The neuroprotective effects of anesthetic preconditioning have been reported in both in vitro and in vivo models.1–4 Sevoflurane (sevo), a popular anesthetic with few clinical side effects, has been shown to reduce cerebral ischemia damage in various experimental models. However, the dose-dependent effects of sevo preconditioning have not been evaluated, and its mechanisms are poorly elucidated.
Nonlethal levels of reactive oxygen species (ROS) generation have been proposed to be initiators or mediators for ischemic tolerance induction.5 It has been demonstrated that the cardioprotective effects of sevo against ischemia– reperfusion injury required ROS generation. Scavenging ROS abolished the preconditioning effects of anesthetics by attenuating mitochondrial uncoupling6 or protein kinases.7 In the central nervous system, ROS was also involved in neuroprotection from sevo preconditioning in mixed cortical neuronal–glial cell cultures against oxygen and glucose deprivation.8 In addition, several studies have shown that endogenous antioxidant enzymes could be induced simultaneously with the development of neuroprotection by different preconditioning methods such as lipopolysaccharide9 and brief ischemia.10 Our previous study on spinal cord ischemia in a rabbit model has demonstrated that hyperbaric oxygen preconditioning activates endogenous antioxidant enzymes through generation of a small amount of ROS.11
Therefore, the current study was designed to test the dose-dependent neuroprotective effects of sevo preconditioning in a focal cerebral ischemia model in rats, and furthermore to verify the oxidant–antioxidant mechanisms of this neuroprotection. We hypothesized that sevo preconditioning could dose-dependently reduce cerebral ischemia–reperfusion injury through releasing ROS, which up-regulates protective antioxidant enzymes before ischemic injury.
The experimental protocol used in this study was approved by the Ethics Committee for Animal Experimentation of the Fourth Military Medical University and was conducted according to the Guidelines for Animal Experimentation of the Fourth Military Medical University (Xi'an, China). Male Sprague–Dawley rats (280 to 300 g, 10-weeks-old) were provided by the Experimental Animal Center of the Fourth Military Medical University and housed under controlled condition with a 12-hour light/dark cycle, temperature at 21°C ± 2°C, and humidity in 60%–70% for at least 1 week before drug treatment or surgery.
To assess the dose-dependent neuroprotective effects of sevo preconditioning on cerebral ischemia–reperfusion injury, we randomly assigned 50 rats to 5 groups: control group; vehicle group, which received 100% oxygen for 1 hour per day; and sevo groups, which received 1%, 2%, or 4% sevo in 100% oxygen for 1 hour per day (Baxter, Guayama, Puerto Rico). Both vehicle and sevo were given for 5 consecutive days. At 24 hours after the last preconditioning, all rats were subjected to middle cerebral artery occlusion (MCAO).
To evaluate the involvement of the initial oxidative stress in the development of ischemic tolerance induced by sevo preconditioning, 70 rats were randomly assigned to 7 groups: control, vehicle, dimethylthiourea (DMTU), N-acetylcysteine (NAC), sevo, sevo + DMTU, and sevo + NAC. Animals in groups sevo, sevo + DMTU, and sevo + NAC were exposed to sevo. The dosage of sevo exposure was screened from experiment 1; vehicle, DMTU, and NAC groups received oxygen preconditioning. Both the free-radical scavenger DMTU and antioxidant NAC were dissolved in saline. DMTU was administered at a dose of 500 mg/kg (i.p.), 1 hour before preconditioning, whereas NAC was supplied at 150 mg/kg (i.p.), 30 minutes in advance (Sigma-Aldrich, St. Louis, MO).12,13 Vehicle and sevo groups received the same volume of saline vehicle.
