Journal of Cerebral Blood Flow & Metabolism (2014) 34, 176–181 & 2014 ISCBFM All rights reserved 0271-678X/14 $32.00 www.jcbfm.com
ORIGINAL ARTICLE
Evolution of intracerebral hemorrhage after intravenous tPA: reversal of harmful effects with mast cell stabilization Ivan Marinkovic1,2, Olli S Mattila3, Daniel Strbian1,2, Atte Meretoja1,4, Shashank Shekhar1,2, Jani Saksi3, Usama Abo-Ramadan1,2, Ville Rantanen5, Perttu J Lindsberg1,3 and Turgut Tatlisumak1,2 Thrombolysis with tissue plasminogen activator (tPA) traditionally demands baseline imaging to rule out intracerebral hemorrhage (ICH), which causes delays in treatment. Preventing possible adverse effects of tPA on ICH would allow rapid on-site thrombolysis in patients with presumed acute ischemic stroke, reducing onset-to-treatment times. We examined how intravenous tPA alters ICH evolution during an extended follow-up, and how mast cell stabilization affects this process. Intracerebral hemorrhage was induced in rats by collagenase injection. Rats received either saline (n ¼ 10), tPA (n ¼ 13), tPA þ low-dose cromoglycate (n ¼ 10), or tPA þ high-dose cromoglycate (n ¼ 10). Magnetic resonance imaging was performed at 24, 48, and 72 hours after ICH induction, together with neurologic evaluations. During 72 hours of follow-up, tPA administration did not significantly increase hematoma volume (mean±s.d. 83.5±14.3 versus 66.7±14.7 mL; P ¼ 0.256) or hemispheric expansion (14.5±5.0 versus 11.5±5.0%; P ¼ 0.457) compared with saline. However, tPA-treated animals had worse neurologic outcomes (Po0.05), and mortality (8/13 versus 3/10). Combining tPA with high-dose cromoglycate mitigated hemispheric expansion (7.4±1.7 versus 14.5±5.0%; P ¼ 0.01), improved neurologic outcome (Po0.001) and decreased mortality (1/10; Po0.05) compared with tPA alone. Our results suggest tPA increases neurologic deficit in ICH, an effect that was abolished by concomitant mast cell stabilization. Further studies are needed to establish the clinical relevance of these findings. Journal of Cerebral Blood Flow & Metabolism (2014) 34, 176–181; doi:10.1038/jcbfm.2013.189; published online 30 October 2013 Keywords: collagenase injection; ICH; mast cells; rat; thrombolysis; tPA
INTRODUCTION Thrombolysis with intravenous recombinant tissue plasminogen activator (tPA) is the only approved medical treatment for acute ischemic stroke. Clinical efficacy of thrombolysis is highly dependent on onset-to-treatment times, with early treatment leading to significantly better outcomes.1 Intracerebral hemorrhage (ICH) must be ruled out before thrombolysis with computed tomography or magnetic resonance imaging, which inevitably leads to treatment delays. In theory, rapid thrombolysis without preceding imaging might be justified in patients with very high likelihood of ischemic stroke based on clinical history and signs, perhaps aided by novel point-of-care ICH biomarkers.2,3 However, at present, there is insufficient knowledge of how thrombolysis alters evolution of ICH to determine risks of such ‘blind’ thrombolysis. To this end, a recent experimental study in mice found that intravenous tPA had no effect on hematoma growth within the first 24 hours after experimental ICH.4 Mast cells (MC) are a class of resident inflammatory cells harboring a wide array of preformed inflammatory and fibrinolytic mediators that are released upon their activation.5 In brain tissue, MCs are located in strategic perivascular positions and are capable of inducing rapid opening of the blood–brain barrier (BBB), simultaneously promoting local inflammatory and hemorrhagic events.5 In earlier reports, our group has shown that MC activation
contributes to hemorrhage formation after thrombolysis in experimental ischemic stroke,6 and that MCs are central mediators of hematoma expansion and edema formation in experimental ICH.7 Using an autologous blood injection model of ICH, we found that intravenous cromoglycate (100 mg/kg) led to significantly better neurologic outcome, decreased mortality, less brain swelling, and smaller hematoma volume growth.7 In addition, tPA is capable of directly inducing MC activation in vitro.7 In the present study, we used the well-characterized collagenase injection model of ICH to determine: (1) how thrombolysis alters hematoma growth, hemispheric expansion, mortality, and neurologic function during a 72-hour follow-up in rats, and (2) whether these effects could be counteracted by intravenous administration of cromoglycate, an MC-stabilizing agent.
MATERIALS AND METHODS Animals All experiments were approved by the local goverment authorities (ELLA animal experiment board, Finland), and conducted in accordance with The Finnish Act on Animal Experimentation (62/2006). Adult male Wistar rats (Harlan Nederland, Horst, The Netherlands), weighing 300 to 350 g, were anesthetized using ketamine hydrochloride (intraperitoneal, 50 mg/kg, Ketalar, Pfizer, Helsinki, Finland, Parke-Davis, Sweden)
1 Department of Neurology, Helsinki University Central Hospital, Helsinki, Finland; 2Experimental MRI Laboratory, Biomedicum, Helsinki, Finland; 3Molecular Neurology, Research Programs Unit, and Department of Clinical Neurosciences, University of Helsinki, Helsinki, Finland; 4Department of Medicine, University of Melbourne, Melbourne, Victoria, Australia and 5Genome-Scale Biology Program, University of Helsinki, Helsinki, Finland. Correspondence: Dr I Marinkovic, Department of Neurology, Helsinki University Central Hospital, Haartmaninkatu 4, 00290 Helsinki, Finland. E-mail: ivan.marinkovic@hus.fi This work was supported by grants from Helsinki Biomedical Graduate School, The Academy of Finland, Sigrid Juse´lius Foundation, Paavo Nurmi Foundation and The Helsinki University Central Hospital (Government subsidiary research funding). Received 7 August 2013; revised 16 September 2013; accepted 5 October 2013; published online 30 October 2013
ICH evolution, intravenous tPA and mast cells I Marinkovic et al
177 and medetomidine hydrochloride (subcutaneously, 0.5 mg/kg, Domitor, Orion, Espoo, Finland). A PE-50 tube was inserted into the left femoral artery for monitoring of blood pressure (Olli BP Meter 533, Kone, Espoo, Finland) and blood gas analysis (subset of animals, AVL OPTI, Rotkreuz, Basel, Switzerland), and into the left femoral vein for intravenous drug administration. Rectal temperature was maintained at 371C by a heating pad during surgical procedures, and with a heating blanket and a thermoregulated heating lamp during magnetic resonance imaging. Body weight and general condition of the animals were observed daily. The analgesic drug buprenorphine (0.05 mg/kg, Temgesic, Schering-Plough, Kenilworth, NJ, USA) was administered every 8 hours after surgery for the first 48 hours. The rats were housed under diurnal lighting conditions and given free access to food and water.
Collagenase-Induced Intracerebral Hemorrhage To mimic ICH, we used the collagenase injection model, as it realistically models expansion of the cerebral hematoma, and is thought to be optimal for studies investigating treatments that effect hemostasis and hematoma development.8 (Supplementary Information is available at the Journal of Cerebral Blood Flow & Metabolism website—http://www.nature.com/ jcbfm). We chose the collagenase dose according to our previous observation that 0.037 IU of collagenase VII-s (Sigma-Aldrich, St Louis, MO, USA) results in hematomas of 70 to 80 mL at 24 hours, a hematoma size that leads to a mortality rate of approximately 50% at 3 days of followup, i.e. similar to the mortality of ICH in human patients.
Pharmacological Protocols After ICH induction, animals were assigned to four treatment groups. All three tPA groups received 10 mg/kg of intravenous tPA (Actilyse, Boehringer-Ingelheim, Ingelheim, Germany, 10% bolus (0.3 mL) and 30minute infusion (2.7 mL), n ¼ 33) immediately after ICH induction. The control group received only saline administered identically (intravenous 0.3 mL bolus and 2.7 mL infusion, n ¼ 10). The tPA þ low-dose cromoglycate group was administered 100 mg/kg of intravenous cromoglycate (Sigma-Aldrich, Steinheim, Germany, 1 mL), a MC stabilizer, directly before the tPA bolus (n ¼ 10), and the high-dose cromoglycate group received the same treatments, but was given a second dose of intravenous cromoglycate (100 mg/kg, 1 mL) 60 minutes later (n ¼ 10). Other groups received equivalent volumes of saline at these time points. The group size of 10 animals was chosen based on earlier reports, together with an increased group size of 13 for the tPA group to compensate for increased mortality, as suggested by pilot experiments.
Magnetic Resonance Imaging Presence of ICH was confirmed at 55 minutes after collagenase injection and thereafter followed up at 2, 24, 48, and 72 hours with magnetic resonance imaging using a 4.7T scanner (PharmaScan, Bruker BioSpin, Rheinstetten, Germany). Please see Supplementary Methods for details.
