Published online before print January 3, 2008, doi: 10.1161/STROKEAHA.107.496059
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Stroke. 2008;39:427.)
© 2008 American Heart Association, Inc.
Original Contributions |
Identification of Variations in Blood-Brain Barrier Opening After Cerebral Ischemia by Dual Contrast-Enhanced Magnetic Resonance Imaging and T1sat Measurements
From the Departments of Anesthesiology (T.N.N., R.L.C.) and Neurology (K.K., J.R.E., R.A.K.), Henry Ford Hospital, Detroit, Mich; and the Department of Physics (K.K., J.R.E., R.A.K.), Oakland University, Rochester, Mich.
Correspondence and reprint requests to Dr Tavarekere N. Nagaraja, Department of Anesthesiology, Henry Ford Hospital, 2799 W Grand Blvd, Detroit, MI 48202. E-mail annta{at}neuro.hfh.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Background and Purpose— Variations in blood-brain barrier (BBB) opening after ischemia have been suggested by some tracer and magnetization transfer studies, although direct in vivo proof is still lacking. Contrast-enhanced magnetic resonance imaging (MRI) is also often used to visualize BBB damage in stroke. We hypothesized that MR contrast agents of different sizes enhance differently when BBB openings vary in size and that magnetization transfer alterations, measured by T1 in the presence of off-resonance radiofrequency saturation (T1sat), in these regions reflect such differences.
Methods— Male Wistar rats (
300 g, n=7) were subjected to 3 hours of suture occlusion of the middle cerebral artery followed by reperfusion. Status of the BBB at 24 hours after the ictus was assessed first by Gd-DTPA (554 Da) MRI and then by Gd–bovine serum albumin linked to Evans blue (Gd-BSA-EB; 68 kDa) MRI for contrast enhancement; T1sat changes, cerebral blood flow, and blood-to-brain transfer constants (Kis) for the 2 contrast agents were measured. After MRI, rats were injected with fluorescent dextran and brains were studied by fluorescence microscopy.
Results— The Gd-BSA-EB–enhancing areas were always smaller (147±80 pixels) than those for Gd-DTPA (308±204 pixels) and were contained within the latter. The difference between the 2 areas was significant (P=0.024). Changes in T1sat were larger in Gd-BSA-EB–enhancing areas (ipsilateral to contralateral [I/C]=1.53±0.20) than in Gd-DTPA–enhancing areas (I/C=1.40±0.24, P=0.005). The differences in cerebral blood flow values between the 2 regions were not significant (P=0.62), but those for the Ki values of the 2 tracers were different (P=0.01 to 0.02). Excellent agreement between regions of Gd-BSA-EB enhancement and EB fluorescence was also observed.
Conclusions— These results substantiate earlier reports of regional differences in BBB opening after stroke and provide the first in vivo evidence for this phenomenon. They also support the possible use of T1sat in quantifying stroke-induced graded BBB damage in the absence of contrast-enhanced MRI.
Key Words: blood-brain barrier • brain drug delivery • brain imaging • middle cerebral artery • neurovascular unit
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Reperfusion after focal cerebral ischemia has been shown to lead to opening of the blood-brain barrier (BBB).1 Such opening is believed to lead to a different type of lesion than permanent arterial occlusion and to play a major role in reperfusion injury.2,3 Interest in acute BBB injury after stroke has recently been heightened since BBB opening was suggested as an avenue for therapeutic access to the brain,4,5 and the size of the BBB opening has been postulated to be predictive of impending hemorrhagic transformation.6 Therefore, maintaining barrier integrity, along with neuroprotective strategies, is suggested to be essential in attenuating ischemic brain injury.7 Evaluation of BBB damage and its response to therapy is also considered important in treating the complete neurovascular unit after cerebral ischemia.8 The opening of the BBB after ischemia is often looked at as an all-or-none phenomenon, but this may not be the case in at least some models of experimental stroke.9,10 Nonetheless, little is known about the physical nature of such openings, and no study so far has investigated them in vivo.
