Hypothermia in acute stroke—Slow versus fast rewarming

Experimental Neurology 204 (2007) 131 – 137 www.elsevier.com/locate/yexnr

Hypothermia in acute stroke—Slow versus fast rewarming An experimental study in rats Christian Berger ⁎, Feng Xia, Martin Köhrmann, Stefan Schwab Department of Neurology, University of Heidelberg, Im Neuenheimer Feld 400, 69120 Heidelberg, Germany Received 6 June 2006; revised 5 September 2006; accepted 4 October 2006 Available online 16 November 2006

Abstract The rewarming phase after therapeutic hypothermia in cerebral ischemia appears crucial as rapid rewarming may lead to rebound phenomena and enhance deleterious ischemic effects. We hypothesized that slow and controlled rewarming after moderate hypothermia is superior to fast rewarming in rats subjected to 90 min temporary middle cerebral artery occlusion (tMCAO). Two experiments were designed: (i) 34 rats were randomly assigned to either normothermic treatment, to hypothermia (33°C) with rapid rewarming within 20 min, or to hypothermia with slow rewarming within 2 h after 4 h of hypothermia starting 2 h after tMCAO. Infarct size, neuroscore, myeloperoxidase and aquaporin 4 (AQP4) positive cells were assessed on day 5 after tMCAO. (ii) In 15 rats, striatal cerebral microdialysis was performed from 1.5 h before until 8 h after tMCAO. Total infarct volume was largest in the normothermic group (89.9 ± 16.8 mm3) followed by the fast rewarming group (69.2 ± 12.6 mm3), and a significantly smaller infarct volume in the slow rewarming group (41.1 ± 6.6 mm3, p < 0.05). Neurological functions improved in both hypothermia groups at day 5 after tMCAO (Neuroscore median 2.5 in normothermia vs. 1.5 in both hypothermia groups) though without any difference between slowly and fast rewarmed animals. Periinfarct expression of AQP4 was less prominent in slowly rewarmed animals as was the count of MPO-positive cells in subcortical regions. Glutamate release was significantly higher at 4 distinct time points in the control group. Slow rewarming after a period of hypothermia is superior to fast rewarming. It may blunt deleterious rebound effects such as overexpression of AQP4, sustain anti-inflammatory mechanisms and thereby preserve the neuroprotection delivered by hypothermia. © 2006 Elsevier Inc. All rights reserved. Keywords: Hypothermia; Glutamate; Middle cerebral artery, stroke; Brain edema; Inflammation

Introduction Hypothermia is a therapeutic strategy used after cardiac arrest and resuscitation to minimize brain damage. It is also applied in traumatic brain injury though a recent large randomized head trauma study brought a negative result (Clifton et al., 2001). For stroke, a number of case series exist suggesting a beneficial effect on mortality in patients with severe types of hemispheric ischemia (Schwab et al., 2001; Schwab et al., 1998b). In these patients, hypothermia significantly reduced the release of excitotoxic glutamate in peri-ischemic brain tissue (Berger et al., 2002). The termination of hypothermia after 2 to 3 days and the subsequent rewarming phase seems crucial for survival (Steiner et al., 2001). During therapeutic hypothermia, many deleterious ⁎ Corresponding author. Fax: +49 6221 565461. E-mail address: [email protected] (C. Berger). 0014-4886/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.expneurol.2006.10.002

ischemic mechanisms are delayed or even stopped. Rapid and uncontrolled rewarming seems to enhance these mechanisms and to allow the infarcted brain to swell, the ICP to increase, and the CPP to decrease. Aoki et al. (2002) demonstrated in cats that a drop of the CPP below 60 mm Hg during rewarming causes relative ischemia. A critically low CPP may in fact be one of the explanations why the large study on hypothermia in traumatic brain injury patients failed to demonstrate a benefit (Clifton et al., 2001), though a subgroup analysis (Clifton et al., 2002) revealed a better outcome in patients less than 45 years of age who were already hypothermic on admission. Prolonged use of hypothermia with durations between 3 and 14 days, depending on intracranial pressure, also improved outcome without higher complication rates (Jiang et al., 2000, for a review see McIntyre et al., 2003). We therefore designed an experiment in rats undergoing temporary middle cerebral artery occlusion (tMCAO) to test the