Calculation of Hematoma Volume and Hemispheric Expansion Hematoma volumes and hemispheric expansion were calculated at all time points from T2*-weighted magnetic resonance imaging images. The borders of the hematoma were outlined manually (Paravision, Bruker BioSpin), and resulting areas were multiplied by slice thickness and summed up to yield the total hematoma volume.7 Thereafter, the volume of both hemispheres was calculated in an identical manner for measurements of hemispheric expansion. The percentage of the hemispheric expansion was calculated as the volumetric increase of the ICH hemisphere compared with the intact hemisphere (% of hemispheric expansion ¼ ((right hemisphere volume/left hemisphere volume)– 1) 100).9,10 Analyses were performed masked to treatment allocation and neurologic outcome.
Neurologic Evaluation and Mortality We scored neurologic performance at days 1, 2, and 3 after ICH induction on an 8-point scale (0: no deficit; 1: contralateral forepaw paresis; 2: 1 þ decreased resistance to lateral push, yet no circling; 3: 2 þ circling to the contralateral side; 4: falling to the contralateral side; 5: rolling; 6: no spontaneous walking with a depressed level of consciousness; 7: death).10 & 2014 ISCBFM
Behavioral Evaluation The assessment of behavioral status was performed based on the SHIRPA protocol11 from video records collected at 24, 48, and 72 hours after ICH (see Supplementary Methods).
Tissue Processing Seventy-two hours after ICH, surviving animals were terminated using 60 mg of sodium barbiturate (intrapertioneally, 60 mg, pentobarbital). After cardiac perfusion (200 mL of ice-cold saline) the brains were dissected into 2 mm thick coronal slices, one slice rostrally and three slices caudally from the site of intracerebral collagenase injection. Each second slice was cut into two 1-mm portions, and the rostral part was embedded into TissueTek (Sakura Finetek, Tokyo, Japan), snap-frozen in liquid nitrogen, and kept thereafter at 801C until 8 mm sections were cut.
In Situ Zymography and Immunohistochemistry Tissue activity of proteolytic gelatinase enzymes (MMP-2 and MMP-9) was visualized using in situ zymography (ISZ) as described previously12 (see Supplementary Methods for details). In situ zymography sections were further stained for the endothelial cell marker von Willebrand factor, and to evaluate microvascular basal lamina disruption, adjacent sections were stained for Collagen IV. We used primary rabbit polyclonal antibodies from Abcam (Cambridge, UK, 19.5 mg/mL and 0.63 mg/mL, respectively) together with Alexafluor secondary antibodies (Invitrogen, Carlsbad, CA, USA). Equivalent concentrations of negative control rabbit immunoglobulin G (Vector Labs, Burlingame, CA, USA) showed no specific immunostaining. For image analysis of the stained sections, four predefined regions of interest were acquired from the perihematomal area and from respective areas of the contralateral hemisphere at 20 magnification with an epifluorescent Axioplan2 microscope (Carl Zeiss, Oberkochen, Germany). Image acquisition was performed masked to treatment group.
Image Analysis Quantification of microvascular properties was performed using the Anduril workflow framework (http://www.csbi.ltdk.helsinki.fi/anduril/) as described before12,13 (see Supplementary Methods).
Statistical Analyses Data are expressed as mean±s.d. of the mean for parametric data, as medians for nonparametric data (also reported individually for each animal), or as proportions. As data sets were nonnormally distributed, comparison of hematoma volumes and hemispheric expansion was performed at each time point using Kruskal–Wallis analysis of variance followed by the Dunn post hoc test. This test was also used for comparing neurologic outcomes and microvascular properties between groups. Kaplan–Meier survival rates were compared with the log-rank test. Pearson correlation determined dependence between variables. A two-tailed Po0.05 was considered significant. Analyses were performed using SPSS v.20.0 (IBM, Armonk, NY, USA).
RESULTS Physiologic Parameters The groups showed no significant differences in the mean arterial blood pressure, pO2, pCO2, or rectal temperature before or after ICH induction, or in body weight at any time point during followup (data not shown). Intracerebral Hemorrhage Volumes The collagenase injection model induced ICH in the basal ganglia, clearly visible in T2*-weighted MR images (Figure 1). No statistically significant differences were seen in hematoma volume between groups at any time point. Adjuvant treatment with cromoglycate did not affect hematoma development (Figure 2). Surviving animals in all groups showed resorption of the hematoma after 24 hours (Supplementary Figure 1). Journal of Cerebral Blood Flow & Metabolism (2014), 176 – 181
ICH evolution, intravenous tPA and mast cells I Marinkovic et al
178 2h
24h
48h
72h
Saline
tPA
tPA + low-dose cromoglycate
tPA + high-dose cromoglycate
180 160
Representative T2*-weighted magnetic resonance images from each group during follow-up.
25
Saline tPA tPA + low-dose cromoglycate tPA + high-dose cromoglycate
Hemispheric expansion (%)
Figure 1.
Hematoma volume (l)
140 120 100 80 60 40
20
Saline tPA tPA + low-dose cromoglycate tPA + high-dose cromoglycate
** * *
**
15 ** 10 NS 5
20 0
0 55min
2h
24h
48h
72h
Figure 2. Development of hematoma volume during follow-up measured from T2*-images. All groups showed equal hematoma volumes at the beginning of the experiment. The most prominent hematoma volumes were present at 24 hours. No significant differences were seen during follow-up (Kruskal–Wallis analysis of variance between-group differences P ¼ 0.523 at 55 minutes, P ¼ 0.189 at 2 hours, P ¼ 0.199 at 24 hours, P ¼ 0.247 at 48 hours, and P ¼ 0.256 at 72 hours). Data as mean±s.d. tPA, tissue plasminogen activator.
Hemispheric Expansion There were no intergroup differences in hemispheric expansion 55 minutes after ICH induction (Figure 3). Treatment with tPA led to statistically nonsignificant increases in hemispheric expansion beginning at 2 hours, whereas both doses of cromoglycate resulted in significantly reduced hemispheric expansion (Figure 3, at 24 hours 11.6±4.8% for saline; 14.1±4.2% for tPA; 7.1±3.3% for tPA þ low-dose cromoglycate; 7.3±3.0% for tPA þ high-dose cromoglycate; post hoc saline versus tPA, P ¼ 0.335; tPA versus tPA þ low-dose cromoglycate, P ¼ 0.003; tPA versus tPA þ high-dose cromoglycate, P ¼ 0.004; saline versus tPA þ low-dose cromoglycate, P ¼ 0.046; saline versus tPA þ high-dose cromoglycate, P ¼ 0.064). Importantly, the effect of high-dose cromoJournal of Cerebral Blood Flow & Metabolism (2014), 176 – 181
55min
2h
24h
48h
72h
Figure 3. Evolution of hemispheric expansion after intracerebral hemorrahge (ICH). No differences were seen 55 minutes after ICH induction. Differences between saline and tissue plasminogen activator (tPA) groups were not statistically different at any time point, whereas cromoglycate groups showed significant reduction in hemispheric expansion throughout follow-up. Kruskal–Wallis analysis of variance between-group differences: P ¼ 0.474 at 55 minutes, P ¼ 0.016 at 2 hours, P ¼ 0.005 at 24 hours, P ¼ 0.029 at 48 hours, and P ¼ 0.041 at 72 hours; post hoc comparisons: * ¼ Po0.05, ** ¼ Po0.01; significant differences were also seen between saline and tPA þ low-dose cromoglycate groups at 24 hours, and saline and tPA þ high-dose cromoglycate groups at 72 hours (Po0.05, not shown in figure). Data as mean±s.d.
glycate on hemispheric expansion lasted longer, and showed a significant reduction compared with both tPA and saline groups at the end of follow-up (Figure 3, at 72 hours 11.5±5.0% for saline; 14.5±5.0% for tPA; 8.0±2.2% for tPA þ low-dose cromoglycate; 7.4±1.7% for tPA þ high-dose cromoglycate; post hoc saline versus tPA, P ¼ 0.457; tPA versus tPA þ low-dose cromoglycate, P ¼ 0.076; tPA versus tPA þ high-dose cromoglycate, P ¼ 0.010; saline versus tPA þ low-dose cromoglycate, P ¼ 0.252; saline versus tPA þ high-dose cromoglycate, P ¼ 0.046). & 2014 ISCBFM
ICH evolution, intravenous tPA and mast cells I Marinkovic et al
179
Mortality No mortality occurred within the first 2 hours after ICH. At the end of follow-up, mortality was 30% (3/10) in the saline-treated group and 62% (8/13) in the tPA-only group (Figure 5). Co-administration of cromoglycate mitigated the effect of tPA, with a mortality of 40% (4/10) in the low-dose group and 10% (1/10) in the high-dose group (Figure 5). Comparison of survival rates showed a statistically significant difference between tPA and tPA þ high-dose cromoglycate-treated groups (P ¼ 0.018, log-rank test).
Behavioral Outcome Overall, behavioral scores were worst at 24 hours, with improvement towards 72 hours (Supplementary Figure 3). At 3 days of follow-up, the tPA group showed impaired scores in most tests compared with the other groups, with cromoglycate treatment partly reversing these detrimental effects of tPA.