Magnetic resonance contrast agent (MRCA)–enhanced MR imaging (MRI) with Gd-DTPA is often used to evaluate BBB opening in stroke.11,12 However, with a molecular weight of 550 Da and a Stokes-Einstein radius of 7 Å, Gd-DTPA cannot differentiate between small and large BBB openings or, in other words, varied BBB damage. Owing probably to this limitation, the acute Gd-DTPA enhancement observed in experimental stroke models has tended to overestimate the area of later hemorrhagic transformation.11,13 Magnetization transfer (MT) is the exchange of spin magnetization between "free" (water) and "bound" (proton) pools, the latter associated with tissue macromolecules; changes in MT represent alterations in the amount of tissue water in contact with tissue macromolecules and other aspects of the interaction between the two. A few MRI studies that included noncontrast-based MT measurements have shown that T1 under an off-resonance saturating radiofrequency field, or T1sat, varies appreciably within the ischemic tissue and is well correlated in space and intensity with the degree of BBB leakage of the MRCA.14,15 MRCA-enhanced MRI permits the quantification of the blood-to-brain distribution of 2 different MRCAs within a short duration of each other and thus, nearly concurrent estimates of BBB opening. On the basis of these observations, we hypothesized that (1) 2 MRCAs of markedly different biophysical properties, Gd-DTPA and Gd-DTPA linked to bovine serum albumin (Gd-BSA), will distribute differently between blood and brain; and (2) that these differences will be correlated in degree and space with T1sat. These hypotheses were tested in a rat model of transient middle cerebral artery (MCA) occlusion that has been shown to reliably result in BBB opening.16 After 3 hours of occlusion followed by reperfusion, the status of the BBB was sequentially examined at 24 hours, first with estimates of noncontrast-based T1sat, then by Gd-DTPA, and finally by Gd-DTPA linked to BSA and Evans blue (Gd-BSA-EB; 68-kDa, 37 Å). Areas of enhancement for the 2 MRCAs were measured and compared with the area of EB leakage visualized by fluorescence microscopy. Quantitative T1sat and blood-to-brain transfer constants (Ki) estimates and cerebral blood flow (CBF) values from the enhancing areas for the 2 MRCAs were compared to each other.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Preparation and Characterization of MRCA
All chemicals used were purchased from Sigma-Aldrich-Fluka (Sigma Chemical Co., St. Louis, Mo) and used as received. The contrast agents Gd-BSA and Gd-DTPA were synthesized and characterized.17,18 The Gd-BSA preparation was then incubated with EB, an often-used, fluorescent, albumin-binding vital dye, and linked to it in a Gd-BSA/EB molar ratio of 1:7 to obtain Gd-BSA-EB. This additional step rendered the molecule with both paramagnetic and fluorescent properties and enabled direct comparison of MR images with those from fluorescence microscopy.
Surgical Procedures
All surgical procedures and experiments were performed under National Institutes of Health guidelines with the approval of the institutional animal care and use committee. Seven male Wistar rats (Charles River Laboratories, Wilmington, Mass) weighing 300 g were used. Under halothane anesthesia, the right MCA was occluded with a 4.0 nylon filament as described in detail elsewhere.16,19 Three hours after MCA occlusion, the filament was withdrawn, initiating reperfusion for 21 hours.
MRI and Fluorescence Microscopy
At 24 hours after MCA occlusion, contrast-enhanced MRI was performed under halothane anesthesia according to published methods15,20 in a 7 Tesla superconducting magnet (Magnex Scientific Inc, Abingdon, UK) interfaced to a Bruker Avance console (Bruker Biospin MRI, Billerica, Mass). One femoral artery and vein were cannulated with PE-50 catheters for blood pressure monitoring and MRCA administration during imaging, respectively. The rats were placed in a supine position in an acrylic holder with a nose cone for administering anesthetic gases and placed inside the magnet.