To study the mechanisms of changes of antioxidant enzyme activity after sevo preconditioning, this experiment consisted of 2 parts. In part 1, 75 rats were grouped into control, vehicle, vehicle + NAC, sevo, and sevo + NAC. At 24 hours after the last preconditioning, all rats were decapitated to evaluate enzyme activity of the right hemisphere (detection kits, Jianchen Biological Institute, Nanjing, China). In part 2, 36 rats were grouped into control, vehicle, and sevo. Blood samples of animals were obtained from femoral arteries to evaluate serum enzyme activity at 24 hours after the last preconditioning. The same volume of warm normal saline was infused through the vena caudalis. Animals in part 2 were then subjected to MCAO.
All rats were acclimated to the animal room for 1 week. Rats in the preconditioning groups were put in a transparent chamber comprising an airtight box (50 × 40 × 30 cm3) with a gas inlet port and an outlet port. During preconditioning, inspired and expired fractions of sevo, oxygen, and carbon dioxide were continuously monitored (MP-60, Phillips Medical Systems, Best, The Netherlands). Carbon dioxide was cleared by using soda lime (Molecular Products Limited, Essex, UK) at the bottom of the container.
Transient Focal Cerebral Ischemia
Focal cerebral ischemia was induced by MCAO in rats using an intraluminal filament technique, as was described previously.14,15 Regional cerebral bloodflow (rCBF) was monitored through a disposable microtip fiberoptic probe (diameter 0.5 mm) connected through a Master Probe to a laser Doppler computerized main unit (PeriFlux 5000, Perimed AB, Stockholm, Sweden). MCAO was considered adequate if rCBF sharply decreased to 30% of the baseline (preischemia) level; otherwise, animals were excluded from analysis. Reperfusion was accomplished by removing the suture after 120 minutes of ischemia, and wounds were sutured.
Arterial Blood Gas Determination
Arterial blood was taken from 20 rats, which were grouped together in experiment 1 (n = 5 for each group). A catheter was inserted to the left femoral artery to draw the blood sample. About 0.3 mL of blood per rat was taken via the femoral artery at the end of the last exposure, onset of MCAO, and reperfusion. Blood gas was immediately analyzed (Rapidlab 1260, Bayer HealthCare, Uxbridge, UK).
Neurobehavioral Evaluation and Infarct Assessment
At 24 hours, 48 hours, and 72 hours after reperfusion, rats' neurologic behaviors were assessed by an observer who was blind to animal groups, according to the method of Garcia et al.16 Animals were then decapitated, and 2-mm-thick coronal sections throughout the brain were stained with 2% 2,3,5-triphenyltetrazolium chloride (TTC, Sigma, St. Louis, MO) to evaluate the infarct volume, as has been described previously.12,14,15 The infarct volume was calculated by Swanson and Sharp's method to correct for edema: 100 × (contralateral hemisphere volume – nonlesioned ipsilateral hemisphere volume)/contralateral hemisphere volume.17
Antioxidant Enzyme Activity Measurement
Samples of the ischemic right hemisphere were homogenized in cold saline with a weight-to-volume ratio of 1:10. The homogenate was centrifuged at 3000 revolutions per minute (rpm) for 15 minutes at 4°C, and the supernatant was removed to a cuvette stored at −80°C. Serum was isolated after blood samples (1 mL) had been centrifuged, and frozen at −80°C until analysis was performed. The measurements of antioxidant enzyme activity were performed according to the technical manuals of the detection kits.11 The activity of superoxide dismutase (SOD), catalase (CAT), and glutathione peroxidase (GSH-px) were respectively surveyed by measuring the absorbance at 550, 240, and 412 nm by using an ultraviolet light spectrophotometer (DU800, Beckman Coulter, Brea, CA).
The software SPSS 11.0 for Windows was used to conduct statistical analyses. All values, except for neurologic scores, were presented as mean ± SD (SD), and analyzed by 1-way analysis of variance (ANOVA). Between-groups differences were detected with a Tukey post hoc test. The neurological scores, presented as median (interquartile range), were analyzed by a nonparametric method (Kruskal–Wallis test) followed by the Mann–Whitney U test with Bonferroni correction. Correlations of the serum antioxidant levels and the percentages of brain infarct volumes were analyzed by using Spearman's rank correlation test. A P value of <0.05 was considered to be statistically significant.
