- Email: [email protected]
Role of single photon emission computed tomography and transcranial doppler ultrasonography in clinical vasospasm
Journal of Clinical Neuroscience (2002) 9(4), 400±403 & 2002 Published by Elsevier Science Ltd. DOI: 10.1054/jocn.2001.1007, available online at http://www.idealibrary.com on
Clinical study
Role of single photon emission computed tomography and transcranial doppler ultrasonography in clinical vasospasm Anthony Jabre1 MD, Viken Babikian2 MD, Rachel A. Powsner3 MD, Edward L. Spatz1 MD Departments of 1Neurosurgery, 2Neurology and 3Radiology, Boston University Medical Center, Boston, Massachusetts, USA
Summary This report presents our experience with Transcranial Doppler (TCD) ultrasonography and Single Photon Emission Computed Tomography (SPECT) in the assessment of patients with aneurysmal subarachnoid haemorrhage (SAH). It was designed to evaluate clinical vasospasm with both TCD and SPECT and determine their diagnostic value. Twenty-eight consecutive patients were examined with both TCD and SPECT, performed within 24 hours of each other. They had a total of 45 TCDs and 46 SPECT scans. Eight patients (29%) developed clinical vasospasm, noted from day 2 to day 11 post subarachnoid haemorrhage; these patients underwent TCDs and SPECT scans when the diagnosis of vasospasm was made. Twenty patients (71%) did not demonstrate clinical vasospasm throughout their hospital stay and underwent TCDs and SPECT scans within the first 2 weeks of their SAH, mostly between day 2 and day 10, the period of greatest risk for vasospasm. TCD and SPECT sensitivity for clinical vasospasm was 100% and 50% respectively, their specificity was only 20% and 60%. TCD sensitivity for symptomatic vasospasm was found to be excellent, whereas SPECT was not found to be as useful. We conclude that TCD is the preferred method in the evaluation of patients with subarachnoid haemorrhage. & 2002 Published by Elsevier Science Ltd. Keywords: SPECT, subarachnoid haemorrhage, TCD, vasospasm
INTRODUCTION Subarachnoid haemorrhage caused by a ruptured berry aneurysm is a devastating event following which only one third of patients recover and two thirds are disabled or die.1 On average there is a yearly incidence of ten to 28 cases per 100 000 inhabitants.2 Rebleeding can be prevented in most cases by early operation; however, vasospasm remains a major cause of morbidity and mortality.3 Vasospasm tends to occur between day 2 and day 17 after subarachnoid haemorrhage, with a peak incidence between days 7 and 12.4±7 The percentage of patients who develop vasospasm is approximately 30%8 and nearly half of those will be inflicted with permanent disability or death.9 Thus, early confirmation and treatment of vasospam may play an important role in alleviating this grim outcome. Cerebral angiography remains an invasive procedure, and therefore great interest has developed in the evaluation of noninvasive tests that could confirm the diagnosis of cerebral vasospasm, once suspected clinically.10±15 This paper presents our experience with brain TCD and SPECT in the evaluation of patients with symptomatic vasospasm, determined by rigorous clinical criteria. MATERIALS AND METHODS A low frequency ultrasonic probe can insonate the basal intracranial vessels and record cerebral flow velocities.16 In this study, all TCDs were performed uniformly by the same
Received 25 May 2001 Accepted 28 August 2001 Correspondence to: Anthony Jabre MD, Boston University Medical Center, 720 Harrison Avenue, Suite 710, Boston, Massachusetts 02118, USA. Tel.: (617) 638 8992; Fax: (617) 638 8979; E-mail: [email protected]
400
neurovascular technologist at the patients' bedsides. Insonation with the TC2-64 monitor, equipped with a 2 MHz probe (Eden Medical Electronics, Uberlingen, Germany) was performed according to a previously described technique.17 Mean flow velocities were considered consistent with vasospasm only if they were higher than 150 cm/s. The TCD studies were interpreted by an examiner who was blinded to the patients' clinical condition and to the SPECT scan results. SPECT scanning provides functional brain imaging and detects alteration in focal cerebral perfusion. In this study, a single headed GE 400 AC/T camera with high resolution collimator was used for SPECT scanning. Tomographic imaging was performed one hour after intravenous injection of 740±1100 MBq of 99mTc-HMPAO. Sixty-four 40-second images were obtained over a 360 degree arc. Images were viewed in one pixel slices (0.5 cm/pixel) and interpreted using a colour scale. The deep cerebellum was assigned a value of 100% perfusion. Cerebral perfusion was considered consistent with vasospasm only when uptake was reduced by at least 50% in comparison with the cerebellum. Defects were registered only if they were transcortical, seen in the axial, coronal and sagittal projections and in at least two slices per projection. All SPECT studies were interpreted after careful review of the Computerized Axial Tomography (CAT) scan and SPECT defects due to nonischemic abnormalities seen on CAT scan, such as parenchymal bleed or postoperative oedema, were discounted. SPECT studies were analyzed by an experienced reader, blinded to the patients' clinical conditions and to the TCD results. Symptoms of vasospasm may include alteration of sensorium, restlessness, disorientation, focal neurological deficit or lethargy.18 In this study, daily detailed neurologic examinations were performed by the same neurosurgery staff, and clinical vasospasm was defined as a new neurological deficit not due to aneurysm rebleed, haematoma, hydrocephalus or metabolic
Evaluation of vasospasm with SPECT and TCD 401 Table 1
Patient data
Case number 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
Age/Sex
Clinical grade @ admission
Aneurysm location
Clinical spasm
SPECT/TCD
Angio spasm
33/F 67/F 49/M 58/F 72/F 55/F 61/M 56/F 57/F 52/F 44/F 49/F 69/F 34/F 41/F 61/M 58/F 51/M 41/M 33/F 32/F 46/F 60/F 37/M 34/F 37/M 58/F 63/F
2 2 2 3 2 3 3 2 3 2 4 2 4 2 2 2 3 4 4 2 3 2 4 3 2 2 3 3
Acom PICA Acom Acom Acom Pcom Acom Acom Pericallos Acom Acom Basilar Pcom MCA Acom Acom Pcom MCA Acom Pcom Basilar MCA Pcom Acom Acom Basilar Pcom Pcom
no no yes yes yes no no no no no no no no yes yes no yes no no yes no no no no yes no no no
ÿ/ ÿ/ ÿ/ / ÿ/ / ÿ/ ÿ/ ÿ/ ÿ/ÿ / ÿ/ÿ ÿ/ÿ / ÿ/ / / / / / ÿ/ / ÿ/ ÿ/ÿ ÿ/ ÿ/ / /
no no no yes no yes no no no no no no yes no no no no yes yes yes no no no no no no no no
Acom: Anterior communicating artery, Pcom: Posterior communicating artery PICA: Posterior inferior cerebellar artery, Pericallos: Pericallosal artery, MCA: Middle cerebral artery, : Vasospasm, ÿ : No vasospasm.
