Cerebral vasospasm evaluated by transcranial Doppler ultrasonography at different intracranial pressures

J Neurosurg 75:752-758, 1991 Cerebral vasospasm evaluated by transcranial Doppler ultrasonography at different intracranial pressures J[)RGEN KLINGELH()FER, M.D., DIRK SANDER, M.D., MANFRED HOLZGRAEFE, M.D., CHRISTIAN BISCHOFF, M.D., AND BASTIAN CONRAD, M.D. Center of Neurological Medicine, Unit;ersity of GOttingen, GOttingen. and Technical University of Munich, Munich, Germany v, The present study evaluates the interdependence of clinical stage, cerebral vasospasm, intracranial pressure (ICP), and transcranial Doppler ultrasonographic parameters. The mean flow velocity of blood in the middle cerebral artery and the index of cerebral circulatory resistance as a measure of the peripheral vascular flow resistance were determined in 76 patients with spontaneous subarachnoid hemorrhage. The ICP was measured using an epidural transducer in 41 patients. There was no case in which both high ICP and a high mean flow velocity were observed simultaneously. The investigations led to the following conclusions. 1) In patients with a resistance index of less than 0.5, changes in the mean flow velocity seem to reflect sufficiently the actual severity and time course of vasospasm. 2) During the time course of vasospasm, an increase in the resistance index above values of 0.6 with a simultaneously decreased mean flow velocity indicates a rise in ICP rather than a reduction in vasospasm. 3) With a pronounced increase in ICP, evaluation of the severity and time course of vasospasm by transcranial Doppler ultrasonography based solely upon the mean flow velocity can lead to false-negative results. KEY WORDS ~ subarachnoid hemorrhage 9 vasospasm 9 intracranial pressure 9 Doppler velocimetry C EREBRAL vasospasm is a complication with major implications for the further therapeutic management of patients with subarachnoid hemorrhage (SAH). Until now, the only way to establish the time course of vasospasm has been by angiography. This is an invasive procedure, however, and cannot be repeated at frequent intervals to monitor the development and resolution of arterial narrowing following SAH.~8.~9 Transcranial Doppler ultrasonography (TCD) has opened up the possibility of noninvasive diagnosis of cerebrovascular spasm and observation of its time course, t, 11,23,24,27 The mean flow velocity has been considered the most important diagnostic parameter in the evaluation of vasospasm by TCD because a clear correlation between the velocity of flow and the diameter of the middle cerebral artery (MCA) measured by angiography has been found in different studies. L~J All patients in whom angiography showed evidence of MCA vasospasm had TCD blood flow velocities of 120 cm/sec or more.~'23 The intracranial flow patterns can also be influenced by other factors, in particular by the intracranial pressure (ICP)/s'~s'16 partial pcoz, :t or the hematocrit. 4 The combined occurrence of vasospasm and increased ICP is of particular importance in the development of neurological deficits, 9'25 thus making the evaluation of both factors essential for prognosis and therapy. 16 The TCD blood flow velocities associated with neurological deficits clearly vary from case to case. t~ Compton, et al., 6 observed that deterioration of clinical condition from Grade I to Grade III (according to the Hunt and Hess scale,4) was accompanied by an increase in the mean flow velocity, whereas patients in clinical Grade IV demonstrated a decrease in mean flow velocity as compared to Grade III patients. Lindegaard, et al, t7 pointed out that classification of MCA spasm by use of flow velocity could be misjudged if blood flow reduction and artery narrowing concur. It was the aim of the present study to evaluate the interdependence of the patient's clinical grade, vasospasm, ICP, and TCD parameters. Clinical Material and Methods Seventy-six patients suffering from spontaneous SAH were included in this study. A clinical summary for this series is shown in Table 1. Angiography was performed 752 J. Neurosurg. / Volume 75/November, 1991