Cerebral Blood Flow
Estimates of CBF were obtained using an arterial spin-labeling technique. Labeling of inflowing arterial water protons was performed with an axial gradient of ±0.3 kHz/mm and a continuous-wave radiofrequency pulse at a power of 0.3 kHz and frequency offset of ±6 kHz, followed by a spin-echo (SE) sequence with a repetition time (TR)/echo time (TE) of 1050 ms/20 ms.21 Four averages of the image were acquired with the gradient polarities and the radiofrequency pulse frequency offsets reversed to remove any gradient asymmetries in the axial direction. The image was acquired over a 32-mm field of view and was reconstructed with a 64x64 image matrix.
T1 and T1sat Measurements
The proton spin-lattice relaxation time, or T1, data were acquired with an imaging variant of the T-one by multiple readout pulses sequence.22 Quantitative estimates of T1sat were also generated with this method.15 This was done by inserting 2 continuous-wave saturation radiofrequency pulses, with an 8-kHz frequency offset, into the Look-Locker (LL) sequence: the first (4.5 seconds) immediately before the inversion pulse and the second (40 ms) after signal acquisition. Initially, the longitudinal magnetization was inverted with an 8-ms nonselective, adiabatic, hyperbolic secant pulse. One phase-encode line of 32 small-tip-angle gradient-echo images (TR/TE=11 seconds/7.0 ms, 32-mm field of view, 128x64 matrix, 2-mm slice thickness) was acquired at 80-ms intervals after each inversion.
BBB Permeability
Estimates of Ki were obtained with a T1-weighted LL sequence (TR/TE=80 ms/4 ms, 128x64 matrix, 24 echoes, 5 slices each, 1.8 mm thick) to produce T1 estimates.15,20 After baseline, precontrast SE T1-weighted images (TR/TE=500 ms/7 ms, 32-mm field of view, 128x128 matrix, 13 slices each, 1.0 mm thick) and LL T1 measurements were obtained, Gd-DTPA (80 µmol/kg IV) was injected, and a series of 10 LL T1-weighted measurements were done at 2.5-minute intervals for 25 minutes. A second set of SE, T1-weighted images were acquired just after this series. The pre- and post-CA SE T1-weighted images, which matched with the T1sat and CBF maps, were used to obtain a subtraction image that showed MRCA enhancement after leakage. The series of LL T1 measurements were processed to obtain a permeability map for Ki measurements. The same procedure was repeated for the Gd-BSA-EB (140 µmol/kg IV) injection after waiting for 30 minutes for the Gd-DTPA signal to decay. Estimates of Ki were produced with MRI Patlak plot methods.15,20
Immediately afterward, the rats were removed from the magnet and injected intravenously with 1 mL fluorescein isothiocyanate–labeled dextran (2000 kDa; 50 mg/mL saline) that was allowed to circulate for 2 minutes. They were then humanely killed by decapitation, and the brains were immersion-fixed in 10% buffered, neutral paraformaldehyde (VWR International, West Chester, Pa) for 48 hours. The fixed brains were cut into 60- to 70-µm-thick coronal sections in a Vibratome (Technical Products International Inc, St. Louis, Mo). The sections were individually mounted onto slides with Glycergel (Dako, Carpenteria, Calif) and coverslipped. The EB and fluorescein isothiocyanate–dextran fluorescences were observed with a fluorescence microscope (Carl Zeiss, Thornwood, NY) equipped with appropriate filters.
