- Email: [email protected]
Xenon-enhanced computed tomography cerebral blood flow measurements in acute cerebral ischemia: Review of 56 cases
Xenon-Enhanced Computed Tomography Cerebral Blood Flow Measurements in Acute Cerebral Ischemia: Review of 56 Cases Giorgio Rubin, MD, Andrew D. Firlik, MD, Elad I. Levy, MD, Ronda R. Pindzola, PhD, and Howard Yonas, MD
Objective: Ischemic stroke must be diagnosed promptly if patients are to be treated with thrombolytic therapy. The diagnosis of acute cerebral ischemia, however, is usually based on clinical and computed tomography (CT) scan findings. CT scans are often normal in the first few hours after stroke. The purpose of this study was to determine whether Xenon-enhanced CT (XeCT) cerebral blood flow (CBF) studies could increase the sensitivity of stroke detection in the acute stage. Methods: CBF studies performed within 8 hours of symptom onset were evaluated in 56 patients who presented with hemispheric stroke symptoms. Mean CBF in the symptomatic vascular territory was calculated and compared with the corresponding contralateral area. CBF values below 18 mL/100g/min on 2 adjacent regions of interest were considered ischemic lesions. CT scans and angiograms were compared with the XeCT findings. Neurological condition on admission and discharge was evaluated by using National Institutes of Health Stroke Scale (NII-ISS) scores. Results: The mean NIHSS score on admission was 12 + 5. Early CT scans were abnormal in 28 (50%) patients. There were 9 (16%) patients who had normal XeCT scans because of spontaneous reperfusion of the ischemic area. XeCT studies showed an ischemic lesion in 47 (84%) patients. In these patients, the mean CBF in the affected vascular territory was 16 -+ 8 mL/100g/min compared with 35 + 13 mL/100g/min in the contralateral specular territory (P < 0.001). There were no false positive or negative XeCT studies, and the location of the perfusion defect corresponded with the CT and/or angiographic findings in all cases. Eight patients died (14%), and the 48 survivors (86%) had a mean NIHSS score of 9 _+ 6 on discharge. Conclusions: CBF measurements were correlated with the CT and angiographic results and greatly assisted in the diagnosis of acute ischemic stroke. XeCT studies used for estimating the location and extent of cerebral ischemia may be important in the triage of patients for acute stroke therapy. Key Words: Cerebral blood flow--Diagnosis--Stroke--Xenon-enhanced computed tomography.
Cerebral blood flow (CBF) is rarely studied in acute stroke victims even though its reduction is the basic pathophysiological process of cerebral ischemia. Cur-
From the University of Pittsburgh Medical Center, Department of Neurosurgery,Pittsburgh, PA. Received January 26,1999;accepted June 2,1999. Dr. Yonasreceivesa research grant from Praxair, Inc. Address reprint requests to Howard Yonas, MD, Department of Neurological Surgery, University of Pittsburgh Medical Center, 200 Lothrop Street, Suite B-400,Pittsburgh, PA 15213. Copyright 9 1999by National Stroke Association 1052-3057/99/0806-000653.00/0
404
rently, the diagnosis of stroke is based on a neurological examination and on computed tomography (CT) findings. CT scans are used to rule out the causes of hemorrhagic stroke and may be useful in excluding patients who should be treated with reperfusion therapy. 1 The CT image, however, is often normal when performed in the first several hours after the ischemic attack. 2"s The European Cooperative Acute Stroke Study (ECASS) and the National Institute of Neurological Disorders and Stroke (NINDS) investigators in the recent reperfusion therapy trial 6,7 have stressed the necessity of better patient selection. Early knowledge of the location and severity of the perfusion defect may help identify stroke patients who
Journal of Stroke and Cerebrovascular Diseases, Vol. 8, No. 6 (November-December), 1999: pp 404-411
405
CBF IN ACUTE STROKE
will benefit from aggressive therapy. Xenon-enhanced CT (XeCT) CBF methodology is a practical, inexpensive, and fast technique that will provide quantitative CBF measurements. In a preliminary report, we evaluated the sensitivity of XeCT for detecting CBF reduction in the first hours after stroke. 8 XeCT proved to be a very promising methodology, but this study was restricted to 20 patients with major stroke and was limited to I territory of the cerebral Circulation. The aim of the present study is to verify the validity of XeCT in detecting acute cerebral ischemia in a larger, heterogeneous stroke population.
Subjects and M e t h o d s Patient Population From January 1, 1996 to June 31, 1998, 74 patients with ischemic stroke underwent XeCT CBF studies at the University of Pittsburgh Medical Center within 8 hours of symptom onset. From this group, 56 patients were selected and retrospectively reviewed. Ischemic stroke was diagnosed when patients presented with acute onset of neurological deficit associated with a CT scan on admission that was either normal or had ischemic changes. The inclusion criteria were (1) hemispheric stroke; (2) absence of hemorrhagic, bilateral, or previous infarction; and (3) CBF map without artifacts. On admission in the emergency room, members of the stroke team, which was composed of neurologists and neurosurgeons, examined all patients. Each patient's neurological deficit was quantified on admission and during their hospital stay by using the National Institutes of Health Stroke Scale (NIHSS). 9 CT scans and XeCT CBF studies were performed as soon as possible after the clinical evaluation. Based on their neurological condition and the results of the radiographic investigation, patients were considered for cerebral angiography and, eventually, for reperfusion therapy. The thrombolytic drug was administered intravenously if the patient was symptomatic for less than 3 hours. Intraarterial treatment was used if the stroke onset was between 3 and 6 hours. During their hospital stay, the neurological condition of every patient was determined daily by using NIHSS scores.
CT Scanning CT images were evaluated for signs of infarction, such as hypodensity or sulcal effacement, and were divided into 4 categories: (1) cortical findings; (2) basal ganglia findings; (3) mass effect on the lateral ventricles; and (4) arterial hyperdensity (as evidence of thrombosis or embolus). For patients who received multiple follow-up CT scans, the scan that most clearly delineated the infarction was used.
XeCT CBF Study The XeCT studies were performed immediately after completion of the first CT scans. A description of the XeCT methodology has been previously reported, l~ Briefly, cortical and subcortical CBF was evaluated by a computer analysis program that divides each of the 3 hemispheric brain slices into 18 to 22 regions of interest (ROIs). Each 2-cm circular ROI was composed of mixed cortical CBF values (gray and white matter). The time required to perform the XeCT study (including image acquisition and CBF measurements) was about 15 minutes. Focal ischemia was defined as an area in which at least 2 or more adjacent ROIs had a CBF below 18 mL/100mg/min. The CBF measurement was ascertained by obtaining the mean value of all the ROIs included in the ischemic vascular territory. The CBF measurement was also ascertained in the homologous contralateral area for comparison. If the radiographic investigations (XeCT, CT, and angiography) did not show any abnormalities, the vascular territory chosen for CBF analysis was established on the basis of the neurological findings. A total of 1878 ROIs were used.
Angiography Digital subtraction angiography was performed directly after the XeCT scan in 33 patients. The angiographic investigation was restricted to the arterial territory responsible for the patient's symptoms. The collateral blood supply through the external carotid artery, cortical anastomosis, and the circle of Willis was evaluated.
Statistical Analysis The CBF change between the 2 vascular territories evaluated was compared by using the two-tailed t-test. The clinical outcome, evaluated by using NIHSS scores, was compared between patients with normal and abnormal XeCT studies by using the analysis of variance (ANOVA).
Results Clinical Findings There were 26 men and 30 women varying in age between 30 to 92 years (mean age, 69 + 15). Thirty-six patients had an ischemic stroke on the left side and 20 on the right. On admission, the mean NIHSS score was 12 5. Seventeen patients underwent thrombolytic therapy, 7 were given neuroprotective drugs, and the remainder were treated with anticoagulants. Eight patients died (14%), and the mean NIHSS score at discharge for the 48 survivors (86%) was 9 --- 6.
CT Scan Findings The mean interval between the onset of symptoms and performance of the first CT scan was 231 -+ 98 minutes.