Table 2
causes, as confirmed by CAT scan or appropriate laboratory tests. Furthermore, for the purpose of this study, a new neurological deficit was attributed to vasospasm only if detected 48 hours or more after the subarachnoid haemorrhage, to eliminate the possibility of deficit directly related to the ictus of the bleed rather than to vasospasm. When a new neurological deficit occurred after surgery, it was attributed to vasospasm only after careful review of the CAT scan and if detected 48 hours or more after craniotomy, to eliminate the possibility of deficit resulting from brain retraction or operative stroke. This study encompasses 28 consecutive patients who presented with subarachnoid haemorrhage and underwent both SPECT scanning and TCD ultrasonography within 24 hours of each other. The patient data is summarised in Table 1. There were a total of 46 SPECT and 45 TCD studies, all of good technical quality. When patients had multiple studies, they were considered to have vasospasm if one SPECT scan showed decreased perfusion or one TCD examination showed increased velocity. All patients were treated with calcium channel blockers and those who developed clinical vasospasm were treated with hypervolemia and hemodilution. RESULTS Eight of the 28 patients (29%) developed clinical vasospasm, noted from day 2 to day 11 post subarachnoid haemorrhage. Their TCDs and SPECT scans were performed when the clinical diagnosis of vasospasm was made. Twenty patients (71%) did not show evidence of clinical vasospasm throughout their hospital stay, and their TCDs and SPECT scans were performed within the first 2 weeks following the subarachnoid haemorrhage and mostly between day 2 and day 10, the period of greatest risk for vasospasm. & 2002 Published by Elsevier Science Ltd.
TCD results
TCD
Clinical vasospasm
Increased velocity Normal velocity Total
Table 3
Yes
No
8 0 8
16 4 20
SPECT results
SPECT
Decreased uptake Normal uptake Total
Clinical vasospasm Yes
No
4 4 8
8 12 20
Four vessel cerebral angiography was performed within 48 hours of patient admission. TCD results are summarised in Table 2. Of the eight patients who developed clinical vasospasm, all had high mean blood flow velocity, indicating a test sensitivity of 100%. On the other hand, of the 20 patients who did not develop clinical vasospasm, four patients had normal mean blood flow velocity, indicating a test specificity of 20%. SPECT results are summarised in Table 3. Of the eight patients who developed clinical vasospasm, four patients had decreased uptake, indicating a test sensitivity of 50%. Of the 20 patients who did not develop clinical vasospasm, 12 patients had normal uptake, indicating a test specificity of 60%. Table 4 summarises the results. Journal of Clinical Neuroscience (2002) 9(4), 400±403
402 Jabre et al. Table 4
Sensitivity Specificity
Summary of TCD and SPECT results TCD (%)
SPECT (%)
100 20
50 60
DISCUSSION TCD is a bedside ultrasound examination capable of measuring blood flow velocity in the basal cerebral arteries.19 It is, however, operator dependent, and can be hindered by several technical obstacles such as adequacy of bone window, anatomical variations in the circle of Willis, angle of insonation, coexistence of increased intracranial pressure, presence of intracranial air, and clip artifact.20 Unlike TCD, SPECT can assess cerebral perfusion at the cellular level,21 it uses radiopharmaceutical agents able to cross the blood brain barrier and distribute to neurons in direct relationship to regional cerebral blood flow.22,23 It can demonstrate cortical ischaemia caused by involvement of small distal vessels that can elude conventional angiography.24 It can attest to the effects of compensatory mechanisms such as autoregulation and collateral circulation to explain instances of significant angiographic spasm without neurologic deficit. Single photon emission computed tomography cannot be performed at the patient's bedside, and is not practical to repeat at frequent intervals. SPECT is generally not operator dependent, but can be subject to technical difficulty related to patient motion, and must be interpreted in light of CAT scan findings capable of noting pathology seen on perfusion images but not directly related to vasospasm, such as haematoma, postoperative oedema or brain retraction effect. Finally, SPECT scan results may not always corroborate with clinical findings, as in cases of luxury perfusion where uptake may be increased in spite of deficiency in metabolism,25 or cases of hypoperfusion with decreased uptake when neuronal function continues to be sustained by increased oxygen extraction and metabolism.26 The optimal threshold of mean flow velocity for the TCD diagnosis of vasospasm is not universally determined. It is clear, however, that lower values would improve sensitivity at the expense of decreasing specificity. In this report, we have chosen a threshold of 150 cm/s. Compton et al., using postoperative TCD in a group of patients with subarachnoid haemorrhage, concluded that high