Microvascular Gelatinase Activity and Basal Lamina Disruption At 72 hours after ICH, tPA-treated groups showed strong in situ zymography activation in both neurons and the microvasculature together with marked loss of collagen IV-positive vessels in the perihematomal area (Supplementary Figure 4). However, no significant differences were seen between the treatment groups in these microvascular properties at this late time point (Supplementary Figure 4, Kruskal–Wallis analysis of variance between-group differences P ¼ 0.21 for gelatinase activity, and P ¼ 0.81 for collagen IV).
DISCUSSION Currently only B5% of acute ischemic stroke patients receive thrombolytic treatment in the United States, largely because of the strict therapeutic time window of 4.5 hours and fear of complications. One potential strategy for increasing availability of thrombolysis could be rapid on-site treatment without prior radiologic exclusion of hemorrhagic stroke in patients deemed to have low probability of primary ICH based on clinical grounds and point-of-care biomarkers.2,3 Importantly, this would eliminate the delays caused by patient transfer to the nearest stroke center and imaging studies.14 Our prior experiments showing beneficial effects of mast cell stabilization in experimental stroke5–7,15 led us to hypothesize that adjuvant treatment with a mast cell stabilizer might improve the safety profile of on-site thrombolysis in case of primary ICH. In line with recent results,4 tPA did not significantly increase hematoma size. However, tPA resulted in detrimental neurologic outcomes during the extended follow-up. Remarkably, co-administration of high-dose cromoglycate, a mast cell stabilizer, led to improved neurologic outcome and showed significantly reduced mortality compared with tPA alone, demonstrating that cromoglycate can reverse adverse effects of tPA in this setting. 100 tPA + high-dose cromoglycate
80 Saline
Survival (%)
Neurologic Outcome No statistically significant differences in neurologic outcome were seen between groups at 24 hours (Figure 4). However, at the end of follow-up, tPA administration resulted in a significantly impaired neurologic outcome, an effect that was fully reversed by high-dose mast cell stabilization (Figure 4; at 72 hours; post hoc saline versus tPA, P ¼ 0.040; tPA versus tPA þ low-dose cromoglycate, P ¼ 0.076; tPA versus tPA þ high-dose cromoglycate, Po0.001). No statistically significant differences in neurologic outcomes were observed between cromoglycate groups compared with saline.
60
tPA + low-dose cromoglycate
40
tPA
20
0 0
24
48
72
Time (hours)
Figure 5. Kaplan–Meier survival curves for each treatment group. The lowest survival rate was present in the tissue plasminogen activator (tPA) only group, whereas it significantly improved by applying cromoglycate, most notably in the high-dose group, where all except one animal survived the whole follow-up period.
*** ***
*
*
7
Neurological score
6 5 3 2 1 0 tPA Saline
24h
tPA + tPA tPA + tPA tPA + high-dose Saline high-dose Saline high-dose cromoglycate cromoglycate cromoglycate tPA + tPA + tPA + 48h 72h low-dose low-dose low-dose cromoglycate cromoglycate cromoglycate
Figure 4. Neurologic scores during follow-up. Gray bars represent median values. Kruskal–Wallis analysis of variance between-group differences: P ¼ 0.053 at 24 hours, P ¼ 0.006 at 48 hours, and P ¼ 0.002 at 72 hours; post hoc comparisons: * ¼ Po0.05, *** ¼ Po0.001. & 2014 ISCBFM
Journal of Cerebral Blood Flow & Metabolism (2014), 176 – 181
ICH evolution, intravenous tPA and mast cells I Marinkovic et al
180 In humans, spontaneous ICH initiates with rapid growth of the hematoma that leads to a damaging mass effect on the surrounding tissue.16 Typically, the majority of hematoma expansion develops early after ictus, with further bleeding occurring in one-third of patients between 1 and 24 hours.17 The collagenase injection model replicated these events (Figure 2), as described previously.8 The fibrinolytic effect of tPA is mediated by florid plasminogen activation and subsequent fibrinolysis, which can result in systemic hypofibrinogenemia, possibly preventing stabilization of the cerebral hematoma.18 In addition, although plasma half-life of tPA is short (t1/2 6 minutes), tPA and plasmin bound to clot fibrin are thought to remain active several hours after treatment.19 Although we administered intravenous tPA immediately after ICH induction to maximize leakage of tPA into the developing hematoma before complete coagulation, there were no significant differences in hematoma volume with this sample size. Our results therefore support the recent notion that intravenous tPA may not break down a sufficient volume of hematoma to cause excess bleeding.4 In comparison, a recent study using a collagenase injection model in mice with equal sample size found that intravenous tPA administered 30 minutes after ICH induction increased hematoma volumes by one-half,20 although the study used a very high dose of collagenase (0.05 IU in mice), which is likely to cause greater vascular damage and increased leakage of tPA into the hematoma (we used 0.037 IU in rats), possibly explaining the differences in results. Interestingly, a recent report from Pfeilschifter et al21 demonstrates that leakage of the BBB marker Evans blue into brain tissue is greatest early (30 minutes) after collagenase-injection induced ICH in mice. This supports the importance of the early time point in our study, and is significant in a clinical sense, as thrombolysis is still rarely administered within the first hour of symptom onset. Perihematomal edema develops in three overlapping phases after ICH,22 beginning as hematoma growth pushes serum into the surrounding tissues.23 The second phase of perihematomal edema is initiated by clotting of the hematoma and release of coagulation factors,24 most importantly thrombin,25 which induce BBB opening and subsequent vasogenic edema within the first few days after ICH.26 Together with the hematoma growth, perihematomal edema can lead to expansion of the affected hemisphere and finally to fatal herniation of vital brain structures. The third and final phase of edema development is caused by erythrocyte degradation products, but occurs late, beyond the follow-up of the present study.22 In theory, several known actions of tPA could lead to increased BBB disruption and subsequently increased hematoma growth, edema formation and subsequent hemispheric expansion in ICH. Interestingly, systemic thrombolysis has been shown to trigger significant paradoxical thrombin activity, peaking 90 minutes after drug administration.27 Furthermore, tPA-induced fibrinolysis is known to release active thrombin from within the clot during lysis.28 Both of these mechanisms could consequently increase thrombin activity within and around the cerebral hematoma, increasing thrombin-mediated BBB opening and brain swelling. Studies of ischemic stroke have revealed several other direct pathways for tPA-induced BBB disruption, which include activation of proteolytic gelatinase enzymes,29 LRP-1,30 PDGFC,31 NMDA-receptor-mediated excitotoxicity,20 and degranulation of MCs.5 Suprisingly, regardless of these known detrimental mechanisms of tPA, our data do not support a significant effect of tPA on ICH-induced hemispheric expansion, an aspect not studied in earlier reports investigating effects of tPA in ICH. Perivascularly positioned inflammatory MCs have been shown to induce rapid opening of the BBB during ischemic and hemorrhagic brain injury via release of a wide array of vasoactive mediators.5 In line with this finding, we have previously shown that MC stabilization with cromoglycate treatment reduces brain swelling by 39% in a middle cerebral artery occlusion model of Journal of Cerebral Blood Flow & Metabolism (2014), 176 – 181
ischemic stroke.15 Similarly, in an autologous blood-injection model of ICH, cromoglycate significantly reduced hematoma growth and brain swelling.7 As tPA has been shown to aggravate MC activation in experimental ischemic stroke,6 we postulated that MCs might also participate in progression of ICH complicated by tPA. Indeed, MC stabilization with cromoglycate significantly inhibited the second phase of perihematomal edema, with both cromoglycate-treated groups showing similar reductions in hemispheric expansion, without changes in hematoma volume. Although the reduction of hemispheric expansion was similar in both cromoglycate-treated groups, the high-dose group showed a superior survival rate and neurologic outcome, suggesting that mast cell stabilization might also have other protective functions in addition to prevention of MC-mediated BBB damage. We attempted to further demonstrate these vasoprotective effects by determining gelatinase activity and concomitant collagen IV degradation of the perihematomal microvasculature in surviving animals at 72 hours, but no significant differences were found at this late time point. At present, we are not able to comment on the relative participation of different intracranial mast cell populations (including meningeal, cortical, and basal ganglia MC) on the observed effects. However, it seems likely that basal ganglia MC of the ipsilateral hemisphere are fully activated early on, leaving the other mast cell populations to activate at later stages of ICH progression. Our study has limitations. The experiments used stereotypic timing of tPA infusion, although in the real world, it is variable within the 4.5 hours time window. As we have only studied ICH, further studies are needed to determine the effect of intravenous tPA and adjuvant cromoglycate treatment on other stroke mimics, such as sub- and epidural hematomas, subarachnoid hemorrhage, and postepileptic conditions, to allow full risk evaluation of ‘blind’ thrombolysis. Importantly, further studies need to demonstrate that the nominal, therapeutic efficacy of tPA remains unaffected when administered with adjuvant mast cell stabilizer. We chose the collagenase injection model as it mimics gradual expansion of the cerebral hematoma,8 and is therefore a suitable starting point for monitoring pharmacological effects on hematoma growth. The findings should be replicated in several additional models to gain credibility to this hypothesis before any transition to clinical settings is possible. In conclusion, our study outlines the evolution of experimental ICH in rats complicated by intravenous tPA, highlighting a detrimental effect on neurologic outcome after longer follow-up. Importantly, this adverse effect of tPA was readily counteracted by cromoglycate treatment. Our results propose the highly provocative paradigm that mast cell stabilization could sufficiently mitigate the risks associated with immediate on-site thrombolysis, thus potentially increasing the availability of effective treatment for acute ischemic stroke. Additional studies are needed to establish clinical safety and applicability of these results. DISCLOSURE/CONFLICT OF INTEREST PJL, TT, and DS are joint patent holders of US Patent No:8,163,734 ‘Use of a mast cell activation or degranulation blocking agent in the manufacture of a medicament for the treatment of a patient subjected to thrombolyses.’ The remaining authors declare no conflict of interest.
