The Effects of Different Anesthesia Techniques on Free Radical Production after Tourniquet-induced Ischemia-reperfusion Injury at Children's Age

Background/Aim. Reperfusion of previously ischemic tissue leads to injuries mediated by reactive oxygen species. The aim of the study was to investigate the effects of different anesthesia techniques on oxidative stress caused by tourniquet-induced ischemia-reperfusion (IR) injury during extremity operations at children's age. Methods. The study included 45 patients American Society of Anesthesiologists (ASA) classification I or II, 8 to 17 years of age, undergoing orthopedic procedures that required bloodless limb surgery. The children were randomized into three groups of 15 patients each: general inhalational anesthesia with sevoflurane (group S), total intravenous anesthesia with propofol (group T) and regional anesthesia (group R). Venous blood samples were obtained at four time points: before peripheral nerve block and induction of general anesthesia (baseline), 1 min before tourniquet release (BTR), 5 and 20 min after tourniquet release (ATR). Postischemic reperfusion injury was estimated by measurement of concentration of malondialdehyde (MDA) in plasma and erythrocytes as well as catalase (CAT) activity. Results. Plasma MDA concentration in the group S was significantly higher at 20 min ATR in comparison with the groups T and R (6.78 ± 0.33 µmolL-1-1 vs 4.07 ± 1.53 and 3.22 ± 0.9. µmolL-1-1 , respectively). There was a significant difference in MDA concentration in erytrocythes between the groups S and T after 5 min of reperfusion (5.88 ± 0.88 vs 4.27 ± 1.04 nmol/mlEr, p < 0.05). Although not statistically significant, CAT activity was slightly increased as compared to baseline in both groups S and R. In the group T, CAT activity decreased at all time points when compared with baseline, but the observed decrease was only statistically significant at BTR Continuous propofol infusion and regional anesthesia techniques attenuate lipid peroxidation and IR injury connected with tourniquet application in pediatric extremity surgery.


Introduction
Pneumatic tourniquets are widely used in pediatric extremity surgery to provide a bloodless field and facilitate dissection.Tourniquet application causes metabolic changes that depend on the tourniquet phase (inflation = ischemia, deflation = reperfusion), the time duration of tourniquet inflation, and the anesthetic technique 1 .Studies have suggested that reperfusion of ischemic tissue may promote potentially harmful pathophysiological reactions 2 .One of these reactions is increased production of free radicals.Free radicals such as superoxide radical (O 2 -•), hydroxyl radical (OH•) and hydrogen peroxide (H 2 0 2 ) are extremely reactive oxygen species (ROS) that have no specific targets and can attack all cellular components and initiate the lipid peroxidation process.Malondialdehyde (MDA) is the end product of lipid peroxidation, and it can be used as a marker of free radical formation.The first line defense mechanism includes antioxidant enzymes such as superoxide dismutase (SOD), catalase (CAT) and glutathione peroxidase (GPx).These enzymes are involved in the clearance of O 2 - • and H 2 O 2 to maintain the structure and function of biological membranes.
The effects of anesthesia on ischemia-reperfusion (IR) injury are of considerable scientific and clinical interest.Propofol, a highly lipid soluble anesthetic, possesses antioxidant activity in vitro [3][4][5][6][7] and in vivo [8][9][10][11] .This property makes it suitable to test for efficacy in clinical and experimental settings of IR injury.Cardioprotection by volatile anesthetics is well established to date 12 .Clinical studies are now warranted to define the role of volatile anesthetic preconditioning in noncardiac tissue which, if conclusive, would be of significant relevance for reduction of perioperative ischemic organ damage and its related morbidity and mortality 13,14 .Some studies reported less hemodynamic and metabolic changes related to tourniquet application in patients undergoing regional anesthesia compared with general anesthesia 1 .Furthermore, local anesthetics inhibit migration, enzyme release and superoxide anion generation of polymorphonuclear leukocytes 15 .To our knowledge, their effects on skeletal muscle ischaemiareperfusion injury in children have not been examined.
In adults various comorbid conditions could overlap and contribute to oxidative stress, thus, studies conducted in children may provide more accurate results and improve our daily practice in a manner to find the best choice of anesthetics for our patient.
The aim of this study was to compare different anesthesia techniques in regard to a possibility to decrease oxidative stress caused by tourniquet-induced IR injury.