flow velocities in excess of 200 cm/s may not be of alarming significance. High flow velocities unaccompanied by neurological symptoms can be seen in early stages of vasospasm when autoregulation is preserved, or when adequate collateral circulation is present.27±32 Sloan et al. presented 34 consecutive patients with subarachnoid haemorrhage, 29 of whom developed evidence of angiographic vasospasm. TCD sensitivity was found to be only 58.6%, whereas TCD specificity was found to be 100%.33 This is in contrast with our own findings where TCD sensitivity was 100% and specificity 20%. However, one notes the unusually high percentage of patients (85%) who developed angiographic spasm in the series of Sloan and his colleagues. Sekhar et al., reporting a series of 21 patients, found eight patients with vasospasm by clinical and xenonCAT scan criteria. Two of these eight patients had false negative results, and none of the 13 patients who did not have vasospasm had a false positive result, indicating a TCD sensitivity of 75%, and an ideal TCD specificity of 100%.34 Other investigators found a correlation between elevation of flow Journal of Clinical Neuroscience (2002) 9(4), 400±403
velocity and the thickness of blood clot in the subarachnoid cisterns as seen on CAT scan.30,32 Soucy and colleagues reported that nine of ten angiographically confirmed cases of vasospasm can be correctly diagnosed with SPECT.15 Lewis et al. evaluated 40 patients using TCD and SPECT, twelve of whom developed clinical vasospasm. They found a TCD sensitivity of 66% and a specificity of 37% and reported an ideal SPECT sensitivity of 100% and specificity of 71%.11 Finally, they noted that of the sixteen patients in whom both TCD and SPECT results were consistent with vasospasm, only eight patients demonstrated evidence of clinical vasospasm. TCD and SPECT diagnostic ability could possibly be enhanced, in selected cases, by CO2 or acetazolamide challenge of the cerebral vascular vasodilatory reserve. Patients with vasospasm and cerebral blood flow compromise can dilate their vascular reserve in an attempt to maintain cerebral perfusion. Administration of CO2 or acetazolamide causes dilatation of the cerebral vasculature. However, patients who have already maximally dilated their cerebral vascular reserve in response to significant ischaemia may not react with further dilation, thus changes from baseline in TCD velocities or SPECT uptake will not be noted. Evaluation of patient's CO2 or acetazolamide reactivity may identify patients at ischaemic threshold, still able to show some changes from baseline in flow velocity or uptake, prompting medical therapy or angioplasty.35,36 CONCLUSION We evaluated 28 patients with subarachnoid haemorrhage using TCD and SPECT studies. Eight patients developed clinical vasospasm, and we found TCD and SPECT sensitivity to be 100% and 50% respectively, and their specificity to be 20% and 60%. It appears that TCD has excellent sensitivity for clinical vasospasm, whereas SPECT has not been shown to be as effective. Thus, we conclude that TCD is the preferred method for the evalution of patients with subarachnoid haemorrhage and symptomatic vasospasm. REFERENCES 1. 2. 3. 4.
5.
6.
7. 8. 9. 10. 11.
Solomon R, Fink M. Current strategies for the management of aneurysmal subarachnoid haemorrhage. Arch Neurol 1987; 44: 769±774. Longstreth WT Jr, Koepsell TD, Yerby MS, Van Bell G. Risk factors for subarachnoid haemorrhage. Stroke 1985; 16: 377±385. Kassell NF, Torner JC. The international cooperative study on timing of aneurysm surgery ± an update. Stroke 1984; 15: 566±570. Kwak R, Niizuma H, Ohi T, Suzuki J. Angiographic study of cerebral vasospasm following rupture of intracranial aneurysms, part I: time of the appearance. Surg Neurol 1979; 11: 257±262. Niizuma H, Kwark R, Otabe K, Suzuki J. Angiographic study of cerebral vasospasm following the rupture of intracranial aneurysm, part II: relation between the site of aneurysm and the occurrence of vasospasm. Surg Neurol 1979; 11: 263±267. Saito I, Sano K. Vasospasm after aneurysm rupture: incidence, onset and course. In Wilkins RH (ed) Cerebral vasospasm (1980) Baltimore, Williams and Wilkins: 294±301. Weir B, Grace M, Hansem J, Rothberg C. Time course of vasospasm in man. J Neurosurg 1978; 48: 173±178. Barker FG, Heros RC. Clinical aspects of vasospasm. Neurosurg Clin North Am 1990; 1: 277±288. Torner JC, Kassell NF, Haley EC. The timing of surgery and vasospasm. Neursurg Clin North Am 1990; 1: 335±347. Aaslid R, Huber P, Nornes H. Evaluation of cerebrovascular spasm with transcranial Doppler ultrasound. J Neurosurg 1984; 60: 37±41. Lewis DH, Hsu S, Eskridge J, Cohen W, Dalley R, Newell DW, Douville C, Pendleton G, Chestnut CH,Winn HR. Brain SPECT and transcranial Doppler ultrasound in vasospasm-induced delayed cerebral ischaemia after subarachnoid haemorrhage. J Stroke Cerebrovasc Dis 1992; 2: 12±21.
& 2002 Published by Elsevier Science Ltd.
Evaluation of vasospasm with SPECT and TCD 403 12.
13.
14.
15.
16.
17.
18.
19. 20.
21.
22.
23.