132

C. Berger et al. / Experimental Neurology 204 (2007) 131–137

hypothesis whether post-ischemic treatment with hypothermia at 33°C followed by slow rewarming reduces infarct size and post-ischemic inflammation and improves neurological outcome compared to animals treated either with normothermia (controls) or with hypothermia and subsequent rapid rewarming. Cerebral microdialysis was used for measurement and comparison of striatal glutamate release. Material and methods Animals and experimental design All animals were housed in individual cages with free access to food and water ad libitum prior to and after surgery. Animal protocols were approved by the Ethical Committee of the University of Heidelberg according to EU regulations. We designed two independent experiments: (i) the outcome experiment focusing on neurological behavior, infarct size, inflammatory response and aquaporin 4 (AQ4) expression 5 days after stroke onset, and (ii) a microdialysis experiment assessing neurochemical changes in the acute phase of stroke before and during treatment. In both experiments, tMCAO for 90 min was performed using the filament method by Longa et al. (1989). Outcome experiment Male Wistar rats (n = 34) weighing 270 to 320 g were randomly assigned to one of three treatment groups: (i) fast rewarming group (n = 12): hypothermia at 33°C for 4 h, rewarming for 20 min; (ii) slow rewarming group (n = 11): hypothermia at 33°C for 4 h, rewarming for 120 min; and (iii) control group (n = 11): normothermia throughout the experiment. Anesthesia was induced with 2% halothane and maintained with 1% halothane in 70% N2O and 30% O2 using a vaporizer and facial mask for rats until 8 h after induction of tMCAO. Treatment commenced 30 min after removal of the filament. To induce hypothermia, the target temperature of the heating pad was decreased to 33°C. Additional ice packs were applied to lower core temperature within 15 min. The heating pad was attached to a rectal probe so that the target temperature could be maintained as long as necessary. Slow rewarming was achieved by gradually increasing the target temperature, while fast rewarming required the support of an infrared light. We used a rectal temperature probe instead of monitoring brain temperature in order to imitate the typical setting of the average intensive care unit when hypothermia is applied though we were aware that brain and rectal temperature may variably differ. The left femoral artery was cannulated with PE-50 polyethylene tubing and served for continuous blood pressure recording and blood gas analysis (AVL 990, Homburg, Germany) until successful rewarming. To assess sensorimotor neurological functions before induction of ischemia, at 24 h after tMCAO, and at 5 days after tMCAO, we used a scoring system as described by Menzies

et al. (1992). The following parameters were scored: forelimb posture, grasping reflex, and spontaneous movements with the following rating scale: 0, no neurological deficit; 1, failure to extend the left forepaw; 2, decreased grip strength of the left forepaw; 3, circling to the left by pulling the tail; 4, spontaneous circling; and 5, death. At day 5, animals were deeply anesthetized with 4% halothane and perfused with 4% paraformaldehyde (PFA). Their brains were incubated in 4% PFA overnight and paraffin embedded. The brains were sliced coronally with a brain slicer at 2-mm intervals, starting at +2.2 mm anterior to − 5.8 mm dorsal of the bregma. Lesion delineation using MAP2 immunohistochemistry As described previously (Kloss et al., 2002), MAP2 (microtubule associated protein 2) immunohistochemical staining is a valid method for assessing infarct size. In brief, 1-μm sections were dewaxed in xylene, pretreated with methanol/H2O2 (0.3%), followed by microwave treatment for antigen retrieval. After blocking (normal swine serum 5% in PBS with 0.2% Triton X-100 for 30 min), sections were incubated with a mouse–anti-rat MAP2 monoclonal antibody (Clone HM-2; Sigma, St. Louis, MO, USA) overnight at 4°C. The sections were then washed (PBST) again, incubated with a biotinylated, rat-absorbed anti-mouse-IgG secondary antibody (Vector, Burlingame, CA, USA), washed, incubated for 1 h with ABC-reagent (Vector), and finally visualized using 0.02% diaminobenzidine (DAB; Sigma) with 0.02% H2O2. The areas of infarction (unstained tissue) were measured on each section using an image analysis system (MCID Elite Version 6.0; Imaging Research, St. Catherines, Canada). To compensate for the effect of brain edema, the corrected infarct size was calculated by the following formula as described previously (Schabitz et al., 2001): corrected infarct area equals left hemisphere area minus (right hemisphere area minus infarct area). The corrected mean total infarct volume then was calculated by multiplying the respective corrected infarct areas by slice thickness. AQP4 and MPO immunohistochemistry Sections (5 μm) were stained using a polyclonal goat antiaquaporin 4–antibody (AQP4 (H-19): sc-9887 Santa Cruz Biotechnology Inc.; 1:50) followed by a biotinylated horse anti-goat–antibody (Vector BA 9500; 1:100) and an avidin– FITC conjugate (Vector A 2001; 1:200). To exclude the possibility of false labeling, the secondary antibody was tested for cross-reactivity by omitting the primary antibody in control experiments. For myeloperoxidase (MPO) staining, we used a polyclonal rabbit anti-MPO (DakoCytomation, Glostrup, Denmark) and incubated 1 μm sections for 60 min at RT. Secondary and tertiary antibodies were used as described for MAP2 immunohistochemistry. To quantify expression of AQP4 and MPO, the cells labeled by the anti-AQP4 or anti-MPO antibody were counted in 2

