vary depending on the hypotensive agents used. Profound hypotension induced with adenosine does not eliminate CO2 reactivity, nor does it lower blood flow to ischemic levels in this model, even in the presence of severe hypocapnia.

Short- and long-term prognosis of patients with a subarachnoid haemorrhage in relation to intra-operative period of hypotension

112 patients operated on for an intracranial aneurysm were surveyed. The immediate (up to 3 days from surgery) and late (from 3 days onwards and up to 2 years) outcome was examined in relation to the level and duration of per-operative hypotension as well as technical difficulties. The risk of both immediate and late post-operative neurological deficit was increased when the systolic blood pressure was pharmacologically reduced below the level of 60 mm Hg. The duration of such hypotension had a similar influence. A proportion of patients, following subarachnoid haemorrhage, respond poorly to hypotension and identification of these "vulnerable perfusers" by bedside cerebral blood flow or other monitoring procedures may be useful.

Cerebral and systemic circulatory effects of arterial hypotension induced by adenosine

In six dogs anesthetized with halothane and nitrous oxide, mean arterial pressure (MAP) was lowered to 40 mm Hg for an average of 90 minutes by intravenous infusion of adenosine. The hypotensive effect of the adenosine was potentiated by administering dipyridamole to block its intravascular inactivation. Blood flow to the brain, spinal cord, heart, kidneys, and skeletal muscle was measured six times in each animal using the radioactive microsphere technique. Determinations were made before, during, and 30 minutes after the hypotensive period. During the hypotensive period, MAP was decreased 61% and was related to a proportional decrease in peripheral vascular resistance. Cardiac index decreased 14%. Total cerebral blood flow (CBF) decreased an average of 28% and cerebral vascular resistance decreased 53%. The reduction in CBF was heterogeneous; the cerebral cortex and corpus callosum were most affected and the brain stem least affected. No change occurred in the cerebral metabolic rate of oxygen usage (CMRO2). Left ventricle flow increased 147% and right ventricle flow increased 271%. Blood flow to the kidneys decreased 70%, and to the liver decreased to 6% of control. Jejunum blood flow increased 138% during recovery, while stomach flow varied but showed no statistical change. There was no tachyphylaxis, rebound hypertension, or toxicity associated with the adenosine-induced hypotension. These properties suggest that adenosine may be a useful agent for inducing arterial hypotension in neurosurgical patients.

Pharmacologically induced profound arterial hypotension in the anesthetized dog

✓ The purpose of this study was to compare the behavioral, hematological, and biochemical effects of profound arterial hypotension induced by adenosine potentiated with dipyridamole with those of hypotension induced by trimethaphan camsylate or sodium nitroprusside. Twenty dogs were anesthetized with halothane and nitrous oxide, paralyzed with pancuronium, and ventilated to an arterial pCO2 of 40 torr. Arterial pressure and heart rate were monitored continuously. The animals were divided into four groups of five dogs each. The first group served as controls, while in the remaining groups the mean arterial pressure (MAP) was lowered to 40 mm Hg with adenosine/dipyridamole, trimethaphan, or nitroprusside for 1 hour, following which the animals were allowed to recover from the anesthetic and observed for 48 hours. Determinations of arterial blood gases and hematological and biochemical parameters were made immediately prior to and at the completion of the 1-hour hypotensive period and 48 hours later.

Reduction of MAP to 40 mm Hg was readily achieved with adenosine/dipyridamole. There was no tachyphylaxis to this drug, and arterial pressure promptly returned toward control levels without overshoot after the infusion was discontinued. In contrast, hypotension of this degree could be produced only with toxic doses of nitroprusside. Trimethaphan was more effective in producing hypotension than nitroprusside, but the dose required was extremely variable, and prolonged intervals were required for pressure to return toward normal after the agent was stopped. The control animals and those that received adenosine/dipyridamole recovered promptly from the anesthetic and were neurologically intact. The animals that received trimethaphan recovered more slowly but were neurologically normal within 12 hours. All of the animals that received nitroprusside died without recovering from the anesthesia. Aside from a mild transient metabolic acidosis and transient elevation of blood urea nitrogen and creatinine in the dogs that received adenosine/dipyridamole or trimethaphan, no specific cardiovascular, hematological, hepatic, or renal toxic effects were noted. Evidence of fatal cyanide toxicity was present in the nitroprusside group.