Doppler evaluation of vasospasm at different ICP's TABLE 1 Etiology of subarachnoid hemorrhage in 76 patients studied by transcranial Doppler ultrasonography* Location of Aneurysm No. of Cases internal carotid artery 11 anterior cerebral artery/anterior communicating artery 28 middle cerebral artery 17 posterior cerebral artery/posterior communicating artery 5 no verified aneurysm 15 * There were 40 females and 36 males. Mean patient age was 49 years (range 21 to 73 years). in all patients. Blood flow velocity in the MCA was measured using a 2-MHz pulsed TCD device.* The instrumentation and techniques used for the TCD examination have been described in detail elsewhere. 2 The time-averaged peak frequencies (mean), converted by the built-in computer into blood flow velocities (cm/sec), were noted for the mean flow velocity. The peak systolic and end-diastolic flow velocities were extracted after manual labeling of the TCD recordings. Flow patterns of the MCA were recorded either intermittently at least once a day or (if the patient's clinical condition deteriorated) continuously by means of a specially designed attached probe holder. Surgery for a verified operable aneurysm was carried out in 26 (34%) of the 76 patients within the first 7 days after SAH. After confirmation of the SAH by bloody or xanthochromic cerebrospinal fluid and/or computerized tomography (CT) scans, the patients received 2 mg/hr nimodipine intravenously for 7 to more than 14 days, depending on the intracranial flow patterns monitored by TCD. The individual timing for changing nimodipine to an oral dose of 60 mg/24 hr was evaluated, again based on the TCD findings. In addition, the usual SAH therapy 3 was performed. The clinical grade according to the classification of Hunt and Hess J4 was evaluated daily by a neurologist experienced in the assessment of cerebrovascular disorders. The Hunt and Hess stages at admission are shown for all patients in Table 2. The degree of SAH on admission CT scans and the resulting distribution of the patients as being at low or high risk for symptomatic vasospasm (Table 2) was determined using the classification of Fisher, et al., s as follows: CT Grade I (thin localized layer); CT Grade II (thick layer in two of the three subarachnoid compartments or one subarachnoid compartment and at the cortical surface); and CT Grade III (severe diffuse SAH with thick layers in all three subarachnoid compartments or in two compartments and at the cortical surface). TABLE 2 Classification on admission of 76 patients with SAH* Patient No. of Classification Cases clinical grade I 16 II 22 III 20 IV 18 SAH grade on CT I 14 II 25 III 37 * SAH = subarachnoid hemorrhage. Clinical grade according to Hunt and Hess;~4 SAH grade on computerized tomography (CT) scans according to Fisher, et al. 8 (see text). Arterial pco2 was monitored using blood gas analysis. Mean arterial blood pressure (MABP) was calculated as: MABP = (systolic pressure - diastolic pressure)/ 3 + diastolic pressure? According to Pourcelot, ~~ the index of cerebral circulatory resistance (R) was calculated from the TCD data as: R = (maximum systolic flow velocity - end-diastolic flow velocity)/maximum systolic flow velocity. The resistance index R is a measure of the peripheral flow resistance; in general, low vascular resistance is characterized by high diastolic flow velocities, whereas low diastolic flow indicates high resistance.~2 In 41 patients, the ICP was also measured, using an epidural transducer.t 7 In all cases with an asymmetrical distribution of the hemorrhage (as determined from the CT scans), the epidural pressure transducer was placed on the side with more blood and the parameters of the ipsilateral MCA were evaluated. If both hemispheres were affected to a similar degree, the parameters of the MCA were evaluated on the side of the implanted epidural device. To achieve a homogeneous study population, we selected 36 of the 41 patients with ICP measurements who fulfilled the following criteria. The patients' clinical status had to allow for a minimum monitoring period of 8 days following SAH, including ICP measurements. Only data taken at time intervals of at least 12 hours were considered. The maximum flow velocities had to be on the side of the implanted epidural device. The hematocrit had to be in the 30% to 40% range; for the TCD recordings to be considered in this analysis, the heart rate had to be rhythmic and range from 60 to 90 beats/see. To minimize the influence of varying PaCO2 on the flow patterns, TCD data were accepted only if the PaCO2 was in the range of 30 to 40 mm Hg at the time of recording. The data from these 36 patients were divided into two groups: Group A consisting of 14 patients with SAH and normal or only slightly increased * Transcranial Doppler device, Model EME TC 2-64, manufactured by Eden Medizinische Elektronik Gmblt, Ueberlingen, Germany. "~ Epidural transducer manufactured by Gaeltec, Ltd., Dunvegan, Isle of Skye, Scotland. J. Neurosurg. / Volume 75 /November, 1991 753