406
G. RUBIN ET AL.
The CT scan on admission, performed in all patients, was abnormal in 28 patients (50%). Signs of infarction were found in the cortex in 17 patients (61%) and in the basal ganglia in 19 (68%). In 11 patients (39%), both of these structures were involved. Mass effect on the ventricles was found in 7 patients (25%), and arterial hyperdensity was shown in 13 (46%). The follow-up CT scan was repeated between 1 and 11 days (mean 2 • 2 days) from the first, depending on the clinical condition of the patient. Forty-one patients were evaluated with a second CT scan. Of these, 4 had a normal study. Among the patients with an abnormal CT scan, 36 (88%) had a sign of infarction in the cortex and 34 (83%) in the basal ganglia; both of these locations were injured in 30 patients (73%). Mass effect on the ventricles was found in 23 patients (56%). Of these, which 5 had a hemorrhagic transformation of the infarction after thrombolysis. Arterial hyperdensity was shown in 14 patients (34%).
Angiographic Findings Angiography, performed in 33 patients (59%), showed an arterial occlusion in 29 patients (30 vessels involved) and was normal in 4. Occlusion of the middle cerebral artery (MCA) was found in 14 patients (47%), 12 in the main trunk and 2 in a peripheral branch. The cervical internal carotid artery (ICA) was occluded in 12 patients (40%), the anterior carotid artery (ACA) in 3 (10%), and the posterior carotid artery (PCA) in 1 (3%). Of the 12 patients with ICA thromboses, 11 maintained perfusion of the ipsilateral ACA via anterior communicating artery (short clot in the ICA and MCA) or from collateral flow through the anterior communicating artery.
XeCT CBF Findings Table 1 provides a summary of the XeCT scan results. The CBF study was normal in 9 patients (16%) and showed evidence of a perfusion defect in 47 (84%). In the group of patients with normal XeCT studies, the mean CBF in the symptomatic vascular territory (38 • 11 mL/100g/min) was comparable with the mean
CBF detected in the asymptomatic side (35 • 9 m L / 1 0 0 g / rain; P = .50). In the group of patients with abnormal XeCT studies (Fig 1), CBF in the ischemic vascular territory was significantly lower (16 • 8 mL/100g/min) than in the homologous contralateral territory (35 __-13 mL/100g/rain; P < .001). Considering only the symptomatic side in patients with normal and abnormal XeCT studies, the mean CBF was significantly higher in the former (38 • 11 mL/100g/min) than in the latter (16 • 8 mL/100g/min; P < .001). The ischemic vascular territories in patients with abnormal XeCT studies were the MCA in 41 patients (87%), the ACA in 2 (4%), the ICA in 3 (6%), and the PCA in 1 (2%). There were 9 patients with normal CBF studies, which were presumably caused by a spontaneous reperfusion of the ischemic area, because these patients had a normal angiogram or follow-up CT scan and greatly improved during their hospital stay (mean NIHSS score at discharge, 2 • 3). Four of these patients had a complete resolution of symptoms within 24 hours. A m o n g the 47 patients with perfusion defects seen on XeCT scans, the CBF maps were compared with the CT a n d / o r angiographic results to confirm the occurrence and location of the cerebral ischemia. XeCT was able to show ischemia in the appropriate vascular territory in all 47 patients studied.
Relationship Between Clinical Findings and CBF There was a clear correlation between the outcome and the XeCT finding in the symptomatic side. A m o n g the 47 patients with an abnormally perfused CBF study, 8 died, and the survivors had a mean NIHSS score of 10 • 6 when discharged. A significantly lower mean NIHSS score (2 • 3, P < .001) was found in patients with a normal XeCT study on admission. The mean CBF on admission of patients with an NIHSS score of 5 or less at discharge (28 • 13 m L / 1 0 0 g / m i n ) w a s higher than patients with a NIHSS score greater than 5 (14 • 8 mL/100g/min). This difference was statistically significant (P < .001). Furthermore, the 8 patients who died during their hospital stay
Table 1. XeCT cerebral blood flow results of 56 patients with acute &chemic stroke
Normal XeCT Abnormal XeCT Vascular region involved MCA ICA ACA PCA
No.
%
Mean CBF symptomatic side
Mean CBF asymptornatic side
P*
9 47
16 84
38 • 11 16 • 8
35 _+ 9 35 -_+ 13
.50 (NS) <.001 (S)
41 3 2 1
87 6 4 2
16 Z 8 10 • 4 11 -_- 11 23
35 • 14 39 • 6 32 • 1 32
<.001 (S) .0027 (S)
Abbreviations: XeCT, Xenon-enhanced computed tomography; CBF, cerebral blood flow; NS, not significant; S, significant; MCA, middle cerebral artery; ICA, internal carotid artery; ACA, anterior carotid artery; PCA, posterior carotid artery. *By two-tailed t-test.
407
CBF IN ACUTE STROKE
Figure 1. CT and XeCT studies pe@n'med o~ admission in a patie*# with a~: acll~e ~;cclusio~ c)f the hlter~ol cer:,ieal c:aroHdaltel y. The CT -ccTItshows o;lly ~z ,did hy~;odensity in the white matter of the l
had significantlylower CBF (mean CBF 12 _+9 mL/] 00g/mhl) compared with the patients who survived (mean CBE 20 -+ 12 m L / l O O g / m i n ; P - .038).
Clinical Results of Patients Treated with Thrombolytic Therapy Among the 17 patients treated with reperfusion therapy, 12 had intraarterial thrombolysis and 5 received thrombolysis intravenously. Their mean NIHSS score on admission was 14 _+ 4. The pretreatrnent CT scan showed early signs of infarction in 11 patients and a hyperdensity of the occluded artery in 8 patients. The admission XeCT study showed an ischemic area in all 17 patients. The mean CBF in the symptomatic vascular region was 16 _+ 8 and in the contralateral asymptomatic territory was 39 -+ 15. This difference was statistically significant (P < .001). Fourteen patients (82%) had a partial or complete recanalization of the occluded artery, and in 3 patients, the reperfusion therapy was unsuccessful. A follow-up CT scan has shown a cerebral infarction in all 17 patients. Further-
more, 7 patients had mass effect on the ventricles, 6 had a residual hyperdensity of the thrombosed artery, and 5 had hem orrh a gic complications. Surgical resection of a hemorrhagic infarction was performed in 4 patients, and 1 patient u n d e r w e n t removal of a swollen, n o n h e m morhagic, infarcted area for decompression. Two patients died, and the mean NIHSS score of the 15 survivors was 9 • 6 at discharge.
Relationship Between CBF and Early CT Changes The CBF data of patients with abnormal XeCT studies were correlated with the CT scans performed on admission (Table 2, Fig 2). The mean CBF of patients with normal and abnormal CT scans were comparable (mean CBF 17 m L / 1 0 0 g / m i n v 15 m L / l O O g / m i n ; P = .36). In the patients with signs of infarction in the basal ganglia, mass effect on the ventricles, and arterial hyperdensity, the mean CBF in the injured vascular territory was significantly lower compared with patients who did not have these findings. On the other hand, the presence of a
408
G. R U B I N ET AL.
Table 2. Mean CBF in patients with abnormal X e C T findings with and without infarction signs in the early CT scan CT
Mean CBF
Symptomatic side
+ Abnormal CT Cortical sign Basal ganglia sign Ventricular mass effect Arterial hyperdensity
15 15 12 9 12
-
• 7 -+_8 • 5 • 5 • 4
Mean CBF
p*
17 • 16 • 18 • 17 • 17 •
9 8 9 8 9
.36 (NS) .78 (NS) .0044 (S) .0044 (S) .018 (S)
Asymptomatic side
+ 36 40 32 3l 3l
• • • z •
14 16 6 7 7
34 33 37 36 37
p*
__+13 • 11 ___16 • 14 • 15
0.58 (NS) 0.1 (NS) 0.19 (NS) 0.2 (NS) 0.087 (NS)
NOTE. + indicates with evidence of infarction; - indicates without evidence of infarction. Abbreviations: XeCT, Xenon-enhanced computed tomography; CT, computed tomography; CBF, cerebral blood flow; NS, not significant; S, significant. *By two-tailed t test. cortical finding on the CT scan was not associated with a significant CBF change. The CBF values in the asymptomatic control area were without any statistical relevance among the different groups.