Methods
After obtaining the Ethical Committee approval (No. 01-1674, March 13 2008, according to the Article 37 of the School of Medicine in the town of Niš Ethical Committee Regulation) and written informed consent from the parents we studied 45 patients ASA classes I or II, 8 to 17 years of age, undergoing orthopedic procedures that required bloodless limb surgery.
The children were randomized to the sevoflurane (n = 15), propofol (n = 15) or peripheral nerve block (n = 15) group.All the patients were premedicated with midazolam.In the sevoflurane group (group S) general anesthesia was induced with thiopental (5 mg/kg) with alfentanyl (25 mcg/kg) and maintained with inhalation of sevoflurane (3-4 vol %).In the propofol group (group T) anaesthesia was induced with propofol followed by continuous infusion of propofol at the rate of 10 mg/kg/h, reducing to 8 and 6 mg/kg/h, at a 10-minute interval.The maintenance dose was adjusted to clinical signs and anticipated demand.Atracurium (0.6 mg/kg) was given to facilitate tracheal intubation in sevoflurane and propofol group and the lungs were ventilated with 65% nitrogen in oxygen.Rescue analgesia was provided by single bolus doses of alfentanyl (10 mcg/kg).In the regional anesthesia group (group R), the patients received peripheral nerve blocks using bupivacaine 0.25% (volume adjusted according to type of block and patient's weight).Peripheral nerves were identified using a peripheral nerve stimulator.During lower limb operations one arm was used for blood sampling and the other for iv fluid infusions.During upper limb operations contralateral arm was used for iv fluid and propofol infusion and the dorsal vein of the foot was used for blood sampling.The tourniquet was applied at a pressure approximately twice the systolic arterial pressure.No blood transfusions were used; the fluid deficits were corrected with Ringer lactate during the operation.Using heparin-locked iv catheters, sequential venous blood samples were obtained at four time points: before peripheral nerve block and induction of general anesthesia (baseline), 1 min before tourniquet release (BTR), 5 and 20 min after tourniquet release (ATR).Blood samples were centrifuged, plasma was carefully removed and erythrocytes (Er) were washed three times in physiological saline.The packed erythrocytes (0.4 mL) were suspended in 1.6 mL phosphate buffer saline pH 7.4 (to obtain 20% suspension of erythrocytes).The received plasma and erythrocytes were stored at -20 °C until assayed.Postischemic reperfusion injury was estimated by measurement of plasma and erythrocytes MDA levels as well as plasma CAT activity.
Concentration of MDA in plasma was determined by modified thiobarbituric acid (TBA) method and products of reaction were measured at 535 nm after adding FeSO 4

16
. The level of MDA was expressed as μmol/L.
Trichloroacetic acid (TCA) was added to 1 mL of suspended packed erythrocytes, a mixture was centrifuged and 1 mL of the supernatant was transferred into another tube.After the addition of disodium ethylene diamine tetra acetate (NaEDTA) and TBA the tubes were mixed and kept in boiling water for 15 min.After the tubes were cooled MDA content of erythrocytes was estimated as thiobarbituric acid reacting substances by spectrophotometric method at 532 nm as described by Jain et al. 17 .The results were expressed as nmol MDA/mL Er.
The activity of CAT in plasma was determined by spectrophotometric method projected by colored complex between H 2 O 2 and ammonium molibdat 18 .The activity was expressed as U/L.The data were expressed as the mean ± SD.Differences among the three groups were analyzed with the Kruskal-Wallis test.Significant differences between two groups were analyzed with the Mann Whitney U test.Changing patterns of plasma MDA, Er MDA and CAT within group were evaluated by the Friedman's two-way ANOVA, paired samples Student's t test and nonparametric Wilcoxon signed rank test.Differences were considered significant for p < 0.05, using SPSS (Version 15).