J. Klingelh6fer, el al. ICP (< 20 mm Hg); Group B including 22 patients with SAH and dearly increased ICP (> 20 mm Hg). Results Clinical Status and TCD Parameters The average values of the maximum mean flow velocities (maximum flow velocities for both sides) as well as the corresponding resistance indices of the 76 patients were compared with the clinical grade according to the classification of Hunt and Hess (Fig. 1). Analysis of these values (expressed as mean _+ standard error of the mean) demonstrated an increase in the mean flow velocity from Grade I (122 _+ 41.8 cm/sec) to Grade III (175 _+ 55.5 cm/sec) and a decrease for Grade IV (93 -+ 30.1 cm/sec). Paralleling these values, the mean resistance index decreased to 0.46 +_ 0.09 in Grade III patients and clearly increased to more than 0.6 in Grade IV patients (results of Student's t-test analysis are given in the legend to Fig. 1). The decrease of the resistance index in Grade III patients is a result of the greater percentage increase in the diastolic flow velocity compared to the systolic flow velocity. Interdependence Between Vaso.spasm, ICP, and TCD Parameters Qualitative Aspects. Figure 2A demonstrates the time course of a typical patient with vasospasm and ICP lower than 20 mm Hg (Group A), and Fig. 2B presents that of a different patient with vasospasm and ICP greater than 20 mm Hg (Group B). The patient in Group A shows an increase in the mean flow velocity from 54 cm/sec (Day 2 after SAH) to a maximum of 190 cm/sec (Day 9 after SAH). The clinical status at this time corresponded to Grade III. In the following days, the mean flow velocity decreased and returned to normal by Day 25 after SAH. An improvement in clinical status to Grade I was also seen in this phase. The resistance index demonstrated a reciprocal development to the mean flow velocity and fluctuated in the phase of high flow velocities between 0.35 and 0.47. With normalization of mean flow velocity, there was an increase in the resistance index to values over 0.5. During the monitoring phase, the resistance index did not rise over 0.58. Angiography performed for diagnostic purposes on Day 17 after SAH revealed an aneurysm of the right internal carotid artery; this vessel as well as the middle and the anterior cerebral arteries still clearly displayed vasospasm, corresponding to the TCD findings at this time (mean flow velocity 152 cm/sec, resistance index 0.37). The ICP of the patient in Group B (Fig. 2B) showed pronounced fluctuations. Clinically, he continually presented a Grade IV status. At nearly constant ICP, the patient showed an increase in the mean flow velocity to 160 cm/sec 12 days after SAH. The increase in ICP to 38 mm Hg and drop in MABP to 79 mm Hg on Day 13 was accompanied by a decrease in the mean flow velocity to 98 cm/sec and the appearance of a readily FlG. 1. Average values of the maximum mean flow velocity (MFV) and the corresponding systolic flow velocity (SFV), diastolic flow velocity (DFV), and resistance index (R) of the 76 patients in this series compared to the clinical grade according to the classification of Hunt and Hess. 14 Statistical significance by Student's t-test (NS = not significant): SFV I vs. II: NS; II vs. III: NS; III vs. IV: NS. DFV I vs. Ih NS; II vs. III: NS; lii vs. IV: p < 0.001. MFV I vs. II: NS; II vs. III: NS; IlI vs. IV: p < 0.00I. R I vs. II: NS; II vs. Ilh NS; lii vs. IV: p < 0.001. discernible "resistance profile" with an increase in the resistance index to 0.80. This constellation, however, is obviously less a product of reduced vasospasm than of an increased ICP. Following successful drug therapy to reduce the high ICP, the mean flow velocity rose again to 142 cm/sec with a concomitant decrease in the resistance index to 0.65 on Day 15, indicating that vasospasm was still present. The resistance index did not fall below 0.5 during the entire monitoring period. Quantitative Aspects. The average values of the resistance