Physiologic variables at the end of the last exposure and during surgery are summarized in Table 1. At the end of the last preconditioning, animals treated with 4% sevo had slight respiratory depression (pH 7.3 ± 0.1; PO2 201.4 ± 6.5 mm Hg; PCO2 56.0 ± 1.5 mm Hg), without hypotension or hypothermia.
The initial CBF before occlusion was recorded as 100%, and subsequent flow changes were expressed in relation to this value. During occlusion, the CBF values remained at <30% of baseline for all rats. At the onset of reperfusion, CBF recovered up to >70% of baseline, and then returned to baseline within 30 minutes (Fig. 1).
Sevoflurane Preconditioning Dose-Dependently Induces Neuroprotection
The neurobehavioral outcomes of groups 4% sevo and 2% sevo were better than those of the control, vehicle, and 1% sevo groups (P < 0.001, for both 4% sevo 14.5 [4.25] and 2% sevo 12.5 [4.5] vs. control 9 [2.5]; P = 0.005, 4% sevo vs. 1% sevo 10 [3.25]; P = 0.029, 2% sevo vs. 1% sevo) at 72 hours after reperfusion. Four percent sevo also induced better neurologic behavior than did 2% sevo (P = 0.044, Fig. 2A).
Two percent sevo and 4% sevo reduced the brain infarct size at 72 hours after stroke (P = 0.035, 2% sevo 37.2 ± 5.9% vs. vehicle 45.8 ± 6.3%; P < 0.001, 4% sevo 29.2 ± 6.7% vs. vehicle). Brain infarct volume percentages in the 4% sevo group were even lower than that in the 1% sevo group (44.4 ± 6.0%, P = 0.003, Fig. 2B). With respect to respiratory depression induced by 4% sevo, we chose 2% as the preconditioning concentration in subsequent experiments.
Neuroprotective Effect of Sevoflurane Preconditioning Is Reversed by DMTU and NAC
As is shown in Figure 3A, sevo preconditioning improved neurologic behavior scores. This beneficial outcome was reversed by preconditioning with DMTU and NAC (P = 0.036 and 0.024, for sevo + DMTU 9 [4.25] and sevo + NAC 9.5 [3.5] vs. sevo 12 [3.5], respectively). Infarct sizes of sevo + DMTU and sevo + NAC groups were larger than those of sevo groups (P = 0.035 and 0.036, for sevo + DMTU 47.6 ± 5.3% and sevo + NAC 48.6 ± 5.0% vs. sevo 38.5 ± 4.9%, respectively, Fig. 3B), but DMTU and NAC had no such effects on brain ischemic outcomes when administered alone.
Antioxidant Enzyme Activities in Brain Tissue Before Ischemia
At 24 hours after the last pretreatment, activities of total SOD (T-SOD) in the right hemispheres in the sevo and O2 vehicle groups were higher than that of the control group (P < 0.001 and P = 0.011 for sevo (4571.9 ± 502.6) and O2 vehicle (4363.0 ± 427.1) vs. control (3654.8 ± 318.6), respectively), and were decreased by NAC (P < 0.001, for both vehicle vs. vehicle + NAC (3333.0 ± 398.2) and sevo vs. sevo + NAC (3471.0 ± 418.2) (Fig. 4A)).
Higher activities of copper (Cu), zinc (Zn)-SOD were found in sevo and vehicle groups (P < 0.001 and 0.001, for sevo (3075.6 ± 466.3) and vehicle (2885.4 ± 251.1) vs. control (2258 ± 208.7), respectively), and were reduced by NAC (P < 0.001, for vehicle vs. vehicle + NAC 2096 ± 98.5 and sevo vs. sevo + NAC 2159.3 ± 185.8, Fig. 4B). However, the activity of manganese (Mn)-SOD was not statistically different among the 5 groups (Fig. 4C).