Relationship Between Angiographic Findings and CBF The relationship between the vascular territory involved in the CBF map and the site of arterial occlusion (and the type of collateral blood supply) found on the angiogram was examined. In particular, if a CBF perfusion defect was found in the MCA region in a patient with an angiographicly determined occlusion of the ICA and MCA b u t with normal blood flow in the ipsilateral ACA, this result was considered concordant. The XeCT ischemic defect on the CBF maps consistently corresponded to the angiographic studies. The 4 patients with normal angiograms did not show any abnormality on XeCT studies. Interestingly, the 11 patients who had an angiographic ICA occlusion, but with blood flow preserved in the ACA, had a perfusion defect limited to the MCA territory on the CBF map. The mean CBF in the symptomatic vascular
territory in patients who had an angiographicly documented arterial occlusion (16 • 8 m L / 1 0 0 g / m i n ) was lower than in patients who had a n o r m a l Study (41 - 13 m L / 1 0 0 g / m i n ; P = .033). This difference was not f o u n d in the contralaterai asymptomatic side (mean CBF 36 + 14 v 35 • 9 m L / 1 0 0 g / m i n ; P = .95).
Discussion This study confirmed that XeCT study is a reliable method to detect cerebral ischemia as early as the first few hours after the onset of symptoms. A perfusion defect on the CBF map, found in 84% of patients, was always concordant with the vascular territory identified on angiograms a n d / o r late CT scans. A normal XeCT study in the remaining patients was presumably related to a spontaneous recanalization of the occluded artery (or to the development of a collateral circulation), which predicted the clinical recovery. In fact, in this subgroup of 9 patients, a normal angiogram or follow-up CT scan was found, and all of them had a good outcome. O n the other hand, lower
o
ll present absent
L~
abnormal CT
cortical sign
basal ganglia venrticular sign mass effect
arterial hyperdensity
Figure2. Mean CBFs in the symptomatic vascular territory in patients with and without abnormalities seen on early CT scans.
CBF IN ACUTE STROKE
CBF values in the abnormally perfused vascular regions were related to higher NIHSS scores and to death. The ability to analyze CBF in the early stages of an ischemic stroke may be important in selecting the patients who should be treated with reperfusion therapy. Our data indicate that patients with normal CBF in the affected vascular territory will show a resolution or significant improvement of the neurological deficit in the next I to 2 days. This is caused by a spontaneous lysis of the offending thrombus or by the development of a compensatory collateral circulation. Thrombolytic therapy is likely to be unnecessary in this subgroup of stroke patients. This study confirmed that the diagnosis of acute cerebral ischemia based only on a CT scan is difficult and that only an abnormal CT scan has important clinical implications, 6,11In our series, a normal CT scan (50%), performed on admission, did not exclude a severe stroke and was associated with a reduced CBF. Most of these patients (above 80%) showed the occlusion of an important cerebral artery on an angiogram and a cerebral infarction on the follow-up CT scan. Furthermore, a third of them had significant neurological deficits (NIHSS scores above 10) at discharge or died. Early changes in the basal ganglia (contrary to the cortex) seen on a CT scan was associated, as in a previous study, s with lower CBF values in the ischemic tissue, which are related to ICA a n d / o r proximal middle cerebral artery occlusion. The deep brain is vascularized by lenticulostriate arteries, which lack a collateral circulation. Therefore, very low levels of CBF are rapidly found after proximal occlusion of the MCA. 12 This fact and the high metabolism of the basal ganglia 12 are responsible for the early damage detected by the CT scan. The relationship between early basal ganglia changes seen on the CT scan and lower levels of CBF is supported by the fact that they were both associated with hemorrhagic complications after thrombolytic therapy. 13-15 It is also not surprising that an early pattern of mass effect on the ventricles, which is an indication of deep a n d / o r severe infarction, seen on a CT scan was associated with lower levels of CBF. The mass effect, through compression of small vessels, could contribute to further decrease the blood flow, aggravating the initial ischemic insult. The vulnerability of the basal ganglia to infarction could also be explained by the toxic effect of dopamine released in massive quantities after the ischemic insult in the extracellular tissue. 16 Drugs that inhibit the release of dopamine have been shown to have a neuroprotective effect after cerebral ischernia.17 We have noted in this review that angiographic evidence of occlusion of the ICA is usually associated (92% of patients) with a XeCT, scan showing a perfusion defect limited to the MCA territory because of the preserved flow in the ipsilateral ACA. These data, not previously reported in the literature, have important clinical implica-
409
tions, because ICA thrombosis has a worse prognosis compared with other anterior circulation arteries ls,19 and is difficult to recanalize with thrombolysis. 2~ The high frequency of collateral blood supply after occlusion of the ICA may be explained by the development of arterial anastomosis within the cerebral circulation when the carotid is only stenotic. Many studies have examined the CBF in cerebral ischemia, 3,8,14,21-36but only a few have investigated CBF within the first hours 3,8,14"22,26"33when treatment decisions must be made. In the subacute phase, transient hyperperfusion of the injured area is not rare and can complicate the interpretation of the CBF map. 27,33,36,37In this series, we did not note any hyperemia (CBF > 70 mL/100mg/min) in the symptomatic hemisphere, even in patients with spontaneous recanalization, presumably because of the acute nature of the CBF studies in this report, in a previous study, however, we have observed luxury perfusion of the ischemic tissue quite frequently (50%) in patients submitted to thrombolytic therapy. 3a Almost all these studies have used single-photon emission computed tomography (SPECT) to evaluate the CBF in patients with ischemic stroke. SPECT, however, has several limitations. It is less sensitive in evaluating white matter ischemia, 24 has poor resolution, and does not provide anatomical reference. Most importantly, SPECT gives only semiquantitative CBF values that are evaluated by comparing the ischemic area with another region of the contralateral hemisphere, assuming that it has normal CBF (about 50 mL/100g/min). In this study, we have confirmed our previous finding s that the CBF in the unaffected hemisphere is reduced by 30% and is highly variable (see Tables 1 and 2). This could be caused by transhemispheric diaschisis, 39although other factors, such as preexisting cerebral arteriosclerosis, increased intracranial pressure, altered autoregulation, and coma, have been suggested to explain the CBF depression in the asymptomatic hemisphere.S~176 Diffusion-weighted and perfusion-weighted magnetic resonance imaging (MRI) are highly sensitive techniques to detect acute cerebral ischemia and are able to identify lesions as small as 4 millimeters. 4244 The imaging of the ischemic area, related to the reduction of the diffusion of water in the brain tissue, is produced by the effect of hypoperfusion but does not give any quantitative information about the CBE 42Therefore, clinically applicable data, such as the severity of the CBF impairment, cannot be assessed. Other limitations of MRI technology are that it is expensive, not practical and must be performed only after the exclusion of an intracranial hemorrhage by CT scanning. The possibility of knowing the size, location, and severity of the ischemic lesion could greatly advance the diagnosis and treatment of stroke. Stroke management could be based on blood flow imaging if standardized
410 CBF data from a multicenter prospective trial could be proven to be effective. Possible clinical applications of CBF measurements are the selection of patients who will benefit from aggressive therapy, 14 prediction of the outcome, 3,22,28 and estimation of the treatment effects.25,45 XeCT is, in our experience, a very practical method to detect cerebral ischemia. It is rapid, inexpensive, and can be a d d e d to a conventional CT scan. XeCT has good anatomical resolution and can evaluate all of the quantitative values of the CBF range in both superficial and deep regions of the brain. We must emphasize that the interpretation of the XeCT image it is not always easy. Investigators m u s t realize that they are assessing circulating fluids and not static, solid tissue. The perfusion defect m a y have a very irregular shape, and normal or even high flow can be found within the ischemic lesion, making its evaluation difficult. Artifacts caused b y head motion m a y alter the CBF values, therefore, a confidence m a p (available from the XeCT software) must always be verified at the same time. Furthermore, a misinterpretation of the XeCT image is a possibility in patients with previous or bilateral infarcts and with severe global hypoperfusion.
Acknowledgment: The authors wish to thank John May for his management of the XeCT computer database and film library.
