Cerebral Blood Flow Restoration and Reperfusion Injury After Ultraviolet Laser–Facilitated Middle Cerebral Artery Recanalization in Rat Thrombotic Stroke
Background and Purpose— A reversible model of focal thrombotic stroke was developed in the rat and examined for histological evidence of reperfusion injury after clinically relevant times of recanalization.
Methods— The distal middle cerebral artery of 28 male Sprague-Dawley rats was occluded by 562-nm laser-driven photothrombosis for 0.5, 2, and 3 hours or permanently (each n=7) and was recanalized by 355-nm UV laser irradiation. Occlusive material was examined by transmission electron microscopy. Cortical cerebral blood flow was monitored by laser-Doppler flowmetry. Brain infarcts were examined histologically at 3 days.
Results— After occlusion, cortical cerebral blood flow was reduced to 33±4% of baseline for all groups and was restored to 82±9%, 75±3%, and 93±7% of baseline for the 0.5-, 2-, and 3-hour groups, respectively, following recanalization after 29±8, 38±20, and 70±33 minutes of UV laser irradiation. The thrombotic occlusion contained compactly aggregated platelets but no fibrin, with length (1.2 to 1.8 mm) proportional to the ischemic period. During recanalization, microchannels containing erythrocytes and scattered leukocytes and bordered by intact disaggregated platelets infiltrated the thrombus. Infarct volumes (mm3) at 3 days were 12±3 for the permanent case and 8±4, 24±3, and 30±9 for the 0.5-, 2-, and 3-hour cases, respectively, thus demonstrating reperfusion injury histologically in the latter 2 groups. No hemorrhage was seen.
Conclusions— UV laser–facilitated dissolution of a conventionally refractory platelet thrombus provides a novel and effective method for restoring blood flow without hemorrhagic complications during thrombotic stroke. This was the first observation of histologically confirmed reperfusion injury in such a model.
Animal models of thrombotic or thromboembolic stroke are becoming increasingly important in basic and preclinical studies of prospective therapies, especially those concerned with thrombolytic1 and, more recently, antiplatelet regimens.2 Occlusive platelet thrombi are of particular concern because they contain little or no fibrin3 and thus are extremely resistant to thrombolytic drugs such as the tissue plasminogen activators approved for acute stroke therapy.3–7⇓⇓⇓⇓ The prevention or removal of refractory platelet thrombi has attracted much interest, as shown by the increasing utilization of arginine-glycine-aspartic acid antagonists2,3,8–10⇓⇓⇓⇓ and thrombin inhibitors11 to inhibit or destabilize intraplatelet fibrinogen cross-links between platelet glycoprotein (GP) IIb-IIIa membrane receptors. Experimentally, such a fibrin-free white thrombus can be produced photochemically in arteries in vivo by a process termed laser-driven photothrombosis.12 Although we often have used this technique to occlude the distal middle cerebral artery (MCA) to simulate thrombotic stroke in the rat,13–15⇓⇓ studies of the consequences of reperfusion after clot lysis in these models have been very few, requiring special conditions. For example, “soft” platelet thrombi transiently occluding an MCA segment can be dissolved hemodynamically after local vasodilation with nimodipine.16
A search for a more focal method of vasodilation revealed that UV light exposure from a filtered arc lamp could relax preconstricted aortic rings prepared in vitro17 and also arteries in vivo by irradiation with pulsed 351-nm XeF18 or 248-nm KrF19 UV lasers. The mechanisms of dilation are attributed to photoacoustic alteration of smooth muscle contractile proteins by shock waves generated during erythrocyte vaporization at high laser-energy density20 or photolytic release of nitric oxide at low energy density,19 respectively. Accordingly, we examined the capability of a 355-nm pulsed, low energy-density UV laser to dilate the patent rat MCA and then to dilate and recanalize the occluded MCA after photothrombosis, concurrent with laser-Doppler measurements of cortical cerebral blood flow (CBF). Full recanalization was complicated by extensive deposits of occlusive material (examined by transmission electron microscopy [TEM]) adjacent to the primary platelet thrombus. Finally, we examined the distal MCA territory 3 days afterward for histological evidence of cerebral reperfusion injury.
