Regional changes in the contralateral healthy Э hemisphere after ischemic lesions evaluated by quantitative T2 parametric maps.код для вставкиСкачать
THE ANATOMICAL RECORD 266:118 –122 (2002) DOI 10.1002/ar.010044 Regional Changes in the Contralateral “Healthy” Hemisphere After Ischemic Lesions Evaluated by Quantitative T2 Parametric Maps ANDREA SBARBATI,1* ANGELO REGGIANI,2 ELENA NICOLATO,1 ROBERTO ARBAN,2 ERNESTO LUNATI,1 AND FRANCESCO OSCULATI1 1 Department of Morphological and Biomedical Sciences, Section of Anatomy and Histology, University of Verona, Verona, Italy 2 GlaxoSmithKline S.p.A., Research Laboratories, Verona, Italy ABSTRACT Modifications in the contralateral “healthy” hemisphere in a population of rats bearing cortical infarction were studied in vivo by magnetic resonance imaging (MRI) with the aim to investigate whether cerebral areas not directly involved in the lesion react at the presence of an ischemic lesion. The study was performed in rats in which a transtemporal approach was adopted to occlude the right middle cerebral artery (MCA). For MRI, the animals were examined at 4.7 Tesla and quantitative T2 parametric images were obtained by a multiecho sequence. Healthy rats and sham-operated animals were used as control groups. The quantitative T2 parametric images showed that in the first week after the ischemia a significant increase in the mean T2 was seen in the lesioned parietal cortex, compared to the corresponding region of healthy rats (106 msec vs. 68 msec, P ⬍ 0.001). The contralateral “healthy” hemisphere showed T2 mean values not significantly different from the corresponding hemisphere of healthy rats (71 msec vs. 70 msec). However, a statistically significant increase in the T2 values was evident in the hypothalamic region (74 msec vs. 66 msec, P ⬍ 0.001). In rats examined 1 month after the ischemia, the T2 values of the hypothalamus were lower than those observed one week after ischemia (69 msec) but remained higher than in controls. The present study demonstrates that after a cerebral ischemia areas of secondary involvement distant from the lesion are present and can be studied in vivo by quantitative MRI. Anat Rec 266:118 –122, 2002. © 2002 Wiley-Liss, Inc. Key words: MRI; rat; hypothalamus; stroke The cellular events taking place around a focus of cerebral ischemia are complex (Garcia et al., 1993; Fisher, 1997). In many animal models, cerebral ischemic lesions can be evidenced by magnetic resonance imaging (MRI) after a few minutes and they enlarge progressively until 24 hr (Rother et al., 1996). The lesion is usually surrounded by an area of penumbra (or salvageable ischemic tissue) in which vasogenic and cytotoxic edema spread and delayed neuronal death may be visible. In addition, secondary changes may gradually develop in ipsilateral areas outside the ischemic region in laboratory animals and in men (Kataoka et al., 1989; Tamura et al., 1990; Nakane et al., 1992). We have recently demonstrated that modifications may also be induced in cerebral areas of the con© 2002 WILEY-LISS, INC. tralateral hemisphere and that they can be evidenced with histochemical, immunohistochemical, and ultrastructural methods (Peng et al., 1996; Sbarbati et al., 1996). These modifications seem to be due, at least in part, to the *Correspondence to: Prof. Andrea Sbarbati, Dept. of Morphological and Biomedical Sciences, Section of Anatomy and Histology, University of Verona, Medical Faculty, Strada Le Grazie 8, 37134 Verona, Italy. Fax: ⫹39-45-8027163. E-mail: SBARBATI@borgoroma.univr.it Received 7 May 2001; Accepted 12 October 2001 SECONDARY INVOLVEMENT IN BRAIN ISCHEMIA development of a collateral circulation. In the days after the development of an ischemic