References 1. Adams HP, Brott TG, Furlan AJ, et al. Guidelines for thrombolytic therapy for acute stroke: A supplement to the guidelines for the management of patients with acute ischemic stroke. Circulation 1996;94:1167-1174. 2. Bryan RN, Levy LM, Whitlow WD, et al. Diagnosis of acute cerebral infarction: Comparison of CT and MR imaging. AJNR 1991;12:611-620. 3. Giubilei F, Lenzi GL, Di Piero V, et al. Predictive value of brain perfusion single-photon emission computed tomography in acute ischemic stroke. Stroke 1990;21:895-900. 4. Horowitz SH, Zito JL, Donnarumma R, et al. Computed tomography-angiographic findings the first five hours of cerebral infarction. Stroke 1991;22:1245-1253. 5. Von Kummer R, Meyding-Lamade U, Forstin M, et al. Sensitivity and prognostic value of early CT in occlusion of the middle cerebral artery trunk. AJNR 1994;15:9-15. 6. Hacke W, Kaste M, Fieschi C, et al, for the ECASS Study Group: Intravenous thrombolysis with recombinant tissue plasminogen activator for acute hemispheric stroke: The European Cooperative Acute Stroke Study (ECASS). JAMA 1995;274:1017-1025. 7. The National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group. Tissue Plasminogen activator for acute ischemic stroke. N Engl J Med 1995;333: 1581-1587. 8. Firlik AD, Kaufmann AM, Wechsler LR, et al. Quantitative cerebral blood flow determinations in acute ischemic stroke: Relationship to computed tomography and angiography. Stroke 1997;28:2208-2213.
G. RUBIN ET AL.
9. Brott TG, Adams HP, Olinger CP, et aL Measurement of acute cerebral infarction: A clinical examination scale. Stroke 1989;20:864-870. 10. Johnson DW, Stringer WA, Marks MP, et al. Stable XeCT cerebral blood flow imaging: Rationale for and role in clinical decision making. AJNR 1991;12:201-213. 11. The NINDS t-PA Stroke Study Group: Intracerebral hemorrhage after intravenous t-PA therapy for ischemic stroke. Stroke 1997;28:2109-2118. 12. Pappata S, Fiorelli M, Rommel T, et al. PET study of changes in local brain hemodynamics and oxygen metabolism after unilateral middle cerebral artery occlusion in baboons. J Cereb Blood Flow Metab 1993;13:416-424. 13. Levy DE, Brott TG, Haley EC, et al. Factors related to intracranial hematoma formation in patients receiving tissue-type plasminogen activator for acute ischemic stroke. Stroke 1994;25:291-297. 14. Ueda T, Hatakeyama T, Kumon Y, et al. Evaluation of risk of hemorrhagic transformation in local intra-arterial thrombolysis in acute ischemic stroke by initial SPECT. Stroke 1994;25:298-303. 15. Yokogami K, Nakano S, Ohta H, et al. Prediction of hemorrhagic complication after thrombolytic therapy for middle cerebral artery occlusion: Value of pre- and post-therapeutic computed tomographic findings and angiographic occlusive site. Neurosurgery 1996;39:11021107. 16. Weinberger J, Thompson L, Samii M. The significance of basal ganglia infarction. J Stroke Cerebrovasc Dis 1995;5: 6-11. 17. Bhardwaj A, Brannan T, Weinberger J. Pentobarbital inhibits extracellular release of dopamine in the ischemic striatum. J Neural Transm 1990;82:111-117. 18. Milandre L, Pestre P, Botti G, et al. Les occlusions carotidielmes revelees par un accident ischemique cerebral. Aspects lesionnels, etiologiques et evolutifs. Ann Med Interne 1990;141:115-122. 19. Toni D, Fiorelli M, Gentile M, et al. Progressing neurological deficit secondary to acute ischemic stroke. Arch Neuro11995;52:670-675. 20. Jansen O, Von Kummer R, Forsting M, et al. Thrombolytic therapy in acute occlusion of the intracranial internal carotid artery bifurcation. AJNR 1995;16:1977-1986. 21. Alexandrov AV, Black SE, Ehrlich LE, et al. Simple visual analysis of brain perfusion on HMPAO SPECT predicts early outcome in acute stroke. Stroke 1996;27:1537-1542. 22. Alexandrov AV, Bladin CF, Elrich LE, et al. Noninvasive assessment of intracranial perfusion in acute cerebral ischemia. J Neuroimaging 1995;5:76-82. 23. Alexandrov AV, Elrich LE, Bladin CF, et al. Cerebral perfusion index. A new marker for clinical outcome in acute stroke. J Neuroimaging 1993;3:209-215. 24. Baird AE, Austin MC, McKay WJ, et al. Sensitivity and specificity of 99mTc-HMPAO SPECT cerebral perfusion measurements during the first 48 hours for the localization of cerebral infarction. Stroke 1997;28:976-980. 25. Baird AE, Donnan GA, Austin MC, et al. Reperfusion after thrombolytic therapy in ischemic stroke measured by single-photon emission computed tomography. Stroke 1994;25:1439-1446. 26. Bowler JV, Wade JPH, Jones BE, et al. Single-photon emission computed tomography using hexamethylpropyleneamhxe oxime in the prognosis of acute cerebral infarction. Stroke 1996;27:82-86. 27. Bushnell DL, Gupta S, Mlcoch AG, et al. Demonstration of focal hyperemia in acute cerebral infarction with
CBF IN ACUTE STROKE
28. 29.
30. 31. 32.
33.
34.
35.
iodine 123 iodoamphetamine. J Nucl Med 1987;28:19201923. Hanson SK, Grotta JC, Rhoades H, et al. Value of single-photon emission-computed tomography in acute stroke therapeutic trials. Stroke 1993;24:1322-1329. Laloux P, Richelle F, Jamart J, et al. Comparative correlation of HMPAO SPECT indices, neurological score, and stroke subtypes with clinical outcome in acute carotid infarcts. Stroke 1995;26:816-821. Limburg M, van Royen EA, Hijdra A, et al. rCBF-SPECT in brain infarction: When does it predict outcome. ] Nucl Med 1991;32:382-387. Limburg M, van Royen EA, Hijdra A, et al. Single-photon emission computed tomography and early death in acute ischemic stroke. Stroke 1990;21:1150-1155. Marchal G, Serrati C, Rioux P, et al. PET imaging of cerebral perfusion and oxygen consumption in acute ischaemic stroke: Relation to outcome. Lancet 1993;341: 925-927. Moretti JL, Defer G, Cinotti L, et al. Luxury perfusion with 99mTc-HMPAO and 123I-IMP SPECT imaging during the subacute phase of stroke. Eur J Nucl Med 1990;16:17-22. Mountz JM, Modell JG, Foster NL, et al. Prognostication of recovery following stroke using the comparison of CT and technetium-99m HM-PAO SPECT. J Nucl Med 1990; 31:61-66. Shimosegawa E, Hatazawa J, Atsusi I, et al. Cerebral infarction within six hours of onset: Prediction of completed infarction with technetium-99m-HMPAO SPECT. J Nucl Med 1994;35:1097-1103.
411
36. Strashun A, Dunn EK, Sarkar SS, et al. Reversible increased technetium-99-HMPAO cerebral cortical activity: A scintigraphic reflection of luxuriant hyperperfusion. J Nucl Med 1992;33:117-119. 37. Olsen TS, Larsen B, Skriver EB, et al. Focal cerebral hyperemia in acute stroke; incidence pathophysiology, and clinical significance. Stroke 1981;12:598-607. 38. Rubin G, Firlik AD, Pindzola RR, et al: The effect of reperfusion therapy on cerebral blood flow in acute stroke. J Stroke Cerebrovasc Dis 1999;8:9-16. 39. Slater R, Reivich M, Goldberg t-I, et al. Diaschisis with cerebral infarction. Stroke 1977;8:684-689. 40. Feeney DM, Baron JC: Diaschisis. Stroke 1986;17:817-830. 41. Lavy S, Melamed E, Portnoy Z. The effects of cerebral infarction on the regional cerebral blood flow of the contralateral hemisphere. Stroke 1975;6:160-163. 42. Sorensen AG, Buonanno FS, Gonzalez RG, et al. Hyperacute stroke: Evaluation with combined multisection diffusion-weighted and hemodynamically weighted echoplanar MR imaging. Radiology 1996;199:391-401. 43. Warach S, Dashe JF, Edelman RR: Clinical outcome in ischemic stroke predicted by early diffusion-weighted and perfusion magnetic resonance imaging: a preliminary analysis. J Cerelo Blood Flow Metab 1996;16:53-59. 44. Warach S, Gaa J, Siewert B, et al: Acute human stroke studied by whole brain echo planar diffusion-weighted magnetic resonance imaging. Ann Neurol 1995;37:231241. 45. Overgaard K, Sperling B, Boysen G, et al. Thrombolytic therapy in acute ischemic stroke: A Danish pilot study. Stroke 1993;24:1439-1446.