ACKNOWLEDGMENTS We thank Taru Puhakka and Nancy Lim for technical assistance.
REFERENCES 1 Lees KR, Bluhmki E, Kummer von R, Brott TG, Toni D, Grotta JC et al. Time to treatment with intravenous alteplase and outcome in stroke: an updated pooled analysis of ECASS, ATLANTIS, NINDS, and EPITHET trials. Lancet 2010; 375: 1695–1703.
& 2014 ISCBFM
ICH evolution, intravenous tPA and mast cells I Marinkovic et al
181 2 Runchey S, McGee S. Does this patient have a hemorrhagic stroke?: clinical findings distinguishing hemorrhagic stroke from ischemic stroke. JAMA 2010; 303: 2280–2286. 3 Foerch C, Niessner M, Back T, Bauerle M, De Marchis GM, Ferbert A et al. Diagnostic accuracy of plasma glial fibrillary acidic protein for differentiating intracerebral hemorrhage and cerebral ischemia in patients with symptoms of acute stroke. Clin Chem 2012; 58: 237–245. 4 Foerch C, Rosidi NL, Schlunk F, Lauer A, Cianchetti FA, Mandeville E et al. Intravenous tPA therapy does not worsen acute intracerebral hemorrhage in mice. PLoS ONE 2013; 8: e54203. 5 Lindsberg PJ, Strbian D, Karjalainen-Lindsberg M-L. Mast cells as early responders in the regulation of acute blood-brain barrier changes after cerebral ischemia and hemorrhage. J Cereb Blood Flow Metab 2010; 30: 689–702. 6 Strbian D, Karjalainen-Lindsberg M-L, Kovanen PT, Tatlisumak T, Lindsberg PJ. Mast cell stabilization reduces hemorrhage formation and mortality after administration of thrombolytics in experimental ischemic stroke. Circulation 2007; 116: 411–418. 7 Strbian D, Tatlisumak T, Ramadan UA, Lindsberg PJ. Mast cell blocking reduces brain edema and hematoma volume and improves outcome after experimental intracerebral hemorrhage. J Cereb Blood Flow Metab 2007; 27: 795–802. 8 MacLellan CL, Silasi G, Poon CC, Edmundson CL, Buist R, Peeling J et al. Intracerebral hemorrhage models in rat: comparing collagenase to blood infusion. J Cereb Blood Flow Metab 2008; 28: 516–525. 9 Gerriets T, Stolz E, Walberer M, Mu¨ller C, Kluge A, Bachmann A et al. Noninvasive quantification of brain edema and the space-occupying effect in rat stroke models using magnetic resonance imaging. Stroke 2004; 35: 566–571. 10 Marinkovic I, Strbian D, Pedrono E, Vekovischeva OY, Shekhar S, Durukan A et al. Decompressive craniectomy for intracerebral hemorrhage. Neurosurgery 2009; 65: 780–786. 11 Rogers DC, Fisher EM, Brown SD, Peters J, Hunter AJ, Martin JE. Behavioral and functional analysis of mouse phenotype: SHIRPA, a proposed protocol for comprehensive phenotype assessment. Mamm Genome 1997; 8: 711–713. 12 Mattila OS, Strbian D, Saksi J, Pikkarainen TO, Rantanen V, Tatlisumak T et al. Cerebral mast cells mediate blood-brain barrier disruption in acute experimental ischemic stroke through perivascular gelatinase activation. Stroke 2011; 42: 3600–3605. 13 Ovaska K, Laakso M, Haapa-Paananen S, Louhimo R, Chen P, Aittoma¨ki V et al. Large-scale data integration framework provides a comprehensive view on glioblastoma multiforme. Genome Med 2010; 2: 65. 14 Weber JE, Ebinger M, Rozanski M, Waldschmidt C, Wendt M, Winter B et al. Prehospital thrombolysis in acute stroke: results of the PHANTOM-S pilot study. Neurology 2013; 80: 163–168. 15 Strbian D, Karjalainen-Lindsberg M-L, Tatlisumak T, Lindsberg PJ. Cerebral mast cells regulate early ischemic brain swelling and neutrophil accumulation. J Cereb Blood Flow Metab 2006; 26: 605–612. 16 Zazulia AR, Diringer MN, Derdeyn CP, Powers WJ. Progression of mass effect after intracerebral hemorrhage. Stroke 1999; 30: 1167–1173.
17 Brott T, Broderick J, Kothari R, Barsan W, Tomsick T, Sauerbeck L et al. Early hemorrhage growth in patients with intracerebral hemorrhage. Stroke 1997; 28: 1–5. 18 Sloan MA, Price TR, Petito CK, Randall AM, Solomon RE, Terrin ML et al. Clinical features and pathogenesis of intracerebral hemorrhage after rt-PA and heparin therapy for acute myocardial infarction: the Thrombolysis in Myocardial Infarction (TIMI) II Pilot and Randomized Clinical Trial combined experience. Neurology 1995; 45: 649–658. 19 Eisenberg PR, Sherman LA, Tiefenbrunn AJ, Ludbrook PA, Sobel BE, Jaffe AS. Sustained fibrinolysis after administration of t-PA despite its short half-life in the circulation. Thromb Haemost 1987; 57: 35–40. 20 Montagne A, He´bert M, Jullienne A, Lesept F, Le Be´hot A, Louessard M et al. Memantine improves safety of thrombolysis for stroke. Stroke 2012; 43: 2774–2781. 21 Pfeilschifter W, Kashefiolasl S, Lauer A, Steinmetz H, Foerch C. Hematoma expansion in experimental intracerebral hemorrhage is not altered by peracute treatment with recombinant tissue plasminogen activator. Neuroscience 2013; 250: 181–188. 22 Xi G, Keep RF, Hoff JT. Mechanisms of brain injury after intracerebral haemorrhage. Lancet Neurol 2006; 5: 53–63. 23 Wagner KR, Xi G, Hua Y, Kleinholz M, de Courten-Myers GM, Myers RE et al. Lobar intracerebral hemorrhage model in pigs: rapid edema development in perihematomal white matter. Stroke 1996; 27: 490–497. 24 Xi G, Wagner KR, Keep RF, Hua Y, de Courten-Myers GM, Broderick JP et al. Role of blood clot formation on early edema development after experimental intracerebral hemorrhage. Stroke 1998; 29: 2580–2586. 25 Lee KR, Kawai N, Kim S, Sagher O, Hoff JT. Mechanisms of edema formation after intracerebral hemorrhage: effects of thrombin on cerebral blood flow, blood-brain barrier permeability, and cell survival in a rat model. J Neurosurg 1997; 86: 272–278. 26 Yang GY, Betz AL, Chenevert TL, Brunberg JA, Hoff JT. Experimental intracerebral hemorrhage: relationship between brain edema, blood flow, and blood-brain barrier permeability in rats. J Neurosurg 1994; 81: 93–102. 27 Merlini PA, Bauer KA, Oltrona L, Ardissino D, Spinola A, Cattaneo M et al. Thrombin generation and activity during thrombolysis and concomitant heparin therapy in patients with acute myocardial infarction. J Am Coll Cardiol 1995; 25: 203–209. 28 Francis CW, Markham RE, Barlow GH, Florack TM, Dobrzynski DM, Marder VJ. Thrombin activity of fibrin thrombi and soluble plasmic derivatives. J Lab Clin Med 1983; 102: 220–230. 29 Tsuji K, Aoki T, Tejima E, Arai K, Lee S-R, Atochin DN et al. Tissue plasminogen activator promotes matrix metalloproteinase-9 upregulation after focal cerebral ischemia. Stroke 2005; 36: 1954–1959. 30 Wang X, Lee S-R, Arai K, Lee S-R, Tsuji K, Rebeck GW et al. Lipoprotein receptormediated induction of matrix metalloproteinase by tissue plasminogen activator. Nat Med 2003; 9: 1313–1317. 31 Su EJ, Fredriksson L, Geyer M, Folestad E, Cale J, Andrae J et al. Activation of PDGFCC by tissue plasminogen activator impairs blood-brain barrier integrity during ischemic stroke. Nat Med 2008; 14: 731–737.
