Original article J. Neuroradiol., 2005, 32, 337-341 © Masson, Paris, 2005 THRESHOLD OF REGIONAL CEREBRAL BLOOD FLOW FOR INFARCTION IN PATIENTS WITH ACUTE CEREBRAL ISCHEMIA M. OHASHI, A. TSUJI, M. KANEKO, M. MATSUDA Department of Neurosurgery, Shiga University of Medical Science, Ohtsu, Shiga 520-2192, Japan. SUMMARY Threshold of regional cerebral blood flow (rCBF) for cerebral tissue survival in relation to time was studied in patients with acute cerebral ischemia with xenon-enhanced computed tomography (XeCT). Case 1: A 58-year-old man with right hemiparesis, total aphasia and a high intensity area of 1cm2 in the left insula on diffusion weighted image underwent XeCT CBF study before and after intra-arterial local thrombolytic therapy (IALT) on the occluded middle cerebral artery (MCA) 4 hours and 7 hours after stroke onset, respectively. Case 2: A 65-year-old woman with recurrent transient ischemic attacks (TIAs) caused by severe stenosis of the left MCA underwent XeCT CBF study 5 hours after onset of the last attack. XeCT was conducted by 5-min wash-in method. In Case 1 the rCBF in the pre-IALT MCA territory was 4 to 19ml/100g/min. The area where rCBF in the post-IALT increased to above 15ml/100g/min were saved, but the other area where it remained in the 9 to 14ml/100g/min evolved into infarct on subsequent CT scan/MR (magnetic resonance) imaging. The patient was discharged with only mild motor dysphasia. In Case 2 the left corona radiata showed rCBF of 7ml/100g/min and this area evolved into infarct on MR imaging. The patient was discharged home with right hemiparesis. Our results showed validity of the rCBF threshold in acute cerebral ischemia reported by Jones et al. Residual rCBF in the acute stage of cerebral ischemic stroke can predict the fate of the lesion. Key words: acute cerebral ischemia, cerebral infarction, flow threshold, ischemic penumbra, xenon CT. RÉSUMÉ Seuil de débit sanguin cérébral correspondant à un infarctus chez les patients ayant une ischémie cérébrale aiguë Le seuil de débit sanguin cérébral (DSC) pour la survie tissulaire en fonction du temps était étudié par le scanner au xénon (XeCT) chez les patients présentant une ischémie cérébrale aiguë. Cas n° 1 : un XeCT a été réalisé chez un homme de 58 ans avec une hémiparésie droite, une aphasie mixte et une zone hyperintense de 1 cm2 au niveau de l’insula gauche sur l’IRM de diffusion. Le XeCT a été effectué avant et après une thrombolyse intra-artérielle (TIA) de l’occlusion de l’artère cérébrale moyenne (ACM) à la 4e et 7e heure de l’accident ischémique. Cas n° 2 : une femme de 65 ans avec une sténose sévère de l’ACM gauche responsable des accidents ischémiques transitoires (AIT) répétés. Le XeCT a été réalisé chez ce patient 5 heure après son accident initial. Dans le cas n° 1, le DSC avant la thrombolyse était de 4 à 19 ml/100g/min. Les zones avec un DSC > 15 ml/ 100g/min ont été récupérées et celles < 9 ml/100g/min ont évoluées vers l’infarctus aux XeCT et IRM de contrôle. Le patient est sortie de l’hôpital avec une aphasie motrice modérée. Dans le cas n° 2, le DSC de la corona radiata gauche était 7 ml/100g/ min avec une évolution vers l’infarctus à l’IRM. La patiente est sortie de l’hôpital avec une hémiparésie droite. En conclusion, nos résultats valident le seuil d’ischémie cérébrale rapporté par Jones et al. Le DSC résiduel dans la phase aiguë de l’ischémie cérébrale peut prédire l’évolution de la lésion. Mots-clés : Ischémie cérébrale aigüe, infarctus cérébral, seuil de perdusion, pénombre ischémique, Xénon CT. INTRODUCTION Since the revolutionary concept of “time is brain” was introduced, whether or not thrombolysis is needed in the treatment of patients with acute cerebral ischemia remains a matter of concern [2, 13]. For decision-making within the therapeutic time window computed tomography (CT) scan, magnetic resonance (MR) imaging, cerebral angiography and regional cerebral blood flow (rCBF) studies such as single photon emission tomography, positron emission tomography and xenon-enhanced CT (XeCT) are used, often in combination. It is well known that recovery of ischemic brain or development of infarction is largely dependent on