Materials and Methods
All animal surgical and maintenance procedures were approved by the Animal Care and Use Committee of the University of Miami in accordance with US Public Health Service regulations. Briefly, male Sprague-Dawley rats (n=34) weighing 250 to 350 g were anesthetized with 2.5% halothane in a 70%/30% mixture of nitrous oxide and oxygen and were maintained on 1% halothane by means of mechanical ventilation. The left distal MCA was exposed via a 3-mm burr hole with the dura intact, essentially as previously described,13,14⇓ with head and body temperatures maintained at 36°C to 37°C. A second burr hole was created between the coronal and parietal sutures for cortical CBF measurement with a laser-Doppler blood perfusion monitor (model 4001, Perimed).
The distal MCA was visualized through an operating microscope (×33, Zeiss), and images of the vascular field were obtained with a 0.5-inch charge-coupled device (CCD) camera (Cohu), observed on a monitor, captured, stored, and analyzed for diameter measurements (160 images per time point) with the use of Image-Pro Plus edge-detection and image-averaging software (Media Cybernetics) interfaced to a dimension analyzer and a personal computer (C Squared).
Distal MCA Photothrombotic Occlusion Facilitated by Dye Laser Irradiation
Laser-driven photothrombosis of the distal MCA was administered as described previously,13 but only a single site ranging from 150 to 250 μm in diameter over all groups was irradiated to the point of occlusion, while all other arteries remained patent. Occlusion was generated within 3 to 5 minutes by the interaction of the 562-nm dye laser beam at powers of 3.5 to 6.5 mW interacting with the intravenously injected photosensitizing dye rose bengal (20 mg/kg) and was accompanied by prominent vasoconstriction and blanching of the distal MCA distally12 and somewhat proximally as well. The lengths of the blanched segments were minimized by using lower irradiation intensities (approximately 130 mW/mm2) than those used previously13–15⇓⇓ and (in the proximal case) by inducing the primary thrombus at a point distal to a well-perfused G1 branch.
Distal MCA Recanalization Facilitated by UV Laser Irradiation
A Q-switched, frequency-tripled neodymium:yttrium-aluminum-garnet (Nd:YAG) laser (Minilase II, New Wave Research) producing 5-ns pulses at a rate of 20 Hz and a UV wavelength of 355 nm was used to irradiate the distal MCA. Laser power was measured with a detector head (model 380101) interfaced to a power meter (model 365, Scientech). The UV laser beam was spatially and optically filtered, expanded by a ×10 beam telescope (model BXUV-4.0-10X, CVI Laser Corp) and then focused by a 25-cm focal length BK7 glass plano-convex lens onto the distal MCA, encompassing its diameter. Dilation of the patent distal MCA with the UV laser beam was examined quantitatively in a closed cranial (fused silica) window preparation (n=7).21
Recanalization of an occluded distal MCA was begun by low-intensity UV laser irradiation (approximately 20 mW/mm2) immediately proximal to the occlusion. This induced dilation of the adjacent thrombosed segment, and thin rivulets of fresh blood were observed to infiltrate the thrombus rapidly. The UV intensity was then increased until the infiltrating blood columns coalesced, indicating disappearance of thrombosed material and recanalization of the segment. The UV beam was then advanced distally (by approximately half the segment diameter), and the process was repeated until complete recanalization was observed visually and verified by normalization of cortical CBF. The duration of irradiation and laser power were then recorded. The primary thrombus segment was the most vasoconstricted and thus required more UV intensity and time. Exposure of dural vessels to UV irradiation was minimized to avoid hemorrhage.
Light Microscopic Analysis of Infarct Volume
Animals that underwent distal MCA occlusion were perfusion-fixed at 3 days in FAM (40% formaldehyde, glacial acetic acid, and methanol, 1:1:8 by volume), and their brains were embedded in paraffin from which 10-μm-thick sections were obtained at 25-μm intervals and stained by hematoxylin and eosin, as described previously.14 For calculating infarct volume, histological sections at 8 coronal levels were digitized at ×1 power by means of a CCD camera (Xillix Technologies Corp), or microscopic images were digitized by means of a Sony CCD color video camera, and data were recorded with an MCID image analysis system (Imaging Research, Inc). Total infarct volume was derived by numerical integration of sequential infarct areas according to Simpson’s rule, with correction for edema.22 Areas bordering the infarct epicenter were evaluated for incomplete infarction (a mixture of normal and necrotic neurons)13 at ×400 magnification.