lesion, scanning electron microscopy observations of vascular cast show a modification of preexisting blood vessels (Coyle and Jokelainen, 1982; Coyle, 1984; Coyle and Heistad, 1987), which allows the development of collateral circles between the two hemispheres (Sbarbati et al., 1996). This involvement of the contralateral hemisphere seems to be protracted for several days, since 1 week after a cerebral lesion increased expression of nitric oxide synthase, an enzyme involved in the production of nitric oxide (Peng et al., 1996), could be immunocytochemically demonstrated in specific regions of the contralateral hemisphere. These data suggest that secondary effects, which may be relevant in the recovery of functional activity, are present in the hemisphere contralateral to an ischemic lesion. The histological analysis is unfit to put in evidence these events; on the contrary, their transient and functional nature is better studied by MRI examination, which has been demonstrated to represent a valid tool to evaluate in vivo areas of secondary changes (see Nakane et al., 1997). Using this technique, several works described the in vivo characteristics of ischemic lesions; however, data about modifications in specific areas of the brain distant from the main lesion and, for this reason, not directly involved in the ischemic process are still scarce and restricted to the ipsilateral hemisphere (Nakane et al., 1992, 1997). In addition, quantitative MRI (see Hoehn-Berlage et al., 1995; Loubinoux et al., 1997), which provides more detailed information than standard T2-weighted images, has never been used to study secondary lesions, necessitating further studies on this topic. In the present work we adopted MRI to study in vivo the presence of modifications in the contralateral “healthy” hemisphere throughout a population of animals bearing cortical infarction. We have used quantitative T2 imaging as a suitable noninvasive method of evaluating vasogenic edema in the brain. The aim of the work was to investigate whether areas of secondary involvement are restricted to the ipsilateral hemisphere or whether they may also be located contralaterally. MATERIALS AND METHODS Male Sprague-Dawley rats, weighing 250 –275 g, were anaesthetized with 400 mg/kg i.p. of chloral hydrate and permanent occlusion of the left middle cerebral artery (MCA) was induced according to the method of Tamura et al. (1981) (slightly modified). Briefly, the rats were anaesthetized with chloral hydrate (Fluka, 300 mg/kg i.p.) and maintained normothermically by means of a homeothermic heating system (IMS “K-temp” control unit) coupled to a rectal thermistor probe. The rats were placed under an operating microscope in the lateral position, and a curved 2-cm skin incision was made in the midpoint between the orbit and the external auditory canal. An incision was made around the superior and posterior margins of the temporalis muscle that was scraped from the lateral aspect of the skull and reflected forwards. A craniotomy was performed by a 018 round burr (Ash) at the junction between the medial wall and the roof of the infratemporal fossa. The position of the skull opening was about 3 mm anterior and 1 mm lateral to the foramen ovale. At this point the dura was opened through a cruciate incision by means of a fine needle. The exposed MCA was coagulated by bipolar diathermy and then the hole was filled with 119 absorbable bone sealant (Absele, Ethicon, Somerville, NJ). The scalp incision was sutured and the animal was placed under a lamp to facilitate recovery from anaesthesia and avoid hypothermia. No signs of hyperthermia were observed during the surgery or MRI procedures. A group of seven healthy male Sprague-Dawley