index in 14 patients from Group A (ICP < 20 mm Hg) and 22 patients in Group B (ICP > 20 mm Hg) are shown in Fig. 3 for the first 12 days after SAH. The highest mean resistance index (0.58 _+ 0.09) was found in Group A patients during the first 2 days of the monitoring period. This decreased to a value below 0.5 between Days 9 and 12. The average mean flow velocity in this phase attained values over 140 cm/sec. The resistance index in Group B was greater than 0.6 (ranging from 0.65 _+ 0.08 to 0.69 + 0.09) during the entire monitoring period, indicating clearly increased peripheral flow resistance in this group. The lowest resistance index values were also observed in the phase of the 754 J. Neurosurg. / Volume 75/November, 1991

Doppler evaluation of vasospasm at different ICP's FIG. 2. Transcranial Doppler ultrasound recordings of the right middle cerebral artery (MCA Rt) of a 5 l- year-old patient suffering from spontaneous subarachnoid hemorrhage (SAH) and normal intracranial pressure (A) and a 34-year-old patient suffering from spontaneous SAH and pathological intracranial pressure (B), registered on different days after SAH. Mean flow velocity (MFV, cm/sec), resistance index (R), mean systemic arterial pressure (MAP, mm Hg), and mean intracranial pressure (ICP, mm Hg) are presented at various recording times. highest flow velocity between Days 7 and 10 after SAH. The lowest mean resistance index was found in both groups on Days 9 and 10. This correlated well with the occurrence of the maximum mean flow velocity (Group A: Day 10.2 _ 3.4 after SAH; Group B: Day 9.8 + 3.2 after SAH). During the critical phase in the development of vasospasm, an angiogram was necessary for diagnostic purposes in four Group A patients (ICP < 20 mm Hg) and in five Group B patients (ICP > 20 mm Hg). The four Group A patients exhibited vasospasm in the corresponding MCA, whereby the mean flow velocities at this point were clearly higher than 120 cm/sec (186 _+ 34.7 cm/sec). Their mean resistance index was less than 0.5 (0.45 _ 0.06) and the mean ICP was in a normal range (11.0 _+ 3.6 mm Hg). In contrast, the five Group B patients in whom the MCA's also showed vasospasm on angiography demonstrated mean corresponding blood flow velocities below 120 cm/sec (97.5 _+ 14.5 cm/sec) and a mean raised resistance index of more FIG. 3. Average values of the resistance index (R) in 14 Group A patients (intracranial pressure (ICP} < 20 mm Hg) and 22 Group B patients (ICP > 20 mm Hg) for the first 12 days after subarachnoid hemorrhage (SAH). Statistical significance (by Student's t-test) of differences between Group A and Group B on Days 1-2: not significant; Days 3-4: p < 0.005; Days 5-6: p < 0.005; Days 7-8: p < 0.01; Days 9-10: p < 0.001; and Days 11-12: p < 0.001. J. Neurosurg. / Volume 75/November, 199I 755

J. Klingelhrfer, et al. an average mean flow velocity of 183 37.5 cm/sec was calculated; Group B showed a 51% slower value (90 + 31.3 cm/sec). Reciprocal behavior in comparison to flow velocity was found in the resistance index: Group B had a 40% higher value (Group A: 0.45 0.06; Group B: 0.63 + 0.06). Group B also had significantly higher CT scores, calculated according to Fisher, et al. (Group A: 2.4 0.63; Group B: 2.8 + 0.39). FIG. 4. Relationship between resistance index (R), mean flow velocity (MFV), and intracranial pressure (ICP). The 219 data sets from 36 patients with ICP measurement were taken from the transcranial Doppler ultrasound recordings between Days 3 and 15 after subarachnoid hemorrhage. than 0.6 (0.69 _+ 0.07). In these Grade IV patients, the mean ICP was 36.5 4.7 mm Hg at the time of TCD recordings. In order to allow a more exact determination of the reciprocal influences of resistance index, mean flow velocity, and ICP, the results from a total of 219 dam sets from the 36 patients with ICP measurement were examined (Fig. 4). The data were taken from TCD recordings between Days 3 and 15 after SAH, a critical phase in regard to vasospasm. Each column shows the resistance index plotted against the mean flow velocity and the ICP. Identical resistance indices with mean flow velocities and ICP's in the same range are represented in one column. With rising ICP, an increase in the resistance index and a decrease in mean flow velocity can be seen. The combination of a high ICP and a high mean flow velocity was never