Animals were transcardially perfused beginning with a 2-minute flush of isotonic saline delivered at a pressure equal to the rat’s mean arterial blood pressure, followed by 500 mL of fixative solution composed of 2% glutaraldehyde and 100 mmol/L sucrose in 0.05 mol/L Na2PO4 buffer. The brain was removed and placed overnight in fresh fixative chilled to 4°C. The desired distal MCA segment was carefully removed from the surface of the brain, placed into a 1% solution of OsO4 in 0.1 mol/L Na2PO4 buffer, and stored overnight at 4°C. The segment was then prepared for TEM essentially as described previously23 and was examined and photographed in a TEM (model CM-10, Philips Research Laboratories) operated at 60 or 80 kV.
Differences in infarct volume and CBF among the treatment groups were determined by ANOVA followed by Newman-Keuls test. All data are expressed as mean±SD.
Physiological variables are shown in Table 1. No significant changes were observed throughout the experiment.
Response of the Patent Distal MCA to UV Laser Irradiation
In the closed cranial window preparation, the patent distal MCA in the initial group of 7 rats was exposed for 2 minutes to 355-nm pulsed UV laser irradiation24; a typical response over time is shown in Figure 1. Dilations with respect to baseline of 64±21%, 64±18%, and 58±24% were observed after 5 seconds, 1 minute, and 2 minutes, respectively, and were essentially unchanged for 30 minutes afterward.
Response of the Distal MCA to Photothrombotic Occlusion
The distal MCA of 4 groups of rats (n=7 each) was then occluded photothrombotically, either permanently or for 0.5, 2, and 3 hours before induction of recanalization. A primary white thrombus formed at the site of dye laser irradiation (Figure 2A) concomitant with severe vasoconstriction to approximately 40% of the baseline diameter (Figure 2B and Table 2). Arterial blanching developed distal to the vasoconstricted region and less so proximally; the thrombus was aged for 30 minutes (Figure 2B) before fixation. The blanched regions contained mostly aggregated platelets with some isolated erythrocytes (Figure 2C). The length of this thrombosed material far exceeded that of the primary thrombus and was proportionate to the ischemic interval (r=0.99, P<0.001) (Table 2). Under TEM, isolated erythrocytes were loosely associated with the thrombus proximally (Figure 2D); they were densely aggregated and attached at the distal end (Figure 2E). In the constricted zone (not shown), the platelets were more compactly aggregated than those in the peripheral regions. No fibrin strands were detected throughout the entire thrombus.
Response of the Occluded Distal MCA to 355-nm UV Laser Irradiation
Recanalization of occluded segments aged for 0.5 to 3 hours was facilitated by 355-nm UV laser irradiation as described in Materials and Methods and was observed in every treated rat. Figure 3 depicts UV treatment of a thrombotic occlusion aged for 3 hours. The original diameter of the entire segment was recovered, but some mural thrombotic material appeared intermittently up to 2 hours after UV treatment. The average UV beam intensity at the onset of recanalization was 48±32 mW/mm2 (n=21). The irradiation time required for recanalization was proportionate to the age of the thrombus (r=0.96, P<0.001; Table 2).
The process of recanalization was examined structurally in another distal MCA segment after an occlusive thrombus was aged for 0.5 hour (Figure 4A) and then underwent UV treatment until blood appeared to infiltrate its proximal half (Figure 4B) and surrounded the thrombus concomitant with distal MCA dilation (Figure 4C). More interesting, however, was the penetration of blood directly into the thrombus owing to the formation of multiple internal channels congruent with and even beyond the UV beam location (Figure 4C). Under TEM, the borders of the proximal infiltrating channels are composed of platelets that appear to have separated cleanly from their cross-channel neighbors (Figures 5A and 5C). In distal channels, however, the borders may be diffuse owing to mechanical damage to platelet membranes (Figure 5B). The channels thus comprise a complicated network filled with erythrocytes, plasma, and occasional leukocytes. No degradation of the infiltrated erythrocytes under 355-nm UV irradiation was seen, thus indicating the absence of destructive photoacoustic effects, but vacuolation that had developed in smooth muscle cells in the primary photothrombosed segment was exacerbated (not shown). No vacuolation was seen in UV laser–irradiated patent segments.