Supplementary Information accompanies the paper on the Journal of Cerebral Blood Flow & Metabolism website (http:// www.nature.com/jcbfm)
& 2014 ISCBFM
Journal of Cerebral Blood Flow & Metabolism (2014), 176 – 181
Copyright of Journal of Cerebral Blood Flow & Metabolism is the property of Nature Publishing Group and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
ORIGINAL ARTICLE
Evolution of intracerebral hemorrhage after intravenous tPA: reversal of harmful effects with mast cell stabilization Ivan Marinkovic1,2, Olli S Mattila3, Daniel Strbian1,2, Atte Meretoja1,4, Shashank Shekhar1,2, Jani Saksi3, Usama Abo-Ramadan1,2, Ville Rantanen5, Perttu J Lindsberg1,3 and Turgut Tatlisumak1,2 Thrombolysis with tissue plasminogen activator (tPA) traditionally demands baseline imaging to rule out intracerebral hemorrhage (ICH), which causes delays in treatment. Preventing possible adverse effects of tPA on ICH would allow rapid on-site thrombolysis in patients with presumed acute ischemic stroke, reducing onset-to-treatment times. We examined how intravenous tPA alters ICH evolution during an extended follow-up, and how mast cell stabilization affects this process. Intracerebral hemorrhage was induced in rats by collagenase injection. Rats received either saline (n ¼ 10), tPA (n ¼ 13), tPA þ low-dose cromoglycate (n ¼ 10), or tPA þ high-dose cromoglycate (n ¼ 10). Magnetic resonance imaging was performed at 24, 48, and 72 hours after ICH induction, together with neurologic evaluations. During 72 hours of follow-up, tPA administration did not significantly increase hematoma volume (mean±s.d. 83.5±14.3 versus 66.7±14.7 mL; P ¼ 0.256) or hemispheric expansion (14.5±5.0 versus 11.5±5.0%; P ¼ 0.457) compared with saline. However, tPA-treated animals had worse neurologic outcomes (Po0.05), and mortality (8/13 versus 3/10). Combining tPA with high-dose cromoglycate mitigated hemispheric expansion (7.4±1.7 versus 14.5±5.0%; P ¼ 0.01), improved neurologic outcome (Po0.001) and decreased mortality (1/10; Po0.05) compared with tPA alone. Our results suggest tPA increases neurologic deficit in ICH, an effect that was abolished by concomitant mast cell stabilization. Further studies are needed to establish the clinical relevance of these findings. Journal of Cerebral Blood Flow & Metabolism (2014) 34, 176–181; doi:10.1038/jcbfm.2013.189; published online 30 October 2013 Keywords: collagenase injection; ICH; mast cells; rat; thrombolysis; tPA
INTRODUCTION Thrombolysis with intravenous recombinant tissue plasminogen activator (tPA) is the only approved medical treatment for acute ischemic stroke. Clinical efficacy of thrombolysis is highly dependent on onset-to-treatment times, with early treatment leading to significantly better outcomes.1 Intracerebral hemorrhage (ICH) must be ruled out before thrombolysis with computed tomography or magnetic resonance imaging, which inevitably leads to treatment delays. In theory, rapid thrombolysis without preceding imaging might be justified in patients with very high likelihood of ischemic stroke based on clinical history and signs, perhaps aided by novel point-of-care ICH biomarkers.2,3 However, at present, there is insufficient knowledge of how thrombolysis alters evolution of ICH to determine risks of such ‘blind’ thrombolysis. To this end, a recent experimental study in mice found that intravenous tPA had no effect on hematoma growth within the first 24 hours after experimental ICH.4 Mast cells (MC) are a class of resident inflammatory cells harboring a wide array of preformed inflammatory and fibrinolytic mediators that are released upon their activation.5 In brain tissue, MCs are located in strategic perivascular positions and are capable of inducing rapid opening of the blood–brain barrier (BBB), simultaneously promoting local inflammatory and hemorrhagic events.5 In earlier reports, our group has shown that MC activation
contributes to hemorrhage formation after thrombolysis in experimental ischemic stroke,6 and that MCs are central mediators of hematoma expansion and edema formation in experimental ICH.7 Using an autologous blood injection model of ICH, we found that intravenous cromoglycate (100 mg/kg) led to significantly better neurologic outcome, decreased mortality, less brain swelling, and smaller hematoma volume growth.7 In addition, tPA is capable of directly inducing MC activation in vitro.7 In the present study, we used the well-characterized collagenase injection model of ICH to determine: (1) how thrombolysis alters hematoma growth, hemispheric expansion, mortality, and neurologic function during a 72-hour follow-up in rats, and (2) whether these effects could be counteracted by intravenous administration of cromoglycate, an MC-stabilizing agent.
MATERIALS AND METHODS Animals All experiments were approved by the local goverment authorities (ELLA animal experiment board, Finland), and conducted in accordance with The Finnish Act on Animal Experimentation (62/2006). Adult male Wistar rats (Harlan Nederland, Horst, The Netherlands), weighing 300 to 350 g, were anesthetized using ketamine hydrochloride (intraperitoneal, 50 mg/kg, Ketalar, Pfizer, Helsinki, Finland, Parke-Davis, Sweden)
1 Department of Neurology, Helsinki University Central Hospital, Helsinki, Finland; 2Experimental MRI Laboratory, Biomedicum, Helsinki, Finland; 3Molecular Neurology, Research Programs Unit, and Department of Clinical Neurosciences, University of Helsinki, Helsinki, Finland; 4Department of Medicine, University of Melbourne, Melbourne, Victoria, Australia and 5Genome-Scale Biology Program, University of Helsinki, Helsinki, Finland. Correspondence: Dr I Marinkovic, Department of Neurology, Helsinki University Central Hospital, Haartmaninkatu 4, 00290 Helsinki, Finland. E-mail: ivan.marinkovic@hus.fi This work was supported by grants from Helsinki Biomedical Graduate School, The Academy of Finland, Sigrid Juse´lius Foundation, Paavo Nurmi Foundation and The Helsinki University Central Hospital (Government subsidiary research funding). Received 7 August 2013; revised 16 September 2013; accepted 5 October 2013; published online 30 October 2013
ICH evolution, intravenous tPA and mast cells I Marinkovic et al
177 and medetomidine hydrochloride (subcutaneously, 0.5 mg/kg, Domitor, Orion, Espoo, Finland). A PE-50 tube was inserted into the left femoral artery for monitoring of blood pressure (Olli BP Meter 533, Kone, Espoo, Finland) and blood gas analysis (subset of animals, AVL OPTI, Rotkreuz, Basel, Switzerland), and into the left femoral vein for intravenous drug administration. Rectal temperature was maintained at 371C by a heating pad during surgical procedures, and with a heating blanket and a thermoregulated heating lamp during magnetic resonance imaging. Body weight and general condition of the animals were observed daily. The analgesic drug buprenorphine (0.05 mg/kg, Temgesic, Schering-Plough, Kenilworth, NJ, USA) was administered every 8 hours after surgery for the first 48 hours. The rats were housed under diurnal lighting conditions and given free access to food and water.
Collagenase-Induced Intracerebral Hemorrhage To mimic ICH, we used the collagenase injection model, as it realistically models expansion of the cerebral hematoma, and is thought to be optimal for studies investigating treatments that effect hemostasis and hematoma development.8 (Supplementary Information is available at the Journal of Cerebral Blood Flow & Metabolism website—http://www.nature.com/ jcbfm). We chose the collagenase dose according to our previous observation that 0.037 IU of collagenase VII-s (Sigma-Aldrich, St Louis, MO, USA) results in hematomas of 70 to 80 mL at 24 hours, a hematoma size that leads to a mortality rate of approximately 50% at 3 days of followup, i.e. similar to the mortality of ICH in human patients.
Pharmacological Protocols After ICH induction, animals were assigned to four treatment groups. All three tPA groups received 10 mg/kg of intravenous tPA (Actilyse, Boehringer-Ingelheim, Ingelheim, Germany, 10% bolus (0.3 mL) and 30minute infusion (2.7 mL), n ¼ 33) immediately after ICH induction. The control group received only saline administered identically (intravenous 0.3 mL bolus and 2.7 mL infusion, n ¼ 10). The tPA þ low-dose cromoglycate group was administered 100 mg/kg of intravenous cromoglycate (Sigma-Aldrich, Steinheim, Germany, 1 mL), a MC stabilizer, directly before the tPA bolus (n ¼ 10), and the high-dose cromoglycate group received the same treatments, but was given a second dose of intravenous cromoglycate (100 mg/kg, 1 mL) 60 minutes later (n ¼ 10). Other groups received equivalent volumes of saline at these time points. The group size of 10 animals was chosen based on earlier reports, together with an increased group size of 13 for the tPA group to compensate for increased mortality, as suggested by pilot experiments.
Magnetic Resonance Imaging Presence of ICH was confirmed at 55 minutes after collagenase injection and thereafter followed up at 2, 24, 48, and 72 hours with magnetic resonance imaging using a 4.7T scanner (PharmaScan, Bruker BioSpin, Rheinstetten, Germany). Please see Supplementary Methods for details.