observed. At an ICP greater than 30 mm Hg, the mean flow velocity was without exception less than 150 cm/sec; flow velocities of more than 200 cm/sec were seen only in cases with ICP values of less than 18 mm Hg. A dependence of the resistance index upon both the mean flow velocity and the magnitude of the ICP was observed: at a given constant mean flow velocity, the resistance index rose with increasing ICP; on the other hand, at a constant ICP, the resistance index decreased with increasing mean flow velocity. A resistance index of less than 0.5 was found only in combination with an ICP less than 20 mm Hg and a mean flow velocity greater than 120 cm/sec. In contrast, a resistance index greater than 0.6 was found only in combination with an ICP greater than 20 mm Hg and a mean flow velocity less than 150 cm/sec. The data for the average maximum mean flow velocities (Fig. 5 upper), the average corresponding resistance index (Fig. 5 center), and the average corresponding degree of SAH on CT scans (Fig. 5 lower) according to the classification of Fisher, et al., 8 are shown for the 36 patients in both Group A and Group B. In Group A, Discussion Flow Velocity vs. Intracranial Pressure The results show that the mean flow velocity is distinctly lower in the majority of Hunt and Hess Grade IV patients than in patients with Grades I to III. These results seem to contrast with various angiographic studies in which a direct correlation was observed between the incidence and severity of vasospasm and an increase in clinical symptoms} 6'2s For example, Voldby, et al.,26 found angiographic evidence of vasospasm in all Grade IV patients. The evaluation of vasospasm with TCD is based on the finding of an inverse proportion between vessel diameter and blood flow velocity. ~ Basically, this means that a reduction in vessel lumen is deduced from an increase in flow velocity. However, other factors beside blood vessel diameter might also have an effect on flow velocity and must be regarded in the evaluation of the severity of vasospasm. For example, Compton, et at., 6 pointed out that a decrease in the mean flow velocity in Grade IV patients compared to Grade III patients may result from the simultaneously observed reduction in cerebral blood flow (CBF). Other research groups 22'26 were able to demonstrate a 20% to 45% reduction in CBF in Grade IV patients versus Grade III patients. Our results indicate that the observed reduction in mean flow velocity in Grade IV patients in the present study is not a result of a less pronounced vasospasm but is instead caused by an increased ICP and reduced cerebral perfusion pressure (CPP), respectively. These findings were directly supported by five patients whose MCA's showed vasospasm on angiography but whose TCD blood flow velocities were clearly less than 120 cm/sec. However, angiography is an invasive procedure and performing it on patients with vasospasm involves an even greater risk of incurring neurological deftcits} 8''9 Thus, during the critical phase in the development of vasospasm, angiography was performed in only nine patients for purposes of special diagnostics. As an indication for the degree of vasospasm to be expected for all patients, the CT scale developed by Fisher, et al., ~ was used. Those and subsequent authors l~ were able to demonstrate a significant correlation between the amount and distribution of subarachnoid blood detected by CT scans early after aneurysmal rupture and the subsequent development of cerebral vasospasm visualized angiographically. Although Group B (ICP > 20 mm Hg) patients showed a significantly higher value for the CT scale score than Group A (ICP < 20 mm 756 J. Neurosurg. / Volume 75/November, 1991

Doppler evaluation of vasospasm at different ICP's Hg), Group B demonstrated a significantly lower mean flow velocity. The significantly lower mean flow velocity in Group B patients is probably a result of two factors: 1) the higher ICP in this group causes a reduction of the CPP and therefore a decrease of the mean flow velocity; and 2) the pressure loss caused by the vasospasm leads to a decrease in CBF 26 due to disturbed autoregulation and to a further reduction in blood flow velocity. Resistance Index The resistance index as a measure of the peripheral vascular flow resistance was reduced in Group A (R < 0.5). This can be a sign of (still) intact autoregulation. Primarily, the vasospasm results in a pressure loss over the spastically narrowed vessel region. In this situation, the decreased resistance index indicates a reactive dilatation in the arterioles. This dilatation allows a compensation for the pressure loss over the narrowed vessel region and subsequently a nearly constant CBF over a large range. In contrast, the elevation of the resistance index of Group B (R > 0.6) can be considered a symptom of raised peripheral vascular flow resistance due to the clearly increased ICP. Under these conditions, autoregulation may be disturbed. 