Vascular Contemporary Management of Subarachnoid Hemorrhage and Vasospasm: The UIC Experience Luke Corsten, M.D., Ali Raja, M.D., Kern Guppy, M.D., Ph.D., Ben Roitberg, M.D., Mukesh Misra, M.D., M. Serdar Alp, M.D., Fady Charbel, M.D., Gerard Debrun, M.D., and James Ausman, M.D., Ph.D. Department of Neurosurgery, The University of Illinois at Chicago, Chicago, Illinois Corsten L, Raja A, Guppy K, Roitberg B, Misra M, Alp MS, Charbel F, Debrun G, Ausman J. Contemporary management of subarachnoid hemorrhage and vasospasm: the UIC experience. Surg Neurol 2001;56:140 –50. BACKGROUND Cerebral vasospasm is a well-known and serious complication of aneurysmal subarachnoid hemorrhage. The means of monitoring and treatment of vasospasm have been widely studied. Each neurosurgical center develops a protocol based on their experience, availability of equipment and personnel, and cost, so as to keep morbidity and mortality rates as low as possible for their patients with vasospasm. METHODS At the University of Illinois at Chicago, we have developed algorithms for the diagnosis and management of cerebral vasospasm based on the experience of the senior authors over the past 25 years. This paper describes in detail our approach to diagnosis and treatment of aneurysmal subarachnoid hemorrhage and vasospasm. Our discussion is highlighted with data from a retrospective analysis of 324 aneurysm patients. RESULTS Over 3 years, 324 aneurysms were treated; 185 (57%) were clipped, 139 (43%) were coiled. The rate of vasospasm for the 324 patients was 27%. The rate of hydrocephalus was 32% for those patients who underwent clipping, and 29% for those coiled. The immediate outcomes for those who underwent clipping was excellent in 35%, good in 38%, poor in 15.5%, vegetative in 3%, and death in 8% of the patients. For those who underwent coiling the immediate outcome was excellent in 64%, good in 14.5%, vegetative in 2.5%, and death in 14.5% of the patients. These statistics include all Hunt and Hess grades. For those patients who underwent clipping, 51% were intact at 6 months follow-up, 15% had a permanent deficit, 10% had a focal cranial nerve deficit, and 2% had died from complications not directly related to the procedure. For those patients who had undergone coiling, 75% were intact at 6 months follow-up, 12.5% had a permanent Address reprint requests to: Dr. Luke A. Corsten, Department of Neurosurgery (MC 799), University of Illinois at Chicago, Neuropsychiatric Institute, 912 S. Wood Street, Chicago, IL 60612-7329. Received Dec. 21, 2000; accepted May 1, 2001. 0090-3019/01/$–see front matter PII S0090-3019(01)00513-4 deficit, and 12.5% had a cranial nerve deficit, with no deaths. CONCLUSIONS The morbidity and mortality of cerebral vasospasm is significant. A good outcome after aneurysmal subarachnoid hemorrhage is dependent upon careful patient management in the preoperative, perioperative, and postoperative periods. The timely work-up and aggressive treatment of neurological deterioration, whether or not it is because of vasospasm, is paramount. © 2001 by Elsevier Science Inc. KEY WORDS Vasospasm, cerebral angioplasty, triple-H therapy, subarachnoid hemorrhage, hydrocephalus. V asospasm is a recognized but poorly understood phenomenon that complicates the course of many patients who suffer aneurysmal subarachnoid hemorrhage (SAH). It is well recognized that vasospasm can lead to delayed ischemic neurological deficit (DIND) and stroke [1,2,24,28,35,36,42]. Although the exact incidence of vasospasm following SAH is not known, it is thought to correlate with the severity of the bleed [28,43]. Since the early 1980s, much has been written about the treatment of vasospasm [28,32,36 –39]. Hypertensive, hemodilution, and hypervolemic (triple-H) therapy in an intensive care unit setting is now a recognized and accepted means of treating vasospasm [1,2,22–25,29,32,42]. Monitoring transcranial doppler (TCD) velocities is an accepted way of monitoring for vasospasm [4,7,12,13, 17,20,26,31,33,35,41]. Within the past decade remarkable advances have been made in the endovascular treatment of vasospasm such that now direct balloon angioplasty of the stenotic segments of vessels can be performed with acceptable risk and often dramatic improvement in the patient’s neurological status [8 –11,15,40,45,46]. Methods for treatment of vasospasm can be © 2001 by Elsevier Science Inc. 655 Avenue of the Americas, New York, NY 10010 The UIC Experience found in many neuro-critical care textbooks and within the neurosurgical literature [1,2,24,28,42,43]. This paper was written with the goal of explaining the protocol used for treatment of SAH and vasospasm at the University of Illinois at Chicago. We will discuss first the algorithm we use for diagnosis of SAH including grading, timing of computed tomography (CT) scan and angiography, as well as the acute management of hydrocephalus. Second, the timing of surgical or endovascular treatment of aneurysmal SAH, including intensive care unit (ICU) management. Third, the means of monitoring for vasospasm including the timing of postoperative angiography as well as the protocols for working up and treating postoperative neurological deterioration. Fourth, the management of vasospasm once the diagnosis is made, and finally, we will discuss some newer modalities being used in treatment and diagnosis of vasospasm. Surg Neurol 141 2001;56:140 –50 Fisher grades of patients with subarachnoid hemorrhage treated with surgical clipping or endovascular coiling. 