These data suggest that hypotension using adenosine/dipyridamole is readily induced, maintained, and reversed, and is not associated with any apparent hematological, or biochemical evidence of toxicity. Further studies leading to a clinical trial of adenosine-induced hypotension appear to be indicated.

Deliberate hypotension for spinal fusion: prospective randomized study with evoked potential monitoring

Twenty-four patients requiring spinal fusion with Harrington rod instrumentation were studied prospectively to determine the effects of moderate hypotension on blood loss, operating conditions, operating time and spinal cord function. Hypotension reduced blood loss and improved operating conditions but did not shorten operating time. Five patients had alterations in somatosensory cortical evoked potentials after straightening of the spine that prompted us to reverse hypotension (when present) and haemodilution, and then to do wake-up tests. All wake-up tests were normal and all evoked potential alterations resolved during operation. Hypotension seems unlikely to increase the risk of neurological damage if spinal cord function is monitored. Our findings suggest that patients subjected to spinal fusion need not be awakened during operation for testing of cord function provided somatosensory evoked potentials are monitored and remain stable.

Effects of profound hypotension on cerebral blood flow during surgery for intracranial aneurysms

The progression of changes in cerebral blood flow (CBF) and neurological status were measured in 12 patients in whom profound hypotension (mean arterial blood pressure (MABP): 30 to 40 mm Hg) was used during intracranial aneurysm surgery. Nine patients (Group I) showed autoregulation of CBF to an MABP of 40 to 50 mm Hg during surgery. None of these patients had arterial spasm preoperatively. Postoperatively, mild flow disturbances were noted at the site of retraction. Three Group I patients developed arterial spasm postoperatively, but there was no associated neurological deterioration. The remaining three patients (Group II) had impaired autoregulation during surgery, and CBF decreased by 35% to 65% at an MABP of 50 mm Hg. Two of these patients had angiography immediately before surgery, and both showed moderate to severe arterial spasm. Relatively severe flow disturbances were noted postoperatively at the site of retraction, and two patients developed ischemic deficits of late onset. Brain retractor pressure and the degree and duration of hypotension were equivalent in the two patient groups. There was no correlation between intraoperative reductions in CBF (to as low as 20 ml/100 gm/min in the unretracted hemisphere) and immediate postoperative neurological deficits. The use of halothane and mannitol and the relatively short duration of the flow reductions were suggested as factors contributing to the protection from ischemia that was observed. Arterial spasm was found to produce hemodynamic instability and reduced CBF, although neurological status was unaffected in the majority of patients. Patients with impaired autoregulation during surgery were at increased risk of delayed ischemic complications postoperatively, and showed characteristic flow disturbances at all three stages of their clinical course.

Early complications and results of surgery for ruptured intracranial aneurysms

A personal prospective study has been carried out on 91 consecutive patients undergoing direct surgery for aneurysms of the anterior Circle of Willis. The protocol particularly involved maintaining normotension for each patient during surgery, and operating on all patients in Grades I, II, and III and on those patients in Grades IV and V suffering from coincidental hydrocephalus or intra-cranial haematoma. The results are discussed with particular regard to earlier complications following surgery.

Results of microsurgical management of intracranial aneurysms

This paper presents the results of a total of 468 aneurysm operations performed on a consecutive series of 463 patients in an 8-year period, using microsurgical techniques. About two-thirds of the patients were operated on within the first 2 weeks after hemorrhage. The postoperative mortality was 4.5%. A good result was obtained in 74.5% of cases, and a fair result in 11.8%, whereas 9.2% were permanently disabled. Hypothermia was used in 142 operations, and 326 were carried out under induced hypotension. Immediate and late results of surgery are discussed with reference to selection of patients, timing of surgery, and the anesthetic and operative procedures.