C. Berger et al. / Experimental Neurology 204 (2007) 131–137 Table 1 Physiologic data

Statistics

Pre-tMCAO

Post-tMCAO

Post-treatment

7.38 ± 0.01 7.41 ± 0.02 7.39 ± 0.01

7.40 ± 0.12 7.43 ± 0.03 7.42 ± 0.01

7.39 ± 0.02 7.40 ± 0.11 7.43 ± 0.02

7.41 ± 0.02 7.42 ± 0.01 7.43 ± 0.02

pCO2(mm Hg) Control 41.2 ± 1.3 Fast R 43.6 ± 2.3 Slow R 42.1 ± 2.6

42.7 ± 3.5 42.5 ± 1.6 44.1 ± 3.6

43.4 ± 3.1 41.2 ± 1.1 43.5 ± 2.5

43.2 ± 1.4 44.6 ± 2.3 44.8 ± 1.6

PO2(mm Hg) Control 103.5 ± 10.4 Fast R 110.2 ± 9.2 Slow R 109.5 ± 8.4

98.7 ± 6.5 100.1 ± 8.6 99.3 ± 9.4

101.2 ± 6.7 99.6 ± 8.5 98.7 ± 11.9

101.2 ± 6.7 99.6 ± 8.5 98.7 ± 11.9

Base excess (mEq/L) Control 0.4 ± 0.1 Fast R 1.2 ± 0.3 Slow R 1.4 ± 0.8

1.0 ± 0.6 − 0.5 ± 0.8 0.4 ± 0.5

0.5 ± 0.6 1.0 ± 0.3 −0.8 ± 0.9

0.5 ± 0.3 1.0 ± 0.5 1.0 ± 1.0

Mean BP (mm Hg) Control 89.3 ± 5.7 Fast R 87.1 ± 3.7 Slow R 88.6 ± 3.5

96.8 ± 5.5 94.3 ± 8.4 97.8 ± 6.1

107 ± 5.7 105 ± 4.4 110 ± 7.5

103 ± 4.2 100 ± 5.7 109 ± 7.8

Temperature Control Fast R Slow R

37.0 ± 0.1 37.1 ± 0.2 37.1 ± 0.1

37.1 ± 0.2 33.3 ± 0.3 33.2 ± 0.2

37.1 ± 0.1 37.3 ± 0.4 37.2 ± 0.2

pH Control Fast R Slow R

(°C) 37.2 ± 0.2 37.1 ± 0.1 37.3 ± 0.2

133

Rewarming

different fields (422 μm × 338 μm) in the border of the infarct both in the cortex and subcortically in the periventricular zone and in matching contralateral fields in 5 animals per treatment group. Cerebral microdialysis Fifteen animals were randomly assigned to one to the three experimental groups. The course was similar to the outcome experiment with two major alterations: two hours prior to MCAO, microdialysis probes (CMA/12, Solna, Sweden) were inserted bilaterally into the striatum according to the coordinates of a rat brain atlas (Paxinos and Watson, 1986) (AP +0.5 mm, ML 3 mm either side, relative to the bregma, 3.5 mm deep). We allowed 1 h to reach equilibrium. Perfusion fluid consisted of artificial CSF solution (NaCl 147 mmol/L, KCl 2.7 mmol/L, CaCl2 1.2 mmol/L, MgCl2 0.85 mmol/L) acidified with 5 mmol/ L H3PO4. Perfusion velocity was 2 μl/min. Vials were replaced every 15 min. Glutamate concentrations were measured enzyme-photometrically (CMA 600 Microdialysis Analyser, CMA, Solna, Sweden). Before and after each experiment, we measured relative recovery and corrected dialysate concentrations accordingly. Microdialysis was performed until 5 h after tMCAO, when animals were sacrificed and their brains were removed immediately and frozen.