Calculation of Hematoma Volume and Hemispheric Expansion Hematoma volumes and hemispheric expansion were calculated at all time points from T2*-weighted magnetic resonance imaging images. The borders of the hematoma were outlined manually (Paravision, Bruker BioSpin), and resulting areas were multiplied by slice thickness and summed up to yield the total hematoma volume.7 Thereafter, the volume of both hemispheres was calculated in an identical manner for measurements of hemispheric expansion. The percentage of the hemispheric expansion was calculated as the volumetric increase of the ICH hemisphere compared with the intact hemisphere (% of hemispheric expansion ¼ ((right hemisphere volume/left hemisphere volume)– 1) 100).9,10 Analyses were performed masked to treatment allocation and neurologic outcome.
Neurologic Evaluation and Mortality We scored neurologic performance at days 1, 2, and 3 after ICH induction on an 8-point scale (0: no deficit; 1: contralateral forepaw paresis; 2: 1 þ decreased resistance to lateral push, yet no circling; 3: 2 þ circling to the contralateral side; 4: falling to the contralateral side; 5: rolling; 6: no spontaneous walking with a depressed level of consciousness; 7: death).10 & 2014 ISCBFM
Behavioral Evaluation The assessment of behavioral status was performed based on the SHIRPA protocol11 from video records collected at 24, 48, and 72 hours after ICH (see Supplementary Methods).
Tissue Processing Seventy-two hours after ICH, surviving animals were terminated using 60 mg of sodium barbiturate (intrapertioneally, 60 mg, pentobarbital). After cardiac perfusion (200 mL of ice-cold saline) the brains were dissected into 2 mm thick coronal slices, one slice rostrally and three slices caudally from the site of intracerebral collagenase injection. Each second slice was cut into two 1-mm portions, and the rostral part was embedded into TissueTek (Sakura Finetek, Tokyo, Japan), snap-frozen in liquid nitrogen, and kept thereafter at 801C until 8 mm sections were cut.
In Situ Zymography and Immunohistochemistry Tissue activity of proteolytic gelatinase enzymes (MMP-2 and MMP-9) was visualized using in situ zymography (ISZ) as described previously12 (see Supplementary Methods for details). In situ zymography sections were further stained for the endothelial cell marker von Willebrand factor, and to evaluate microvascular basal lamina disruption, adjacent sections were stained for Collagen IV. We used primary rabbit polyclonal antibodies from Abcam (Cambridge, UK, 19.5 mg/mL and 0.63 mg/mL, respectively) together with Alexafluor secondary antibodies (Invitrogen, Carlsbad, CA, USA). Equivalent concentrations of negative control rabbit immunoglobulin G (Vector Labs, Burlingame, CA, USA) showed no specific immunostaining. For image analysis of the stained sections, four predefined regions of interest were acquired from the perihematomal area and from respective areas of the contralateral hemisphere at 20 magnification with an epifluorescent Axioplan2 microscope (Carl Zeiss, Oberkochen, Germany). Image acquisition was performed masked to treatment group.
Image Analysis Quantification of microvascular properties was performed using the Anduril workflow framework (http://www.csbi.ltdk.helsinki.fi/anduril/) as described before12,13 (see Supplementary Methods).
Statistical Analyses Data are expressed as mean±s.d. of the mean for parametric data, as medians for nonparametric data (also reported individually for each animal), or as proportions. As data sets were nonnormally distributed, comparison of hematoma volumes and hemispheric expansion was performed at each time point using Kruskal–Wallis analysis of variance followed by the Dunn post hoc test. This test was also used for comparing neurologic outcomes and microvascular properties between groups. Kaplan–Meier survival rates were compared with the log-rank test. Pearson correlation determined dependence between variables. A two-tailed Po0.05 was considered significant. Analyses were performed using SPSS v.20.0 (IBM, Armonk, NY, USA).
RESULTS Physiologic Parameters The groups showed no significant differences in the mean arterial blood pressure, pO2, pCO2, or rectal temperature before or after ICH induction, or in body weight at any time point during followup (data not shown). Intracerebral Hemorrhage Volumes The collagenase injection model induced ICH in the basal ganglia, clearly visible in T2*-weighted MR images (Figure 1). No statistically significant differences were seen in hematoma volume between groups at any time point. Adjuvant treatment with cromoglycate did not affect hematoma development (Figure 2). Surviving animals in all groups showed resorption of the hematoma after 24 hours (Supplementary Figure 1). Journal of Cerebral Blood Flow & Metabolism (2014), 176 – 181
ICH evolution, intravenous tPA and mast cells I Marinkovic et al
178 2h
24h
48h
72h
Saline
tPA
tPA + low-dose cromoglycate
tPA + high-dose cromoglycate
180 160
Representative T2*-weighted magnetic resonance images from each group during follow-up.
25
Saline tPA tPA + low-dose cromoglycate tPA + high-dose cromoglycate
Hemispheric expansion (%)
Figure 1.
Hematoma volume (l)
140 120 100 80 60 40
20
Saline tPA tPA + low-dose cromoglycate tPA + high-dose cromoglycate
** * *
**
15 ** 10 NS 5
20 0
0 55min
2h
24h
48h
72h
Figure 2. Development of hematoma volume during follow-up measured from T2*-images. All groups showed equal hematoma volumes at the beginning of the experiment. The most prominent hematoma volumes were present at 24 hours. No significant differences were seen during follow-up (Kruskal–Wallis analysis of variance between-group differences P ¼ 0.523 at 55 minutes, P ¼ 0.189 at 2 hours, P ¼ 0.199 at 24 hours, P ¼ 0.247 at 48 hours, and P ¼ 0.256 at 72 hours). Data as mean±s.d. tPA, tissue plasminogen activator.
Hemispheric Expansion There were no intergroup differences in hemispheric expansion 55 minutes after ICH induction (Figure 3). Treatment with tPA led to statistically nonsignificant increases in hemispheric expansion beginning at 2 hours, whereas both doses of cromoglycate resulted in significantly reduced hemispheric expansion (Figure 3, at 24 hours 11.6±4.8% for saline; 14.1±4.2% for tPA; 7.1±3.3% for tPA þ low-dose cromoglycate; 7.3±3.0% for tPA þ high-dose cromoglycate; post hoc saline versus tPA, P ¼ 0.335; tPA versus tPA þ low-dose cromoglycate, P ¼ 0.003; tPA versus tPA þ high-dose cromoglycate, P ¼ 0.004; saline versus tPA þ low-dose cromoglycate, P ¼ 0.046; saline versus tPA þ high-dose cromoglycate, P ¼ 0.064). Importantly, the effect of high-dose cromoJournal of Cerebral Blood Flow & Metabolism (2014), 176 – 181
55min
2h
24h
48h
72h
Figure 3. Evolution of hemispheric expansion after intracerebral hemorrahge (ICH). No differences were seen 55 minutes after ICH induction. Differences between saline and tissue plasminogen activator (tPA) groups were not statistically different at any time point, whereas cromoglycate groups showed significant reduction in hemispheric expansion throughout follow-up. Kruskal–Wallis analysis of variance between-group differences: P ¼ 0.474 at 55 minutes, P ¼ 0.016 at 2 hours, P ¼ 0.005 at 24 hours, P ¼ 0.029 at 48 hours, and P ¼ 0.041 at 72 hours; post hoc comparisons: * ¼ Po0.05, ** ¼ Po0.01; significant differences were also seen between saline and tPA þ low-dose cromoglycate groups at 24 hours, and saline and tPA þ high-dose cromoglycate groups at 72 hours (Po0.05, not shown in figure). Data as mean±s.d.
glycate on hemispheric expansion lasted longer, and showed a significant reduction compared with both tPA and saline groups at the end of follow-up (Figure 3, at 72 hours 11.5±5.0% for saline; 14.5±5.0% for tPA; 8.0±2.2% for tPA þ low-dose cromoglycate; 7.4±1.7% for tPA þ high-dose cromoglycate; post hoc saline versus tPA, P ¼ 0.457; tPA versus tPA þ low-dose cromoglycate, P ¼ 0.076; tPA versus tPA þ high-dose cromoglycate, P ¼ 0.010; saline versus tPA þ low-dose cromoglycate, P ¼ 0.252; saline versus tPA þ high-dose cromoglycate, P ¼ 0.046). & 2014 ISCBFM
ICH evolution, intravenous tPA and mast cells I Marinkovic et al
179
Mortality No mortality occurred within the first 2 hours after ICH. At the end of follow-up, mortality was 30% (3/10) in the saline-treated group and 62% (8/13) in the tPA-only group (Figure 5). Co-administration of cromoglycate mitigated the effect of tPA, with a mortality of 40% (4/10) in the low-dose group and 10% (1/10) in the high-dose group (Figure 5). Comparison of survival rates showed a statistically significant difference between tPA and tPA þ high-dose cromoglycate-treated groups (P ¼ 0.018, log-rank test).
Behavioral Outcome Overall, behavioral scores were worst at 24 hours, with improvement towards 72 hours (Supplementary Figure 3). At 3 days of follow-up, the tPA group showed impaired scores in most tests compared with the other groups, with cromoglycate treatment partly reversing these detrimental effects of tPA.