Objective: Ischemic stroke must be diagnosed promptly if patients are to be treated with thrombolytic therapy. The diagnosis of acute cerebral ischemia, however, is usually based on clinical and computed tomography (CT) scan findings. CT scans are often normal in the first few hours after stroke. The purpose of this study was to determine whether Xenon-enhanced CT (XeCT) cerebral blood flow (CBF) studies could increase the sensitivity of stroke detection in the acute stage. Methods: CBF studies performed within 8 hours of symptom onset were evaluated in 56 patients who presented with hemispheric stroke symptoms. Mean CBF in the symptomatic vascular territory was calculated and compared with the corresponding contralateral area. CBF values below 18 mL/100g/min on 2 adjacent regions of interest were considered ischemic lesions. CT scans and angiograms were compared with the XeCT findings. Neurological condition on admission and discharge was evaluated by using National Institutes of Health Stroke Scale (NII-ISS) scores. Results: The mean NIHSS score on admission was 12 + 5. Early CT scans were abnormal in 28 (50%) patients. There were 9 (16%) patients who had normal XeCT scans because of spontaneous reperfusion of the ischemic area. XeCT studies showed an ischemic lesion in 47 (84%) patients. In these patients, the mean CBF in the affected vascular territory was 16 -+ 8 mL/100g/min compared with 35 + 13 mL/100g/min in the contralateral specular territory (P < 0.001). There were no false positive or negative XeCT studies, and the location of the perfusion defect corresponded with the CT and/or angiographic findings in all cases. Eight patients died (14%), and the 48 survivors (86%) had a mean NIHSS score of 9 _+ 6 on discharge. Conclusions: CBF measurements were correlated with the CT and angiographic results and greatly assisted in the diagnosis of acute ischemic stroke. XeCT studies used for estimating the location and extent of cerebral ischemia may be important in the triage of patients for acute stroke therapy. Key Words: Cerebral blood flow--Diagnosis--Stroke--Xenon-enhanced computed tomography.
Cerebral blood flow (CBF) is rarely studied in acute stroke victims even though its reduction is the basic pathophysiological process of cerebral ischemia. Cur-
From the University of Pittsburgh Medical Center, Department of Neurosurgery,Pittsburgh, PA. Received January 26,1999;accepted June 2,1999. Dr. Yonasreceivesa research grant from Praxair, Inc. Address reprint requests to Howard Yonas, MD, Department of Neurological Surgery, University of Pittsburgh Medical Center, 200 Lothrop Street, Suite B-400,Pittsburgh, PA 15213. Copyright 9 1999by National Stroke Association 1052-3057/99/0806-000653.00/0
404
rently, the diagnosis of stroke is based on a neurological examination and on computed tomography (CT) findings. CT scans are used to rule out the causes of hemorrhagic stroke and may be useful in excluding patients who should be treated with reperfusion therapy. 1 The CT image, however, is often normal when performed in the first several hours after the ischemic attack. 2"s The European Cooperative Acute Stroke Study (ECASS) and the National Institute of Neurological Disorders and Stroke (NINDS) investigators in the recent reperfusion therapy trial 6,7 have stressed the necessity of better patient selection. Early knowledge of the location and severity of the perfusion defect may help identify stroke patients who
Journal of Stroke and Cerebrovascular Diseases, Vol. 8, No. 6 (November-December), 1999: pp 404-411
405
CBF IN ACUTE STROKE
will benefit from aggressive therapy. Xenon-enhanced CT (XeCT) CBF methodology is a practical, inexpensive, and fast technique that will provide quantitative CBF measurements. In a preliminary report, we evaluated the sensitivity of XeCT for detecting CBF reduction in the first hours after stroke. 8 XeCT proved to be a very promising methodology, but this study was restricted to 20 patients with major stroke and was limited to I territory of the cerebral Circulation. The aim of the present study is to verify the validity of XeCT in detecting acute cerebral ischemia in a larger, heterogeneous stroke population.
Subjects and M e t h o d s Patient Population From January 1, 1996 to June 31, 1998, 74 patients with ischemic stroke underwent XeCT CBF studies at the University of Pittsburgh Medical Center within 8 hours of symptom onset. From this group, 56 patients were selected and retrospectively reviewed. Ischemic stroke was diagnosed when patients presented with acute onset of neurological deficit associated with a CT scan on admission that was either normal or had ischemic changes. The inclusion criteria were (1) hemispheric stroke; (2) absence of hemorrhagic, bilateral, or previous infarction; and (3) CBF map without artifacts. On admission in the emergency room, members of the stroke team, which was composed of neurologists and neurosurgeons, examined all patients. Each patient's neurological deficit was quantified on admission and during their hospital stay by using the National Institutes of Health Stroke Scale (NIHSS). 9 CT scans and XeCT CBF studies were performed as soon as possible after the clinical evaluation. Based on their neurological condition and the results of the radiographic investigation, patients were considered for cerebral angiography and, eventually, for reperfusion therapy. The thrombolytic drug was administered intravenously if the patient was symptomatic for less than 3 hours. Intraarterial treatment was used if the stroke onset was between 3 and 6 hours. During their hospital stay, the neurological condition of every patient was determined daily by using NIHSS scores.
CT Scanning CT images were evaluated for signs of infarction, such as hypodensity or sulcal effacement, and were divided into 4 categories: (1) cortical findings; (2) basal ganglia findings; (3) mass effect on the lateral ventricles; and (4) arterial hyperdensity (as evidence of thrombosis or embolus). For patients who received multiple follow-up CT scans, the scan that most clearly delineated the infarction was used.
XeCT CBF Study The XeCT studies were performed immediately after completion of the first CT scans. A description of the XeCT methodology has been previously reported, l~ Briefly, cortical and subcortical CBF was evaluated by a computer analysis program that divides each of the 3 hemispheric brain slices into 18 to 22 regions of interest (ROIs). Each 2-cm circular ROI was composed of mixed cortical CBF values (gray and white matter). The time required to perform the XeCT study (including image acquisition and CBF measurements) was about 15 minutes. Focal ischemia was defined as an area in which at least 2 or more adjacent ROIs had a CBF below 18 mL/100mg/min. The CBF measurement was ascertained by obtaining the mean value of all the ROIs included in the ischemic vascular territory. The CBF measurement was also ascertained in the homologous contralateral area for comparison. If the radiographic investigations (XeCT, CT, and angiography) did not show any abnormalities, the vascular territory chosen for CBF analysis was established on the basis of the neurological findings. A total of 1878 ROIs were used.
Angiography Digital subtraction angiography was performed directly after the XeCT scan in 33 patients. The angiographic investigation was restricted to the arterial territory responsible for the patient's symptoms. The collateral blood supply through the external carotid artery, cortical anastomosis, and the circle of Willis was evaluated.
Statistical Analysis The CBF change between the 2 vascular territories evaluated was compared by using the two-tailed t-test. The clinical outcome, evaluated by using NIHSS scores, was compared between patients with normal and abnormal XeCT studies by using the analysis of variance (ANOVA).
Results Clinical Findings There were 26 men and 30 women varying in age between 30 to 92 years (mean age, 69 + 15). Thirty-six patients had an ischemic stroke on the left side and 20 on the right. On admission, the mean NIHSS score was 12 5. Seventeen patients underwent thrombolytic therapy, 7 were given neuroprotective drugs, and the remainder were treated with anticoagulants. Eight patients died (14%), and the mean NIHSS score at discharge for the 48 survivors (86%) was 9 --- 6.
CT Scan Findings The mean interval between the onset of symptoms and performance of the first CT scan was 231 -+ 98 minutes.
406
G. RUBIN ET AL.
The CT scan on admission, performed in all patients, was abnormal in 28 patients (50%). Signs of infarction were found in the cortex in 17 patients (61%) and in the basal ganglia in 19 (68%). In 11 patients (39%), both of these structures were involved. Mass effect on the ventricles was found in 7 patients (25%), and arterial hyperdensity was shown in 13 (46%). The follow-up CT scan was repeated between 1 and 11 days (mean 2 • 2 days) from the first, depending on the clinical condition of the patient. Forty-one patients were evaluated with a second CT scan. Of these, 4 had a normal study. Among the patients with an abnormal CT scan, 36 (88%) had a sign of infarction in the cortex and 34 (83%) in the basal ganglia; both of these locations were injured in 30 patients (73%). Mass effect on the ventricles was found in 23 patients (56%). Of these, which 5 had a hemorrhagic transformation of the infarction after thrombolysis. Arterial hyperdensity was shown in 14 patients (34%).