Mickey B, Vorstrup S, Voldby B, Lindwald H, Harmsen A, Lassen NA. Serial measurements of regional cerebral blood flow with subarachnoid haemorrhage using Xe-133 inhalation and emission computed tomography. J Neurosurg 1984; 60: 916±922. Powsner RA, O'Tuama LA, Jabre A, Melhem ER. SPECT imaging in cerebral vasospasm following subarachnoid haemorrhage: A predictor of clinical outcome. J Nucl Med 1998; 39: 765±769. Sloan MA. Detection of vasospasm following subarachnoid haemorrhage, in Babikian VL, Wechsler LR (eds) Transcranial Doppler Ultrasonography (1993) St. Louis, Mosby Publishing Co: 105±127. Soucy JP, McNamara D, Mohr G, Lamoureux J, Danais S. Evaluation of vasospasm secondary to subarachnoid haemorrhage with Tc 99m-HMPAO tomoscintigraphy. J Nucl Med 1990; 31: 972±977. Aaslid R, Markwalder TM, Nornes H. Non invasive transcranial Doppler ultrasound recording of flow velocity in basal cerebral arteries. J Neurosurg 1982; 57: 769±774. Saver JL, Feldmann E. Basic transcranial Doppler examination: Technique and anatomy, in Babikian VL, Wechsler LR (eds) Transcranial Doppler Ultrasonography (1993) St Louis, Mosby Publishing Co: 11±28. Barnwell SL, Higashida RT, Halbach VV, Dowd CF, Wilson CB, Hieshima GB. Transluminal angioplasty of intracerebral vessels for delayed cerebral arterial spasm. Reversal of neurological deficits after delayed treatment. Neurosurgery 1989; 25: 424±429. Newell DW, Winn HR. Transcranial Doppler in cerebral vasospasm. Neurosurg Clin North Am 1990; 1: 319±328. Klingelhofer J, Sander D, Holzgrafe M, Biscoff C, Conrad B. Cerebral vasospasm evaluated by transcranial Doppler at different intracranial pressures. J Neurosurg 1991; 75: 752±758. Johsita H, Kassell NF, Sasaki T, Ogawa H. Impaired capillary perfusion and brain edema following experimental subarachnoid haemorrhage: A morphometric study. J Neurosurg 1990; 73: 410±417. Kuhl D, Barrio J, Huang S, Selin C, Ackermann RF, Lear JL, WY JL, Lin TH, Phelps ME. Quantifying local cerebral blood flow in N-isopropyl-p-I-123-iodoamphetamine (IMP) tomography. J Nucl Med 1982; 23: 196±203. O'Leary DH, Hill TC, Lee RG, Clouse ME, Holman BL. The use of 123-iodoamphetamines and single photon emission computed tomography to assess local cerebral blood flow. AJNR 1983; 4: 547±549.
& 2002 Published by Elsevier Science Ltd.
24.
25. 26.
27.
28.
29. 30.
31. 32.
33.
34.
35.
36.
Rawluk D, Smith FW, Deans HE, Gemmel HG, MacDonald AF. Technetium 99m HMPAO scanning in patients with subarachnoid haemorrhage: A preliminary study. Br J Neurosurg 1988; 61: 26±29. Lassen NA. The luxury-perfusion syndrome and its possible relation to metabolic acidosis located with the brain. Lancet 1966; 2: 1113±1115. Powers WJ, Grubb RL Jr, Baker RP, Mintun MA, Raichle ME. Regional cerebral blood flow and metabolism in reversible ischemia due to vasospasm. Determination by positron emission tomography. J Neurosurg 1985; 62: 539±546. Compton JS, Redmond S, Symon L. Cerebral blood flow velocity in subarachnoid haemorrhage: a transcranial Doppler study. J Neurol Neurosurg Psychiatry 1987; 50: 1499±1503. Caplan LR, Brass LM, De Witt LD, Adams RJ, Gomez C, Otis S, Wechsler LR, Von Reutern G-M. Transcranial Doppler ultrasound: Present status. Neurology 1990; 40: 696±700. De Witt LD, Wechsler LR. Transcranial Doppler: Curr concepts Cerebrovasc Dis Stroke 1987; 22: 31±36. Harders AG, Gilsbach JML. Time course of blood velocity changes related to vasospasm in the circle of Willis measured by transcranial Doppler ultrasound. J Neurosurg 1987; 66: 718±728. Heros RC, Zervas NT, Varsos V. Cerebral vasospasm after subarachnoid heamorrhage: An update. Ann Neurol 1983; 14: 599±608. Seiler RW, Grolimund P. Aaslid A, Huber P, Nornes H. Cerebral vasospasm evaluated by transcranial ultrasound correlated with clinical grade and CT visualized subarachnoid haemorrhage. J Neurosurg 1986; 64: 594±600. Sloan MA, Haley EC Jr, Kassell NF, Henry ML, Stewart SR, Beskin RR, Sevilla EA, Torner JC. Sensitivity and specificity of transcranial Doppler ultrasonography in the diagnosis of vasospasm following subarachnoid haemorrhage. Neurology 1989; 39: 1514±1518. Sekhar LN, Wechsler LR, Yonas H. Luyckx K, Obrist W. Value of Transcranial Doppler examination in the diagnosis of cerebral vasospasm after subarachnoid haemorrhage. Neurosurgery 1988; 22: 813±821. Devous MD Sr, Batjer HH, Ajmani AK, Samson DS, Bonte FJ. SPECT measurements of cerebrovascular reserve in patients with hemodynamic risk. J Nucl Med 1988 (abstr); 29: 911. Hassler W, Chioffi F. CO2 reactivity of cerebral vasospasm after aneurysmal subarachnoid hemorrhage. Acta Neurochir 1989; 98: 167±175.