1 prevention of re-bleeding [5]. For this reason, the blood pressure parameter is maintained throughout the preoperative course, including angiography, until the aneurysm is secured. Diagnosis and Initial Management Nimodipine The heralding symptoms of SAH include sudden headache, nuchal rigidity, mental status changes, and focal cranial nerve or motor deficits, and are indications for emergency CT scan of the brain. At our institution we perform a plain (nonenhanced) CT scan. If CT scan is negative for hemorrhage, lumbar puncture (LP) is performed and the cerebrospinal fluid (CSF) is analyzed for xanthrochromia. Any patient with the diagnosis of SAH, suspected SAH despite negative CT scan or LP, or suspicion of an expanding (symptomatic) but unruptured aneurysm is admitted to our neurosurgical intensive care unit (NSICU). Patients in the NSICU receive neurological examinations and have their vital signs (including CVP and ICP) measured every hour because of the rapid changes in neurological status that can occur in SAH patients. Patients on any vasoactive drips including nitroprusside have their vital signs measured every 15 minutes. All patients are started on oral nimodipine (60 mg every 4 hours). This calcium channel blocker has been shown to improve longterm neurological outcome in SAH patients who suffer vasospasm [3]. A 21-day course of nimodipine is given. Anti-epileptics are not routinely given but are used as needed if seizures should develop. Initial Studies and Blood Pressure Control Once in the NSICU, central venous and arterial lines are placed for fluid and blood pressure management. Routine blood tests are performed including arterial blood gasses, complete blood count, serum electrolytes, and coagulation studies, as well as a base-line electrocardiogram and chest X-ray. If the patient is known not to have hypertension, a nitroprusside drip is started and the patient’s systolic blood pressure is maintained below 110 mmHg. If the patient has a history of hypertension with an elevated baseline blood pressure the peak systolic blood pressure is maintained 20% below baseline. The rationale of tight blood pressure control is Fisher Grading and HuntHess Grading Based upon the patient’s admission CT scan the patient is assigned a Fisher grade. Based upon their neurological status, they are assigned a Hunt–Hess grade. Figures 1 and 2 display the Fisher grades and Hunt–Hess grades for those patients presenting with SAH who underwent clipping or coiling. These data are derived from a 3-year retrospective analysis of 324 aneurysm patients. Angiography As soon as the patient is stabilized, urgent cerebral angiography is performed. At our institution a four- 142 Surg Neurol 2001;56:140 –50 Corsten et al 1 Statistics for External Ventricular Drainage Total EVDs EVDs for SAH Average duration of EVD Deaths Aneurysm rebleeds Infections Intracerebral hemotomas 103 56 (54%) 10.7 days (range 1–28 days) 0 0 2 (2%) 1 (1%) EVD, external ventricular drain; SAH, subarachnoid hemorrhage. Hunt–Hess grades of patients with subarachnoid hemorrhage treated with surgical clipping or endovascular coiling. 2 vessel cerebral angiogram with digital subtraction including views of both extracranial and intracranial vessels is performed on all patients presenting with SAH. This is performed as soon as possible after the diagnosis of SAH, usually within the first 6 to 12 hours after hemorrhage. Ventricular Drainage A ventricular drainage catheter is often placed in those patients who have evidence of hydrocephalus on CT scan. If a patient who does not have significantly enlarged ventricles needs ICP monitoring we still advocate the use of a ventriculostomy catheter because it affords a means of monitoring intracranial pressure (ICP) as well as lowering it (CSF drainage). This principle also becomes important in the management and optimization of cerebral perfusion pressure (CPP). The judgment for the timing of external ventricular drain (EVD) placement is guided most strongly by the patient’s overall neurological status. Generally, even mild ventricular enlargement is treated with an EVD if the patient’s Hunt–Hess grade is 2 or greater. EVDs are placed at the bedside in the NSICU using local anesthetic and IV sedation with midazolam or propofol. A standard ventriculostomy kit with a hand-held twist drill is used. EVDs are placed almost exclusively in the frontal position using the standard landmarks of a point 2.5 to 3.0 cm off midline and a point 1 to 2 cm anterior to the coronal suture for the location of the entry point. A trajectory is used with the goal being placement of the catheter tip in the third ventricle via the foramen of Monro. Placing the catheter tip in the third ventricle helps avoid the complication of the catheter tip becoming entangled in the choroid plexus. The EVD is leveled 10 cm above the