Angiographic Findings Angiography, performed in 33 patients (59%), showed an arterial occlusion in 29 patients (30 vessels involved) and was normal in 4. Occlusion of the middle cerebral artery (MCA) was found in 14 patients (47%), 12 in the main trunk and 2 in a peripheral branch. The cervical internal carotid artery (ICA) was occluded in 12 patients (40%), the anterior carotid artery (ACA) in 3 (10%), and the posterior carotid artery (PCA) in 1 (3%). Of the 12 patients with ICA thromboses, 11 maintained perfusion of the ipsilateral ACA via anterior communicating artery (short clot in the ICA and MCA) or from collateral flow through the anterior communicating artery.
XeCT CBF Findings Table 1 provides a summary of the XeCT scan results. The CBF study was normal in 9 patients (16%) and showed evidence of a perfusion defect in 47 (84%). In the group of patients with normal XeCT studies, the mean CBF in the symptomatic vascular territory (38 • 11 mL/100g/min) was comparable with the mean
CBF detected in the asymptomatic side (35 • 9 m L / 1 0 0 g / rain; P = .50). In the group of patients with abnormal XeCT studies (Fig 1), CBF in the ischemic vascular territory was significantly lower (16 • 8 mL/100g/min) than in the homologous contralateral territory (35 __-13 mL/100g/rain; P < .001). Considering only the symptomatic side in patients with normal and abnormal XeCT studies, the mean CBF was significantly higher in the former (38 • 11 mL/100g/min) than in the latter (16 • 8 mL/100g/min; P < .001). The ischemic vascular territories in patients with abnormal XeCT studies were the MCA in 41 patients (87%), the ACA in 2 (4%), the ICA in 3 (6%), and the PCA in 1 (2%). There were 9 patients with normal CBF studies, which were presumably caused by a spontaneous reperfusion of the ischemic area, because these patients had a normal angiogram or follow-up CT scan and greatly improved during their hospital stay (mean NIHSS score at discharge, 2 • 3). Four of these patients had a complete resolution of symptoms within 24 hours. A m o n g the 47 patients with perfusion defects seen on XeCT scans, the CBF maps were compared with the CT a n d / o r angiographic results to confirm the occurrence and location of the cerebral ischemia. XeCT was able to show ischemia in the appropriate vascular territory in all 47 patients studied.
Relationship Between Clinical Findings and CBF There was a clear correlation between the outcome and the XeCT finding in the symptomatic side. A m o n g the 47 patients with an abnormally perfused CBF study, 8 died, and the survivors had a mean NIHSS score of 10 • 6 when discharged. A significantly lower mean NIHSS score (2 • 3, P < .001) was found in patients with a normal XeCT study on admission. The mean CBF on admission of patients with an NIHSS score of 5 or less at discharge (28 • 13 m L / 1 0 0 g / m i n ) w a s higher than patients with a NIHSS score greater than 5 (14 • 8 mL/100g/min). This difference was statistically significant (P < .001). Furthermore, the 8 patients who died during their hospital stay
Table 1. XeCT cerebral blood flow results of 56 patients with acute &chemic stroke
Normal XeCT Abnormal XeCT Vascular region involved MCA ICA ACA PCA
No.
%
Mean CBF symptomatic side
Mean CBF asymptornatic side
P*
9 47
16 84
38 • 11 16 • 8
35 _+ 9 35 -_+ 13
.50 (NS) <.001 (S)
41 3 2 1
87 6 4 2
16 Z 8 10 • 4 11 -_- 11 23
35 • 14 39 • 6 32 • 1 32
<.001 (S) .0027 (S)
Abbreviations: XeCT, Xenon-enhanced computed tomography; CBF, cerebral blood flow; NS, not significant; S, significant; MCA, middle cerebral artery; ICA, internal carotid artery; ACA, anterior carotid artery; PCA, posterior carotid artery. *By two-tailed t-test.
407
CBF IN ACUTE STROKE
Figure 1. CT and XeCT studies pe@n'med o~ admission in a patie*# with a~: acll~e ~;cclusio~ c)f the hlter~ol cer:,ieal c:aroHdaltel y. The CT -ccTItshows o;lly ~z ,did hy~;odensity in the white matter of the l
had significantlylower CBF (mean CBF 12 _+9 mL/] 00g/mhl) compared with the patients who survived (mean CBE 20 -+ 12 m L / l O O g / m i n ; P - .038).
Clinical Results of Patients Treated with Thrombolytic Therapy Among the 17 patients treated with reperfusion therapy, 12 had intraarterial thrombolysis and 5 received thrombolysis intravenously. Their mean NIHSS score on admission was 14 _+ 4. The pretreatrnent CT scan showed early signs of infarction in 11 patients and a hyperdensity of the occluded artery in 8 patients. The admission XeCT study showed an ischemic area in all 17 patients. The mean CBF in the symptomatic vascular region was 16 _+ 8 and in the contralateral asymptomatic territory was 39 -+ 15. This difference was statistically significant (P < .001). Fourteen patients (82%) had a partial or complete recanalization of the occluded artery, and in 3 patients, the reperfusion therapy was unsuccessful. A follow-up CT scan has shown a cerebral infarction in all 17 patients. Further-
more, 7 patients had mass effect on the ventricles, 6 had a residual hyperdensity of the thrombosed artery, and 5 had hem orrh a gic complications. Surgical resection of a hemorrhagic infarction was performed in 4 patients, and 1 patient u n d e r w e n t removal of a swollen, n o n h e m morhagic, infarcted area for decompression. Two patients died, and the mean NIHSS score of the 15 survivors was 9 • 6 at discharge.
Relationship Between CBF and Early CT Changes The CBF data of patients with abnormal XeCT studies were correlated with the CT scans performed on admission (Table 2, Fig 2). The mean CBF of patients with normal and abnormal CT scans were comparable (mean CBF 17 m L / 1 0 0 g / m i n v 15 m L / l O O g / m i n ; P = .36). In the patients with signs of infarction in the basal ganglia, mass effect on the ventricles, and arterial hyperdensity, the mean CBF in the injured vascular territory was significantly lower compared with patients who did not have these findings. On the other hand, the presence of a
408
G. R U B I N ET AL.
Table 2. Mean CBF in patients with abnormal X e C T findings with and without infarction signs in the early CT scan CT
Mean CBF
Symptomatic side
+ Abnormal CT Cortical sign Basal ganglia sign Ventricular mass effect Arterial hyperdensity
15 15 12 9 12
-
• 7 -+_8 • 5 • 5 • 4
Mean CBF
p*
17 • 16 • 18 • 17 • 17 •
9 8 9 8 9
.36 (NS) .78 (NS) .0044 (S) .0044 (S) .018 (S)
Asymptomatic side
+ 36 40 32 3l 3l
• • • z •
14 16 6 7 7
34 33 37 36 37
p*
__+13 • 11 ___16 • 14 • 15
0.58 (NS) 0.1 (NS) 0.19 (NS) 0.2 (NS) 0.087 (NS)
NOTE. + indicates with evidence of infarction; - indicates without evidence of infarction. Abbreviations: XeCT, Xenon-enhanced computed tomography; CT, computed tomography; CBF, cerebral blood flow; NS, not significant; S, significant. *By two-tailed t test. cortical finding on the CT scan was not associated with a significant CBF change. The CBF values in the asymptomatic control area were without any statistical relevance among the different groups.