Journal of Clinical Neuroscience (2002) 9(4), 400±403
Clinical study
Role of single photon emission computed tomography and transcranial doppler ultrasonography in clinical vasospasm Anthony Jabre1 MD, Viken Babikian2 MD, Rachel A. Powsner3 MD, Edward L. Spatz1 MD Departments of 1Neurosurgery, 2Neurology and 3Radiology, Boston University Medical Center, Boston, Massachusetts, USA
Summary This report presents our experience with Transcranial Doppler (TCD) ultrasonography and Single Photon Emission Computed Tomography (SPECT) in the assessment of patients with aneurysmal subarachnoid haemorrhage (SAH). It was designed to evaluate clinical vasospasm with both TCD and SPECT and determine their diagnostic value. Twenty-eight consecutive patients were examined with both TCD and SPECT, performed within 24 hours of each other. They had a total of 45 TCDs and 46 SPECT scans. Eight patients (29%) developed clinical vasospasm, noted from day 2 to day 11 post subarachnoid haemorrhage; these patients underwent TCDs and SPECT scans when the diagnosis of vasospasm was made. Twenty patients (71%) did not demonstrate clinical vasospasm throughout their hospital stay and underwent TCDs and SPECT scans within the first 2 weeks of their SAH, mostly between day 2 and day 10, the period of greatest risk for vasospasm. TCD and SPECT sensitivity for clinical vasospasm was 100% and 50% respectively, their specificity was only 20% and 60%. TCD sensitivity for symptomatic vasospasm was found to be excellent, whereas SPECT was not found to be as useful. We conclude that TCD is the preferred method in the evaluation of patients with subarachnoid haemorrhage. & 2002 Published by Elsevier Science Ltd. Keywords: SPECT, subarachnoid haemorrhage, TCD, vasospasm
INTRODUCTION Subarachnoid haemorrhage caused by a ruptured berry aneurysm is a devastating event following which only one third of patients recover and two thirds are disabled or die.1 On average there is a yearly incidence of ten to 28 cases per 100 000 inhabitants.2 Rebleeding can be prevented in most cases by early operation; however, vasospasm remains a major cause of morbidity and mortality.3 Vasospasm tends to occur between day 2 and day 17 after subarachnoid haemorrhage, with a peak incidence between days 7 and 12.4±7 The percentage of patients who develop vasospasm is approximately 30%8 and nearly half of those will be inflicted with permanent disability or death.9 Thus, early confirmation and treatment of vasospam may play an important role in alleviating this grim outcome. Cerebral angiography remains an invasive procedure, and therefore great interest has developed in the evaluation of noninvasive tests that could confirm the diagnosis of cerebral vasospasm, once suspected clinically.10±15 This paper presents our experience with brain TCD and SPECT in the evaluation of patients with symptomatic vasospasm, determined by rigorous clinical criteria. MATERIALS AND METHODS A low frequency ultrasonic probe can insonate the basal intracranial vessels and record cerebral flow velocities.16 In this study, all TCDs were performed uniformly by the same
Received 25 May 2001 Accepted 28 August 2001 Correspondence to: Anthony Jabre MD, Boston University Medical Center, 720 Harrison Avenue, Suite 710, Boston, Massachusetts 02118, USA. Tel.: (617) 638 8992; Fax: (617) 638 8979; E-mail: [email protected]
400
neurovascular technologist at the patients' bedsides. Insonation with the TC2-64 monitor, equipped with a 2 MHz probe (Eden Medical Electronics, Uberlingen, Germany) was performed according to a previously described technique.17 Mean flow velocities were considered consistent with vasospasm only if they were higher than 150 cm/s. The TCD studies were interpreted by an examiner who was blinded to the patients' clinical condition and to the SPECT scan results. SPECT scanning provides functional brain imaging and detects alteration in focal cerebral perfusion. In this study, a single headed GE 400 AC/T camera with high resolution collimator was used for SPECT scanning. Tomographic imaging was performed one hour after intravenous injection of 740±1100 MBq of 99mTc-HMPAO. Sixty-four 40-second images were obtained over a 360 degree arc. Images were viewed in one pixel slices (0.5 cm/pixel) and interpreted using a colour scale. The deep cerebellum was assigned a value of 100% perfusion. Cerebral perfusion was considered consistent with vasospasm only when uptake was reduced by at least 50% in comparison with the cerebellum. Defects were registered only if they were transcortical, seen in the axial, coronal and sagittal projections and in at least two slices per projection. All SPECT studies were interpreted after careful review of the Computerized Axial Tomography (CAT) scan and SPECT defects due to nonischemic abnormalities seen on CAT scan, such as parenchymal bleed or postoperative oedema, were discounted. SPECT studies were analyzed by an experienced reader, blinded to the patients' clinical conditions and to the TCD results. Symptoms of vasospasm may include alteration of sensorium, restlessness, disorientation, focal neurological deficit or lethargy.18 In this study, daily detailed neurologic examinations were performed by the same neurosurgery staff, and clinical vasospasm was defined as a new neurological deficit not due to aneurysm rebleed, haematoma, hydrocephalus or metabolic
Evaluation of vasospasm with SPECT and TCD 401 Table 1
Patient data
Case number 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
Age/Sex
Clinical grade @ admission
Aneurysm location
Clinical spasm
SPECT/TCD
Angio spasm
33/F 67/F 49/M 58/F 72/F 55/F 61/M 56/F 57/F 52/F 44/F 49/F 69/F 34/F 41/F 61/M 58/F 51/M 41/M 33/F 32/F 46/F 60/F 37/M 34/F 37/M 58/F 63/F
2 2 2 3 2 3 3 2 3 2 4 2 4 2 2 2 3 4 4 2 3 2 4 3 2 2 3 3
Acom PICA Acom Acom Acom Pcom Acom Acom Pericallos Acom Acom Basilar Pcom MCA Acom Acom Pcom MCA Acom Pcom Basilar MCA Pcom Acom Acom Basilar Pcom Pcom
no no yes yes yes no no no no no no no no yes yes no yes no no yes no no no no yes no no no
ÿ/ ÿ/ ÿ/ / ÿ/ / ÿ/ ÿ/ ÿ/ ÿ/ÿ / ÿ/ÿ ÿ/ÿ / ÿ/ / / / / / ÿ/ / ÿ/ ÿ/ÿ ÿ/ ÿ/ / /
no no no yes no yes no no no no no no yes no no no no yes yes yes no no no no no no no no
Acom: Anterior communicating artery, Pcom: Posterior communicating artery PICA: Posterior inferior cerebellar artery, Pericallos: Pericallosal artery, MCA: Middle cerebral artery, : Vasospasm, ÿ : No vasospasm.