Microvascular Gelatinase Activity and Basal Lamina Disruption At 72 hours after ICH, tPA-treated groups showed strong in situ zymography activation in both neurons and the microvasculature together with marked loss of collagen IV-positive vessels in the perihematomal area (Supplementary Figure 4). However, no significant differences were seen between the treatment groups in these microvascular properties at this late time point (Supplementary Figure 4, Kruskal–Wallis analysis of variance between-group differences P ¼ 0.21 for gelatinase activity, and P ¼ 0.81 for collagen IV).
DISCUSSION Currently only B5% of acute ischemic stroke patients receive thrombolytic treatment in the United States, largely because of the strict therapeutic time window of 4.5 hours and fear of complications. One potential strategy for increasing availability of thrombolysis could be rapid on-site treatment without prior radiologic exclusion of hemorrhagic stroke in patients deemed to have low probability of primary ICH based on clinical grounds and point-of-care biomarkers.2,3 Importantly, this would eliminate the delays caused by patient transfer to the nearest stroke center and imaging studies.14 Our prior experiments showing beneficial effects of mast cell stabilization in experimental stroke5–7,15 led us to hypothesize that adjuvant treatment with a mast cell stabilizer might improve the safety profile of on-site thrombolysis in case of primary ICH. In line with recent results,4 tPA did not significantly increase hematoma size. However, tPA resulted in detrimental neurologic outcomes during the extended follow-up. Remarkably, co-administration of high-dose cromoglycate, a mast cell stabilizer, led to improved neurologic outcome and showed significantly reduced mortality compared with tPA alone, demonstrating that cromoglycate can reverse adverse effects of tPA in this setting. 100 tPA + high-dose cromoglycate
80 Saline
Survival (%)
Neurologic Outcome No statistically significant differences in neurologic outcome were seen between groups at 24 hours (Figure 4). However, at the end of follow-up, tPA administration resulted in a significantly impaired neurologic outcome, an effect that was fully reversed by high-dose mast cell stabilization (Figure 4; at 72 hours; post hoc saline versus tPA, P ¼ 0.040; tPA versus tPA þ low-dose cromoglycate, P ¼ 0.076; tPA versus tPA þ high-dose cromoglycate, Po0.001). No statistically significant differences in neurologic outcomes were observed between cromoglycate groups compared with saline.
60
tPA + low-dose cromoglycate
40
tPA
20
0 0
24
48
72
Time (hours)
Figure 5. Kaplan–Meier survival curves for each treatment group. The lowest survival rate was present in the tissue plasminogen activator (tPA) only group, whereas it significantly improved by applying cromoglycate, most notably in the high-dose group, where all except one animal survived the whole follow-up period.
*** ***
*
*
7
Neurological score
6 5 3 2 1 0 tPA Saline
24h
tPA + tPA tPA + tPA tPA + high-dose Saline high-dose Saline high-dose cromoglycate cromoglycate cromoglycate tPA + tPA + tPA + 48h 72h low-dose low-dose low-dose cromoglycate cromoglycate cromoglycate
Figure 4. Neurologic scores during follow-up. Gray bars represent median values. Kruskal–Wallis analysis of variance between-group differences: P ¼ 0.053 at 24 hours, P ¼ 0.006 at 48 hours, and P ¼ 0.002 at 72 hours; post hoc comparisons: * ¼ Po0.05, *** ¼ Po0.001. & 2014 ISCBFM
Journal of Cerebral Blood Flow & Metabolism (2014), 176 – 181
ICH evolution, intravenous tPA and mast cells I Marinkovic et al
180 In humans, spontaneous ICH initiates with rapid growth of the hematoma that leads to a damaging mass effect on the surrounding tissue.16 Typically, the majority of hematoma expansion develops early after ictus, with further bleeding occurring in one-third of patients between 1 and 24 hours.17 The collagenase injection model replicated these events (Figure 2), as described previously.8 The fibrinolytic effect of tPA is mediated by florid plasminogen activation and subsequent fibrinolysis, which can result in systemic hypofibrinogenemia, possibly preventing stabilization of the cerebral hematoma.18 In addition, although plasma half-life of tPA is short (t1/2 6 minutes), tPA and plasmin bound to clot fibrin are thought to remain active several hours after treatment.19 Although we administered intravenous tPA immediately after ICH induction to maximize leakage of tPA into the developing hematoma before complete coagulation, there were no significant differences in hematoma volume with this sample size. Our results therefore support the recent notion that intravenous tPA may not break down a sufficient volume of hematoma to cause excess bleeding.4 In comparison, a recent study using a collagenase injection model in mice with equal sample size found that intravenous tPA administered 30 minutes after ICH induction increased hematoma volumes by one-half,20 although the study used a very high dose of collagenase (0.05 IU in mice), which is likely to cause greater vascular damage and increased leakage of tPA into the hematoma (we used 0.037 IU in rats), possibly explaining the differences in results. Interestingly, a recent report from Pfeilschifter et al21 demonstrates that leakage of the BBB marker Evans blue into brain tissue is greatest early (30 minutes) after collagenase-injection induced ICH in mice. This supports the importance of the early time point in our study, and is significant in a clinical sense, as thrombolysis is still rarely administered within the first hour of symptom onset. Perihematomal edema develops in three overlapping phases after ICH,22 beginning as hematoma growth pushes serum into the surrounding tissues.23 The second phase of perihematomal edema is initiated by clotting of the hematoma and release of coagulation factors,24 most importantly thrombin,25 which induce BBB opening and subsequent vasogenic edema within the first few days after ICH.26 Together with the hematoma growth, perihematomal edema can lead to expansion of the affected hemisphere and finally to fatal herniation of vital brain structures. The third and final phase of edema development is caused by erythrocyte degradation products, but occurs late, beyond the follow-up of the present study.22 In theory, several known actions of tPA could lead to increased BBB disruption and subsequently increased hematoma growth, edema formation and subsequent hemispheric expansion in ICH. Interestingly, systemic thrombolysis has been shown to trigger significant paradoxical thrombin activity, peaking 90 minutes after drug administration.27 Furthermore, tPA-induced fibrinolysis is known to release active thrombin from within the clot during lysis.28 Both of these mechanisms could consequently increase thrombin activity within and around the cerebral hematoma, increasing thrombin-mediated BBB opening and brain swelling. Studies of ischemic stroke have revealed several other direct pathways for tPA-induced BBB disruption, which include activation of proteolytic gelatinase enzymes,29 LRP-1,30 PDGFC,31 NMDA-receptor-mediated excitotoxicity,20 and degranulation of MCs.5 Suprisingly, regardless of these known detrimental mechanisms of tPA, our data do not support a significant effect of tPA on ICH-induced hemispheric expansion, an aspect not studied in earlier reports investigating effects of tPA in ICH. Perivascularly positioned inflammatory MCs have been shown to induce rapid opening of the BBB during ischemic and hemorrhagic brain injury via release of a wide array of vasoactive mediators.5 In line with this finding, we have previously shown that MC stabilization with cromoglycate treatment reduces brain swelling by 39% in a middle cerebral artery occlusion model of Journal of Cerebral Blood Flow & Metabolism (2014), 176 – 181
ischemic stroke.15 Similarly, in an autologous blood-injection model of ICH, cromoglycate significantly reduced hematoma growth and brain swelling.7 As tPA has been shown to aggravate MC activation in experimental ischemic stroke,6 we postulated that MCs might also participate in progression of ICH complicated by tPA. Indeed, MC stabilization with cromoglycate significantly inhibited the second phase of perihematomal edema, with both cromoglycate-treated groups showing similar reductions in hemispheric expansion, without changes in hematoma volume. Although the reduction of hemispheric expansion was similar in both cromoglycate-treated groups, the high-dose group showed a superior survival rate and neurologic outcome, suggesting that mast cell stabilization might also have other protective functions in addition to prevention of MC-mediated BBB damage. We attempted to further demonstrate these vasoprotective effects by determining gelatinase activity and concomitant collagen IV degradation of the perihematomal microvasculature in surviving animals at 72 hours, but no significant differences were found at this late time point. At present, we are not able to comment on the relative participation of different intracranial mast cell populations (including meningeal, cortical, and basal ganglia MC) on the observed effects. However, it seems likely that basal ganglia MC of the ipsilateral hemisphere are fully activated early on, leaving the other mast cell populations to activate at later stages of ICH progression. Our study has limitations. The experiments used stereotypic timing of tPA infusion, although in the real world, it is variable within the 4.5 hours time window. As we have only studied ICH, further studies are needed to determine the effect of intravenous tPA and adjuvant cromoglycate treatment on other stroke mimics, such as sub- and epidural hematomas, subarachnoid hemorrhage, and postepileptic conditions, to allow full risk evaluation of ‘blind’ thrombolysis. Importantly, further studies need to demonstrate that the nominal, therapeutic efficacy of tPA remains unaffected when administered with adjuvant mast cell stabilizer. We chose the collagenase injection model as it mimics gradual expansion of the cerebral hematoma,8 and is therefore a suitable starting point for monitoring pharmacological effects on hematoma growth. The findings should be replicated in several additional models to gain credibility to this hypothesis before any transition to clinical settings is possible. In conclusion, our study outlines the evolution of experimental ICH in rats complicated by intravenous tPA, highlighting a detrimental effect on neurologic outcome after longer follow-up. Importantly, this adverse effect of tPA was readily counteracted by cromoglycate treatment. Our results propose the highly provocative paradigm that mast cell stabilization could sufficiently mitigate the risks associated with immediate on-site thrombolysis, thus potentially increasing the availability of effective treatment for acute ischemic stroke. Additional studies are needed to establish clinical safety and applicability of these results. DISCLOSURE/CONFLICT OF INTEREST PJL, TT, and DS are joint patent holders of US Patent No:8,163,734 ‘Use of a mast cell activation or degranulation blocking agent in the manufacture of a medicament for the treatment of a patient subjected to thrombolyses.’ The remaining authors declare no conflict of interest.