Response of Cortical CBF to Occlusion and Subsequent Recanalization
The time histories of cortical CBF response to the permanent occlusion experiment and to the experiments based on occlusion for 0.5, 2, and 3 hours followed by recanalization are shown in Figure 6. Laser-Doppler cortical CBF was reduced to 30±3% of baseline for permanent occlusion and 36±4%, 37±2%, and 35±2% of baseline for the 0.5-, 2-, and 3-hour groups, respectively. The ischemic periods were followed by time intervals representing the average times for UV treatment for each group (Table 2) until resumption of cortical CBF to near-normal levels, amounting to 82±9%, 75±3%, and 93±7% of baseline for the 0.5-, 2-, and 3-hour groups, respectively, during 2 hours of observation after full recanalization.
Note that the true ischemic interval is the sum of the predesignated occlusion time, plus the time taken to recanalize the total occlusion length (Figure 6 and Table 2): approximately 1 hour for the 0.5-hour occlusion group, 2 hours 40 minutes for the 2-hour group, and 4 hours 10 minutes for the 3-hour group. Recovery of flow at the later times in a thrombotic animal model of stroke is quite unusual but is consonant with the time scale for stroke patients undergoing acute thrombolytic therapy.
Reperfusion Injury in the Distal MCA Territory
Brain infarcts in the 4 groups at 3 days appeared only in the ipsilateral hemisphere, were confined to the dorsolateral cerebral cortex, and displayed typical cellular alterations consistent with ischemic necrosis. Infarct volumes were 12±3 mm3 for the permanent cases and 8±4, 24±3, and 30±9 mm3 for the cases in which recanalization was begun after 0.5, 2, and 3 hours of thrombosis-induced ischemia (Figure 7). Because the infarct volumes for the 2- and 3-hour cases far exceeded that for the permanently occluded cases (P<0.001), reperfusion injury is thus demonstrated. No hemorrhages were observed. In addition, incomplete infarction13 was observed sporadically at the periphery of the infarct epicenter as a small percentage of its area. For the permanent, 0.5-, 2-, and 3-hour groups, this was 4.3%, 0.45%, 4.0%, and 1.5%, respectively. The frequency of occurrence in each group was 5 of 7, 1 of 7, 4 of 7, and 3 of 7, respectively.
These studies have revealed that (1) a cerebral artery occluded by an extensive, fibrin-free platelet thrombus can be recanalized by a novel laser method15 that instigates the formation of multiple, progressively enlarging channels in the thrombus interior and (2) CBF can be restored to nearly normal levels at clinically relevant times, only to engender reperfusion injury. This occlusive platelet thrombus is intractable to conventional lytic methods, but the UV laser method overcomes these limitations and thus provides an unusual and effective means to restore blood flow during thrombotic stroke. Direct UV irradiation of the stabilized platelet mass is evidently unnecessary, because the intrathrombus channels can still form distal to that point (Figure 4C). It may also be possible to recanalize cerebral arterial branches indirectly owing to the diffusible nature of the dilation effect (Figure 1). Finally, because cerebral ischemia in this model is induced by occlusive thrombosis only, collateral pathways are not compromised by concurrent mechanical ligature (eg, of the common carotid arteries) and are thus available as pharmacological conduits. This model of reversible thrombotic stroke is therefore reduced to its most essential aspects.