26 Aaslid, et al., j observed neurological deficits solely in patients with SAH in whom the flow velocities in the extracranial internal carotid artery were simultaneously reduced during the phase of increased mean flow velocity of the MCA. In our study, no irreversible neurological deficits were observed in the examined patients whose resistance index fell below 0.5 in the phase of increased mean flow velocity; on the other hand, in those patients who developed irreversible neurological deficits a resistance index of more than 0.6 was always present during the vasospastic phase. A decrease in CBF obviously occurred in these patients as a result of vasospasm associated with increased ICP. The resistance index in patients with SAH and vasospasm thus delivers valuable information about ICP alterations and about the functional capacity and "vasomotor reserve capacity" of the autoregulatory system. False Negatives in TCD Group B patients (ICP > 20 mm Hg) demonstrated a significantly lower mean flow velocity although its CT scale score was significantly higher than that of Group A patients (ICP < 20 mm Hg). These results indicate that a TCD evaluation of the severity of vasospasm based solely upon the mean flow velocity in patients with raised ICP (usually Grade IV patients) can lead to false-negative results (blood flow velocities too low in the presence of a distinct vasospasm). The angiographic findings in the Group B patients on whom an angiography was performed for diagnostic purposes during the critical phase in the development of vasospasm supported this statement. Thus, besides the critical evaluation of the resistance index, an ICP measurement is especially important, particularly in Grade IV patients, FIG. 5. Average maximum mean flow velocities (MFV, Group A vs. B: p < 0.001, Student's t-test) (upper), average corresponding resistance index (R, Group A vs. B: p < 0.001) (center), and average corresponding degree of subaraehnoid hemorrhage on computerized tomography (CT) scans (lower) according to the classification of Fisher, etal., 8 (CT scale score; Group A vs. B: p < 0.02) for the 14 Group A patients (intracranial pressure (ICP) < 20 mm Hg) and the 22 Group B patients (ICP > 20 mm Hg). The mean PaCO2 of both groups (+ standard error of the mean) did not differ significantly (Group A: 36.5 _+ 2.0 mm Hg, Group B: 35.1 _+ 3.2 mm Hg). to decide whether the decrease in flow velocity occurring during SAH with vasospasm is a result of the decrease in vasospasm or an increase in ICP. Conclusions The following statements can be made on the basis of the present study. 1) In patients with a resistance index of less than 0.5, changes in the mean flow velocity seem to reflect sufficiently the actual severity and time course of vasospasm. 2) During the time course of vasospasm, an increase in the resistance index above values of 0.6 with a simultaneously decreased mean flow velocity indicates a rise in ICP or drop in CPP, J. Neurosurg. / Volume 75/November, 1991 757

J. Klingelhrfer, et al. respectively, rather than a reduction in vasospasm. 3) With a pronounced increase in ICP, an evaluation of the severity and time course of vasospasm by TCD based solely upon the mean flow velocity can lead to false-negative results. References 1. Aaslid R, Huber P, Nornes H: A transcranial Doppler method in the evaluation of cerebrovascular spasm. Neuroradiology 28:11-16, 1986 2. Aaslid R, Markwalder TM, Nornes H: Noninvasive transcranial Doppler ultrasound recording of flow velocity in basal cerebral arteries. J Neurosurg 57:769-774,! 