foramen of Monro and allowed to drain in order relieve the symptoms of hydrocephalus, while preventing ventricular collapse. Care must be taken during EVD insertion to avoid excessive drainage of CSF as this can alter the transmural pressure of the aneurysm [5]. We do not routinely change drains without evidence of colonization or infection. In our 3-year retrospective analysis, we found the rate of hydrocephalus to be 32% for those patients who underwent surgical clipping of an aneurysm, and 29% for those who underwent endovascular coiling. We recently performed a second retrospective analysis of 103 EVD insertions in our NSICU over 1 year. EVDs were inserted using the protocol described in the previous paragraph. The data for these patients are displayed in Table 1. Of 103 EVDs, 56 were placed for SAH. The average length of treatment with external ventricular drainage was 10.7 days (range 1–28 days). Timing of Surgery or Coiling Once an aneurysm is identified, the patient usually undergoes craniotomy and clipping or endovascular coiling as soon as possible, usually within 24 hours. If for some reason it is necessary to postpone surgery and it is felt that the aneurysm is at risk for re-rupture because of its size or geometry, the patient is started on an infusion of epsilon amino-caproic acid (IV bolus of 5 grams in 100 ml saline over 1 hour, followed by a continuous infusion of 1 gram per hour). Although reports 20 years ago did not demonstrate a benefit from epsilon amino-caproic acid, more recently this antifibrinolytic has been shown to reduce the incidence of rebleeding in early aneurysm surgery patients [21]. Coiling is performed in the angiography suite under general anesthesia. Because the patient needs to be heparinized during the coiling, we believe it is prudent to wait at least 6 hours after an EVD is placed to avoid the risk of intracerebral or subdural The UIC Experience 2 Surg Neurol 143 2001;56:140 –50 Location of Aneurysms in Patients with Subarachnoid Hemorrhage LOCATION Anterior communicating Paraophthalmic Posterior communicating Internal carotid-other Anterior choroidal Middle cerebral Internal carotid-cavernous Pericallosal Basilar tip Posterior inferior cerebellar Posterior cerebral Vertebrobasilar junction Basilar-trunk Superior cerebellar Anterior inferior cerebellar ANEURYSMS ANEURYSMS CLIPPED COILED (%) (%) 27 10 17 8 5.5 8 2 5.5 8 4.5 0 0 0 4.5 0 30 9 12 0 3 0 0 3 28 0 3 6 3 0 0 bleeding. If there is evidence of vasospasm at the time of coiling, it is usually treated during the same session with balloon angioplasty once the aneurysm is secured. In our retrospective analysis of 324 aneurysm patients, 185 (57%) were treated with clipping and 139 (43%) were treated with coiling. Table 2 displays the percent of aneurysms at a given location in those patients that presented with a ruptured aneurysm. As can be seen from Table 2, certain aneurysm locations are more amenable to endovascular coiling, especially anterior communicating, basilar tip, paraclinoid, and carotid terminus. Likewise, more distal aneurysms, such as middle cerebral or posterior inferior cerebellar (PICA) aneurysms necessitate a surgical approach. Figure 3 displays the immediate outcomes for the set of patients discussed above. Immediate outcome refers to the patient’s status upon discharge from the hospital after treatment. This figure includes those patients who suffered vasospasm. An excellent outcome means there was no focal cognitive, motor, or sensory deficit. A good outcome means the patient had a mild or moderate cognitive, motor, or sensory deficit. This deficit would limit the patient’s ability to make a full recovery; some rehabilitation may have been needed. A poor outcome means the patient suffered a severe deficit and may have needed a feeding gastrostomy tube and/or a tracheostomy; inpatient rehabilitation was usually indicated. A vegetative outcome refers to persistent vegetative state. We believe that our mortality rate is reflective of those patients who underwent treatment despite a severe or morbid presentation (Hunt–Hess grade 4 or 5). Immediate outcomes of patients with subarachnoid hemorrhage after undergoing surgical clipping or endovascular coiling. 3 Figure 4 displays the outcomes at 6 months’ follow-up for the same set of patients. These percentages account for those patients who were followed after discharge from the hospital after the primary treatment; therefore, they do not include the perioperative deaths displayed in Figure 3. The patient was considered intact if he or she was able to return to normal preoperative activities. A permanent deficit refers to some form of cognitive, motor, or sensory deficit that limits a patient’s ability to function at a normal level. A cranial nerve deficit is a focal deficit of cranial nerve function that might limit the patient’s ability to drive or return to work. There were no deaths at 6 months’ follow-up for this group. Postoperative Management A postoperative CT scan is performed on all surgical patients usually within 1 to 4 hours of clipping Outcomes at 6 months’ follow-up for patients with subarachnoid hemorrhage undergoing surgical clipping or endovascular coiling. 