Oxidative stress and antioxidant status in patients undergoing prolonged exposure to hyperbaric oxygen

Clinical Biochemistry 37 (2004) 312 – 317

Oxidative stress and antioxidant status in patients undergoing prolonged exposure to hyperbaric oxygen Serena Benedetti, a,* Antonio Lamorgese, b Michele Piersantelli, a Silvia Pagliarani, a Francesca Benvenuti, a and Franco Canestrari a a

Institute of Biological Chemistry ‘‘G.Fornaini’’, University of Urbino ‘‘Carlo Bo’’, 2-61029 Urbino (PU), Italy b Hyperbaric Medicine, Therapy and Research Center, Fano (PU), Italy Received 14 October 2003; received in revised form 10 December 2003; accepted 12 December 2003

Abstract Objectives: To evaluate the condition of oxidative stress in patients undergoing prolonged exposure to hyperbaric oxygen (HBO) and the possible modifications of the antioxidant defense systems in the absence of antioxidant supplementation. Design and methods: Twelve patients exposed to 15 HBO treatments for pathological conditions related to hypoxia were included in the study. Oxidative stress indices as well as plasma and erythrocyte antioxidant levels were measured in blood samples collected both at the 1st and 15th HBO session. Results: The repeated exposures to HBO led to a significant accumulation of plasmatic reactive oxygen metabolites (ROM) and malondialdehyde (MDA). After 15 HBO sessions, no relevant differences were detected for reduced glutathione (GSH), a-tocopherol, and retinol plasma levels; however, a significant decrease in erythrocyte superoxide dismutase (SOD) and catalase (CAT) activity was observed when compared to the 1st HBO exposure; glutathione peroxidase (GPx) activity remained almost unchanged. Conclusions: In the absence of antioxidant supplementation, the prolonged HBO treatment leads to a condition of oxidative stress that seems to affect in particular the response of the enzymatic antioxidant defense system; the possible relationship between the chemical modifications of the enzymes caused by oxygen reactive species and the consequent inactivation of the proteins is under investigation. D 2004 The Canadian Society of Clinical Chemists. All rights reserved. Keywords: Hyperbaric oxygen therapy; Oxidative stress; Antioxidant status; Reactive oxygen metabolites

Introduction Hyperbaric oxygen (HBO) therapy is achieved by exposing the patient to a barometric pressure higher than ambient pressure (1 atm absolute pressure or 1 ATA), while the patient breathes 100% O2 [1,2]. Commonly used therapeutic pressures range from 2 to 2.8 ATA, although lower pressures, for example, 1.8 ATA, have also been used clinically. In general, HBO exposure is repeated daily, or at most twice daily in urgent or emergency situations, for a duration of 60 –90 min. Exposure to HBO favorably leads to an increase of dissolved oxygen in the blood, and it has been successfully

* Corresponding author. Institute of Biological Chemistry ‘‘G.Fornaini’’, University of Urbino ‘‘Carlo Bo’’, Via Saffi, 2-61029 Urbino (PU), Italy. Fax: +39-0722-320188. E-mail address: [email protected] (S. Benedetti).

used for the treatment of a variety of clinical conditions related to hypoxia, including decompression sickness, acute carbon monoxide intoxication, air embolism, soft tissue infections, radiation necrosis, and impaired wound healing [3]. However, it has been demonstrated that HBO also leads to increased reactive oxygen species formation that can cause cellular damage with lipid, protein, and DNA oxidation [4,5]. For this reason, HBO therapy seems to be an excellent model system for the investigation of oxidative stress and its biological consequences. To date, the cellular response to oxidative stress has been principally investigated in animal models [6,7], and only a few studies have evaluated the effects of HBO in humans. Recently, Dennog et al. [8] determined the in vivo antioxidant response to a single HBO exposure (3
20-min periods) in healthy subjects, finding no significant differences in antioxidant levels and antioxidant enzymes involved in the primary defense against oxidative damage

0009-9120/$ - see front matter D 2004 The Canadian Society of Clinical Chemists. All rights reserved. doi:10.1016/j.clinbiochem.2003.12.001

S. Benedetti et al. / Clinical Biochemistry 37 (2004) 312–317

before and 24 h after the HBO session. However, these study conditions are quite different and not applicable during an HBO therapeutic protocol, when the patient is exposed to repeated HBO treatments depending on the underlying disease. Here, a clinical study involving 12 patients exposed to 15 consecutive HBO treatments (1 session/day) for pathological conditions related to hypoxia is reported. None of the patients received antioxidant supplementation before the beginning and during the therapy. To evaluate the in vivo antioxidant response to the prolonged HBO exposure, we measured some oxidative stress indices as well as plasma and cellular antioxidant levels in blood samples collected both at the 1st and 15th HBO session.