Image Analysis and Statistics
The precontrast SE T1-weighted image was subtracted from the postcontrast SE T1-weighted images for both tracers to generate regions of interest with extravascular enhancement of Gd-DTPA and Gd-BSA-EB. The mean intensity plus 3 SDs taken from 3 random fields on the contralateral side was chosen to define the upper boundary of normal gray-scale intensity to demarcate baseline pixels.23 The total area of BBB damage from each rat was calculated by counting the numbers of enhancing pixels above this baseline within the MR-defined ischemic lesion from each rat. The brain sections with EB leakage were digitized and reconstructed to show such regions. The total numbers of enhancing pixels from the Gd-DTPA and Gd-BSA-EB maps were compared with each other, and the Gd-DTPA and Gd-BSA-EB images were visually compared with EB fluorescence images. The CBF and T1sat values from the 2 enhancing areas from each rat were calculated15 and expressed as I/C. For each experiment, from the T1-weighted LL maps and Patlak plots, Ki was also calculated for Gd-DTPA separately in Gd-DTPA- and Gd-BSA-EB–enhancing regions and for Gd-BSA-EB in the Gd-BSA-EB–enhancing region. All values are reported as mean±SD. Two-tailed, paired and Student’s t tests were used for statistical comparisons, and significance was inferred at P<0.05.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The Gd-DTPA and Gd-BSA preparations were found to contain no unbound gadolinium or DTPA. At 7 T and 25°C, a 20% solution of Gd-BSA exhibited T1 and T2 times of 128.3 and 48.9 ms, respectively, whereas Gd-DTPA gave T1 and T2 times of 33.2 and 32.6 ms, respectively, at a comparable concentration.18
All 7 rats lost some body weight during the 24-hour poststroke period. Average weight at surgery was 298±15 g and at 24 hours after the ictus, 258±11 g. The CBF I/C values in the stroke-affected brain regions were 0.64±0.15 in the Gd-DTPA– and 0.66±0.14 in the Gd-BSA-EB–enhancing areas (P=0.62).
Extravascular Gd-DTPA enhancement indicating BBB opening was observed in parts of the ischemic tissue in all rats. Subsequent enhancement of Gd-BSA-EB was also observed, although in every instance, this region was always smaller than that of Gd-DTPA (the Table) and contained within the latter (Figures 1A, 1B, 2A, and 2
B). However, it should be noted that Gd-BSA-EB–enhancing pixels had also shown Gd-DTPA enhancement previously. However, for simplicity sake, they will be referred to as Gd-BSA-EB–enhancing areas. The number of pixels with Gd-DTPA enhancement was 308±204 and for Gd-BSA-EB, 147±80. The difference in the numbers of enhancing pixels for the 2 MRCAs was significant (P=0.02). For each animal, the changes in T1sat clustered in 2 ranges for these 2 areas (the Table). The normalized mean T1sat value from Gd-DTPA–enhancing areas was 1.40±0.24 and from those with Gd-BSA-EB (ie, both Gd-DTPA and Gd-BSA-EB) enhancement, 1.52±0.20 (P=0.005). Three different comparisons were made among the Ki values (min–1): (1) between Gd-DTPA Kis from the outer enhancing area (0.0016±0.001) and the smaller inner area, which showed Gd-BSA-EB enhancement later on (0.0017±0.0010; P=0.59); (2) between Gd-DTPA Ki from the outer Gd-DTPA area and Gd-BSA-EB Ki (0.0004±0.0002; P=0.02); and (3) between Gd-DTPA Ki from the inner area and Gd-BSA-EB Ki (P=0.01).
|
|
|
Distributions of Gd-DTPA and Gd-BSA were reflected by quantitative T1sat maps. The regions with higher T1sat values were seen as brighter, hyperintense areas, and those with lower values, as darker gray (Figures 1C and 2
C). Good visual agreement between the regions of Gd-BSA-EB enhancement and extravascular EB fluorescence was also observed in all cases (eg, Figures 1B and 1D, 2B and 2D). Cellular uptake of extravascular EB-labeled albumin in ipsilateral, BBB damaged regions was evident in the higher-magnification photomicrographs (data not shown).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
These data support the hypotheses and suggest regional differences in the degree of BBB opening within ischemia/reperfusion-damaged tissue and that T1sat may be sensitive to such differences, which may be the result of water shifts among blood and other tissue compartments and changes in associated tissue proteins.15,24 Supporting our first hypothesis, the smaller Gd-DTPA was observed to enhance across a much greater area than the larger Gd-BSA-EB. The alterations in T1sat were larger in the Gd-BSA-EB–enhancing area, supporting our second hypothesis. To the best of our knowledge, these are the first reported in vivo MRI observations of this phenomenon in stroke.