4 144 Surg Neurol 2001;56:140 –50 or coiling. This scan is performed to rule out any peri-procedural hemorrhage, contusion, or hydrocephalus, and at the same time to establish a baseline examination for future reference. Once an aneurysm is secured, the patient’s blood pressure parameters are liberalized. If there is any immediate evidence of unexpected postoperative deficit that cannot be explained based upon CT findings, or if there are any symptoms to suggest vessel occlusion or vasospasm, immediate postoperative angiography is performed to confirm integrity of vascular anatomy. We rarely find it necessary to perform immediate postoperative angiography. The patient’s neurological status, vital signs, and ICP are assessed every hour and variation is noted. Postoperative serum electrolytes, complete blood count, and coagulation studies are checked every 8 or 12 hours to monitor for hyponatremia, anemia, or coagulopathy. Maintenance IV fluids are given. Blood transfusion is given if necessary to maintain the serum hemoglobin at least 11 g/dl to optimize blood viscosity and oxygen delivery [19]. A patient with cerebral edema or problems with increased ICP can be managed with intermittent boluses of mannitol. TCDs are measured daily in all patients with SAH and recorded on a bedside flowsheet. We have found that the day-to-day trend in TCD velocities is more important than the absolute peak or mean velocities when monitoring for vasospasm. This is because a patient developing vasospasm may show a sudden increase in TCD velocity when comparing one day to the next, even though the absolute peak velocity alone might not appear to be critically elevated. Evaluation of Neurological Deterioration It is important to detect vasospasm before the patient suffers delayed ischemic neurological deficit or stroke. Any patient who is at risk for postoperative vasospasm and has symptoms of neurological deterioration or mental status changes undergoes emergency cerebral angiogram to rule out vasospasm. Attention is paid to the daily TCD velocities as well as to the other possibilities within the differential diagnosis such as hyponatremia, seizure, cerebral edema, and hydrocephalus, but until proven otherwise, or unless there is convincing evidence to the contrary, the patient is assumed to have vasospasm. Although the peak incidence of vasospasm is Day 5 to Day 7 after SAH, we have Corsten et al 3 Differential Diagnosis of Neurological Deterioration Focal causes ● Vasospasm (especially within carotid circulation) ● Hemorrhage ● Infarct ● Contusion (surgical) ● Focal Seizures Generalized causes ● Vasospasm (especially within vertebrobasilar circulation) ● Infarct ● Elevated intracranial pressure (edema, hydrocephalus) ● Seizures ● Metabolic (hyponatremia, alcohol withdrawal, thyroid dysfunction, hepatic dysfunction) ● Drug toxicity (antiepileptics, corticosteroids) ● Hypoxemia ● Shock (hypovolemic, cardiogenic, neurogenic, septic) ● Fever (infection, drug-related or allergic, central or hypothalamic) ● Degenerative (pre-morbid dementia such as Alzheimer’s or Korsakoff’s) found vasospasm in patients as early as 3 days to as late as 21 days post-hemorrhage. We suspect that vasospasm may occur even earlier. When in doubt, rule out vasospasm. If there is no evidence of vasospasm on angiography, the work-up of the differential diagnosis is pursued until the problem is found. The general principles of neurological localization are applied. A patient is diagnosed as having either a focal or a diffuse neurological change. A diffuse neurological change might suggest seizure, hyponatremia, or some other metabolic abnormality, infection, hypoxemia, elevation of ICP, or drug toxicity. A focal neurological change is more suggestive of an ischemic event, hemorrhage, or possibly a focal seizure. Table 3 displays a list of the differential diagnoses for a postoperative patient with neurological deterioration. An electrical encephalogram (EEG) should be performed as soon as possible to rule out seizures. If the patient has a ventricular drain, laboratory tests of CSF, including cultures and Gram stain, should be performed to rule out meningitis. If the patient is febrile or appears septic, cultures of blood, CSF, urine, and sputum should be performed. Chest X-ray should be performed, and any central venous lines should be changed. This work-up should be successful in identifying those diagnoses that might pose a direct or acute threat to the patient. Vasospasm, hydrocephalus, intracranial bleed, seizures, sepsis, and acute hypoxemia The UIC Experience are all problems that need immediate identification and treatment to minimize the patient’s risk of neurological injury. If angiography shows no evidence of vasospasm, CT scan is normal or unchanged, serum sodium is normal, EEG shows no evidence of seizure, ICP is within normal range, there is no evidence of drug toxicity, and there is no evidence of meningitis or other infection, further work-up is warranted. Magnetic resonance imaging (MRI) may be indicated to evaluate for brainstem pathology, and 24 hour EEG monitoring may be indicated, as well as further tests such as brainstem auditory evoked