All data were subjected to analysis of variance (ANOVA) and Bonferroni error protection to confirm significant group effects. Unpaired comparisons between two groups were performed using the Student's t test. An alpha error rate of 0.05 was taken as the criterion for significance for all tests. All data are presented as mean ± SD. Analyses were performed with StatView® statistical software (SAS Institute Inc., 1998). Results Of a total of 34 rats in the first experimental part (outcome experiment), 30 rats survived until day 5 after tMCAO. Four rats died within the first 24 h. Mortality rate was statistically not different between the three treatment groups (control, fast, slow rewarming: 1/11, 2/12, 1/11). Blood gas results (pH, pO2, pCO2, base excess) and mean arterial blood pressure also did not differ between treatment groups at distinct time points (Table 1). Total infarct volume was largest in the control group (89.9 ± 16.8 mm3) followed by the fast rewarming group (69.2 ± 12.6 mm3), and a significantly smaller infarct volume in the slow rewarming group (41.1 ± 6.6 mm3; Fig. 1). The neuroscore demonstrated worsening in all three groups 24 h after tMCAO. However, rats in both groups treated with hypothermia had better neurological functions than control animals at day 5 (Fig. 2). In the subcortical periventricular infarct zone, MPO-positive cell count was significantly lower in those animals treated with slow rewarming after hypothermia (108 ± 13) than in normothermic rats (141 ± 14) as well as in rapidly rewarmed animals (140 ± 14; Fig. 3). In slowly rewarmed animals, MPO cell count was similar to subcortical zones of the non-infarcted hemisphere, whereas it was significantly higher in the other two treatment arms both in the cortical and the subcortical region. AQP4 labeling was induced in regions bordering the infarct both in cortical and in subcortical sections compared to contralateral matching fields (data not shown). The number of positive cells labeled by AQP4 antibody was 21 ± 6 cells per field in the infarct border of normothermic animals and 19 ± 7

Fig. 1. Total infarct volume in all 3 treatment groups as assessed by MAP2 staining. Infarct volume in the slow rewarming group was reduced by about 50% compared to control group. * indicates significant difference between treatment groups in an overall statistical test.

134

C. Berger et al. / Experimental Neurology 204 (2007) 131–137

Fig. 2. Assessment of neuroscore before tMCAO and at days 1 and 5 after tMCAO: both hypothermia groups improved significantly on day 5 after tMCAO.

cells of animals treated with rapid rewarming after hypothermia versus 6 ± 3 in slowly rewarmed animals (p < 0.05). Regional distribution of AQP4 labeling was observed around the infarcted area, mainly in the cortex and in paraventricular nuclei of the hypothalamus (Fig. 4). In the microdialysis subexperiment, striatal glutamate release increased dramatically in all groups during the MCA occlusion for 90 min (Fig. 5). This sharp rise was followed by a fairly rapid decrease of the glutamate concentration in all animals treated with hypothermia while glutamate in the control group only gradually decreased. There was a trend for glutamate to be lower in the slow rewarming group, however this was not significant. Discussion These data confirm previous stroke experiments in that early post-ischemic treatment of rats with hypothermia improves outcome. In addition, they demonstrate the superiority of a subsequent slow and controlled rewarming phase versus an uncontrolled and rapid rewarming phase with reduction of the infarct volume by approximately 50%. The specific experimental setting in this study may have contributed to some of the differences observed between the slow and the rapid rewarming group: (i) rapid active rewarming using an infrared light may have exacerbated brain temperature as we only controlled rectal temperature and not brain temperature directly. (ii) Total time with temperatures below normothermia was longer in slowly rewarmed animals than in rats rapidly rewarmed. Thus, the experimental setting favored the hypothermia group with slow rewarming. However, this setting more accurately reflects the real situation for stroke patients treated with hypothermia (Schwab et al., 1998a; Steiner et al., 2001). A vast body of experimental literature established that mild hypothermia reduces cerebral injury both in traumatic brain injury and in cerebral ischemia. Several experimental studies exist demonstrating neuronal recovery after both focal and global brain ischemia by hypothermia (Coimbra and Wieloch,