Relationship Between Angiographic Findings and CBF The relationship between the vascular territory involved in the CBF map and the site of arterial occlusion (and the type of collateral blood supply) found on the angiogram was examined. In particular, if a CBF perfusion defect was found in the MCA region in a patient with an angiographicly determined occlusion of the ICA and MCA b u t with normal blood flow in the ipsilateral ACA, this result was considered concordant. The XeCT ischemic defect on the CBF maps consistently corresponded to the angiographic studies. The 4 patients with normal angiograms did not show any abnormality on XeCT studies. Interestingly, the 11 patients who had an angiographic ICA occlusion, but with blood flow preserved in the ACA, had a perfusion defect limited to the MCA territory on the CBF map. The mean CBF in the symptomatic vascular
territory in patients who had an angiographicly documented arterial occlusion (16 • 8 m L / 1 0 0 g / m i n ) was lower than in patients who had a n o r m a l Study (41 - 13 m L / 1 0 0 g / m i n ; P = .033). This difference was not f o u n d in the contralaterai asymptomatic side (mean CBF 36 + 14 v 35 • 9 m L / 1 0 0 g / m i n ; P = .95).
Discussion This study confirmed that XeCT study is a reliable method to detect cerebral ischemia as early as the first few hours after the onset of symptoms. A perfusion defect on the CBF map, found in 84% of patients, was always concordant with the vascular territory identified on angiograms a n d / o r late CT scans. A normal XeCT study in the remaining patients was presumably related to a spontaneous recanalization of the occluded artery (or to the development of a collateral circulation), which predicted the clinical recovery. In fact, in this subgroup of 9 patients, a normal angiogram or follow-up CT scan was found, and all of them had a good outcome. O n the other hand, lower
o
ll present absent
L~
abnormal CT
cortical sign
basal ganglia venrticular sign mass effect
arterial hyperdensity
Figure2. Mean CBFs in the symptomatic vascular territory in patients with and without abnormalities seen on early CT scans.
CBF IN ACUTE STROKE
CBF values in the abnormally perfused vascular regions were related to higher NIHSS scores and to death. The ability to analyze CBF in the early stages of an ischemic stroke may be important in selecting the patients who should be treated with reperfusion therapy. Our data indicate that patients with normal CBF in the affected vascular territory will show a resolution or significant improvement of the neurological deficit in the next I to 2 days. This is caused by a spontaneous lysis of the offending thrombus or by the development of a compensatory collateral circulation. Thrombolytic therapy is likely to be unnecessary in this subgroup of stroke patients. This study confirmed that the diagnosis of acute cerebral ischemia based only on a CT scan is difficult and that only an abnormal CT scan has important clinical implications, 6,11In our series, a normal CT scan (50%), performed on admission, did not exclude a severe stroke and was associated with a reduced CBF. Most of these patients (above 80%) showed the occlusion of an important cerebral artery on an angiogram and a cerebral infarction on the follow-up CT scan. Furthermore, a third of them had significant neurological deficits (NIHSS scores above 10) at discharge or died. Early changes in the basal ganglia (contrary to the cortex) seen on a CT scan was associated, as in a previous study, s with lower CBF values in the ischemic tissue, which are related to ICA a n d / o r proximal middle cerebral artery occlusion. The deep brain is vascularized by lenticulostriate arteries, which lack a collateral circulation. Therefore, very low levels of CBF are rapidly found after proximal occlusion of the MCA. 12 This fact and the high metabolism of the basal ganglia 12 are responsible for the early damage detected by the CT scan. The relationship between early basal ganglia changes seen on the CT scan and lower levels of CBF is supported by the fact that they were both associated with hemorrhagic complications after thrombolytic therapy. 13-15 It is also not surprising that an early pattern of mass effect on the ventricles, which is an indication of deep a n d / o r severe infarction, seen on a CT scan was associated with lower levels of CBF. The mass effect, through compression of small vessels, could contribute to further decrease the blood flow, aggravating the initial ischemic insult. The vulnerability of the basal ganglia to infarction could also be explained by the toxic effect of dopamine released in massive quantities after the ischemic insult in the extracellular tissue. 16 Drugs that inhibit the release of dopamine have been shown to have a neuroprotective effect after cerebral ischernia.17 We have noted in this review that angiographic evidence of occlusion of the ICA is usually associated (92% of patients) with a XeCT, scan showing a perfusion defect limited to the MCA territory because of the preserved flow in the ipsilateral ACA. These data, not previously reported in the literature, have important clinical implica-
409
tions, because ICA thrombosis has a worse prognosis compared with other anterior circulation arteries ls,19 and is difficult to recanalize with thrombolysis. 2~ The high frequency of collateral blood supply after occlusion of the ICA may be explained by the development of arterial anastomosis within the cerebral circulation when the carotid is only stenotic. Many studies have examined the CBF in cerebral ischemia, 3,8,14,21-36but only a few have investigated CBF within the first hours 3,8,14"22,26"33when treatment decisions must be made. In the subacute phase, transient hyperperfusion of the injured area is not rare and can complicate the interpretation of the CBF map. 27,33,36,37In this series, we did not note any hyperemia (CBF > 70 mL/100mg/min) in the symptomatic hemisphere, even in patients with spontaneous recanalization, presumably because of the acute nature of the CBF studies in this report, in a previous study, however, we have observed luxury perfusion of the ischemic tissue quite frequently (50%) in patients submitted to thrombolytic therapy. 3a Almost all these studies have used single-photon emission computed tomography (SPECT) to evaluate the CBF in patients with ischemic stroke. SPECT, however, has several limitations. It is less sensitive in evaluating white matter ischemia, 24 has poor resolution, and does not provide anatomical reference. Most importantly, SPECT gives only semiquantitative CBF values that are evaluated by comparing the ischemic area with another region of the contralateral hemisphere, assuming that it has normal CBF (about 50 mL/100g/min). In this study, we have confirmed our previous finding s that the CBF in the unaffected hemisphere is reduced by 30% and is highly variable (see Tables 1 and 2). This could be caused by transhemispheric diaschisis, 39although other factors, such as preexisting cerebral arteriosclerosis, increased intracranial pressure, altered autoregulation, and coma, have been suggested to explain the CBF depression in the asymptomatic hemisphere.S~176 Diffusion-weighted and perfusion-weighted magnetic resonance imaging (MRI) are highly sensitive techniques to detect acute cerebral ischemia and are able to identify lesions as small as 4 millimeters. 4244 The imaging of the ischemic area, related to the reduction of the diffusion of water in the brain tissue, is produced by the effect of hypoperfusion but does not give any quantitative information about the CBE 42Therefore, clinically applicable data, such as the severity of the CBF impairment, cannot be assessed. Other limitations of MRI technology are that it is expensive, not practical and must be performed only after the exclusion of an intracranial hemorrhage by CT scanning. The possibility of knowing the size, location, and severity of the ischemic lesion could greatly advance the diagnosis and treatment of stroke. Stroke management could be based on blood flow imaging if standardized
410 CBF data from a multicenter prospective trial could be proven to be effective. Possible clinical applications of CBF measurements are the selection of patients who will benefit from aggressive therapy, 14 prediction of the outcome, 3,22,28 and estimation of the treatment effects.25,45 XeCT is, in our experience, a very practical method to detect cerebral ischemia. It is rapid, inexpensive, and can be a d d e d to a conventional CT scan. XeCT has good anatomical resolution and can evaluate all of the quantitative values of the CBF range in both superficial and deep regions of the brain. We must emphasize that the interpretation of the XeCT image it is not always easy. Investigators m u s t realize that they are assessing circulating fluids and not static, solid tissue. The perfusion defect m a y have a very irregular shape, and normal or even high flow can be found within the ischemic lesion, making its evaluation difficult. Artifacts caused b y head motion m a y alter the CBF values, therefore, a confidence m a p (available from the XeCT software) must always be verified at the same time. Furthermore, a misinterpretation of the XeCT image is a possibility in patients with previous or bilateral infarcts and with severe global hypoperfusion.
Acknowledgment: The authors wish to thank John May for his management of the XeCT computer database and film library.