rats, with the same weight, and a further group composed of sham-operated rats (n ⫽ 3) were used as controls. For MRI examination, the rats underwent gas anaesthesia (O2 containing 0.5%–1% halothane) and were positioned, within an 8-cm-diameter saddle coil, in a SIS 200/330 imager spectrometer (SIS Co., Freemont, CA) equipped with a 4.7Tesla horizontal magnet (Oxford Ltd., Oxford, UK) and 33-cm bore. Part of the MRI experiments were carried out using a Biospec System (Bruker, Karlsruhe, Germany) equipped with the same magnet and a SMIS (Surrey Medical Imaging Systems, Ltd., Surrey, UK) gradient insert. A 72-mm-i.d. transmitter/receiver birdcage coil was used. Proton MRI (at 200 MHz) was performed. In each animal, a multislice scout spin-echo sequence was performed with the following parameters: orientation ⫽ sagittal, thickness ⫽ 2 mm, time of repetition (TR) ⫽ 600 msec, time of echo (TE) ⫽ 18 msec, number of experiments (NEX) ⫽ 1, field of view (FOV) ⫽ 10 cm, and matrix size ⫽ 256 ⫻ 192. Then rapidacquisition relaxation-enhanced (RARE) (orientation ⫽ axial, thickness ⫽ 2 mm, TR ⫽ 5,000 msec, TE ⫽ 85 msec, NEX ⫽ 1, FOV ⫽ 10 cm, and matrix size ⫽ 256 ⫻ 256) and diffusion-weighted (orientation ⫽ axial, thickness ⫽ 2 mm, FOV ⫽ 6 cm, NEX ⫽ 1, and matrix size ⫽ 128 ⫻ 128) images were acquired. Quantitative T2 parametric images were obtained by a multiecho sequence (orientation ⫽ axial, TR ⫽ 1,200 msec, TE ⫽ 20 msec, number of echoes ⫽ 6, slice thickness ⫽ 2 mm, FOV ⫽ 8 cm, acquisition time ⫽ 150 sec, and matrix size ⫽ 128 ⫻ 128). Such T2 maps were calculated pixel-by-pixel by fitting the logarithms of the expression S ⫽ S0 exp(–nTE/T2), obtained with the different values n ⫽ 1– 6 from the multiecho sequence. Likewise, T1 maps were produced from spin-echo sequences with different values of TR. During the acquisition the temperature of the animals was not controlled; however, clinical examination performed at the end of the experiment in the present and in a previous study (Sbarbati et al., 2000) revealed that in none of the rats was hypothermia induced by gas anaesthesia. Quantitative parametric T2 maps were obtained in ischemic rats 1 week (7–9 days) and 1 month (28 –31 days) after the MCA occlusion (MCAO), while the rats of control (nonoperated or sham-operated) were observed once. In sham-operated rats, parametric maps were obtained 7 days after MCAO. Quantitative T2 data were obtained by region of interest (ROI) analysis of parametric images (axial sections). Operator-defined ROIs were manually traced on cerebral areas identified according to Paxinos and Watson (1986) approximately at the following coordinates: interaural 7.2 mm/ bregma –1.8 mm. For each rat, 12 areas were evaluated and the mean value was calculated. The number of pixels involved in the ROIs was optimized over the shape of the corresponding structure and ranged from 3– 6. The intraROI variability, evaluated as the semidispersion, was about 5%. Two operators performed the measures: one of them was blinded to the disease or control group. The variability between operators (interrater) was less than 0.4%, while that of the same operator (intrarater) was less than 0.1%. Periventricular areas were excluded from calculation to avoid partial volume effects. For statistical analysis, single- 120 SBARBATI ET AL. demonstrated regional differences with respect to the true normal cerebral hemisphere of control animals (Peng et al., 1996; Sbarbati et al., 1996). This secondary postischemic involvement has not been studied in vivo by quantitative MRI, probably because previous MRI studies have mainly focused on the delimitation of the lesion and its penumbra (Roberts et al., 1993; Kohno et al., 1995; Busch et al., 1996; Dreher et