There is no established mechanism to explain how UV laser–induced dilation facilitates formation of the microscopic intrathrombus channels that lead to recanalization of a platelet-occluded artery (Figure 4C). This process nonetheless engenders a decrease in thrombus density and thus provides pathways for blood permeation along with enhancement of surface area available for plasma protein or exogenous drug interaction at enhanced efficacy. To illustrate, during channel formation the platelets appear to separate cleanly (Figure 5A and 5C) as if by dethrombosis.10,11⇓ This process presumes that platelets bound by platelet GPIIb-IIIa receptor-fibrinogen cross-links can be dissociated by thrombin inhibitor factors, provided that the thrombus contains channels for blood infiltration.11 Treatment with an antibody to the platelet GPIIb-IIIa receptor alone was correlated with channel formation in coronary thrombi in dogs and establishment of reflow in dogs and in humans as well,8 possibly owing to enlargement of preexisting seepage channels.11 Arterial flow pulsations may further erode the integrity of interplatelet binding. Mechanical25 and ultrasonic26 means have also been used to destabilize a thrombus to enhance the potency of thrombolytic agents (eg, recombinant tissue plasminogen activator). Alternatively, local intra-arterial infusion of urokinase achieved acceptable recanalization of the MCA trunk in up to 66% of clinical cases after 2 hours of treatment, although the frequency of symptomatic intracranial hemorrhage was 10%.6 Recently, a percutaneously transmitted high energy-density dye laser beam has been used clinically to disintegrate brain artery occlusions photoacoustically.27 With similar optical fiber technology, percutaneous UV laser irradiation at much lower energy densities may possibly augment or substitute for intra-arterial thrombolytic interventions. Finally, although platelet arginine-glycine–aspartic acid inhibitors have augmented thrombolytic therapy clinically7 and have suppressed platelet accumulation in ischemic microvessels in baboons,2 they cannot be used in the rat because a conformational change in the rat platelet GPIIb-IIIa receptor after binding of the inhibitor drug allows fibrinogen to bind as well.9 The UV laser method nonetheless facilitates the dissociation of intraplatelet binding forces, with apparently endogenous factors or their derivatives19,20⇓ that are insensitive to this limitation.
However, despite near-complete restoration of cortical CBF in the reversible thrombotic stroke groups, histological infarct volumes at 3 days after reperfusion at the relatively late absolute times of 3 to 4 hours were enhanced by 2- to 2.5-fold compared with the permanently occluded group. The chief criterion for reperfusion injury, ie, conversion of tissue from a reversible to an irreversible state of injury, is thus fulfilled.28,29⇓ Such direct evidence for reperfusion injury in thrombotic stroke has not been reported previously. Possible potentiators of thrombotic reperfusion injury include platelet secretions that evidently damage the blood-brain barrier downstream and presage severe morphological changes in neurons, astrocytes, and endothelium.16 The participation of acute inflammation is indicated by the intense platelet aggregation distal to the primary thrombus, which likely elicits a similar response in the ischemic microvasculature2 and also by the acute association of leukocytes30 with the recanalizing thrombus (Figure 5A). Previously, reperfusion injury was demonstrated in the brains of rats mechanically recirculated between 2 and 5 hours after dual-ligature occlusion (MCA and ipsilateral common carotid artery) as approximately 3-fold enhancement of infarct volume at 1 day compared with the permanently occluded group.28 Our cortical CBF results (Figure 5) indicate that reperfusion injury is potentiated by intermediate, but not severe, levels of ischemia in cerebral tissue28 and by delayed cortical CBF restoration29 and blood-brain barrier damage.16,29⇓ Identification of the processes contributing to reperfusion injury in our reversible model of thrombotic stroke now appears feasible, as does study of their eventual mitigation.
This work was supported by US Public Health Service grants RO1 NS23244 (Dr Watson) and RO1 NS27127 (Dr Dietrich) from the National Institute of Neurological Disorders and Stroke. The authors thank Rob Camarena for photography, the electron microscopy core of the Miami Project to Cure Paralysis for tissue analysis, Guillermo Hernandez for histology, David Smith and Weizhao Zhao for image processing, Professor F.W. Flitney for discussions on vasodilation, and Dr Per Wester for critical reading of the manuscript and specific suggestions for improvement. Technical information concerning UV laser–facilitated dissolution of platelet occlusions was included in a US patent application filed June 11, 2000, to preserve the possibility of its extension to clinical use.
- Received September 11, 2001.
- Revision received November 7, 2001.
- Accepted November 8, 2001.