Table 2
causes, as confirmed by CAT scan or appropriate laboratory tests. Furthermore, for the purpose of this study, a new neurological deficit was attributed to vasospasm only if detected 48 hours or more after the subarachnoid haemorrhage, to eliminate the possibility of deficit directly related to the ictus of the bleed rather than to vasospasm. When a new neurological deficit occurred after surgery, it was attributed to vasospasm only after careful review of the CAT scan and if detected 48 hours or more after craniotomy, to eliminate the possibility of deficit resulting from brain retraction or operative stroke. This study encompasses 28 consecutive patients who presented with subarachnoid haemorrhage and underwent both SPECT scanning and TCD ultrasonography within 24 hours of each other. The patient data is summarised in Table 1. There were a total of 46 SPECT and 45 TCD studies, all of good technical quality. When patients had multiple studies, they were considered to have vasospasm if one SPECT scan showed decreased perfusion or one TCD examination showed increased velocity. All patients were treated with calcium channel blockers and those who developed clinical vasospasm were treated with hypervolemia and hemodilution. RESULTS Eight of the 28 patients (29%) developed clinical vasospasm, noted from day 2 to day 11 post subarachnoid haemorrhage. Their TCDs and SPECT scans were performed when the clinical diagnosis of vasospasm was made. Twenty patients (71%) did not show evidence of clinical vasospasm throughout their hospital stay, and their TCDs and SPECT scans were performed within the first 2 weeks following the subarachnoid haemorrhage and mostly between day 2 and day 10, the period of greatest risk for vasospasm. & 2002 Published by Elsevier Science Ltd.
TCD results
TCD
Clinical vasospasm
Increased velocity Normal velocity Total
Table 3
Yes
No
8 0 8
16 4 20
SPECT results
SPECT
Decreased uptake Normal uptake Total
Clinical vasospasm Yes
No
4 4 8
8 12 20
Four vessel cerebral angiography was performed within 48 hours of patient admission. TCD results are summarised in Table 2. Of the eight patients who developed clinical vasospasm, all had high mean blood flow velocity, indicating a test sensitivity of 100%. On the other hand, of the 20 patients who did not develop clinical vasospasm, four patients had normal mean blood flow velocity, indicating a test specificity of 20%. SPECT results are summarised in Table 3. Of the eight patients who developed clinical vasospasm, four patients had decreased uptake, indicating a test sensitivity of 50%. Of the 20 patients who did not develop clinical vasospasm, 12 patients had normal uptake, indicating a test specificity of 60%. Table 4 summarises the results. Journal of Clinical Neuroscience (2002) 9(4), 400±403
402 Jabre et al. Table 4
Sensitivity Specificity
Summary of TCD and SPECT results TCD (%)
SPECT (%)
100 20
50 60
DISCUSSION TCD is a bedside ultrasound examination capable of measuring blood flow velocity in the basal cerebral arteries.19 It is, however, operator dependent, and can be hindered by several technical obstacles such as adequacy of bone window, anatomical variations in the circle of Willis, angle of insonation, coexistence of increased intracranial pressure, presence of intracranial air, and clip artifact.20 Unlike TCD, SPECT can assess cerebral perfusion at the cellular level,21 it uses radiopharmaceutical agents able to cross the blood brain barrier and distribute to neurons in direct relationship to regional cerebral blood flow.22,23 It can demonstrate cortical ischaemia caused by involvement of small distal vessels that can elude conventional angiography.24 It can attest to the effects of compensatory mechanisms such as autoregulation and collateral circulation to explain instances of significant angiographic spasm without neurologic deficit. Single photon emission computed tomography cannot be performed at the patient's bedside, and is not practical to repeat at frequent intervals. SPECT is generally not operator dependent, but can be subject to technical difficulty related to patient motion, and must be interpreted in light of CAT scan findings capable of noting pathology seen on perfusion images but not directly related to vasospasm, such as haematoma, postoperative oedema or brain retraction effect. Finally, SPECT scan results may not always corroborate with clinical findings, as in cases of luxury perfusion where uptake may be increased in spite of deficiency in metabolism,25 or cases of hypoperfusion with decreased uptake when neuronal function continues to be sustained by increased oxygen extraction and metabolism.26 The optimal threshold of mean flow velocity for the TCD diagnosis of vasospasm is not universally determined. It is clear, however, that lower values would improve sensitivity at the expense of decreasing specificity. In this report, we have chosen a threshold of 150 cm/s. Compton et al., using postoperative TCD in a group of patients with subarachnoid haemorrhage, concluded that high