1994; Colbourne et al., 1999; Kawai et al., 2000). Not only preischemic initiation of hypothermia is neuroprotective (Chen et al., 1992; Dietrich et al., 1993), but also post-ischemic application (Coimbra and Wieloch, 1994; Kawai et al., 2000; Kollmar et al., 2002) even when started up to 6 h after the onset of ischemia (Colbourne et al., 1999). The duration of the rewarming phase following hypothermia was discussed in none of these studies. As early as 1968, Michenfelder and Theye observed that 48 h of moderate hypothermia (29°C) were consistently lethal in primates subjected to MCA infarction due to massive cerebral edema and other systemic complications that occurred within the first 3 h of active rewarming (Michenfelder and Theye, 1968). Similarly, Steen et al. (1979) described severe systemic complications and an inadequate and inhomogeneous perfusion of the brain if cats were rapidly rewarmed after 24 h of hypothermia. This led to a depletion of the brain's energy stores (ATP) due to an overly increased cerebral metabolism with insufficient brain perfusion. A rebound brain edema, hypotension, and death have been reported also for dogs (Blair et al., 1956) and rats (Popovic, 1959). The protection of brain tissue delivered by hypothermia has been attributed to decreases in excitatory amino acid release and preservation of metabolic stores. In this study, release of glutamate in the infarct core was reduced immediately after the filament occluding the MCA was removed in hypothermic animals. This confirms previous animal studies (Boris-Moller and Wieloch, 1998; Busto et al., 1989; Fujisawa et al., 1999; Kawai et al., 2000; Nakashima and Todd, 1996; Ooboshi et al., 2000; Winfree et al., 1996) as well as results obtained from cerebral microdialysis in stroke patients treated with hypothermia (Berger et al., 2002). However, though we observed a trend for lower glutamate concentrations in the slow rewarming group, both hypothermia groups did not differ significantly during the rewarming phase. Other mechanisms may equally contribute to neuroprotection by hypothermia as a neuroprotective effect can be observed

Fig. 3. Count of MPO-positive cells in a defined cortical and subcortical area of the infarcted hemisphere and its contralateral counterpart. † indicates significant difference between treatment groups in the same area; * indicates significant difference between infarcted and non-infarcted hemisphere in the same area.

C. Berger et al. / Experimental Neurology 204 (2007) 131–137

135

Fig. 4. Immunofluorescent staining of aquaporin 4 positive cells: expression is prominent in the cortical (left column) as well as in the subcortical periinfarct region (right column) in controls (A) and the fast rewarming treatment group (B) but much less in the slow rewarming group (C).

Fig. 5. Striatal glutamate concentrations, measured every 30 min, starting 1.5 h before MCAO until 8 h after tMCAO. Note a sharp increase in glutamate during tMCAO in all groups and a relatively rapid decrease after reperfusion in both hypothermia groups. * indicates significant differences between treatment groups at the same time points.

even when hypothermia was delayed for hours after onset of ischemia at a time when glutamate has been released and energy stores are depleted (Colbourne et al., 1999; Maier et al., 2001). These mechanisms include processes attenuating acute inflammatory responses. Infiltrating leukocytes are thought to contribute to secondary ischemic brain damage by producing toxic substances and disrupting the blood–brain barrier (del Zoppo et al., 2000; Loddick and Rothwell, 1996). Infiltration of leukocytes is mediated by binding to endothelial intercellular adhesion molecule-1 (ICAM-1) (Zhang et al., 1995). Both early and delayed hypothermia reduce the accumulation of polymorph-nuclear leukocytes and inhibit the overexpression of intercellular adhesion molecule-1 (ICAM-1) (Deng et al., 2003; Inamasu et al., 2001; Ishikawa et al., 1999; Kawai et al., 2000). In line with the above observations, our study demonstrated that at day 5 after tMCAO the number of MPO-positive cells was significantly lower in subcortical regions of the infarct in those animals treated with hypothermia followed by slow rewarming. No such effect was seen in normothermic animals and in those rapidly rewarmed. This result is surprising as it demonstrates a sustained anti-inflammatory effect of hypothermia when it is followed by slow rewarming.