Sodium nitroprusside as a hypotensive agent in intracranial aneurysm surgery

Sodium nitroprusside (SNP) was used to induce hypotension during intracranial aneurysm surgery in 67 patients. The effects of SNP infusion (0.1 mg/ml) on blood pressure were rapid and it was easy to adjust blood pressure to desired levels in most patients. When SNP was stopped, the blood pressure returned instantly to the initial level. In eight patients an increase to about 25% or more above prehypotensive level was seen, counteracted in two patients by administration of small doses of halothane. There was a mean increase of 36% in heart rate. Total doses of SNP were 0.05--120 mg (mean: 10.8), corresponding to 0.08--6.8 micrograms/kg/min (mean: 1.9). No metabolic acidosis indicating cyanide intoxication was observed. Tachyphylaxis was seen in three patients, and SNP had to be discontinued in one. It is concluded that SNP gives a rapid and effective hypotension but tachyphylaxis and subsequent danger of cyanide intoxication exist. Therefore, in some cases SNP has to be replaced by or combined with some other hypotensive agent to achieve the desired effect. As there is a risk of impairment of cerebral autoregulation after the use of SNP, it is important to avoid sudden and prolonged blood pressure fluctuations, and to continue with controlled hyperventilation in the postoperative period to reduce the risk of brain oedema and high intracranial pressure.

Clinical evaluation of intravenous nitroglycerin for neurosurgery

Moment-to-moment control of blood pressure is important in the management of the neurosurgical patient. The ideal agent to control blood pressure or induce hypotension should be non-toxic, maintain cerebrovascular autoregulation, and not alter cardiac output or change intracranial pressure. Intravenous nitroglycerin has been used to control blood pressure in 54 neurosurgical cases. This agent produces a rapid, controllable, but not precipitous fall in blood pressure without rebound, is non-toxic, may not alter cerebrovascular autoregulation, and does not raise intracranial pressure. Our clinical experience with intravenous nitroglycerin indicates that it has an important role as a hypotensive agent for the neurosurgical patient.

Cerebral arterial blood flow and aneurysm surgery. Part 2: Induced hypotension and autoregulatory capacity

A study of 21 patients was conducted to clarify the autoregulatory capacity in patients subjected to induced hypotension during intracranial surgery for saccular aneurysms. Trimethaphan camsylate (Arfonad) was used for induced hypotension and arterial blood flow was measured with an electromagnetic flow probe on the internal carotid artery or one of its main intracranial branches. In Grade I and II patients the control arterial blood pressure (ABP) ranged from a mean of 90 to 135 mm Hg (average 110 mm Hg), with a lower level of autoregulation (LLAR) from 35 to 85 mm Hg (average 62 mm Hg). Grade III patients had a control ABP of between 105 and 145 mm Hg (average 124 mm Hg) and the LLAR was found to be between 60 and 95 mm Hg (average 76 mm Hg). There was a significant difference between the two groups with regard to both the control ABP and the LLAR. A surprising result obtained from these data was that the average lower autoregulatory range (the difference between control ABP and LLAR) is practically the same in the two groups. A systematic investigation of the upper limit of autoregulation was not possible for ethical reasons. In those few patients in whom spontaneous increase in the ABP made such observations possible, upper limits up to 150 mm Hg with a total autoregulatory capacity of about 75 mm Hg were observed. In some patients, however, lower limits and cing that the upper limit of autoregulation is markedly influenced by several factors.

Evoked potential changes during brain retraction in dogs

Brain retraction and induced hypotension are surgical adjuncts capable of compromising cerebral blood flow. To evaluate their effects upon brain function, cortical evoked potentials, neurological status and cortical histological changes were determined as a function of graded levels of brain retractor and systemic perfusion pressure in the dog. Somatosensory evoked potentials recorded from the site of application of brain retraction showed a decrement as a function of both the amount of retraction pressure and the systemic perfusion pressure. An electrode distant from the retractor site showed similar, though reduced and more variable changes in amplitude. For higher levels of brain retractor pressure, induced hypotension to 50 mm Hg systemic perfusion pressure produced greater reductions in evoked potentials than in normotensive subjects. It was demonstrated that a reduction of 50% of the evoked potential amplitude after sixty minutes brain retraction signaled, with high probability, the occurrence of postoperative sensory and/or motor deficits and cortical histopathology. It was concluded that cortical evoked potentials represent a reliable indicator of the functional effects produced by applied cortical retraction pressure at several levels of systemic perfusion pressure. It was suggested that the recording of evoked potentials would prove most useful during neurosurgical procedures employing induced hypotension and brain retraction.