982 3. Billet J, Godersky JC, Adams HP Jr: Management of aneurysmal subarachnoid hemorrhage. Stroke 19: 1300-1305, 1988 4. Brass LM, Pavlakis SG, DeVivo D, et al: Transcranial Doppler measurements of the middle cerebral artery. Effect of hematocrit. Stroke 19:1466-1469, 1988 5. Bruce DA (ed): The Pathophysiology of Increased Intracranial Pressure. Current Concepts. Philadelphia: Upjohn, 1978 6. Compton JS, Redmond S, Symon L: Cerebral blood velocity in subarachnoid haemorrhage. A transcranial Doppler study. J Neurol Neurosurg Psychiatry 50: 1499-1503, 1987 7. Dietrich K, Gaab M, Knoblich OF, et al: A new miniaturized system for monitoring the epidural pressure in children and adults. Neuropadiatrie 8:21-28, 1977 8. Fisher CM, Kistler.IP, Davis JM: Relation of cerebral vasospasm to subarachnoid hemorrhage visualized by computerized tomographic scanning. Neurosurgery 6: 1-9, 1980 9. Gelmers H J, Beks JWF, Journe6 HL: Regional cerebral blood flow in patients with subarachnoid haemorrhage. Aeta Neurochir 47:245-251, 1979 10. Gurusinghe Nq', Richardson AE: The value of computerized tomography in aneurysmal subarachnoid hemorrhage. The concept of the CT score. J Neurosurg 60: 763-770, 1984 11 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 66: 718-728, 1987 12. Hassler W: Hemodynamic Aspects of Cerebral Angiomas. New York: Springer-Verlag, 1987, pp 9-10 13. Hassler W, Steinmetz H, Gawlowski J: Transcranial Doppler ultrasonography in raised intracranial pressure and in intracranial circulatory arrest. J Neurosurg 68: 745-751, 1988 14. Hunt WE, Hess RM: Surgical r~sk as related to time of intervention in the repair of intracranial aneurysms. J Neurosurg 28:14-20, 1968 15. Klingelhbfer J, Conrad 13, Benecke R, et al: Evaluation of intracranial pressure from transcranial Doppler studies in cerebral disease. J Neurol 235:159-162, 1988 16. Klingelhbfer J, Conrad B, Sander D, et al: Evaluation of vasospasm at different cerebral perfusion pressures. J Cardiovase Ultrasonogr 7:107-108, 1988 (Abstract) 17. Lindegaard KF, Nomes H, Baake S J, et al: Cerebral vasospasm diagnosis by means of angiography and blood velocity measurements. Aeta Neurochir 100:12-24, 1989 18. Mani RL, Eisenberg RL: Complications of catheter cerebral arteriography: analysis of 5000 procedures. Ih Relation of complication rates to clinical and arteriographic diagnosis. A JR 131:867-869, 1978 19. Perret G, Nishioka H: Report of the Cooperative Study of Intracranial Aneurysms and Subarachnoid Hemorrhage. Cerebral angiography. An analysis of the diagnostic value and complications of carotid and vertebral angiography in 5,484 patients. J Neurnsurg 25:98-114, 1966 20. Pourcelot L: Diagnostic ultrasound for cerebral vascular diseases, in Donald I, Levi S (eds): Present and Future of Diagnostic Ultrasound. Rotterdam: Kooyker, 1976, pp 141-147 21. Ringelstein EB, Sievers C, Ecker S, et al: Noninvasive assessment of CO:-induced cerebral vasomotor response in normal individuals and patients with internal carotid artery occlusion. Stroke 19:963-969, 1988 22. Rosenstein J, Wang ADJ, Symon L, et al: Relationship between hemispheric cerebral blood flow, central conduction time, and clinical grade in aneurysmal subarachnoid hemorrhage. J Neurosurg 62:25-30, 1985 23. Seiler RW, Grolimund P, Aaslid R, et al: Cerebral vasospasm evaluated by transcranial ultrasound correlated with clinical grade and CT-visualized subarachnoid hemorrhage. J Neurosurg 64:594-600, 1986 24. Sekhar LN, Wechsler LR, Yonas H, et al: Value of transcranial Doppler examination in the diagnosis of cerebral vasospasm after subarachnoid hemorrhage. Neurosurgery 22:813-821, 1988 25. Voldby B, Enevoldsen EM: Intracranial pressure changes following aneurysm rupture. Part h Clinical and angiographic correlations. J Neurosurg 56:186-196, 1982 26. Voldby B, Enevoldsen EM, Jensen FT: Regional CBF, intraventricular pressure, and cerebral metabolism in patients with ruptured intracranial aneurysms. J Neurosurg 62:48-58, 1985 27. Wechsler LR, Ropper AH, Kistler JP: Transcranial Doppler in cerebrovascular disease. Stroke 17:905-915, 1986 28. Weir B, Grace M, Hansen J, et al: Time course of vasospasm in man. J Neurosurg 48:173-178, 1978 Manuscript received May 17, 1990. Accepted in final form March 25, 199l. This study was supported in part by Grant SFB 220 from the Deutsche Forschungsgemeinschaft. Address reprint requests to: Jfirgen Klingelh6fer, M.D., Neurologische Klinik und Poliklinik der Technischen Universit/it M(inchen, M6hlstrasse 28, W-8000 Miinchen 80, Germany. 758 J. Neurosurg. / Volume 75/November, 1991