136

C. Berger et al. / Experimental Neurology 204 (2007) 131–137

Another sustained effect was observed on AQP4 positive cells in the slow rewarming group. Aquaporin 4 is present on astrocytic end-feet in contact with brain vessels and is upregulated after ischemia (Badaut et al., 2002). It is associated with a disruption of the brain water homeostasis and with the development of brain edema. Both in the cortex and in white matter adjacent to the infarct, AQP4 expression was increased in the infarcted hemisphere as compared to the corresponding contralateral area. However, this increase was less pronounced in slowly rewarmed animals. Less brain edema formation may have enhanced the infarct size reduction by hypothermia. Beside clinical investigations emphasizing the importance of a controlled rewarming scheme following hypothermia, there is growing evidence from other conditions than cerebral ischemia that slow rewarming is superior fast rewarming after hypothermia: in liver transplantation, the use of rapid rewarming has been recognized to be damaging probably due to rapid ATP depletion and energy failure associated with mitochondrial dysfunction (Leducq et al., 1998; Marsh et al., 1991; Wakiyama et al., 1997). In a model for traumatic brain injury, axonal injury was exacerbated by rapid rewarming (20 min) as compared to slow rewarming (90 min) after a 1-hour period of 32°C hypothermia postinjury (Suehiro and Povlishock, 2001). The same group demonstrated that an important mechanism for this superiority of slow rewarming is the preserved arteriolar vascular responses in pial vessels which was impaired in animals rapidly rewarmed (Suehiro et al., 2003). Uncoupling of cerebral blood flow and metabolism during rapid rewarming was also observed in a gerbil model for transient forebrain ischemia (Nakamura et al., 1999), though the effect was more marked the deeper the target temperature of hypothermia was (24 vs. 30.5°C). The common result of these studies was that controlled slow rewarming could preserve the protective effects of hypothermia while rapid rewarming often leads to a reversion of these effects. Thus, various mechanisms may contribute to the beneficial effects of slow rewarming such as preserved vasoreactivity and metabolic coupling, sustained anti-inflammatory effects and attenuated brain edema formation. In summary, controlled slow rewarming after hypothermia in experimental stroke is crucial to sustain any neuroprotective effects of hypothermia. The rewarming modality, therefore, seems to be of great importance for any clinical trial using hypothermia in stroke or in traumatic brain injury. Acknowledgment This study was in part supported by the BmBF (Bundesministerium für Bildung und Forschung) as part of the competence network stroke (Teilprojekt B7). References Aoki, A., Mori, K., Maeda, M., 2002. Adequate cerebral perfusion pressure during rewarming to prevent ischemic deterioration after therapeutic hypothermia. Neurol. Res. 24, 271–280. Badaut, J., Lasbennes, F., Magistretti, P., Regli, J., 2002. Aquaporins in brain: distribution, physiology, and pathophysiology. J. Cereb. Blood Flow Metab. 22, 367–378.