the residual blood flow and the duration of ischemia [1, 7]. However, Reprint request: M. OHASHI e-mail: mmrm@belle.shiga-med.ac.jp reports on the critical flow and the period of ischemia that can be tolerated in humans are scarce. We report two cases of acute ischemic stroke in which XeCT was used to assess the rCBF and correlated with cerebral angiography, CT scan and MR imaging. METHODS XeCT was performed with a 5-min wash-in protocol. After base scan the patient started to inhale a mixture of gas (28% Xe, 30-50% oxygen and balance air) made in the stable Xe gas rebreathing system (AZ-725, Anzai Medical) and delivered by a mask, and underwent CT scans (Asteion, Toshiba) every one min for 5-min inhalation period. Quantification of rCBF was achieved by the image processing software (AZ-7000W, Anzai Medical). 338 M. OHASHI et al. FIG. 1. – a: CT scan at 2.5 hours after onset. Hyperdense middle cerebral artery sign (arrow), and blurring of insular ribbon (arrow head) are observed. b: MR angiography at 3 hours after onset. Left middle cerebral artery is not visualized at the distal trunk. FIG. 1. – a : scanner réalisé 2,5 heures après le debut de la symptomatologie. Hyperdensité de l’artère cérébrale moyenne (flèche), et hypodensité autour de l’insula (tête de flèche) sont observées. b : L’angioIRM réalisé à la 3e heure n’a pas visualisé le tronc distal de l’artère cérébrale moyenne gauche. FIG. 2. – a: DWI at 3 hours after onset. Hyperintense lesion is localized in the left insula. b: FLAIR image at 3 hours after onset. Mild hyperintensity in the left insular cortex and close to the sylvian fissure. c: XeCT at 4 hours after onset. Hypoperfusion is far more extensive than the lesion observed on MR imaging. numerals: rCBF expressed in ml/100g/min. FIG. 2. – a : IRM de diffusion réalisee 3 heures après la symptomatologie. Lésion hyperintense localisée à l’insula gauche. b : Le FLAIR correspondant. Hyperintensité modérée dans le cortex insulaire et à proximité de la valée sylvienne. c : Le TDM au xénon à la 4e heure. Hypoperfusion plus étendue que la lésion observée à l’IRM. Les chiffres correspondent au débit sanguin cérébrale exprimé en ml/100g/min. Case reports Case 1 A 58-year-old man developed right arm weakness and difficulty in speech and was admitted to our department 2 and a half hours after onset. On admission he had motor and sensory aphasia (jargon) and right hemiparesis. The National Institute of Health Stroke Scale (NIHSS) score [3] was 12. He was on medication for diabetes mellitus and had a habit of smoking. CT showed the early signs of ischemic infarction such as hyperdense middle cerebral artery (MCA) sign and blurring of insular ribbon on the left (figure 1a). MR imaging at 3 hours after symptom onset revealed a small distinctly hyperintense lesion in the left insula on diffusion weighted image (DWI) and mild hyperintensity in the insula and closing of the sylvian fissure on fluid-attenuated inversion-recovery (FLAIR) image (figures 2a and 2b). MR angiography disclosed MCA occlusion at its distal trunk (figure 1b). XeCT CBF study at 4 hours after onset showed a wider area of decreased perfusion than DWI in the left frontoparietotemporal region corresponding to the territories supplied by the operculofrontal, posterior parietal, angular and posterior temporal arteries (figures 2c, 3a and 3b). The rCBF values in these areas were 4.3 to 18.9 ml/100g/min while those in the corresponding areas in the contralateral hemisphere were 29.6 to 43.7 ml/100g/min. The MCA occlusion at the distal segment of the main trunk was confirmed by cerebral angiography (figures 3a and 3b). Five hours passed since the symptom onset. Neurological deficits did not show any improvement. We decided to go on to intra-arterial thrombolysis after informed consent. Urokinase, 120,000 U, was directly infused into the occluded segment of the MCA. It was repeated two more times for a total dose of 360,000 U. The posterior parietal and posterior temporal arteries and the proximal part of the angular artery were recanalized (figures 3c and 3d). No further thrombolysis was attempted because 6 and a half hours, longer than the therapeutic time window