References 1. Adams HP, Brott TG, Furlan AJ, et al. Guidelines for thrombolytic therapy for acute stroke: A supplement to the guidelines for the management of patients with acute ischemic stroke. Circulation 1996;94:1167-1174. 2. Bryan RN, Levy LM, Whitlow WD, et al. Diagnosis of acute cerebral infarction: Comparison of CT and MR imaging. AJNR 1991;12:611-620. 3. Giubilei F, Lenzi GL, Di Piero V, et al. Predictive value of brain perfusion single-photon emission computed tomography in acute ischemic stroke. Stroke 1990;21:895-900. 4. Horowitz SH, Zito JL, Donnarumma R, et al. Computed tomography-angiographic findings the first five hours of cerebral infarction. Stroke 1991;22:1245-1253. 5. Von Kummer R, Meyding-Lamade U, Forstin M, et al. Sensitivity and prognostic value of early CT in occlusion of the middle cerebral artery trunk. AJNR 1994;15:9-15. 6. Hacke W, Kaste M, Fieschi C, et al, for the ECASS Study Group: Intravenous thrombolysis with recombinant tissue plasminogen activator for acute hemispheric stroke: The European Cooperative Acute Stroke Study (ECASS). JAMA 1995;274:1017-1025. 7. The National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group. Tissue Plasminogen activator for acute ischemic stroke. N Engl J Med 1995;333: 1581-1587. 8. Firlik AD, Kaufmann AM, Wechsler LR, et al. Quantitative cerebral blood flow determinations in acute ischemic stroke: Relationship to computed tomography and angiography. Stroke 1997;28:2208-2213.
G. RUBIN ET AL.
9. Brott TG, Adams HP, Olinger CP, et aL Measurement of acute cerebral infarction: A clinical examination scale. Stroke 1989;20:864-870. 10. Johnson DW, Stringer WA, Marks MP, et al. Stable XeCT cerebral blood flow imaging: Rationale for and role in clinical decision making. AJNR 1991;12:201-213. 11. The NINDS t-PA Stroke Study Group: Intracerebral hemorrhage after intravenous t-PA therapy for ischemic stroke. Stroke 1997;28:2109-2118. 12. Pappata S, Fiorelli M, Rommel T, et al. PET study of changes in local brain hemodynamics and oxygen metabolism after unilateral middle cerebral artery occlusion in baboons. J Cereb Blood Flow Metab 1993;13:416-424. 13. Levy DE, Brott TG, Haley EC, et al. Factors related to intracranial hematoma formation in patients receiving tissue-type plasminogen activator for acute ischemic stroke. Stroke 1994;25:291-297. 14. Ueda T, Hatakeyama T, Kumon Y, et al. Evaluation of risk of hemorrhagic transformation in local intra-arterial thrombolysis in acute ischemic stroke by initial SPECT. Stroke 1994;25:298-303. 15. Yokogami K, Nakano S, Ohta H, et al. Prediction of hemorrhagic complication after thrombolytic therapy for middle cerebral artery occlusion: Value of pre- and post-therapeutic computed tomographic findings and angiographic occlusive site. Neurosurgery 1996;39:11021107. 16. Weinberger J, Thompson L, Samii M. The significance of basal ganglia infarction. J Stroke Cerebrovasc Dis 1995;5: 6-11. 17. Bhardwaj A, Brannan T, Weinberger J. Pentobarbital inhibits extracellular release of dopamine in the ischemic striatum. J Neural Transm 1990;82:111-117. 18. Milandre L, Pestre P, Botti G, et al. Les occlusions carotidielmes revelees par un accident ischemique cerebral. Aspects lesionnels, etiologiques et evolutifs. Ann Med Interne 1990;141:115-122. 19. Toni D, Fiorelli M, Gentile M, et al. Progressing neurological deficit secondary to acute ischemic stroke. Arch Neuro11995;52:670-675. 20. Jansen O, Von Kummer R, Forsting M, et al. Thrombolytic therapy in acute occlusion of the intracranial internal carotid artery bifurcation. AJNR 1995;16:1977-1986. 21. Alexandrov AV, Black SE, Ehrlich LE, et al. Simple visual analysis of brain perfusion on HMPAO SPECT predicts early outcome in acute stroke. Stroke 1996;27:1537-1542. 22. Alexandrov AV, Bladin CF, Elrich LE, et al. Noninvasive assessment of intracranial perfusion in acute cerebral ischemia. J Neuroimaging 1995;5:76-82. 23. Alexandrov AV, Elrich LE, Bladin CF, et al. Cerebral perfusion index. A new marker for clinical outcome in acute stroke. J Neuroimaging 1993;3:209-215. 24. Baird AE, Austin MC, McKay WJ, et al. Sensitivity and specificity of 99mTc-HMPAO SPECT cerebral perfusion measurements during the first 48 hours for the localization of cerebral infarction. Stroke 1997;28:976-980. 25. Baird AE, Donnan GA, Austin MC, et al. Reperfusion after thrombolytic therapy in ischemic stroke measured by single-photon emission computed tomography. Stroke 1994;25:1439-1446. 26. Bowler JV, Wade JPH, Jones BE, et al. Single-photon emission computed tomography using hexamethylpropyleneamhxe oxime in the prognosis of acute cerebral infarction. Stroke 1996;27:82-86. 27. Bushnell DL, Gupta S, Mlcoch AG, et al. Demonstration of focal hyperemia in acute cerebral infarction with
CBF IN ACUTE STROKE
28. 29.
30. 31. 32.
33.
34.
35.
iodine 123 iodoamphetamine. J Nucl Med 1987;28:19201923. Hanson SK, Grotta JC, Rhoades H, et al. Value of single-photon emission-computed tomography in acute stroke therapeutic trials. Stroke 1993;24:1322-1329. Laloux P, Richelle F, Jamart J, et al. Comparative correlation of HMPAO SPECT indices, neurological score, and stroke subtypes with clinical outcome in acute carotid infarcts. Stroke 1995;26:816-821. Limburg M, van Royen EA, Hijdra A, et al. rCBF-SPECT in brain infarction: When does it predict outcome. ] Nucl Med 1991;32:382-387. Limburg M, van Royen EA, Hijdra A, et al. Single-photon emission computed tomography and early death in acute ischemic stroke. Stroke 1990;21:1150-1155. Marchal G, Serrati C, Rioux P, et al. PET imaging of cerebral perfusion and oxygen consumption in acute ischaemic stroke: Relation to outcome. Lancet 1993;341: 925-927. Moretti JL, Defer G, Cinotti L, et al. Luxury perfusion with 99mTc-HMPAO and 123I-IMP SPECT imaging during the subacute phase of stroke. Eur J Nucl Med 1990;16:17-22. Mountz JM, Modell JG, Foster NL, et al. Prognostication of recovery following stroke using the comparison of CT and technetium-99m HM-PAO SPECT. J Nucl Med 1990; 31:61-66. Shimosegawa E, Hatazawa J, Atsusi I, et al. Cerebral infarction within six hours of onset: Prediction of completed infarction with technetium-99m-HMPAO SPECT. J Nucl Med 1994;35:1097-1103.
411
36. Strashun A, Dunn EK, Sarkar SS, et al. Reversible increased technetium-99-HMPAO cerebral cortical activity: A scintigraphic reflection of luxuriant hyperperfusion. J Nucl Med 1992;33:117-119. 37. Olsen TS, Larsen B, Skriver EB, et al. Focal cerebral hyperemia in acute stroke; incidence pathophysiology, and clinical significance. Stroke 1981;12:598-607. 38. Rubin G, Firlik AD, Pindzola RR, et al: The effect of reperfusion therapy on cerebral blood flow in acute stroke. J Stroke Cerebrovasc Dis 1999;8:9-16. 39. Slater R, Reivich M, Goldberg t-I, et al. Diaschisis with cerebral infarction. Stroke 1977;8:684-689. 40. Feeney DM, Baron JC: Diaschisis. Stroke 1986;17:817-830. 41. Lavy S, Melamed E, Portnoy Z. The effects of cerebral infarction on the regional cerebral blood flow of the contralateral hemisphere. Stroke 1975;6:160-163. 42. Sorensen AG, Buonanno FS, Gonzalez RG, et al. Hyperacute stroke: Evaluation with combined multisection diffusion-weighted and hemodynamically weighted echoplanar MR imaging. Radiology 1996;199:391-401. 43. Warach S, Dashe JF, Edelman RR: Clinical outcome in ischemic stroke predicted by early diffusion-weighted and perfusion magnetic resonance imaging: a preliminary analysis. J Cerelo Blood Flow Metab 1996;16:53-59. 44. Warach S, Gaa J, Siewert B, et al: Acute human stroke studied by whole brain echo planar diffusion-weighted magnetic resonance imaging. Ann Neurol 1995;37:231241. 45. Overgaard K, Sperling B, Boysen G, et al. Thrombolytic therapy in acute ischemic stroke: A Danish pilot study. Stroke 1993;24:1439-1446.