Berger, C., Schabitz, W.R., Georgiadis, D., Steiner, T., Aschoff, A., Schwab, S., 2002. Effects of hypothermia on excitatory amino acids and metabolism in stroke patients: a microdialysis study. Stroke 33, 519–524. Blair, E., Montgomery, A.V., Swan, H., 1956. Post-hypothermia circulatory failure: I. Physiologic observations on the circulation. Circulation 13, 909–915. Boris-Moller, F., Wieloch, T., 1998. Changes in the extracellular levels of glutamate and aspartate during ischemia and hypoglycemia. Effects of hypothermia. Exp. Brain Res. 121, 277–284. Busto, R., Globus, M.Y., Dietrich, W.D., Martinez, E., Valdes, I., Ginsberg, M.D., 1989. Effect of mild hypothermia on ischemia-induced release of neurotransmitters and free fatty acids in rat brain. Stroke 20, 904–910. Chen, H., Chopp, M., Zhang, Z.G., Garcia, J.H., 1992. The effect of hypothermia on transient middle cerebral artery occlusion in the rat. J. Cereb. Blood Flow Metab. 12, 621–628. Clifton, G.L., Miller, E.R., Choi, S.C., Levin, H.S., McCauley, S., Smith Jr., K.R., Muizelaar, J.P., Wagner Jr., F.C., Marion, D.W., Luerssen, T.G., Chesnut, R.M., Schwartz, M., 2001. Lack of effect of induction of hypothermia after acute brain injury. N. Engl. J. Med. 344, 556–563. Clifton, G.L., Miller, E.R., Choi, S.C., Levin, H.S., McCauley, S., Smith Jr., K.R., Muizelaar, J.P., Marion, D.W., Luerssen, T.G., 2002. Hypothermia on admission in patients with severe brain injury. J. Neurotrauma 19, 293–301. Coimbra, C., Wieloch, T., 1994. Moderate hypothermia mitigates neuronal damage in the rat brain when initiated several hours following transient cerebral ischemia. Acta Neuropathol. 87, 325–331. Colbourne, F., Li, H., Buchan, A.M., 1999. Indefatigable CA1 sector neuroprotection with mild hypothermia induced 6 hours after severe forebrain ischemia in rats. J. Cereb. Blood Flow Metab. 19, 742–749. del Zoppo, G., Ginis, I., Hallenbeck, J.M., Iadecola, C., Wang, X., Feuerstein, G.Z., 2000. Inflammation and stroke: putative role for cytokines, adhesion molecules and iNOS in brain response to ischemia. Brain Pathol. 10, 95–112. Deng, H., Han, H.S., Cheng, D., Sun, G.H., Yenari, M.A., 2003. Mild hypothermia inhibits inflammation after experimental stroke and brain inflammation. Stroke 34, 2495–2501. Dietrich, W.D., Busto, R., Alonso, O., Globus, M.Y., Ginsberg, M.D., 1993. Intraischemic but not postischemic brain hypothermia protects chronically following global forebrain ischemia in rats. J. Cereb. Blood Flow Metab. 13, 541–549. Fujisawa, H., Koizumi, H., Ito, H., Yamashita, K., Maekawa, T., 1999. Effects of mild hypothermia on the cortical release of excitatory amino acids and nitric oxide synthesis following hypoxia. J. Neurotrauma 16, 1083–1093. Inamasu, J., Suga, S., Sato, S., Horiguchi, T., Akaji, K., Mayanagi, K., Kawase, T., 2001. Intra-ischemic hypothermia attenuates intercellular adhesion molecule-1 (ICAM-1) and migration of neutrophil. Neurol. Res. 23, 105–111. Ishikawa, M., Sekizuka, E., Sato, S., Yamaguchi, N., Inamasu, J., Bertalanffy, H., Kawase, T., Iadecola, C., 1999. Effects of moderate hypothermia on leukocyte–endothelium interaction in the rat pial microvasculature after transient middle cerebral artery occlusion. Stroke 30, 1679–1686. Jiang, J., Yu, M., Zhu, C., 2000. Effect of long-term mild hypothermia therapy in patients with severe traumatic brain injury: 1-year follow-up review of 87 cases. J. Neurosurg. 93, 546–549. Kawai, N., Okauchi, M., Morisaki, K., Nagao, S., 2000. Effects of delayed intraischemic and postischemic hypothermia on a focal model of transient cerebral ischemia in rats. Stroke 31, 1982–1989 (discussion 1989). Kloss, C.U., Thomassen, N., Fesl, G., Martens, K.H., Yousri, T.A., Hamann, G.F., 2002. Tissue-saving infarct volumetry using histochemistry validated by MRI in rat focal ischemia. Neurol. Res. 24, 713–718. Kollmar, R., Schabitz, W.R., Heiland, S., Georgiadis, D., Schellinger, P.D., Bardutzky, J., Schwab, S., 2002. Neuroprotective effect of delayed moderate hypothermia after focal cerebral ischemia: an MRI study. Stroke 33, 1899–1904. Leducq, N., Delmas-Beauvieux, M.C., Bourdel-Marchasson, I., Dufour, S., Gallis, J.L., Canioni, P., Diolez, P., 1998. Mitochondrial permeability transition during hypothermic to normothermic reperfusion in rat liver demonstrated by the protective effect of cyclosporin A. Biochem. J. 336 (Pt 2), 501–506.