of 6 hours, had passed. XeCT at 7 hours after onset showed that the central portion of the ischemic area regained flow of 36.6 to 47 ml/100g/min, mildly hyperemic compared to the corresponding area in the contralateral hemisphere (figure 4a). This hyperemic area corresponded to the reperfused posterior parietal artery territory. The other areas supplied by the non-recanalized arteries remained at low flow level of 8.8 to 17.9 ml/100g/min. On Day 3, CT showed a low density area in the posterior insula and the adjacent areas where rCBF remained at low level, 8.8 to 14ml/100g/min (figures 4a and 4b). On Day 6, FLAIR image showed a hyperintense lesion in the same area and it was a little larger with mixed intensity in the upper level where thrombolysis failed and rCBF remained at 14 ml/100g/min (figures 4a and 4c). The patient was THRESHOLD OF REGIONAL CEREBRAL BLOOD FLOW FOR INFARCTION IN PATIENTS 339 FIG. 3. – a and b: Cerebral angiography before intra-arterial thrombolysis, frontal view (a) and lateral view (b). The main branches of the middle cerebral artery are occluded. c and d: Cerebral angiography after intra-arterial thrombolysis, frontal view (c) and lateral view (d). The posterior parietal and posterior temporal arteries and the proximal portion of the angular artery are recanalized. FIG. 3. – Angiographie cérébrale de face (a) et profile (b) avant la thrombolyse intra-artérielle. Occlusion des principales brances de l’artère cérébrale moyenne. Angiographie cérébrale de face (c) et profile (d) après la thrombolyse intra-artérielle. Révascularisation des artères temporales, pariétales postérieures et la portion de l’artère angulaire. FIG. 4. – a: XeCT at 7 hours after onset. numerals: rCBF expressed in ml/100g/min. b: CT scan on Day 3. c: FLAIR image on Day 6. The left insular area with 8.8 ml/100g/min and the posterior frontal subcortical region with 7.4 ml/100g/min evolved into infarction on later CT scan and MR imaging. FIG. 4. – a : Le TDM au xénon à la 7e du début de la symptomatologie. Les chiffres correspondent au débit sanguin cérébrale exprimé en ml/100g/min. b : Le scanner réalisé au 3e jour. c : L’IRM en FLAIR 6e jour. L’insula gauche avec un débit de 8,8 ml/100g/min et la région frontale postérieure et souscorticale avec un débit de 7,4 ml/100 g/min ont évolué vers l’infarctus visibles au scanner et IRM réalisés plus tard. FIG. 5. – a: XeCT at 5 hours after onset of the last attack. numerals: rCBF expressed in ml/100g/min. b: DWI at 4 hours after onset of the last attack. c: MR imaging 6 month after onset. Left corona radiata with hyperintsnsity on DWI has a flow of 6.8 ml/100g/min. FIG. 5. – a : Le TDM au xénon 5 heures apèrs le dernière AVC. Les chiffres correspondent au débit sanguin cérébrale exprimé en ml/100 g/min. b : L’IRM de diffusion à la 4e heure. c : IRM à la 6e heure. Le débit de la capsule interne (hyperintensité à l’IRM de diffusion) est 6,8 ml/100g/min. discharged with modified Rankin Scale [14] score of 1 on Day 21. Case 2 A 65-year-old woman was admitted with recurrent TIAs over a 3-day period. On admission, 4 hours after onset of the last attack, she was aphasic and hemiparetic on the right side. The NIHSS score was 5. DWI showed a hyperintense lesion in the left putamen extending to the corona radiata (figure 5b). MR angiography revealed no visualization of the left MCA bifurcation and distal branches. The rCBF values at 5 hours after the last symptom onset were 6.8 ml/100g/min in the left corona radiata, 17 ml/100g/min in the surrounding white matter and 25 ml/100g/min in the cortical/sub- 340 M. OHASHI et al. FIG. 6. – The rCBF values in ischemic tissue and time from symptom onset. The ischemic tissue with less than 10 ml/100g min evolved to infarction (black). Tissue with more than 15 ml/100g/min at 4 to 7 hours can survive (white), but some may not (gray). Arrows indicate changes of flow from pre- to postthrombolysis in the designated region of interest. The dotted line indicates the threshold curve reported by Jones et al. cir cles: Case 1, triangles: Case 2. FIG. 6. – Évolution du débit sanguin cérébral des zones ischémiques dans le temps. Les zones ischémiques avec un débi < 10 