- ↵Zivin JA, Fisher M, DeGirolami U, Hemenway CC, Stashak JA. Tissue plasminogen activator reduces neurological damage after cerebral embolism. Science. 1985; 230: 1289–1292.
- ↵Abumiya T, Fitridge R, Mazur C, Copeland BR, Koziol JA, Tschopp JF, Pierschbacher MD, del Zoppo GJ. Integrin αIIbβ3 inhibitor preserves microvascular patency in experimental acute focal cerebral ischemia. Stroke. 2000; 31: 1402–1409.
- ↵Topol EJ. Toward a new frontier in myocardial reperfusion therapy: emerging platelet preeminence. Circulation. 1998; 97: 211–218.
- ↵del Zoppo GJ, Poeck K, Pessin MS, Wolpert SM, Furlan AJ, Ferbert A, Alberts MJ, Zivin JA, Wechsler L, Busse O, Greenlee R Jr, Brass L, Mohr JP, Feldmann E, Hacke W, Kase CS, Biller J, Gress D, Otis SM. Recombinant tissue plasminogen activator in acute thrombotic and embolic stroke. Ann Neurol. 1992; 32: 78–86.
- ↵Furlan A, Higashida R, Wechsler L, Gent M, Rowley H, Kase C, Pessin M, Ahuja A, Callahan F, Clark WM, Silver F, Rivera F. Intra-arterial prourokinase for acute ischemic stroke: the PROACT II study: a randomized controlled trial: Prolyse in Acute Cerebral Thromboembolism. JAMA. 1999; 282: 2003–2011.
- ↵Gold HK, Garabedian HD, Dinsmore RE, Guerrero LJ, Cigarros JE, Palacios IF, Leinbach RC. Restoration of coronary flow in myocardial infarction by intravenous chimeric 7E3 antibody without exogenous plasminogen activators. Circulation. 1997; 95: 1755–1759.
- ↵Basani RB, D’Andrea G, Mitra N, Vilaire G, Richberg M, Kowalska MA, Bennet JS, Ponez M. RGD-containing peptides inhibit fibrinogen binding to platelet αIIbβ3 by inducing an allosteric change in the amino-terminal portion of αIIb. J Biol Chem. 2001; 276: 13975–13981.
- ↵Muhlestein JB. Fibrin-fibrinogen and dethrombosis. Blood Coag Fibrinol. 1999; 10: S59–S62.
- ↵Markgraf CG, Kraydieh S, Prado R, Watson BD, Dietrich WD, Ginsberg MD. Comparative histopathologic consequences of photothrombotic occlusion of the distal middle cerebral artery in Sprague-Dawley and Wistar rats. Stroke. 1993; 24: 286–293.
- ↵Watson BD, Prado R, Liu Y, Fernandez G. Reperfusion and its injury characteristics in an ultraviolet laser-mediated reversible model of photothrombotic stroke. J Cereb Blood Flow Metab. 1999; 19 (suppl 1): S579.Abstract.
- ↵Furchgott RF, Ehrreich SJ, Greenblatt E. The photoactivated relaxation of smooth muscle of rabbit aorta. J Gen Physiol. 1961; 44: 499–519.
- ↵Gal D, Chokshi SK, Mosseri M, Clarke RH, Isner JM. Percutaneous delivery of low-level laser energy reverses histamine-induced spasm in atherosclerotic Yucatan microswine. Circulation. 1992; 85: 756–768.
- ↵Watson BD, Prado R, Dietrich WD. Q-switched Nd:YAG laser irradiation at 355 nm mediates rapid focal dilation of rat middle cerebral artery. Soc Neurosci Abstr. 1998; 24(pt 1): 1171.Abstract.
- ↵Lutsep HL, Campbell M, Clark WM, Jansen O, Lefkowitz MA, Marks MP, Norbash A, Putnam C, Schumacher M. EPAR therapy system for treatment of acute stroke: safety study results. Stroke. 2001; 32: 319.Abstract.
- ↵Yang GY, Betz AL. Reperfusion-induced injury to the blood-brain barrier after middle cerebral artery occlusion in rats. Stroke. 1994; 25: 1658–1664.