C. Berger et al. / Experimental Neurology 204 (2007) 131–137 Loddick, S.A., Rothwell, N.J., 1996. Neuroprotective effects of human recombinant interleukin-1 receptor antagonist in focal cerebral ischaemia in the rat. J. Cereb. Blood Flow Metab. 16, 932–940. Longa, E.Z., Weinstein, P.R., Carlson, S., Cummins, R., 1989. Reversible middle cerebral artery occlusion without craniectomy in rats. Stroke 20, 84–91. Maier, C.M., Sun, G.H., Kunis, D., Yenari, M.A., Steinberg, G.K., 2001. Delayed induction and long-term effects of mild hypothermia in a focal model of transient cerebral ischemia: neurological outcome and infarct size. J. Neurosurg. 94, 90–96. Marsh, D.C., Hjelmhaug, J.A., Vreugdenhil, P.K., Kerr, J.A., Rice, M.J., Belzer, F.O., Southard, J.H., 1991. Hypothermic preservation of hepatocytes. III. Effects of resuspension media on viability after up to 7 days of storage. Hepatology 13, 500–508. McIntyre, L.A., Fergusson, D.A., Hebert, P.C., Moher, D., Hutchison, J.S., 2003. Prolonged therapeutic hypothermia after traumatic brain injury in adults: a systematic review. JAMA 289, 2992–2999. Menzies, S.A., Hoff, J.T., Betz, A.L., 1992. Middle cerebral artery occlusion in rats: a neurological and pathological evaluation of a reproducible model. Neurosurgery 31, 100–107. Michenfelder, J.D., Theye, R.A., 1968. Hypothermia: effect on canine brain and whole-body metabolism. Anesthesiology 29, 1107–1112. Nakamura, T., Miyamoto, O., Yamagami, S., Hayashida, Y., Itano, T., Nagao, S., 1999. Influence of rewarming conditions after hypothermia in gerbils with transient forebrain ischemia. J. Neurosurg. 91, 114–120. Nakashima, K., Todd, M.M., 1996. Effects of hypothermia on the rate of excitatory amino acid release after ischemic depolarization. Stroke 27, 913–918. Ooboshi, H., Ibayashi, S., Takano, K., Sadoshima, S., Kondo, A., Uchimura, H., Fujishima, M., 2000. Hypothermia inhibits ischemia-induced efflux of amino acids and neuronal damage in the hippocampus of aged rats. Brain Res. 884, 23–30. Paxinos, G., Watson, C., 1986. The Rat Brain in Stereotaxic Coordinates. Academic Press, San Diego, CA. Popovic, V., 1959. Lethargic hypothermia in hibernators and non-hibernators. Ann. N. Y. Acad. Sci. 80, 320–331.

137

Schabitz, W.R., Hoffmann, T.T., Heiland, S., Kollmar, R., Bardutzky, J., Sommer, C., Schwab, S., 2001. Delayed neuroprotective effect of insulinlike growth factor-i after experimental transient focal cerebral ischemia monitored with mri. Stroke 32, 1226–1233. Schwab, S., Schwarz, S., Aschoff, A., Keller, E., Hacke, W., 1998a. Moderate hypothermia and brain temperature in patients with severe middle cerebral artery infarction. Acta Neurochir., Suppl. (Wien) 71, 131–134. Schwab, S., Schwarz, S., Spranger, M., Keller, E., Bertram, M., Hacke, W., 1998b. Moderate hypothermia in the treatment of patients with severe middle cerebral artery infarction. Stroke 29, 2461–2466. Schwab, S., Georgiadis, D., Berrouschot, J., Schellinger, P.D., Graffagnino, C., Mayer, S.A., 2001. Feasibility and safety of moderate hypothermia after massive hemispheric infarction. Stroke 32, 2033–2035. Steen, P.A., Soule, E.H., Michenfelder, J.D., 1979. Detrimental effect of prolonged hypothermia in cats and monkeys with and without regional cerebral ischemia. Stroke 10, 522–529. Steiner, T., Friede, T., Aschoff, A., Schellinger, P.D., Schwab, S., Hacke, W., 2001. Effect and feasibility of controlled rewarming after moderate hypothermia in stroke patients with malignant infarction of the middle cerebral artery. Stroke 32, 2833–2835. Suehiro, E., Povlishock, J.T., 2001. Exacerbation of traumatically induced axonal injury by rapid posthypothermic rewarming and attenuation of axonal change by cyclosporin A. J. Neurosurg. 94, 493–498. Suehiro, E., Ueda, Y., Wei, E.P., Kontos, H.A., Povlishock, J.T., 2003. Posttraumatic hypothermia followed by slow rewarming protects the cerebral microcirculation. J. Neurotrauma 20, 381–390. Wakiyama, S., Yanaga, K., Soejima, Y., Nishizaki, T., Sugimachi, K., 1997. Reduction of rewarming injury of the hepatic graft by a heat insulator. Br. J. Surg. 84, 459–463. Winfree, C.J., Baker, C.J., Connolly Jr., E.S., Fiore, A.J., Solomon, R.A., 1996. Mild hypothermia reduces penumbral glutamate levels in the rat permanent focal cerebral ischemia model. Neurosurgery 38, 1216–1222. Zhang, R.L., Chopp, M., Zaloga, C., Zhang, Z.G., Jiang, N., Gautam, S.C., Tang, W.X., Tsang, W., Anderson, D.C., Manning, A.M., 1995. The temporal profiles of ICAM-1 protein and mRNA expression after transient MCA occlusion in the rat. Brain Res. 682, 182–188.