CAT activity in the sevo group (1.1 ± 0.1) was higher than that in the control (0.7 ± 0.1, P < 0.001), which was decreased by NAC (0.6 ± 0.1, P < 0.001, Fig. 4D).
GSH-px activity in the sevo groups (951.1 ± 96.8) was higher than that in the control (704.9 ± 100.4, P < 0.001) and vehicle (855.8 ± 75.6, P < 0.001) groups, and was significantly decreased in the sevo + NAC group (540.1 ± 77.5, P < 0.001, Fig. 4E).
On the basis of the above results, CAT activity and G SH-px activity, which was increased in the sevo preconditioning groups, were chosen to be further tested in serum.
Serum Antioxidant Activities
As is shown in Figure 5A, no significant CAT activity was observed in serum among the control, vehicle, and sevo groups (Fig. 5A). However, serum GSH-px activity increased significantly in sevo groups (sevo 211.1 ± 29.3 vs. control 172.6 ± 30.2, P = 0.039; Fig. 5B).
Additionally, serum GSH-pX activity negatively correlated with the infarct volume of ischemic brain (correlation coefficient = −0.73, 95% confidence interval [CI] (−1.36–−0.51), Fig. 5C).
Anesthetic preconditioning may prevent or delay neurological complications such as perioperative brain ischemic injury, which can occur during or after surgical procedures, such as carotid endarterectomy and aortic repair.18 The optimal dose to maximize neuroprotection is not clearly established for sevo, a potent neuroprotective drug, and varies widely in the literatures. The sevo preconditioning protocol and dose gradient used in this study were chosen on the basis of previous reports.14,19 In the present study, we verified that 2% sevo pretreatment was sufficient to reduce infarct volume and neurologic deficits, whereas 1% was not, which suggests that a threshold concentration of sevo is needed to induce ischemic tolerance. Although 4% sevo preconditioning induced a better neuroprotective outcome than did 2% sevo, it also contributed to hypercapnia at the end of preconditioning. Hence, it is more reasonable to use 2% to study the mechanisms of sevo preconditioning-induced neuroprotection.
As has been shown, sevo preconditioning-induced tolerance to focal cerebral ischemia–reperfusion injury was reversed by DMTU and NAC. DMTU, a potent free-radical scavenger, mainly scavenges hydroxyl radicals20,21; NAC, a GSH-reducing substrate, is widely used as an antioxidant to eliminate lipid peroxidation and H2O2.22 This paradigm with low dosages was supposed to reduce the ROS induced by preconditioning while not affecting the ROS produced by subsequent ischemia–reperfusion injury.12,13 Our results strongly suggest that an initial oxidative stress generated by sevo preconditioning may trigger cascades that finally lead to ischemic tolerance. This is also consistent with previous publications that the ROS level was elevated immediately after isoflurane preconditioning.23,24
Thereafter, our results showed that sevoflurane preconditioning induced an increase in activities of T-SOD, Cu, Zn-SOD, CAT, and GSH-px in the brain tissue. NAC reduced the elevation of antioxidant enzyme activity, which illustrates a direct relationship between ROS generation and the increase of antioxidant enzyme activity after sevo preconditioning.
However, in the current study, the increase of T-SOD and Cu, Zn-SOD activity was reasonably attributed to the effects of O2 vehicle, because there were no differences between the sevo and control groups. Thus, CAT and GSH-px may play much more important roles than do other antioxidant enzymes in brain ischemic tolerance induced by sevo preconditioning. In previous studies, there was no consistency in the types of increased antioxidant enzymes after different preconditioning. The increase of Cu, Zn-SOD and GSH-px activity was identified after hypoxic preconditioning by Arthur et al.,25 but they failed to identify the increase in CAT activity after hypoxic preconditioning. The up-regulation of CAT and SOD, but not GSH-px, was found in our previous study after hyperbaric oxygen preconditioning. It has also been shown that Mn-SOD activity was increased by some preconditioning.26,27 The discrepancies between our current results and the previous findings could be due to the use of different types of animal models or preconditioning stimuli.