potentials (BAERs) or somatosensory evoked potentials (SSEPs). Flow studies such as xenon-CT or SPECT may also be helpful, as well as thorough metabolic and endocrine testing to rule out problems such as hypothyroidism or adrenal insufficiency. Treatment of Vasospasm Angiographic evidence of vasospasm is treated immediately with balloon angioplasty of all accessible arterial segments that demonstrate evidence of spasm. These include the proximal and supraclinoid internal carotid arteries, vertebral and basilar arteries, and proximal segments of the middle cerebral arteries. We have found it technically difficult to navigate an angioplasty catheter into the anterior cerebral arteries and so it is frequently not possible to treat the A1 segment of the anterior cerebral artery. Distal middle cerebral arteries (M2 or smaller) and posterior cerebral arteries also can rarely be accessed for balloon angioplasty. Angioplasty is performed under general anesthesia. It is the opinion of the neuroradiologists and neurosurgeons at our institution that intra-arterial infusion of papaverine has no long-term benefit in the treatment of vasospasm, and this therapy is no longer used. Our 3 year analysis of 324 aneurysm patients treated at the UIC Medical Center revealed an incidence of vasospasm of 27%. We consider balloon angioplasty of the narrowed vessel segments to be the primary treatment for vasospasm because it directly addresses the anatomic basis of the disease. Often diffuse vasospasm is identified where the larger vessels can be opened with angioplasty, but the more distal vessels are unable to be accessed. In this case, aggressive medical management becomes the only alternative. Subsequent increase in TCD velocities or further neurological deterioration are indications for repeated angiography with angioplasty. Although it is unusual for a segment of vessel to re-stenose after angioplasty, it has been Surg Neurol 145 2001;56:140 –50 observed, and these patients undergo additional sessions of angioplasty. The mainstay of medical treatment of vasospasm is triple-H therapy [1,2,22–25,29,32,42]. This is employed aggressively as soon as vasospasm is diagnosed. All patients with vasospasm undergo immediate pulmonary artery (Swan–Ganz) catheter placement. The rationale for this is that in attempting to modulate cerebral perfusion with aggressive volume expansion and vasopressors, significant stress may be placed on the heart. The patient may be at risk for myocardial infarction or pulmonary edema. The pulmonary capillary wedge pressure (PCWP) and other cardiac parameters need to be monitored carefully when triple-H therapy is applied. Triple-H therapy is usually applied in a standard fashion. Swan–Ganz monitoring is always used. The patient is first given aggressive volume expansion. Attention is paid to the patient’s baseline PCWP and cardiac output. Colloids such as hetastarch or albumin may be used. As the process of volume expansion is carried out, careful attention is paid to the patient’s neurological status to see if there is improvement. The hemoglobin is maintained at 11 g/dl either by blood transfusion or by hemodilution as necessary [19]. Volume expansion is continued until the patient’s PCWP is maximized without compromise of cardiopulmonary function or there is neurological improvement. This is usually not higher than a PCWP of 16 mmHg to 18 mmHg. We feel it a good practice to calculate the patient’s serum colloid oncotic pressure (COP) as this can help determine how much hydration the patient will tolerate before developing complications such as pulmonary edema. COP (in mmHg) can be estimated easily by measuring the total serum protein (TP) in g/dl and using the equation: COP ⫽ 2.1 (TP) ⫹ 0.16 (TP2) ⫹ 0.009 (TP3)[27]. As a general rule, the PCWP should not be allowed to exceed the calculated colloid oncotic pressure. If there is improvement in the patient’s condition, the volume expansion is continued. If, after pushing a patient’s PCWP to 18 mmHg or the calculated COP, there is either no improvement or only slight improvement, hypertensive therapy with a vasopressor drip such as dopamine or norepinephrine is begun. The rationale for performing volume expansion before starting pressor drips is that vasopressors may not be successful in improving cerebral blood flow if the patient is dehydrated. Also, systemic vasoconstriction in a patient who is not well hydrated puts end organs such as the kidneys and intestines at risk for ischemia. If a patient is found to be euvolemic after Swan–Ganz placement, volume expansion and vaso- 146 Surg Neurol 2001;56:140 –50 pressor therapies are usually instituted simultaneously. Hypertensive therapy is titrated according to a patient’s baseline systolic and mean arterial blood pressure as well as the baseline cardiac output. The blood pressure is raised to 20% to 30% above baseline by titrating the vasopressor drips while the patient is observed for signs of neurological improvement. If improvement is noted, the vasopressor therapy is continued; if there is no neurological improvement after blood pressure elevation and optimization of cardiac output, the vasopressor infusion is stopped