Results
There were no significant differences among the groups in age, weight and height (Table 1).There was also no sig- µmolL-1).The concentration of MDA in plasma in the group S 20 min ATR was also significantly higher in comparison with the groups T and R at the same time point (4.07 ± 1.53 and 3.22 ± 0.9 µmolL -1 , respectively).Within the group T a significantly decreased level of plasma MDA was observed 5 min ATR (2.21 ± 1.10 vs 2.77 ± 1.02, p < 0.05).The concentration of MDA in erythrocyte (Figure 2) after 20 min of reperfusion was significantly higher than concentration after 5 min of reperfusion in the group T (4.95 ± 1.39 vs 4.27 ± 1.04 nmol/mL Er), while 5 min ATR values (5.88 ± 0.88 nmol/mL Er) in the group S were significantly higher (p < 0.05) than baseline (5.11 ± 1.53), BTR (5.02 ± 1.10), and 20 min ATR (4.98 ± 0.80) values.There was a significant difference between the groups S and T after 5 min of reperfusion (5.88 ± 0.88 vs 4.27 ± 1.04 nmol/mL Er, p < 0.05).There were no significant differences in the group R although a slight decrease in MDA erythrocyte values was noted in all time points compared to the baseline.Although no statistically significant, CAT activity (Figure 3) was slightly increased compared to baseline in both groups S and R (p > 0.05).In the group T, CAT activity decreased at BTR and 5 and 20 min ATR when compared with the baseline, but the observed decrease was only statistically significant at BTR (34.70 ± 9.27 vs 39.69 ± 12.91 UL -1 , p < 0.05).Group S -sevoflurane group; Group T -propofol group; Group R -regional anestesia group; BTR -before tourniguet release; ATR -after tournignet release