al., 1998; Rordorf et al., 1998). Study Limitations Fig. 1. a: RARE axial image of an ischemic rat brain. b: T2 map obtained from the same animal. factor one-way analysis of variance (ANOVA) was used to compare the mean values of the groups. Bonferroni correction was used for multiple comparisons and differences at a P value of ⬍0.001; both vs. healthy rats and vs. the sham group were considered significant. RESULTS In the first week after the MCAO, in vivo MRI examination showed in all studied animals a hyperintense area in the right fronto-parietal cortex, which was visible, both in the T2- and diffusion-weighted pulse sequences. Figure 1 shows a RARE axial image of an ischemic rat brain and the T2 map obtained from the same animal. Figure 2 depicts the evolution of the ischemic lesion in another subject. One month after the ischemia the lesion is reduced in size but still visible. The quantitative T2 parametric images show (Table 1) a significant increase in T2 in the ischemic region the first week after MCAO, in comparison with the corresponding region (parietal cortex) of healthy or sham rats (106.1 msec vs. 67.5 msec healthy, vs. 69.0 msec sham; P ⬍ 0.001). The contralateral, “healthy” hemisphere of the ischemic rats showed a global mean value of T2 not significantly different from the corresponding hemisphere of healthy rats (70.7 msec vs. 69.5 msec). However, when different areas of the contralateral hemisphere were examined, a statistically significant increase in the T2 values in the hypothalamic region (73.7 msec vs. 66.3 msec healthy, vs. 69.4 msec sham; P ⬍ 0.001) was evident. The sampled hypothalamic region is evidenced in Figure 3. The T2 values of this region of the ischemic rats ranged from 71–78 msec and an increase was visible in all examined rats, compared to the controls. In other areas of the contralateral hemisphere, nonstatistically significant alterations of the T2 values were evident (Table 1). In rats examined 1 month after ischemia, T2 values of the hypothalamus were lower than those recorded 1 week after ischemia (68.7 msec) but remained higher than those in healthy controls (66.3 msec). The T2 values obtained from the contralateral hypothalamic region or whole contralateral hemisphere in each rat are reported in Figure 4. These data reveal a significant increase in hypothalamic T2 values 1 week after MCAO. DISCUSSION Secondary Postischemic Involvement The cerebral hemisphere contralateral to an ischemic lesion is widely considered “normal” at MRI, and it is free of systemic effects on T2 values originating from the lesion (Hoehn-Berlage et al., 1995). However, other techniques In vivo paradigms provide anatomical information that is not available by traditional histological methods. Major limitations are the partial volume effect and the inaccurate sampling of the hypothalamic ROI. In fact, a single hypothalamic nucleus of a rat cannot be visualized with MRI. However, the technique of ROI definition is widely used and turned out to be a sound method for our measurements: the interrater variability is scarce and the intrarater variability is quite negligible, allowing statistically significant differences between the groups in particular areas to be obtained. Another drawback of analyzing in vivo data pertains to a certain degree of variability between anatomical regions and groups. This variability in SDs found by other groups (Loubinoux et al., 1997) is confirmed in the present work, suggesting that it does not originate from the sampling method but is intrinsically linked to in vivo MRI evaluations. Effect of Ischemia on T2 Values In the first week after a permanent ischemic lesion, T2 increases on the whole contralateral hemisphere, but this change