after a trial of 3 to 4 hours. During hypertensive treatment for vasospasm, careful attention is paid to the patient’s cerebral perfusion pressure (CPP) if a ventriculostomy or ICP monitor is in place. Every effort is made to maintain an optimum CPP (greater than 70 mmHg), especially in those patients who have problems with elevated ICP. The endpoint of triple-H therapy is clinical improvement. Every effort is made to achieve this endpoint; triple-H therapy is attenuated or discontinued if there is evidence of myocardial, pulmonary, hepatic, gastro-intestinal, or renal failure as a result of the therapy. New Modalities for Monitoring Vasospasm Patients Because of the poor understanding of the pathophysiologic mechanism of vasospasm, further investigation is needed. New ways of treating vasospasm and new methods for diagnosing vasospasm are yet to be discovered. The use of the INVOS transcutaneous cerebral oximeter (Somanetics Corporation, Troy, MI), a cutaneous sensor capable of measuring cortical oxygen saturation, is a new means of monitoring for vasospasm [14,30]. We apply the cutaneous sensor pads on both sides of the forehead. Although the numerical value displayed by the monitor is not always an accurate indicator of cerebral oxygenation, we have found that the minute-to-minute saturation trend can be helpful in identifying those patients developing cortical ischemia from vasospasm [14,30]. A number of trials have been conducted at our institution utilizing a Neurotrend probe (Codman/ Johnson&Johnson, New Brunswick, NJ) which is placed directly into the brain parenchyma by way of a burr hole and bolt at the bedside, or inserted directly at surgery. The probe is able to measure cerebral pH, temperature, oxygen, and CO2 ten- Corsten et al sions in a continuous fashion. We have found in trials both during surgery and during postoperative monitoring that the intracerebral oxygen probe reliably shows a significant decrease in pH and oxygen content, with accompanying increase in CO2 tension, in the areas of ischemic cortex [6,16,18]. The limitations of the cerebral oxygen probe are that while the measurements are accurate, they are only significant for the tissue immediately surrounding the probe. This means that if the probe is in the anterior region of the frontal lobe, ischemia within the posterior-frontal, temporal, or parietal lobes would not necessarily be detected. Whether this modality is a clinically useful way to monitor for cerebral ischemia or vasospasm remains to be seen. Recent developments in phase contrast magnetic resonance angiography (PCMRA) technology have made it possible to obtain quantitative flow values for all major vessels in the cerebral circulation. Besides vessel diameter, flow rate and wall shear stress values can be measured. The error in this noninvasive technique has been calculated to be less than 7% [44]. This technology is currently being applied to a wide population of patients with various neurovascular disorders at our institution. We are not currently using PCMRA quantitative flow measurements as a standard means of screening for vasospasm; however, this technology is an innovative way of studying the blood flow and wall shear stress in patients both pre- and post-angioplasty. Conclusions Obtaining good outcomes in patients presenting with subarachnoid hemorrhage is dependent upon careful and aggressive preoperative, perioperative, and postoperative care. Often a patient who presents with a low-grade bleed and who undergoes successful aneurysm clipping can suffer severe postoperative complications leading to a dismal outcome. Patients with subarachnoid hemorrhage and vasospasm can be the sickest patients a neurosurgeon might encounter. Careful attention to detail in every aspect of the patient’s care is paramount. Timely and aggressive work-up and treatment of postoperative neurological deterioration, whether or not it is because of vasospasm, can often mean the difference between a good outcome and a poor outcome after aneurysmal SAH. We hope that the algorithms and strategies discussed above will give some insight to the reader. The UIC Experience Surg Neurol 147 2001;56:140 –50 REFERENCES 1. Awad IA, Carter LP, Spetzler RF, Medina M, Williams Jr FC. Clinical vasospasm after subarachnoid hemorrhage: response to hypervolemic hemodilution and arterial hypertension. Stroke 1987;18:365–72. 2. Barker II FG, Heros RC. Clinical aspects of vasospasm. Neurosurg Clin N Am 1990;1:277– 88. 3. Barker II FG, Ogilvy, CS. Efficacy of prophylactic nimodipine for delayed ischemic deficit after subarachnoid hemorrhage: a metaanalysis. J Neurosurg 1996; 84:405–14. 4. Bartels RH, Verhagen WI, Van der Spek JA, Grotenhuis JA, Brandsma E, Notermans SL. Transcranial Doppler ultrasonography: influence on scheduling of angiography and delayed surgery for ruptured intracranial aneurysms. J Neurosurg Sci 1994;38:21–7. 5. Canham, PB, Ferguson, GG. A mathematical model for the mechanics of saccular aneurysms. Neurosurgery 1985;17:291–5. 6. Charbel FT, Du X, Hoffman WE, Ausman JI. Brain tissue PO2, PCO2 and pH during cerebral vasospasm. Surg Neurol 2000;54:432– 8. 