These results confirm the presence of diverse BBB openings in stroke, as suggested by earlier studies that used radiolabeled and fluorescent tracers of different sizes.9,10,25 One study reported that the transfer constants for sucrose (342 Da, 7 Å) were nearly 3 times higher than that of inulin (5000 Da, 13.5 Å) at 6 hours after a 10-minute period of global ischemia.25 Similar results were also found in a transient focal ischemia model at 24 hours after 2 hours of ischemia.10 Suggesting size-limited extravasation of plasma-borne substrates during acute reperfusion after 3 hours of MCA occlusion, EB was found to be leaking, whereas a large dextran (2000 kDa) was not.9 One study investigated the nature of BBB openings induced either osmotically or by hypertension26 and showed that tracers of different sizes such as sucrose, inulin, and fluorescent dextrans extravasated in a size-dependent fashion. The dissimilar MRCA leakage patterns observed in this study and the molecular size differences between Gd-DTPA and Gd-BSA-EB suggested variations in the size of stroke-induced BBB openings in this model.
Other contributing factors for the restricted distribution of Gd-BSA-EB may be the lower diffusivity of this larger CA in the relatively small and tortuous extracellular space of the brain27 and/or cellular uptake of extravascular albumin.28 The extracellular space in the ischemic brain can be further compressed by astrocytic swelling.29 However, as discussed next, concurrent changes in T1sat were larger in the Gd-BSA-EB–enhancing areas than in the Gd-DTPA-enhancing areas, suggesting that greater BBB damage was a major factor contributing to the observed differences in contrast enhancement. Therefore, we conclude that regions with Gd-BSA-EB enhancement may have larger openings and hence, greater BBB damage.
One possible shortcoming of this study was that changing the order of Gd-DTPA and Gd-BSA-EB administration was not attempted. This experimental design was not possible because unlike Gd-DTPA (plasma half-life of 18 minutes), albumin and albumin-bound substrates have a plasma half-life of 3 hours and even with a bolus injection, are known to maintain relatively constant plasma levels for several hours, by which time BBB pathology might change. These factors would not have allowed Gd-DTPA imaging in rapid succession with Gd-BSA-EB.
These differences in relaxivity, plasma half-life, and distribution patterns of the 2 CAs are also suspected to have resulted in the different enhancement patterns. However, use of the Patlak plot takes into account the blood levels of the tracer in question in estimates of Ki, thus facilitating direct quantitative comparisons irrespective of tracer properties.
Opening of the BBB has been noted in several disorders. In a growing tumor, it is believed to indicate ongoing angiogenesis, and its absence is considered an indicator of treatment efficacy. On the contrary, in acute stroke, transient BBB opening has also been reported.4,30 Such openings have been suggested as predictors of looming pathology6 and also as possible conduits to deliver therapeutics to the injured brain, which is otherwise not very accessible.4,5,31 However, the putative drugs may be limited by their own effective diameters (eg, presence of hydration rings that increase the native molecular size) and/or the size of available openings. Therefore, quantitative, noninvasive analysis of the size of BBB openings may help determine an optimal drug delivery vehicle to the injured brain via transient BBB openings. Characterizing the severity and nature of neuronal injury associated with such varied BBB openings may help determine their role in predicting impending tissue damage, as suggested by some investigators.6
Contrast-enhanced detection of BBB damage is dependent on the delivery of the MRCA to the affected region(s). In stroke, CBF is the first parameter to be affected, and in areas of very low CBF, BBB damage may go undetected due to inefficient MRCA delivery. In fact, it has been shown that owing to the steep drop in CBF, the severity of ischemic damage may sometimes be inversely related to the degree of MR contrast enhancement.32 Additionally, as noted before, Gd-DTPA enhancement alone may not be a spatially accurate predictor of ensuing hemorrhagic transformation after stroke.11,13 Therefore, noncontrast-based MRI techniques as surrogate markers to detect BBB injury after stroke are also being sought. Some MT parameters have shown promise in this regard. In particular, T1sat was shown to be sensitive to changes in free and protein-bound water pools as a result of BBB damage and consequent edema.15,24,33 Significant BBB damage after stroke, often predictive of impending hemorrhagic transformation, was reported to be indicated by higher I/C ratios of T1sat.24 In the present study, I/C ratios of T1sat were incrementally elevated in Gd-BSA-EB–enhancing regions, apparently with more BBB damage, when compared with Gd-DTPA–enhancing regions or regions with less BBB damage. These observations provide further evidence for the predictive ability of T1sat to track variations in BBB damage and suggest its possible clinical use to quantify graded BBB damage in the absence of contrast-enhanced MRI.