is not statistically significant. This finding seems to be in accordance with results obtained by Schuier and Hossmann (1980), who demonstrated in cats that massive edema is absent in the contralateral hemisphere of animals with cerebral ischemia. However, analysis of the different regions demonstrated that significant differences with the homotopic regions of nonischemic animals can be observed by quantitative MRI. In control regions our quantitative evaluations are generally in accordance with those calculated by Loubinoux et al. (1997), who found that the T2 values are 65.9 msec for the cortex, 63.7 msec for the striatum, 66.0 msec for the hypothalamus, and 55.8 msec for the thalamus. Data of previous studies (Loubinoux et al., 1997) and the present findings demonstrate that the cortex and the hypothalamus have more or less similar values in normal conditions. In the week after an ischemic lesion, the hypothalamus in the contralateral “normal” hemisphere was found to have a significantly higher T2 than the cortex. The high T2 values in the hypothalamus are difficult to explain because in our experimental model this region is not primarily involved in the ischemic process (Tamura et al., 1981). A long-range interaction with the contralateral hemisphere has been described in other types of lesions, such as lesions induced by kainic acid injection (Zaczek et al., 1980). In our opinion, the T2 increase detected in the contralateral hemisphere is not a direct indicator of local lesion, since the values observed are clearly lower than those of the ischemic zone. A high relaxation time T2 is linked to changes in the physical state of the tissutal free water and is generally considered to result from the presence of excessive amounts of extracellular liquid (vasogenic edema). However, alterations in the activity status of the tissue, influencing the cerebral blood volume, could 121 SECONDARY INVOLVEMENT IN BRAIN ISCHEMIA Fig. 2. Evolution of an ischemic lesion in a rat brain. RARE image of the brain 2 days after MCAO (a), 9 days after MCAO (b), and 30 days after MCAO (c). TABLE 1. Quantitative T2 data obtained by ROI analysis of parametric images Hypothalamus (ipsilateral) Parietal cortex (ipsilateral) Cingulate cortex (ipsilateral) Pyriform cortex (ipsilateral) Basal ganglia (ipsilateral) Ipsilateral hemisphere Hypothalamus (contralateral) Parietal cortex (contralateral) Cingulate cortex (contralateral) Pyriform cortex (contralateral) Basal ganglia (contralateral) Contralateral hemisphere Healthy (n ⫽ 7) Sham (n ⫽ 3) One week after MCAO (n ⫽ 8) 67.0 ⫾ 1.4 67.5 ⫾ 1.0 66.7 ⫾ 1.0 78.3 ⫾ 1.0 64.3 ⫾ 1.4 69.2 ⫾ 0.3 66.3 ⫾ 0.7 66.4 ⫾ 0.4 67.4 ⫾ 1.2 75.3 ⫾ 1.9 62.2 ⫾ 0.6 69.5 ⫾ 0.4 69.0 ⫾ 2.0 69.0 ⫾ 1.7 67.0 ⫾ 2.6 82.7 ⫾ 2.5 67.0 ⫾ 2.6 71.3 ⫾ 1.0 69.4 ⫾ 0.7 70.1 ⫾ 1.7 67.5 ⫾ 1.0 77.9 ⫾ 1.9 67.4 ⫾ 3.2 72.7 ⫾ 1.1 73.4 ⫾ 1.8 106.1 ⫾ 12.6 67.5 ⫾ 1.5 80.1 ⫾ 2.4 65.4 ⫾ 1.5 79.5 ⫾ 4.7 73.7 ⫾ 2.1 68.9 ⫾ 1.4 69.8 ⫾ 1.5 77.7 ⫾ 2 65.8 ⫾ 0.9 70.7 ⫾ 1.3 One month after MCAO (n ⫽ 8) (P⬍0.001) (P⬍0.001) 68.7 ⫾ 3.6 70.1 ⫾ 4.9 67.1 ⫾ 1.9 77.9 ⫾ 3.0 64.3 ⫾ 1.4 74.7 ⫾ 6.1 68.7 ⫾ 1.1 67.8 ⫾ 0.7 65.3 ⫾ 1.4 77.6 ⫾ 1.5 63.4 ⫾ 1.0 70.1 ⫾ 1.6 Values of T2 are expressed in milliseconds (mean ⫾ SD). The measurements of the experimental groups with significant difference vs. both healthy and sham groups are indicated (P ⬍ 0.001, one-way single-factor ANOVA). also cause bigger changes in the T2 values (Cohen and Bookheimer, 1994). The high signal intensity observed in nonquantitative T2-weighted images was found in areas of secondary involvement (substantia nigra) ipsilateral to an ischemic lesion (Nakane et al., 1997). These secondary changes have been suggested to be due to a transsynaptic, neurotransmitter-mediated