flow velocities in excess of 200 cm/s may not be of alarming significance. High flow velocities unaccompanied by neurological symptoms can be seen in early stages of vasospasm when autoregulation is preserved, or when adequate collateral circulation is present.27±32 Sloan et al. presented 34 consecutive patients with subarachnoid haemorrhage, 29 of whom developed evidence of angiographic vasospasm. TCD sensitivity was found to be only 58.6%, whereas TCD specificity was found to be 100%.33 This is in contrast with our own findings where TCD sensitivity was 100% and specificity 20%. However, one notes the unusually high percentage of patients (85%) who developed angiographic spasm in the series of Sloan and his colleagues. Sekhar et al., reporting a series of 21 patients, found eight patients with vasospasm by clinical and xenonCAT scan criteria. Two of these eight patients had false negative results, and none of the 13 patients who did not have vasospasm had a false positive result, indicating a TCD sensitivity of 75%, and an ideal TCD specificity of 100%.34 Other investigators found a correlation between elevation of flow Journal of Clinical Neuroscience (2002) 9(4), 400±403
velocity and the thickness of blood clot in the subarachnoid cisterns as seen on CAT scan.30,32 Soucy and colleagues reported that nine of ten angiographically confirmed cases of vasospasm can be correctly diagnosed with SPECT.15 Lewis et al. evaluated 40 patients using TCD and SPECT, twelve of whom developed clinical vasospasm. They found a TCD sensitivity of 66% and a specificity of 37% and reported an ideal SPECT sensitivity of 100% and specificity of 71%.11 Finally, they noted that of the sixteen patients in whom both TCD and SPECT results were consistent with vasospasm, only eight patients demonstrated evidence of clinical vasospasm. TCD and SPECT diagnostic ability could possibly be enhanced, in selected cases, by CO2 or acetazolamide challenge of the cerebral vascular vasodilatory reserve. Patients with vasospasm and cerebral blood flow compromise can dilate their vascular reserve in an attempt to maintain cerebral perfusion. Administration of CO2 or acetazolamide causes dilatation of the cerebral vasculature. However, patients who have already maximally dilated their cerebral vascular reserve in response to significant ischaemia may not react with further dilation, thus changes from baseline in TCD velocities or SPECT uptake will not be noted. Evaluation of patient's CO2 or acetazolamide reactivity may identify patients at ischaemic threshold, still able to show some changes from baseline in flow velocity or uptake, prompting medical therapy or angioplasty.35,36 CONCLUSION We evaluated 28 patients with subarachnoid haemorrhage using TCD and SPECT studies. Eight patients developed clinical vasospasm, and we found TCD and SPECT sensitivity to be 100% and 50% respectively, and their specificity to be 20% and 60%. It appears that TCD has excellent sensitivity for clinical vasospasm, whereas SPECT has not been shown to be as effective. Thus, we conclude that TCD is the preferred method for the evalution of patients with subarachnoid haemorrhage and symptomatic vasospasm. REFERENCES 1. 2. 3. 4.
5.
6.
7. 8. 9. 10. 11.
Solomon R, Fink M. Current strategies for the management of aneurysmal subarachnoid haemorrhage. Arch Neurol 1987; 44: 769±774. Longstreth WT Jr, Koepsell TD, Yerby MS, Van Bell G. Risk factors for subarachnoid haemorrhage. Stroke 1985; 16: 377±385. Kassell NF, Torner JC. The international cooperative study on timing of aneurysm surgery ± an update. Stroke 1984; 15: 566±570. Kwak R, Niizuma H, Ohi T, Suzuki J. Angiographic study of cerebral vasospasm following rupture of intracranial aneurysms, part I: time of the appearance. Surg Neurol 1979; 11: 257±262. Niizuma H, Kwark R, Otabe K, Suzuki J. Angiographic study of cerebral vasospasm following the rupture of intracranial aneurysm, part II: relation between the site of aneurysm and the occurrence of vasospasm. Surg Neurol 1979; 11: 263±267. Saito I, Sano K. Vasospasm after aneurysm rupture: incidence, onset and course. In Wilkins RH (ed) Cerebral vasospasm (1980) Baltimore, Williams and Wilkins: 294±301. Weir B, Grace M, Hansem J, Rothberg C. Time course of vasospasm in man. J Neurosurg 1978; 48: 173±178. Barker FG, Heros RC. Clinical aspects of vasospasm. Neurosurg Clin North Am 1990; 1: 277±288. Torner JC, Kassell NF, Haley EC. The timing of surgery and vasospasm. Neursurg Clin North Am 1990; 1: 335±347. Aaslid R, Huber P, Nornes H. Evaluation of cerebrovascular spasm with transcranial Doppler ultrasound. J Neurosurg 1984; 60: 37±41. Lewis DH, Hsu S, Eskridge J, Cohen W, Dalley R, Newell DW, Douville C, Pendleton G, Chestnut CH,Winn HR. Brain SPECT and transcranial Doppler ultrasound in vasospasm-induced delayed cerebral ischaemia after subarachnoid haemorrhage. J Stroke Cerebrovasc Dis 1992; 2: 12±21.
& 2002 Published by Elsevier Science Ltd.
Evaluation of vasospasm with SPECT and TCD 403 12.
13.
14.
15.
16.
17.
18.
19. 20.
21.
22.
23.