Summary
Opening of the BBB after stroke and reperfusion is a well-known phenomenon. Such opening is varied and may even be biphasic. Neuroprotection offered by some therapies is suggested to be partially due to vasoprotective actions.7 Moreover, vasoprotective drugs are being sought for combination therapy with tissue plasminogen activator to make the latter safer.34 Yet quantitative evaluation of BBB opening after stroke and its response to putative therapies is not being done regularly. It seems crucial to perform such studies to understand the effects of temporal and spatial changes in BBB permeability on the progression of stroke injury, in mediating the therapeutic efficacy of putative neuroprotective agents, and to evaluate vasoprotective drugs for potential combination therapy with tissue plasminogen activator. Noninvasive, repeatable measures such as T1sat may be useful for such purposes.
|
Acknowledgments |
|---|
We are grateful to Dr J.D. Fenstermacher for helpful suggestions, and we thank Jun Xu, Kelly A. Keenan, and Polly A. Whitton for technical assistance. K.K. thanks Dr M. Chopp, Department of Neurology, Henry Ford Hospital, for financial support.
Sources of Funding
This study was supported by the American Heart Association–Bugher Foundation Award 0270176N and National Institutes of Health grant 1RO1NS38540–01 (to R.A.K.); National Institutes of Health grant 1RO1HL70023–01 (to J.R.E.); and American Heart Association Scientist Development grant 0635403N (to T.N.N.).
Disclosures
None.
|
Footnotes |
|---|
Portions of these data were presented in abstract form at the 31st International Stroke Conference, 2006, Kissimmee, Fla.
Received June 12, 2007; accepted July 3, 2007.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1. Yang GY, Betz AL. Reperfusion-induced injury to the blood-brain barrier after middle cerebral artery occlusion in rats. Stroke. 1994; 25: 1658–1664;discussion, 1664–1665.
2. Aronowski J, Labiche LA. Perspectives on reperfusion-induced damage in rodent models of experimental focal ischemia and role of 
-protein kinase C. Ilar J. 2003; 44: 105–109.[Medline]
[Order article via Infotrieve]
3. Aronowski J, Strong R, Grotta JC. Reperfusion injury: demonstration of brain damage produced by reperfusion after transient focal ischemia in rats. J Cereb Blood Flow Metab. 1997; 17: 1048–1056.[CrossRef][Medline] [Order article via Infotrieve]
4. Huang ZG, Xue D, Preston E, Karbalai H, Buchan AM. Biphasic opening of the blood-brain barrier following transient focal ischemia: effects of hypothermia. Can J Neurol Sci. 1999; 26: 298–304.[Medline] [Order article via Infotrieve]
5. Lo EH, Singhal AB, Torchilin VP, Abbott NJ. Drug delivery to damaged brain. Brain Res Rev. 2001; 38: 140–148.[CrossRef][Medline] [Order article via Infotrieve]
6. Jiang Q, Zhang RL, Zhang ZG, Knight RA, Ewing JR, Ding G, Lu M, Arniego P, Zhang L, Hu J, Li Q, Chopp M. Magnetic resonance imaging characterization of hemorrhagic transformation of embolic stroke in the rat. J Cereb Blood Flow Metab. 2002; 22: 559–568.[Medline] [Order article via Infotrieve]
7. Fagan SC, Hess DC, Hohnadel EJ, Pollock DM, Ergul A. Targets for vascular protection after acute ischemic stroke. Stroke. 2004; 35: 2220–2225.