disinhibition as a result of loss of GABAergic input resulting in a reversible edema (Nakane et al., 1997). Therefore, our results seem to indicate that even if not directly involved in the cerebral ischemia areas of the contralateral hemisphere can be secondarily involved. This involvement could be due to vasogenic edema or to an alteration of the functional activity of the tissue. Both aspects are difficult to be demonstrated by conventional histological methods. It has been proved that MRI is a more suitable tool to estimate edema than histology, which might explain why hypothalamic involvement has not been detected in postmortem examination. Also the transient nature of this event, which is reduced 1 month after the MCAO, may have hampered postmortem demonstration. A further hypothesis is that the contralateral involvement could be related to the increased permeability of the cerebral blood vessels described in the days after an ischemia (Brightman et al., 1970; Preston et al., 1993). The hypothalamus could be particularly sensitive to this vascular condition due to the absence of the blood-brain barrier in the area of the median eminence (Brightman et al., 1970; Sage and Wilson, 1994). CONCLUSION In conclusion, the present study demonstrates that in brain ischemia involvement of hypothalamic areas out of Fig. 3. MRI image (left) and anatomical map (right) showing the hypothalamic region and the ROI used for MRI evaluations: 1, paraventricular nucleus hypothalamus; 2, anterior hypothalamic nucleus; 3, ventromedial nucleus hypothalamus. the lesion can be present and studied in vivo by quantitative MRI. This finding confirms the hypothesis that, in addition to the primarily involved areas (the lesion and its penumbra), areas of secondary, transient involvement 122 SBARBATI ET AL. Fig. 4. Comparison of T2 values in MCAO-operated, sham-operated, and healthy control rats. a: Contralateral hypothalamic region. b: Contralateral “healthy” hemisphere. may occur after a cerebral ischemia. These areas also develop at a distance from the lesion and may be located in the contralateral hemisphere. In our opinion, the presence of secondary involvement areas deserves further study since they might play an important role in the recovery process following an ischemic lesion and as such could be the target of pharmacological treatments. LITERATURE CITED Brightman MW, Klatzo T, Olsson Y, Reese T. 1970. The blood-brain barrier to proteins under normal and pathologic conditions. J Neurol Sci 10:215–239. Busch E, Gyngell ML, Eis M, Hoehn Berlage M, Hossmann KA. 1996. Potassium-induced cortical spreading depressions during focal cerebral ischemia in rats: contribution to lesion growth assessed by diffusion-weighted NMR and biochemical imaging. J Cereb Blood Flow Metab 16:1090 –1099. Cohen MS, Bookheimer SY. 1994. Localization of brain function using magnetic resonance imaging. Trends Neurosci 17:268 –277. Coyle P. 1984. Diameter and length changes in cerebral collaterals after middle cerebral artery occlusion in the young rat. Anat Rec 210:357–364. Coyle P, Jokelainen PT. 1982. Dorsal cerebral arterial collaterals of the rat. Anat Rec 203:397– 404. Coyle P, Heistad DD. 1987. Blood flow through cerebral collateral vessels one month after middle cerebral artery occlusion. Stroke 18:407– 411. Dreher W, Kuhn B, Gyngell ML, Busch E, Niendorf T, Hossmann KA, Leibfritz D. 1998. Temporal and regional changes during focal ischemia in rat brain studied by proton spectroscopic imaging and quantitative diffusion NMR imaging. Magn Reson Med 39:878 – 888. Fisher M. 1997. Characterizing the target of acute stroke therapy. Stroke 28:866 – 872. Garcia JH, Yoshida Y, Cheng H, Li Y, Zhang ZG, Lian J, Chen S, Chopp