7. Clyde BL, Resnick DK, Yonas H, Smith HA, Kaufmann AM. The relationship of blood velocity as measured by transcranial doppler ultrasonography to cerebral blood flow as determined by stable xenon computed tomographic studies after aneurysmal subarachnoid hemorrhage. Neurosurgery 1996;38:896 –905. 8. Elliott JP, Newell DW, Lam DJ, Eskridge JM, Douville CM, Le Roux PD, Lewis DH, Mayberg MR, Grady MS, Winn HR. Comparison of balloon angioplasty and papaverine infusion for the treatment of vasospasm following aneurysmal subarachnoid hemorrhage. J Neurosurg 1998;88:277– 84. 9. Eskridge JM, Newell DW, Pendleton GA. Transluminal angioplasty for treatment of vasospasm. Neurosurg Clin N Am 1990;1:387–99. 10. Eskridge JM, McAuliffe W, Song JK, Deliganis AV, Newell DW, Lewis DH, Mayberg MR, Winn HR. Balloon angioplasty for the treatment of vasospasm: results of first 50 cases. Neurosurgery 1998;42:510 –17. 11. Firlik AD, Kaufmann AM, Jungreis CA, Yonas H. Effect of transluminal angioplasty on cerebral blood flow in the management of symptomatic vasospasm following aneurysmal subarachnoid hemorrhage. J Neurosurg 1997;86:830 –9. 12. Grosset DG, Straiton J, du Trevou M, Bullock R. Prediction of symptomatic vasospasm after subarachnoid hemorrhage by rapidly increasing transcranial Doppler velocity and cerebral blood flow changes. Stroke 1992;23:674 –9. 13. Harders AG, Gilsbach JM. Time course of blood velocity changes related to vasospasm in the circle of Willis measured by transcranial Doppler ultrasound. J Neurosurg 1987;66:718 –28. 14. Hernandez–Avila G, Dujovny M, Slavin KV, Luer MS, Nijensohn E, Geremia G, Ausman JI. Use of transcranial cerebral oximetry to monitor regional cerebral oxygen saturation during neuroendovascular procedures. AJNR Am J Neuroradiol 1995;16:1618 –25. 15. Higashida RT, Halbach VV, Cahan LD, Brant– Zawadzki M, Barnwell S, Dowd C, Hieshima GB. Transluminal angioplasty for treatment of intracranial arterial vasospasm. J Neurosurg 1989;71:648 –53. 16. Hoffman WE, Wheeler P, Edelman G, Charbel FT, 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. Torres NJ, Ausman JI. Hypoxic brain tissue following subarachnoid hemorrhage. Anesthesiology 2000;92: 442– 6. Hurst RW, Schnee C, Raps EC, Farber R, Flamm ES. Role of transcranial Doppler in neuroradiological treatment of intracranial vasospasm. Stroke 1993;24: 299 –303. Johnson WD, Bolognese P, Miller JI, Heger IM, Liker MA, Milhorat TH. Continuous postoperative ICBF monitoring in aneurysmal SAH patients using a combined ICP-laser Doppler fiberoptic probe. J Neurosurg Anesthesiol 1996;8:199 –207. Kusunoki M, Kimura K, Nakamura M, Isaka Y, Yoneda S, Abe H. Effects of hematocrit variations on cerebral blood flow and oxygen transport in ischemic cerebrovascular disease. J Cereb Blood Flow Metab 1981;1: 413–7. Laumer R, Steinmeier R, Gonner F, Vogtmann T, Priem R, Fahlbusch R. Cerebral hemodynamics in subarachnoid hemorrhage evaluated by transcranial Doppler sonography. Part 1. Reliability of flow velocities in clinical management. Neurosurgery 1993;33:1–9. Leipzig TJ, Redelman K, Horner TG, Indianapolis Neurosurgical Group. Reducing the risk of rebleeding before early aneurysm surgery: a possible role for antifibrinolytic therapy. J Neurosurg 1997;86:220 –5. Lennihan L, Mayer SA, Fink ME, Beckford A, Paik MC, Zhang H, Wu YC, Klebanoff LM, Raps EC, Solomon RA. Effect of hypervolemic therapy on cerebral blood flow after subarachnoid hemorrhage: a randomized controlled trial. Stroke 2000;31:383–91. Levy ML, Giannotta SL. Cardiac performance indices during hypervolemic therapy for cerebral vasospasm. J Neurosurg 1991;75:27–31. Levy ML, Giannotta SL. Induced hypertension and hypervolemia for treatment of cerebral vasospasm. Neurosurg Clin N Am 1990;1:357– 65. Levy ML, Rabb CH, Zelman V, Giannotta SL. Cardiac performance enhancement from dobutamine in patients refractory to hypervolemic therapy for cerebral vasospasm. J Neurosurg 1993;79:494 –9. Lindegaard KF. The role of transcranial Doppler in the management of patients with subarachnoid haemorrhage—a review. Acta Neurochir Suppl (Wien) 1999; 72:59 –71. Marino PL. Acute respiratory distress syndrome. In: Zinner SR ed. The ICU Book. 2nd Edition. Baltimore. Williams and Wilkins 1998:379. Martin N, Khanna R, Rodts G. The intensive care management of patients with subarachnoid hemorrhage. In: Andrews BT, eds. Neurosurgical intensive care. San Francisco: McGraw Hill, 1993:291–310. Medlock MD, Dulebohn SC, Elwood PW. Prophylactic hypervolemia without calcium channel blockers in early aneurysm surgery. Neurosurgery 1992;30:12– 6. Misra M, Dujovny M, Aletich V, Alp MS, Debrun GM, Ausman JI. Transcranial cerebral oximetry in endovascular treatment of carotid-cavernous fistula. Neuroradiology 1996;38:590 – 4. Newell DW, Winn HR. Transcranial Doppler in cerebral vasospasm. Neurosurg Clin N Am 1990;1:319 –28. Origitano TC, Wascher TM, Reichman OH, Anderson DE. Sustained increased cerebral blood flow with prophylactic hypertensive hypervolemic hemodilution 148 Surg Neurol 2001;56:140 –50 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. (“triple-H” therapy) after subarachnoid hemorrhage. Neurosurgery 1990;27:729 –39;739 – 40. Proust F, Callonec F, Clavier E, Lestrat