8. Lo EH, Moskowitz MA, Jacobs TP. Exciting, radical, suicidal: how brain cells die after stroke. Stroke. 2005; 36: 189–192.
9. Nagaraja TN, Keenan KA, Brown SL, Fenstermacher JD, Knight RA. Relative distribution of plasma flow markers and red blood cells across BBB openings in acute cerebral ischemia. Neurol Res. 2007; 29: 78–80.[Medline] [Order article via Infotrieve]
10. Preston E, Webster J. Differential passage of [14C]sucrose and [3H]inulin across rat blood-brain barrier after cerebral ischemia. Acta Neuropathol. 2002; 103: 237–242.[CrossRef][Medline] [Order article via Infotrieve]
11. Knight RA, Barker PB, Fagan SC, Li Y, Jacobs MA, Welch KM. Prediction of impending hemorrhagic transformation in ischemic stroke using magnetic resonance imaging in rats. Stroke. 1998; 29: 144–151.
12. Warach S, Latour LL. Evidence of reperfusion injury, exacerbated by thrombolytic therapy, in human focal brain ischemia using a novel imaging marker of early blood-brain barrier disruption. Stroke. 2004; 35 (suppl 1): 2659–2661.
13. Yenari M, Beaulieu C, Iwai K, Seri S, Steinberg G, Moseley ME Characterization of hemorrhagic embolic infarction using diffusion-weighted imaging. Fourth Scientific Meeting of the International Society for Magnetic Resonance in Medicine. 1996: 487. Abstract.
14. Ewing JR, Jiang Q, Boska M, Zhang L, Zhang ZG, Brown SL, Li GH, Divine GW, Chopp M. T1 and magnetization transfer at 7 Tesla in acute ischemic infarct in the rat. Magn Reson Med. 1999; 41: 696–705.[CrossRef][Medline] [Order article via Infotrieve]
15. Knight RA, Nagesh V, Nagaraja TN, Ewing JR, Whitton PA, Bershad E, Fagan SC, Fenstermacher JD. Acute BBB opening in experimentally induced focal cerebral ischemia is preferentially identified by quantitative magnetization transfer imaging. Magn Reson Med. 2005; 54: 822–832.[CrossRef][Medline] [Order article via Infotrieve]
16. Fagan SC, Nagaraja TN, Fenstermacher JD, Zheng J, Johnson M, Knight RA. Hemorrhagic transformation is related to the duration of occlusion and treatment with tissue plasminogen activator in a non-embolic stroke model. Neurol Res. 2003; 25: 377–382.[CrossRef][Medline] [Order article via Infotrieve]
17. Strich G, Hagan PL, Gerber KH, Slutsky RA. Tissue distribution and magnetic resonance spin lattice relaxation effects of gadolinium-DTPA. Radiology. 1985; 154: 723–726.