In the serum study, GSH-px activity was significantly increased by sevo rather than CAT, which could be explained in 2 ways: first, the oxidation–antioxidation system in the periphery and central nervous system may have different sensitivities to anesthetic preconditioning. Second, the normal concentration of GSH-px in blood is much higher than that of CAT and may be much easier to detect when it changes. However, these conjectures should be verified by further experiments.
Serum GSH-px activity conferred by sevo preconditioning before brain ischemia had a close negative relationship with infarct volume percentage after reperfusion, which provides a potential method to predict the efficiency of sevo preconditioning before the advent of stroke lesion. This result was in line with the previous finding that GSH-px activity in erythrocytes of stroke-prone spontaneously hypertensive rats decreased significantly at the onset of stroke, and was shown to be useful as an index for judging the progress of stroke biochemically.28 Even so, future studies should be performed to develop GSH-px to be a candidate indicator for predicting the effectiveness of anesthetic preconditioning and the prognosis of stroke lesions.
Several limitations of this study deserve comment. First, among 4 antioxidant enzymes in brain tissue, which we tested before ischemia, there was no difference in T-SOD and its 2 subtypes between the sevo and vehicle preconditioning groups. Because only 1 time point (24 hours after last preconditioning) was tested in our study, we cannot exclude the possibility that SOD participates in the preconditioning effect of sevo. Therefore, the time course of antioxidant enzyme changes after sevo preconditioning during normoxia deserves further study. Second, the ROS-mediated signaling cascades in sevo preconditoning are quite complicated. One mechanism proposed in Riess et al.'s study of sevo preconditioning on guinea pig isolated hearts was that ROS was related to the changes in the KATP channel, an important agent involved in anesthetic preconditioning neuroprotection.29 We have also evaluated previously the results of the role of KATP in isoflurane preconditioning.14 The relationship between ROS and KATP in the mechanisms of sevo preconditioning in the rat brain needs further elucidation.
In conclusion, this study demonstrates for the first time that sevo preconditioning induced cerebral ischemic tolerance in a dose-dependent manner. DMTU and NAC, 2 potent free-radical scavengers, reversed the neuroprotection of sevo preconditioning. NAC also abolished the increase of antioxidant enzyme activity induced by sevo exposure in brain tissue before ischemia. The increased activity of GSH-px in serum conferred by sevo was negatively correlated to the degree of ischemic injury in terms of infarct volumes. These data indicate that sevo preconditioning induces cerebral ischemic tolerance in a dose– response manner through ROS-mediated up-regulation of antioxidant enzyme activities in rats. Our data support the idea that in the future, it might be possible to use serum GSH-px activity as a marker to assess the effectiveness of sevo preconditioning before high-risk surgeries, such as carotid endarterectomy or repair of the ascending aorta, for the alleviation of perioperative cerebral ischemic injury.18
Name: Qianzi Yang, MD.
Contribution: Conduct of study, data interpretation, and manuscript preparation.
Name: Hui Dong, MD, PhD.
Contribution: Study design and manuscript preparation.
Name: Jiao Deng, MD.
Contribution: Conduct of study.
Name: Qiang Wang, MD, PhD.
Contribution: Manuscript preparation.
Name: Ruidong Ye, MD.
Contribution: Conduct of study.
Name: Xuying Li, MD.
Contribution: Data analysis.
Name: Sheng Hu, MD, PhD.
Contribution: Data analysis.
Name: Hailong Dong, MD, PhD.
Contribution: Manuscript review.
Name: Lize Xiong, MD, PhD.
Contribution: Study design, data interpretation, and manuscript review.
We thank Dr. Yan Lu (Department of Anesthesiology, Xijing Hospital, Fourth Military Medical University) for his critical reading of the current version of the manuscript.