ACKNOWLEDGMENTS We thank Taru Puhakka and Nancy Lim for technical assistance.
REFERENCES 1 Lees KR, Bluhmki E, Kummer von R, Brott TG, Toni D, Grotta JC et al. Time to treatment with intravenous alteplase and outcome in stroke: an updated pooled analysis of ECASS, ATLANTIS, NINDS, and EPITHET trials. Lancet 2010; 375: 1695–1703.
& 2014 ISCBFM
ICH evolution, intravenous tPA and mast cells I Marinkovic et al
181 2 Runchey S, McGee S. Does this patient have a hemorrhagic stroke?: clinical findings distinguishing hemorrhagic stroke from ischemic stroke. JAMA 2010; 303: 2280–2286. 3 Foerch C, Niessner M, Back T, Bauerle M, De Marchis GM, Ferbert A et al. Diagnostic accuracy of plasma glial fibrillary acidic protein for differentiating intracerebral hemorrhage and cerebral ischemia in patients with symptoms of acute stroke. Clin Chem 2012; 58: 237–245. 4 Foerch C, Rosidi NL, Schlunk F, Lauer A, Cianchetti FA, Mandeville E et al. Intravenous tPA therapy does not worsen acute intracerebral hemorrhage in mice. PLoS ONE 2013; 8: e54203. 5 Lindsberg PJ, Strbian D, Karjalainen-Lindsberg M-L. Mast cells as early responders in the regulation of acute blood-brain barrier changes after cerebral ischemia and hemorrhage. J Cereb Blood Flow Metab 2010; 30: 689–702. 6 Strbian D, Karjalainen-Lindsberg M-L, Kovanen PT, Tatlisumak T, Lindsberg PJ. Mast cell stabilization reduces hemorrhage formation and mortality after administration of thrombolytics in experimental ischemic stroke. Circulation 2007; 116: 411–418. 7 Strbian D, Tatlisumak T, Ramadan UA, Lindsberg PJ. Mast cell blocking reduces brain edema and hematoma volume and improves outcome after experimental intracerebral hemorrhage. J Cereb Blood Flow Metab 2007; 27: 795–802. 8 MacLellan CL, Silasi G, Poon CC, Edmundson CL, Buist R, Peeling J et al. Intracerebral hemorrhage models in rat: comparing collagenase to blood infusion. J Cereb Blood Flow Metab 2008; 28: 516–525. 9 Gerriets T, Stolz E, Walberer M, Mu¨ller C, Kluge A, Bachmann A et al. Noninvasive quantification of brain edema and the space-occupying effect in rat stroke models using magnetic resonance imaging. Stroke 2004; 35: 566–571. 10 Marinkovic I, Strbian D, Pedrono E, Vekovischeva OY, Shekhar S, Durukan A et al. Decompressive craniectomy for intracerebral hemorrhage. Neurosurgery 2009; 65: 780–786. 11 Rogers DC, Fisher EM, Brown SD, Peters J, Hunter AJ, Martin JE. Behavioral and functional analysis of mouse phenotype: SHIRPA, a proposed protocol for comprehensive phenotype assessment. Mamm Genome 1997; 8: 711–713. 12 Mattila OS, Strbian D, Saksi J, Pikkarainen TO, Rantanen V, Tatlisumak T et al. Cerebral mast cells mediate blood-brain barrier disruption in acute experimental ischemic stroke through perivascular gelatinase activation. Stroke 2011; 42: 3600–3605. 13 Ovaska K, Laakso M, Haapa-Paananen S, Louhimo R, Chen P, Aittoma¨ki V et al. Large-scale data integration framework provides a comprehensive view on glioblastoma multiforme. Genome Med 2010; 2: 65. 14 Weber JE, Ebinger M, Rozanski M, Waldschmidt C, Wendt M, Winter B et al. Prehospital thrombolysis in acute stroke: results of the PHANTOM-S pilot study. Neurology 2013; 80: 163–168. 15 Strbian D, Karjalainen-Lindsberg M-L, Tatlisumak T, Lindsberg PJ. Cerebral mast cells regulate early ischemic brain swelling and neutrophil accumulation. J Cereb Blood Flow Metab 2006; 26: 605–612. 16 Zazulia AR, Diringer MN, Derdeyn CP, Powers WJ. Progression of mass effect after intracerebral hemorrhage. Stroke 1999; 30: 1167–1173.
17 Brott T, Broderick J, Kothari R, Barsan W, Tomsick T, Sauerbeck L et al. Early hemorrhage growth in patients with intracerebral hemorrhage. Stroke 1997; 28: 1–5. 18 Sloan MA, Price TR, Petito CK, Randall AM, Solomon RE, Terrin ML et al. Clinical features and pathogenesis of intracerebral hemorrhage after rt-PA and heparin therapy for acute myocardial infarction: the Thrombolysis in Myocardial Infarction (TIMI) II Pilot and Randomized Clinical Trial combined experience. Neurology 1995; 45: 649–658. 19 Eisenberg PR, Sherman LA, Tiefenbrunn AJ, Ludbrook PA, Sobel BE, Jaffe AS. Sustained fibrinolysis after administration of t-PA despite its short half-life in the circulation. Thromb Haemost 1987; 57: 35–40. 20 Montagne A, He´bert M, Jullienne A, Lesept F, Le Be´hot A, Louessard M et al. Memantine improves safety of thrombolysis for stroke. Stroke 2012; 43: 2774–2781. 21 Pfeilschifter W, Kashefiolasl S, Lauer A, Steinmetz H, Foerch C. Hematoma expansion in experimental intracerebral hemorrhage is not altered by peracute treatment with recombinant tissue plasminogen activator. Neuroscience 2013; 250: 181–188. 22 Xi G, Keep RF, Hoff JT. Mechanisms of brain injury after intracerebral haemorrhage. Lancet Neurol 2006; 5: 53–63. 23 Wagner KR, Xi G, Hua Y, Kleinholz M, de Courten-Myers GM, Myers RE et al. Lobar intracerebral hemorrhage model in pigs: rapid edema development in perihematomal white matter. Stroke 1996; 27: 490–497. 24 Xi G, Wagner KR, Keep RF, Hua Y, de Courten-Myers GM, Broderick JP et al. Role of blood clot formation on early edema development after experimental intracerebral hemorrhage. Stroke 1998; 29: 2580–2586. 25 Lee KR, Kawai N, Kim S, Sagher O, Hoff JT. Mechanisms of edema formation after intracerebral hemorrhage: effects of thrombin on cerebral blood flow, blood-brain barrier permeability, and cell survival in a rat model. J Neurosurg 1997; 86: 272–278. 26 Yang GY, Betz AL, Chenevert TL, Brunberg JA, Hoff JT. Experimental intracerebral hemorrhage: relationship between brain edema, blood flow, and blood-brain barrier permeability in rats. J Neurosurg 1994; 81: 93–102. 27 Merlini PA, Bauer KA, Oltrona L, Ardissino D, Spinola A, Cattaneo M et al. Thrombin generation and activity during thrombolysis and concomitant heparin therapy in patients with acute myocardial infarction. J Am Coll Cardiol 1995; 25: 203–209. 28 Francis CW, Markham RE, Barlow GH, Florack TM, Dobrzynski DM, Marder VJ. Thrombin activity of fibrin thrombi and soluble plasmic derivatives. J Lab Clin Med 1983; 102: 220–230. 29 Tsuji K, Aoki T, Tejima E, Arai K, Lee S-R, Atochin DN et al. Tissue plasminogen activator promotes matrix metalloproteinase-9 upregulation after focal cerebral ischemia. Stroke 2005; 36: 1954–1959. 30 Wang X, Lee S-R, Arai K, Lee S-R, Tsuji K, Rebeck GW et al. Lipoprotein receptormediated induction of matrix metalloproteinase by tissue plasminogen activator. Nat Med 2003; 9: 1313–1317. 31 Su EJ, Fredriksson L, Geyer M, Folestad E, Cale J, Andrae J et al. Activation of PDGFCC by tissue plasminogen activator impairs blood-brain barrier integrity during ischemic stroke. Nat Med 2008; 14: 731–737.
Supplementary Information accompanies the paper on the Journal of Cerebral Blood Flow & Metabolism website (http:// www.nature.com/jcbfm)
& 2014 ISCBFM
Journal of Cerebral Blood Flow & Metabolism (2014), 176 – 181
Copyright of Journal of Cerebral Blood Flow & Metabolism is the property of Nature Publishing Group and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