Discussion
Prolonged ischemia with tourniquet inflation and subsequent reperfusion causes lipid peroxidation resulting in tissue injury.Although skeletal muscle is thought to be relatively insensitive to the deleterious effects of ischemia and subsequent reperfusion, injury can occur as a result of ischemia such as tourniquet application 9,19 .
Liposoluble anaesthetic drug propofol (2,6diisopropylphenol) shares a similar structure with phenolic antioxidants like the endogenous alpha-tocopherol (vitamin E) which has been shown to protect cellular membranes against lipid peroxidation processes induced by ROS 20 .A commercial form of propofol (Recofol ® : propofol 10 mg/mL) is formulated in intralipid, lipid vehicle emulsion (10% soya bean emulsion, egg phosphatides and glycerol).Direct effects of propofol on ROS are unknown.As propofol did not enter the cells, the drug would act by scavenging the active oxygen species released in extracellular medium.Intralipids could also act directly on cell membranes, and induce structure alterations leading to a decrease of ROS release in the extracellular medium 5 .Numerous experimental studies suggest that volatile anesthetics may protect beyond the heart various tissues and organs subjected to ischemic insult 13 .A mechanism by which potent inhalation anesthetics may inhibit free radicals is not known but may involve reducing intracellular calcium concentrations and enhancing availability of interstitial glycolysis metabolites (glucose, lactate and pyruvate) in the skeletal muscle during ischemia and reperfusion 21 .A difficulty in studying the effects of anesthetic agents, specifically in vivo, is a frequent lack of comparison with a non-anaesthetized control group, as many studies have been conducted using an acutely instrumented model requiring at least basal anaesthesia 2 .Although iv midazolam, given as a premedication agent to all patients, seems to be an advantageous sedative with amnesic effect, it has been reported as having no effect on ROS production 22 .The use of alfentanyl can contribute to a decrease in the surgical stress factor, as opioids minimize intraoperative he-modinamic changes 23 , but there are some limited reports of its antioxidative activity 24 .Other drugs utilized in our study, a muscle relaxant, as well as atropine and neostigmine, did not exert any antioxidant activity in vitro 25 .Regional anesthesia cases (group R) were included in our study for two reasons: to investigate the effects of regional anesthesia on ROS formation and to use this group as a control one for the groups S and T. The pattern of ROS formation from surgical stress is different from that of reperfusion injury.Release of a tourniquet causes an abruptly massive production of ROS and starts oxidative damage 11 .
In our study, concentration of MDA in plasma decreased nonsignificantly in the groups S and R before reperfusion (BTR) compared to the baseline, which might had been the result of the effects of sedation from releasing the stress in the operating room (group R) and from the effect of volatile anesthetic (group S).Allaouchiche et al. 10 reported that sevoflurane did not induce a chemical reaction sequence leading to the generation of oxygen free radicals in their study.It was shown also that maintenance of general anesthesia with clinically relevant concentrations of nitrous oxide or volatile anesthetic was not associated with an increase in toxic oxygen metabolites.The pre-reperfusion (BTR) plasma concentration of MDA in the group T seems to be influenced by the tourniquet time which was significantly longer in this group compared to the group R. In the same group T we observed the most prominent decrease in plasma MDA levels 5 min ATR.This reduction may be explained by the fact that tissues below the tourniquet were saturated with propofol.Concentration of MDA in plasma in the group S 20 min ATR was significantly higher in comparison with the groups T and R at the same time point.Kotani et al. 26 found that sevoflurane induced an inflammatory response, whereas the changes in expression of proinflammatory cytokines occured during exposure to sevoflurane.Turan et al. 27 observed similar changes in MDA concentration among general, total intravenous and regional anesthesia groups in their study.This may be due to effects of both propofol to reduce oxidative stress as free radical scavenger and regional anesthesia which reduce the stress-inducing hormones, such as adrenaline, noradrenaline and cortisol [28][29][30] .Moreover, Erturk et al. 31 concluded that even low-dose infusion of propofol (2 mg/kg/h after a 0.2 mg/kg bolus) offered protection from tourniquet-induced IR injury in arthroscopic knee surgery under spinal anesthesia.These findings were not confirmed in pateints submitted to elective nonischaemia-nonreperfusion surgery 23 .
Concentration of MDA in erytrocytes after 20 min of reperfusion was significantly higher than concentration after 5 min of reperfusion in the group T. This increase is in accordance with Cheng et al. 11 study in which the highest ROS production measured by chemiluminescence happened at 20 min after reperfusion.On the contrary, 5 min ATR values in the group S were significantly higher (p < 0.05) than baseline and 20 min ATR values possibly as a result of sudden oxidative burst after tourniquet release caused by high oxygen concentration in inhaled gas mixture (35%).A significant difference in erythocyte MDA between the groups S and T after 5 min of reperfusion seems to be in line with this figure reported earlier in the literature that propofol protect erythrocytes against oxidative stress 3 .There were no significant differences within the group R although slight decrease in MDA erythrocyte values was noted at all time points compared to baseline, possibly as a result of bupivacaine property to inhibit surface receptor expression, phagocytosis, and oxidative burst in a time-and concentration-dependent manner 15 .
A role of CAT in oxidative stress adaptation is incompletely understood as compared to small molecular weight antioxidants.Literature data concerning the activities of enzymatic antioxidants in IR patients are contradictory 32,33 .This could be explained by adaptation specificity, depending on tissue type, which, probably, is linked with the tissue metabolism character and function.It is also possible that changes in CAT activity are a consequence of apparent posttranslational modification of the enzyme.CAT is an enzyme exerting a dual function; it catalyzes decomposition of hydrogen peroxide (H 2 O 2 ) to produce water and oxygen (catalytic activity) or oxidation of H donors (peroxidatic activity) and under normal circumstances their ratio is 30 : 70.Another mechanism of the enzyme specific activity alterations following exposure to oxidative stress might be a change of the ratio of these two reactions 34 .Al-Abrash et al. 35 found an increase in catalase activity in all studied patients suffering from oxidative stress (cardiovascular diseases, diabetes, tumor, inflammation, dermatological diseases, anemia and Wilson's disease).In our study, CAT activity 20 min ATR was increased compared to the baseline in the groups S and R, and decreased in the group T.These alterations were out statistically significant.The only statistically significant decrease was observed 1 min before the reperfusion in the group T compared with the baseline value.Moreover, this reduction in CAT activity was accompanied by a concomitant increase of MDA production.Turan et al. 27 observed significantly lower CAT activity 5 and 20 min ATR in patients receiving general anesthesia maintained with halothane and supposed that this effect was related with increased H 2 0 2 production.On the contrary, increased plasma CAT activity was noted after reperfusion in experimental model of skeletal muscle IR injuries conducted by Bosco et al. 36 and unchanged CAT activity was found in rats anesthetized with either thiopental, etomidate, ketamine or propofol given in an anesthetic dose 37 .

Conclusion
This study demonstrates that total intravenous anesthesia with propofol and regional anesthesia techniques provide better antioxidant defense than general inhalational anesthesia with sevoflurane against IR injury related to tourniquet application in pediatric extremity surgery.

Table 1 Characteristics of the patients
*Data are expresed like mean±standard deviation or number of patient Group S -sevoflurane group; Group T -propofol group; Group R -regional anestesia group; UEX -upper extremities; LEX -lower lextremities *Grup T vs Group R