1. Warner DS, McFarlane C, Todd MM, Ludwig P, McAllister AM. Sevoflurane and halothane reduce focal ischemic brain damage in the rat. Possible influence on thermoregulation. Anesthesiology 1993;79:985–92
2. Zhang HP, Yuan LB, Zhao RN, Tong L, Ma R, Dong HL, Xiong L. Isoflurane preconditioning induces neuroprotection by attenuating ubiquitin-conjugated protein aggregation in a mouse model of transient global cerebral ischemia. Anesth Analg 2010;111:506–14
3. Limatola V, Ward P, Cattano D, Gu J, Giunta F, Maze M, Ma D. Xenon preconditioning confers neuroprotection regardless of gender in a mouse model of transient middle cerebral artery occlusion. Neuroscience 2010;165:874–81
4. Kapinya KJ, Löwl D, Fütterer C, Maurer M, Waschke KF, Isaev NK, Dirnagl U. Tolerance against ischemic neuronal injury can be induced by volatile anesthetics and is inducible NO synthase dependent. Stroke 2002;33:1889–98
5. Ohtsuki T, Matsumoto M, Kuwabara K, Kitagawa K, Suzuki K, Taniguchi N, Kamada T. Influence of oxidative stress on induced tolerance to ischemia in gerbil hippocampal neurons. Brain Res 1992;25:246–52
6. Sedlic F, Pravdic D, Ljubkovic M, Marinovic J, Stadnicka A, Bosnjak ZJ. Differences in production of reactive oxygen species and mitochondrial uncoupling as events in the preconditioning signaling cascade between desflurane and sevoflurane. Anesth Analg 2009;109:405–11
7. Lamberts RR, Onderwater G, Hamdani N, Vreden MJ, Steenhuisen J, Eringa EC, Loer SA, Stienen GJ, Bouwman RA. Reactive oxygen species-induced stimulation of 5′AMP-activated protein kinase mediates sevoflurane-induced cardioprotection. Circulation 2009;15:10–5
8. Velly LJ, Canas PT, Guillet BA, Labrande CN, Masmejean FM, Nieoullon AL, Gouin FM, Bruder NJ, Pisano PS. Early anesthetic preconditioning in mixed cortical neuronal–glial cell cultures subjected to oxygen–glucose deprivation: the role of adenosine triphosphate dependent potassium channels and reactive oxygen species in sevoflurane-induced neuroprotection. Anesth Analg 2009;108:955–63
9. Orio M, Kunz A, Kawano T, Anrather J, Zhou P, Iadecola C. Lipopolysaccharide induces early tolerance to excitotoxicity via nitric oxide and cGMP. Stroke 2007;38:2812–7
10. Choi YS, Cho KO, Kim EJ, Sung KW, Kim SY. Ischemic preconditioning in the rat hippocampus increases antioxidant activities but does not affect the level of hydroxyl radicals during subsequent severe ischemia. Exp Mol Med 2007;39: 556–63
11. Nie H, Xiong L, Lao N, Chen S, Xu N, Zhu Z. Hyperbaric oxygen preconditioning induces tolerance against spinal cord ischemia by upregulation of antioxidant enzymes in rabbits. J Cereb Blood Flow Metab 2006;26:666–74
12. Zhang X, Xiong L, Hu W, Zheng Y, Zhu Z, Liu Y, Chen S, Wang X. Preconditioning with prolonged oxygen exposure induces ischemic tolerance in the brain via oxygen free radical formation. Can J Anaesth 2004;51:258–63
13. Simerabet M, Robin E, Aristi I, Adamczyk S, Tavernier B, Vallet B, Bordet R, Lebuffe G. Preconditioning by an in situ administration of hydrogen peroxide: involvement of reactive oxygen species and mitochondrial ATP-dependent potassium channel in a cerebral ischemia–reperfusion model. Brain Res 2008;1240:177–84