Mickey B, Vorstrup S, Voldby B, Lindwald H, Harmsen A, Lassen NA. Serial measurements of regional cerebral blood flow with subarachnoid haemorrhage using Xe-133 inhalation and emission computed tomography. J Neurosurg 1984; 60: 916±922. Powsner RA, O'Tuama LA, Jabre A, Melhem ER. SPECT imaging in cerebral vasospasm following subarachnoid haemorrhage: A predictor of clinical outcome. J Nucl Med 1998; 39: 765±769. Sloan MA. Detection of vasospasm following subarachnoid haemorrhage, in Babikian VL, Wechsler LR (eds) Transcranial Doppler Ultrasonography (1993) St. Louis, Mosby Publishing Co: 105±127. Soucy JP, McNamara D, Mohr G, Lamoureux J, Danais S. Evaluation of vasospasm secondary to subarachnoid haemorrhage with Tc 99m-HMPAO tomoscintigraphy. J Nucl Med 1990; 31: 972±977. Aaslid R, Markwalder TM, Nornes H. Non invasive transcranial Doppler ultrasound recording of flow velocity in basal cerebral arteries. J Neurosurg 1982; 57: 769±774. Saver JL, Feldmann E. Basic transcranial Doppler examination: Technique and anatomy, in Babikian VL, Wechsler LR (eds) Transcranial Doppler Ultrasonography (1993) St Louis, Mosby Publishing Co: 11±28. Barnwell SL, Higashida RT, Halbach VV, Dowd CF, Wilson CB, Hieshima GB. Transluminal angioplasty of intracerebral vessels for delayed cerebral arterial spasm. Reversal of neurological deficits after delayed treatment. Neurosurgery 1989; 25: 424±429. Newell DW, Winn HR. Transcranial Doppler in cerebral vasospasm. Neurosurg Clin North Am 1990; 1: 319±328. Klingelhofer J, Sander D, Holzgrafe M, Biscoff C, Conrad B. Cerebral vasospasm evaluated by transcranial Doppler at different intracranial pressures. J Neurosurg 1991; 75: 752±758. Johsita H, Kassell NF, Sasaki T, Ogawa H. Impaired capillary perfusion and brain edema following experimental subarachnoid haemorrhage: A morphometric study. J Neurosurg 1990; 73: 410±417. Kuhl D, Barrio J, Huang S, Selin C, Ackermann RF, Lear JL, WY JL, Lin TH, Phelps ME. Quantifying local cerebral blood flow in N-isopropyl-p-I-123-iodoamphetamine (IMP) tomography. J Nucl Med 1982; 23: 196±203. O'Leary DH, Hill TC, Lee RG, Clouse ME, Holman BL. The use of 123-iodoamphetamines and single photon emission computed tomography to assess local cerebral blood flow. AJNR 1983; 4: 547±549.
& 2002 Published by Elsevier Science Ltd.
24.
25. 26.
27.
28.
29. 30.
31. 32.
33.
34.
35.
36.
Rawluk D, Smith FW, Deans HE, Gemmel HG, MacDonald AF. Technetium 99m HMPAO scanning in patients with subarachnoid haemorrhage: A preliminary study. Br J Neurosurg 1988; 61: 26±29. Lassen NA. The luxury-perfusion syndrome and its possible relation to metabolic acidosis located with the brain. Lancet 1966; 2: 1113±1115. Powers WJ, Grubb RL Jr, Baker RP, Mintun MA, Raichle ME. Regional cerebral blood flow and metabolism in reversible ischemia due to vasospasm. Determination by positron emission tomography. J Neurosurg 1985; 62: 539±546. Compton JS, Redmond S, Symon L. Cerebral blood flow velocity in subarachnoid haemorrhage: a transcranial Doppler study. J Neurol Neurosurg Psychiatry 1987; 50: 1499±1503. Caplan LR, Brass LM, De Witt LD, Adams RJ, Gomez C, Otis S, Wechsler LR, Von Reutern G-M. Transcranial Doppler ultrasound: Present status. Neurology 1990; 40: 696±700. De Witt LD, Wechsler LR. Transcranial Doppler: Curr concepts Cerebrovasc Dis Stroke 1987; 22: 31±36. Harders AG, Gilsbach JML. Time course of blood velocity changes related to vasospasm in the circle of Willis measured by transcranial Doppler ultrasound. J Neurosurg 1987; 66: 718±728. Heros RC, Zervas NT, Varsos V. Cerebral vasospasm after subarachnoid heamorrhage: An update. Ann Neurol 1983; 14: 599±608. Seiler RW, Grolimund P. Aaslid A, Huber P, Nornes H. Cerebral vasospasm evaluated by transcranial ultrasound correlated with clinical grade and CT visualized subarachnoid haemorrhage. J Neurosurg 1986; 64: 594±600. Sloan MA, Haley EC Jr, Kassell NF, Henry ML, Stewart SR, Beskin RR, Sevilla EA, Torner JC. Sensitivity and specificity of transcranial Doppler ultrasonography in the diagnosis of vasospasm following subarachnoid haemorrhage. Neurology 1989; 39: 1514±1518. Sekhar LN, Wechsler LR, Yonas H. Luyckx K, Obrist W. Value of Transcranial Doppler examination in the diagnosis of cerebral vasospasm after subarachnoid haemorrhage. Neurosurgery 1988; 22: 813±821. Devous MD Sr, Batjer HH, Ajmani AK, Samson DS, Bonte FJ. SPECT measurements of cerebrovascular reserve in patients with hemodynamic risk. J Nucl Med 1988 (abstr); 29: 911. Hassler W, Chioffi F. CO2 reactivity of cerebral vasospasm after aneurysmal subarachnoid hemorrhage. Acta Neurochir 1989; 98: 167±175.
Journal of Clinical Neuroscience (2002) 9(4), 400±403