ml/100g/min ont évoluées vers l’infarctus (a). Quelques zones avec un débit > 15 ml/100g/min à la 4 e jusqu’à la 7e heure peuvent survire (b) et d’autres non (c). Les flèches indiquent les changement avant et après la thrombolyse dans les régions d’intérêt définies. La courbe (ligne discontinue) indique le seui rapporté dans la litérature par Jones et al. Les circles représen le cas clinique n° 1 et les triangles le cas n° 2. cortical area of the left MCA territories (figure 5a). Cerebral angiography revealed severe stenosis of the left MCA trunk and no filling of the major branches, but fairly good peripheral filling through leptomeningeal collaterals. An unruptured aneurysm was found at the contralateral MCA bifurcation. Thrombolysis was not attempted because of mild neurological deficits, being close to the end of therapeutic time window and the angiographic findings. After returning to the ward her neurological status worsened to a NIHSS score of 15 in spite of intravenous anticoagulation and antiplatelet treatment. MR imaging 24 hours later showed that the hyperintense lesion was slightly larger in the left corona radiata that evolved into infarction on follow-up MR imaging (figure 5c). The white matter with rCBF of 6.8ml/100g/min evolved into infarction, but brain tissue with more than 17ml/100g/min was saved. Her neurological deficits started to improve in 4 days. Two months later she was discharged with NIHSS score of 4. DISCUSSION Our results show that the ischemic area measured by XeCT is more extensive than that indicated by CT scan or MR imaging in the early hours after symptom onset. Perfusion CT or perfusion MR could detect it as XeCT did. However, those modalities have difficulties in quantification of rCBF. In the clinical setting where prompt decision-making is required whether going on to thrombolysis or not, XeCT CBF measurement is superior to CT and MR perfusion. XeCT enables quantitative measurement of rCBF with relatively high spatial resolution and correlation to anatomical structures [5, 15], although some drawbacks [11, 15, 16] are reported. Intra-arterial delivery of thrombolytics has been shown to be effective in recanalizing occlusions of the MCA [4]. But only a select group of patients falls into this category. XeCT not only can determine whether or not the collateral supply is present but also can distinguish which patients have reversible or irreversible flow levels [9]. Retrospective analysis of Case 1 indicated that the 4-hour-old ischemic area with rCBF of 15.5 to 18.9 ml/ 100g/min was partly salvaged by intra-arterial thrombolysis. While the ischemic areas with pre- and postthrombolysis flows of 4.3 and 7.4 ml/100g/min, respectively, and of 10.6 and 8.8 ml/100g/min, respectively, evolved into dense infarction, the areas with post-thrombolysis flow of 15.5 ml/100g/min were salvaged. This flow level falls in the border zone of the threshold curve between the penumbra and infarction [7]. The ischemic tissue that remained in low rCBF level of 14 ml/100g/min at 7 hours evolved into mottled pattern infarction on later imaging. Here, the rCBF value of 14 ml/100g/min was an averaged figure in data processing from a rather wider region of interest that should be a mixture of the tissue with flow as low as 6 ml/100g/min destined to infarction [8] and the salvaged tissue with higher flow than 14 ml/100g/min. Functional recovery of the patient was eventually very good. Thus, rCBF of approximately15 ml/100g/min may be the threshold for reversibility of the ischemic mixed cortical gray and white matter at 4 to 7 hours after symptom onset. In Case 2 the white matter with rCBF of 6.8 ml/ 100g/min at 5 hours after symptom onset evolved into infarction. This result is contrary to the report that white matter has greater tolerance to low flow level like this than gray matter [8], although the value itself fits well to the Jones’curve [7]. Since this patient had repeated TIAs before the last stroke, low perfusion could have persisted much longer than 5 hours. We observed mild hyperemia after thrombolysis in Case 1. This area did not show abnormalities on later imaging. The prognostic significance of hyperemia