M. 1993. Progression from ischemic injury to infarct following middle cerebral artery occlusion in the rat. Am J Pathol 142: 623– 635. Hoehn-Berlage M, Eis M, Back T, Kohno K, Yamashita K. 1995. Changes of relaxation times (T1, T2) and apparent diffusion coefficient after permanent middle cerebral artery occlusion in the rat: temporal evolution, regional extent, and comparison with histology. Magn Reson Med 34:824 – 834. Kataoka K, Hayakawa T, Yamada K, Mushiroi T, Kuroda R, Mogami H. 1989. Neural network disturbance after focal cerebral ischemia in rats. Stroke 20:1226 –1235. Kohno K, Back T, Hoehn Berlage M, Hossmann KA. 1995. A modified rat model of middle cerebral artery thread occlusion under electrophysiological control for magnetic resonance investigations. Magn Reson Imaging 13:65–71. Loubinoux I, Volk A, Borredon J, Guirimand S, Tiffon B, Seylaz J, Méric P. 1997. Spreading of vasogenic edema and cytotoxic edema assessed by quantitative diffusion and T2 magnetic resonance imaging. Stroke 28:419 – 427. Nakane M, Teraoka A, Asato R, Tamura A. 1992. Degeneration of the ipsilateral substantia nigra following cerebral infarction in the striatum. Stroke 23:328 –332. Nakane M, Tamura A, Nagaoka T, Hirakawa K. 1997. MR detection of secondary changes remote from ischemia: preliminary observations after occlusion of the middle cerebral artery in rats. Am J Neuroradiol 18:945–950. Paxinos G, Watson C. 1986. The rat brain in stereotaxic coordinates. Sydney: Academic Press. p 25–30. Peng ZC, Pietra C, Sbarbati A, Ziviani L, Yan XB, Osculati F, Bentivoglio M. 1996. Induction of NADPH-diaphorase activity in the rat forebrain after middle cerebral artery occlusion. Exp Neurol 138: 105–120. Preston E, Sutherland G, Finsten A. 1993. Three openings of the blood-brain barrier produced by forebrain ischemia in the rat. Neurosci Lett 149:75–78. Roberts TP, Vexler Z, Derugin N, Moseley ME, Kucharczyk J. 1993. High-speed MR imaging of ischemic brain injury following stenosis of the middle cerebral artery. J Cereb Blood Flow Metab 13:940 – 946. Rordorf G, Koroshetz WJ, Copen WA, Cramer SC, Schaefer PW, Budzik Jr RF, Schwamm LH, Buonanno F, Sorensen AG, Gonzalez G. 1998. Regional ischemia and ischemic injury in patients with acute middle cerebral artery stroke as defined by early diffusionweighted and perfusion-weighted MRI. Stroke 29:939 –943. Rother J, de Crespigny AJ, D’Arceuil H, Mosley ME. 1996. MR detection of cortical spreading depression immediately after focal ischemia in the rat. J Cereb Blood Flow Metab 16:214 –220. Sage MR, Wilson AJ. 1994. The blood-brain barrier: an important concept in neuroimaging. AJNR Am J Neuroradiol 15:601– 622. Sbarbati A, Pietra C, Baldassarri AM, Guerrini U, Ziviani L, Reggiani A, Boicelli A, Osculati F. 1996. The microvascular system in ischemic cortical lesions. Acta Neuropathol (Berl) 92:56 – 63. Sbarbati A, Reggiani A, Lunati E, Arban R, Nicolato E, Marzola P, Asperio RM, Bernardi P, Osculati F. 2000. Regional cerebral blood volume mapping after ischemic lesions. Neuroimage 2000 12:418–424. Schuier FJ, Hossmann KA. 1980. Experimental brain infarcts in cats. II. Ischemic brain edema. Stroke 11:593– 601. Tamura A, Graham DI, McCulloch J, Teasdale GM. 1981. Focal cerebral ischaemia in the rat: 1. Description of technique and early neuropathological consequences following middle cerebral artery occlusion. J Cereb Blood Flow Metab 1:43– 60. Tamura A, Kirino T, Sano K, Takagi K, Oka H. 1990. Atrophy of the ipsilateral substantia nigra following middle cerebral artery occlusion in the rat. Brain Res 510:154 –157. Zaczek R, Simonton S, Coyle JT. 1980. Local and distant neuronal degeneration following intrastriatal injection. J Neuropathol Exp Neurol 39:245–264.