JP, Hannequin D, Thiebot J, Freger P. Usefulness of transcranial color-coded sonography in the diagnosis of cerebral vasospasm. Stroke 1999;30:1091– 8. Samson D, Batjer HH, Bowman G, Mootz L, Krippner WJ Jr, Meyer YJ, Allen BC. A clinical study of the parameters and effects of temporary arterial occlusion in the management of intracranial aneurysms. Neurosurgery 1994;34:22–9. Sekhar LN, Wechsler LR, Yonas H, Luyckx K, Obrist W. Value of transcranial Doppler examination in the diagnosis of cerebral vasospasm after subarachnoid hemorrhage. Neurosurgery 1988;22:813–21. Shimoda M, Oda S, Tsugane R, Sato O. Intracranial complications of hypervolemic therapy in patients with a delayed ischemic deficit attributed to vasospasm. J Neurosurg 1993;78:423–9. Solomon RA, Fink ME, Lennihan L. Early aneurysm surgery and prophylactic hypervolemic hypertensive therapy for the treatment of aneurysmal subarachnoid hemorrhage. Neurosurgery 1988;23:699 –704. Solomon RA, Onesti ST, Klebanoff L. Relationship between the timing of aneurysm surgery and the development of delayed cerebral ischemia. J Neurosurg 1991;75:56 – 61. Sundt Jr TM, Kobayashi S, Fode NC, Whisnant JP. Results and complications of surgical management of 809 intracranial aneurysms in 722 cases. Related and unrelated to grade of patient, type of aneurysm, and timing of surgery. J Neurosurg 1982;56:753– 65. Touho H. Percutaneous transluminal angioplasty in the treatment of atherosclerotic disease of the anterior cerebral circulation and hemodynamic evaluation. J Neurosurg 1995;82:953– 60. Vora YY, Suarez–Almazor M, Steinke DE, Martin ML, Findlay JM. Role of transcranial Doppler monitoring in the diagnosis of cerebral vasospasm after subarachnoid hemorrhage. Neurosurgery 1999;44:1237– 48. Weir B. The history of cerebral vasospasm. Neurosurg Clin N Am 1990;1:265–76. Zabramski JM, Hamilton MG. Cerebral vasospasm. In: Carter LP, Spetzler RF, eds. Neurovascular Surgery. New York: McGraw-Hill Inc. 1993;583– 601. Zhao M, Charbel FT, Alperin N. Improved phasecontrast flow quantification by 3-dimensional vessel localization. Mag Reson Imag 2000;18:697–706. Zubkov AY, Lewis AI, Scalzo D. Transluminal angioplasty and intra-arterial papaverine for the treatment of cerebral vasospasm after ruptured arteriovenous malformations. Surg Neurol 1999;51:75– 80. Zubkov AY, Lewis AI, Scalzo D, Bernanke DH, Harkey HL. Morphological changes after percutaneous transluminal angioplasty. Surg Neurol 1999;51:399 – 403. COMMENTARY The authors nicely present their own management protocol for subarachnoid hemorrhage and vasospasm. This article is valuable to neurosurgeons because it demonstrates a practical management strategy for SAH; however, there is no worldwide Corsten et al consensus for treating vasospasm following SAH. We look forward to future contributions to this problem from the authors’ institution and others. Shigeaki Kobayashi, M.D. Tetsuyoshi Horiuchi, M.D. Department of Neurosurgery Shinshu University School of Medicine Matsumoto, Japan Dr Corsten and his colleagues have put together a very useful summary of the current strategies for the management of aneurysmal subarachnoid hemorrhage and cerebral vasospasm. For the most part, we manage patients at Columbia in a very similar fashion. Therefore, I will not dwell on the similarities of treatment protocols but rather on areas where there is still some discussion. The tail end of the risk period for delayed cerebral ischemia remains somewhat controversial. In my experience of over 8,000 patients with aneurysmal SAH, I have never seen a case of new onset delayed cerebral ischemia first occurring more than 14 days after SAH in a patient who was operated on within the first 48 hours after the initial rupture. Similarly, no patient had new onset delayed cerebral ischemia occurring more than 14 days postcraniotomy. Therefore, I wonder if some of the reported cases of late onset delayed cerebral ischemia occurring in the 14 to 21 day period might be related to bleeding at the time of surgery and not necessarily to the after-effects of the initial hemorrhage. In otherwise straightforward cases of SAH with early aneurysm surgery, we do not continue vasospasm prophylaxis past the 14th day after SAH. It is also questionable whether nimodipine needs to be continued past the 14th day. It is evident that patients who develop delayed cerebral ischemia less than 14 days after SAH often have symptoms that persist past the 14th day, but I have yet to see a case with an initial presentation of delayed cerebral ischemia past the 14th day post-SAH in otherwise straightforward cases. Ventricular drainage is also an area of controversy. The downside of ventricular drainage includes the precipitation of rebleeding of the aneurysms and the possible introduction of bacterial colonization and ventriculitis. The authors report that in their series there was only one case of colonization of an EVD catheter and one case of ventriculitis. Our experience has shown that there is a significantly higher risk of infection when ventriculostomy catheters are placed in patients with SAH, particularly with intraventricular hemorrhage. Therefore, if the patient is stable, even with mild