18. Nagaraja TN, Croxen RL, Panda S, Knight RA, Keenan KA, Brown SL, Fenstermacher JD, Ewing JR. Application of arsenazo III in the preparation and characterization of an albumin-linked, gadolinium-based macromolecular magnetic resonance contrast agent. J Neurosci Methods. 2006; 157: 238–245.[CrossRef][Medline] [Order article via Infotrieve]
19. Knight RA, Nagaraja TN, Ewing JR, Nagesh V, Whitton PA, Bershad E, Fagan SC, Fenstermacher JD. Quantitation and localization of blood-to-brain influx by MRI and quantitative autoradiography in a model of transient focal ischemia. Magn Reson Med. 2005; 54: 813–821.[CrossRef][Medline] [Order article via Infotrieve]
20. Ewing JR, Knight RA, Nagaraja TN, Yee JS, Nagesh V, Whitton PA, Li L, Fenstermacher JD. Patlak plots of Gd-DTPA MRI data yield blood-brain transfer constants concordant with those of 14C-sucrose in areas of blood-brain opening. Magn Reson Med. 2003; 50: 283–292.[CrossRef][Medline] [Order article via Infotrieve]
21. Ewing JR, Wei L, Knight RA, Pawa S, Nagaraja TN, Brusca T, Divine GW, Fenstermacher JD. Direct comparison of local cerebral blood flow rates measured by MRI arterial spin-tagging and quantitative autoradiography in a rat model of experimental cerebral ischemia. J Cereb Blood Flow Metab. 2003; 23: 198–209.[CrossRef][Medline] [Order article via Infotrieve]
22. Brix G, Schad LR, Deimling M, Lorenz WJ. Fast and precise T1 imaging using a TOMROP sequence. Magn Reson Imaging. 1990; 8: 351–356.[CrossRef][Medline] [Order article via Infotrieve]
23. Nagaraja TN, Nagesh V, Ewing JR, Whitton PA, Fenstermacher JD, Knight RA. Step-down infusions of Gd-DTPA yield greater contrast-enhanced magnetic resonance images of BBB damage in acute stroke than bolus injections. Magn Reson Imaging. 2007; 25: 311–318.[CrossRef][Medline] [Order article via Infotrieve]
24. Ding G, Nagesh V, Jiang Q, Zhang L, Zhang ZG, Li L, Knight RA, Li Q, Ewing JR, Chopp M. Early prediction of gross hemorrhagic transformation by noncontrast agent MRI cluster analysis after embolic stroke in rat. Stroke. 2005; 36: 1247–1252.
25. Preston E, Foster DO. Evidence for pore-like opening of the blood-brain barrier following forebrain ischemia in rats. Brain Res. 1997; 761: 4–10.[CrossRef][Medline] [Order article via Infotrieve]
26. Mayhan WG, Heistad DD. Permeability of blood-brain barrier to various sized molecules. Am J Physiol Heart Circ Physiol. 1985; 248: H712–H718.
27. Nicholson C, Sykova E. Extracellular space structure revealed by diffusion analysis. Trends Neurosci. 1998; 21: 207–215.[CrossRef][Medline] [Order article via Infotrieve]
28. Murakami K, Kawase M, Kondo T, Chan PH. Cellular accumulation of extravasated serum protein and DNA fragmentation following vasogenic edema. J Neurotrauma. 1998; 15: 825–835.[Medline] [Order article via Infotrieve]
29. Liu KF, Li F, Tatlisumak T, Garcia JH, Sotak CH, Fisher M, Fenstermacher JD. Regional variations in the apparent diffusion coefficient and the intracellular distribution of water in rat brain during acute focal ischemia. Stroke. 2001; 32: 1897–1905.
30. Rosenberg GS, Estrada EY, Dencoff JE. Matrix metalloproteinases and TIMPs are associated with blood-brain barrier opening after reperfusion in rat brains. Stroke. 1998; 29: 2189–2195.
31. Habgood MD, Bye N, Dziegielewska KM, Ek CJ, Lane MA, Potter A, Morganti-Kossman C, Saunders NR. Changes in blood-brain barrier permeability to large and small molecules following traumatic brain injury in mice. Eur J Neurosci. 2007; 25: 231–238.[Medline] [Order article via Infotrieve]
32. Lo EH, Pan Y, Matsumoto K, Kowall NW. Blood-brain barrier disruption in experimental focal ischemia: comparison between in vivo MRI and immunocytochemistry. Magn Reson Imaging. 1994; 12: 403–411.[CrossRef][Medline] [Order article via Infotrieve]
33. Jiang Q, Ewing JR, Ding GL, Zhang L, Zhang ZG, Li L, Whitton P, Lu M, Hu J, Li QJ, Knight RA, Chopp M. Quantitative evaluation of BBB permeability after embolic stroke in rat using MRI. J Cereb Blood Flow Metab. 2005; 25: 583–592.[CrossRef][Medline] [Order article via Infotrieve]
34. Kaur J, Zhao Z, Klein GM, Lo EH, Buchan AM. The neurotoxicity of tissue plasminogen activator? J Cereb Blood Flow Metab. 2004; 24: 945–963.[Medline] [Order article via Infotrieve]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||