remains unclear, although it is reported that postischemic hyperemia resulted in neurological deficit [6]. Hyperemia per se may not carry a uniformly consistent prognosis [17]. Although we have to be careful to apply the experimental results to clinical practice and data are scarce, our results seem to agree well with the threshold curve reported by Jones et al. [7] (figure 6). Levy et al. [10]suggested that the posterior CBF of 6ml/100g/min at 90 min of basilar artery occlusion might be the threshold for reversible ischemia. We need to accumulate more data on this critical issue. CONCLUSION We suggest that a rCBF of 15ml/100g/min at 4 to 7 hours after symptom onset may be the threshold of mixed cortical gray and white matter for reversible ischemia. Detection of irreversibly damaged tissue and salvageable tissue in the early hours of acute THRESHOLD OF REGIONAL CEREBRAL BLOOD FLOW FOR INFARCTION IN PATIENTS cerebral ischemia is important for therapeutic interventions. XeCT is a powerful tool for this purpose. REFERENCES [1] ASTRUP J, SIESJO BK, AND SYMON L. Thresholds in cerebral ischemia – The ischemic penumbra. Stroke 1981; 12: 723-725. [2] BARON JC, VON KUMMER R, DEL ZOPPO GJ. Treatment of acute ischemic stroke. Challenging the concept of a rigid and universal time window. Stroke 1995; 26: 2219-2221. [3] BROTT T, MARLER JR, OLINGER CP, ADAMS HP JR, TOMSICK T, BARSAN WG, BILLER J, EBERLE R, HERTZBERG V, WALKER M. Measurements of acute cerebral infarction: Lesion size by computed tomography. Stroke 1989; 20: 871875. [4] DEL ZOPPO GJ, HIGASHIDA RT, FURLAN AJ, PESSIN MS, ROWLEY HA, GENT M. PROACT: a phase II randomized trial of recombinant pro-urokinase by direct arterial delivery in acute middle cerebral artery stroke: PROACT Investigators: Prolyse in Acute Cerebral Thromboembolism. Stroke 1998; 29: 4-11. [5] GUR D, GOOD WF, WOLFSON SK JR, YONAS H, SHABASON L. In vivo mapping of local cerebral blood flow by xenon-enhanced computed tomography. Science 1982; 215: 1267-1268. [6] HUGHES RL, YONAS H, GUR D, LATCHAW R. Cerebral blood flow determination within the first 8 hours of cerebral infarction using stable xenon-enhanced computed tomography. Stroke 1989; 20: 754-760. [7] JONES TH, MORAWETZ RB, CROWELL RM, MARCOUX FW, FITZGIBBON SJ, DEGIROLAMI U, OJEMANN RG. Thresholds of focal cerebral ischemia in awake monkeys. J Neurosurg 1981; 54: 773-782. [8] KAUFMANN AM, FIRLIK AD, FUKUI MB, WECHSLER LR, JUNGRIES CA, YONAS H. Ischemic Core and Penumbra in Human Stroke. Stroke 1999; 30: 93-99. 341 [9] KILPATRICK MM, YONAS H, GOLDSTEIN S, KASSAM AB, GEBEL JM JR, WECHSLER LR, JUNGREIS CJ, FUKUI MB. CTbased assessment of acute stroke. CT, CT angiography, and xenon-enhanced CT cerebral blood flow. Stroke 2001; 32: 2543-2549. [10] LEVY EI, SCARROW AM, KANAL E, RUBIN G, YONAS H, KIRBY L. Reversible ischemia determined by xenon-enhanced CT after 90 minutes of complete basilar artery occlusion. AJNR Am J Neuroradiol 1998; 19: 1943-1946. [11] MATSUDA M, LEE H, KURIBAYASHI K, SHIINO A, HANDA J. Effect of xenon gas inhalation on intracranial pressure during xenon CT CBF measurement. Intracranial Pressure IX, Springer, Tokyo, 1994. [12] MEYER JS, HAYMAN LA, AMANO T, NAKAJIMA S, SHAW T, LAUZON P, DERMAN S, KARACAN I, HARATI Y: Mapping local blood flow of human brain by CT scanning during stable xenon inhalation. Stroke 1981;12: 426-436. [13] 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; 333:1 581-1587. [14] VAN SWIETEN JC, KOUDSTAAL PJ, VISSER MC, SCHOUTEN HJ, VAN GIJN J: Interobserver agreement for the assessment of handicap in stroke patients. Stroke 1988: 19: 604-607. [15] WINKLER S, TURKS P, HOLDEN J, KOEPPE R, RUSY B, GARBER E. Xenon effects on CNS control of respiratory rate and tidal volume – The danger of apnea. Cerebral Blood Flow and Metabolism Measurement, Springer, Berlin Heidelberg New York Tokyo, 1985. [16] YONAS H, GRUNDY B, GUR D, SHABASON L, WOLFSON SK JR, COOK E. Side effects of xenon inhalation. J Comput Assist Tomogr 1981; 5: 591-592. [17] YONAS H, GUR D, CLAASSEN D, WOLFSON SK JR, MOOSSY J. Stable xenon-enhanced CT measurement of cerebral blood flow in reversible focal ischemia in baboons. J Neurosurg 1990; 73: 266-273.