14. Xiong L, Zheng Y, Wu M, Hou L, Zhu Z, Zhang X, Lu Z. Preconditioning with isoflurane produces dose-dependent neuroprotection via activation of adenosine triphosphate-regulated potassium channels after focal cerebral ischemia in rats. Anesth Analg 2003;96:233–7
15. Wang Q, Peng Y, Chen S, Gou X, Hu B, Du J, Lu Y, Xiong L. Pretreatment with electroacupuncture induces rapid tolerance to focal cerebral ischemia through regulation of endocannabinoid system. Stroke 2009;40:2157–64
16. Garcia JH, Wagner S, Liu KF, Hu XJ. Neurological deficit and extent of neuronal necrosis attributable to middle cerebral artery occlusion in rats. Stroke 1995;26:627–34
17. Swanson RA, Sharp FR. Infarct measurement methodology. J Cereb Blood Flow Metab 1994;14:697–8
18. Selim M. Perioperative stroke. N Engl J Med 2007;356:706–13
19. Dahmani S, Tesnière A, Rouelle D, Desmonts JM, Mantz J. Thiopental and isoflurane attenuate the decrease in hippocampal phosphorylated Focal Adhesion Kinase (pp125FAK) content induced by oxygen–glucose deprivation. Br J Anaesth 2004;93:270–4
20. Wiegand F, Liao W, Busch C, Castell S, Knapp F, Lindauer U, Megow D, Meisel A, Redetzky A, Rusher K, Trendelenburg G, Victorov I, Riepe M, Diener C, Dirnagl U. Respiratory chain inhibition induces tolerance to focal cerebral ischemia. J Cereb Blood Flow Metab 1999;19:1229–37
21. Puisieux F, Deplanque D, Bulckaen H, Maboudou P, Gelé P, Lhermitte M, Lebuffe G, Bordet R. Brain ischemic preconditioning is abolished by antioxidant drugs but does not up-regulate superoxide dismutase and glutathion peroxidase. Brain Res 2004;1027:30–7
22. Abu-Kishk I, Kozer E, Goldstein LH, Weinbaum S, Bar-Haim A, Alkan Y, Petrov I, Evans S, Siman-Tov Y, Berkovitch M. Oral N-acetylcysteine has a deleterious effect in acute iron intoxication in rats. Am J Emerg Med 2010;28:8–12
23. Nguyen LT, Rebecchi MJ, Moore LC, Glass PS, Brink PR, Liu L. Attenuation of isoflurane-induced preconditioning and reactive oxygen species production in the senescent rat heart. Anesth Analg 2008;107:776–82
24. Sang H, Cao L, Qiu P, Xiong L, Wang R, Yan G. Isoflurane produces delayed preconditioning against spinal cord ischemic injury via release of free radicals in rabbits. Anesthesiology 2006;105:953–60
25. Arthur PG, Lim SC, Meloni BP, Munns SE, Chan A, Knuckey NW. The protective effect of hypoxic preconditioning on cortical neuronal cultures is associated with increases in the activity of several antioxidant enzymes. Brain Res 2004;1017: 146–54
26. Garnier P, Demougeot C, Bertrand N, Prigent-Tessier A, Marie C, Beley A. Stress response to hypoxia in gerbil brain: HO-1 and Mn-SOD expression and glial activation. Brain Res 2001;893:301–9
27. Ravati A, Ahlemeyer B, Becker A, Klumpp S, Krieglstein J. Preconditioning-induced neuroprotection is mediated by reactive oxygen species and activation of the transcription factor nuclear factor-kappaB. J Neurochem 2001;78:909–19
28. Murakami T, Takemori K, Yoshizumi H. Prediction of stroke lesions in stroke-prone spontaneously hypertensive rats by glutathione peroxidase in erythrocytes. Biosci Biotechnol Biochem 1995;59:1459–63
29. Riess ML, Kevin LG, McCormick J, Jiang MT, Rhodes SS, Stowe DF. Anesthetic preconditioning: the role of free radicals in sevoflurane-induced attenuation of mitochondrial electron transport in Guinea pig isolated hearts. Anesth Analg 2005;100:46–53