Materials and methods Subjects Twelve patients (M = 8, F = 4, mean age of 57.5 F 15.0 years) at the Hyperbaric Therapy Center of Fano (Italy) were enrolled in the clinical study after giving informed consent. Pathologies treated by HBO therapy were diabetic feet (n = 4), refractory chronic osteomyelitis (n = 4) and aseptic osteonecrosis (n = 4). Specific exclusion criteria considered for the present study were smoking, malnutrition (serum albumin cut-off <3.3 g/dl), and oral antioxidant supplementation at the moment of the enrollment. Due to their clinical condition, none of the patients practiced physical training. Eight volunteers, belonging to the medical assistance staff, were also included in the study as a reference group. Hyperbaric protocol Patients were exposed to 15 consecutive HBO treatments (1 session/day) according to a routine therapy protocol: the multiplace chamber (Sistemi Iperbarici, Ardea, Italy) was pressurized with compressed air while the patients breathed 100% O2 using a mask at a pressure of 2.5 ATA for a total of two 30-min periods, with a 3-min interval during which the patients breathed air. Sample collection From each patient, venous heparinized blood samples were taken before the beginning of the HBO session and immediately on exit from the chamber, both at the 1st and 15th HBO treatment. Heparinized blood was also collected from the reference group. Tubes were centrifuged and plasma aliquots were taken. After the removal of the buffy coat, erythrocytes were washed three times in buffered saline; hemolysate was obtained by adding 19 vol of ice-cold distilled water to 1 vol of packed erythrocytes. Hemoglobin content in erythro-

313

cyte lysate was determined at 540 nm after addition of the Drabkins solution. Biochemical analysis The following parameters were monitored during the study: reactive oxygen metabolites (ROM) and malondialdehyde (MDA) as indices of oxidative stress; reduced glutathione (GSH), a-tocopherol, and retinol as non-enzymatic antioxidants; erythrocyte superoxide dismutase (SOD), catalase (CAT), and glutathione peroxidase (GPx) as enzymatic antioxidants. Colorimetric determination of ROM ROM were evaluated in plasma samples by the use of the ‘‘d-ROM test’’ (Diacron s.r.l., Grosseto, Italy) [9,10]. In this test, plasmatic ROM (mainly hydroperoxides), in presence of iron (that is released from plasma proteins by an acidic buffer, the R2 reagent of the kit), are able to generate alkoxyl and peroxyl radicals, according to the Fenton’s reaction. Such radicals, in turn, are able to oxidize an alkyl-substituted aromatic amine (that is dissolved in a chromogenic mixture, the R1 reagent of the kit), thus transforming them into a pink derivative photometrically quantified at 505 nm. The intensity of the developed color is directly proportional to the concentration of ROM, according to the Lambert – Beer’s law. Briefly, 10 Al of chromogenic substrate (R1) and 1 ml of buffer, pH 4.8 (R2), were mixed with 10 Al of plasma. A blank reagent obtained by replacing plasma with distilled water and a standard with assigned value were included for each series of assays. After 1 min of incubation at 37jC, each sample underwent photometric reading by measuring their absorbance at 505 nm immediately and after 1, 2, and 3 min. The results of d-ROM test are expressed in arbitrary units called ‘‘Carratelli Units’’ (CARR U), according to the following formula: CARR U ¼ F
ðDAbs=minÞ where F is a correction factor with an assigned value (approximately 9000 at 37jC according to the results obtained with the standard); (DAbs/min) are the mean differences of the absorbances recorded at 1, 2, and 3 min. As reported in Ref. [11], the linearity range of d-ROM test is between 50 and 500 CARR U; for values up to 500 CARR U, a sample dilution is required. Intra-assay coefficient of variation is 2.1%, while inter-assay is 3.1%. It has been experimentally established that 1 CARR U corresponds to 0.08 mg of H2O2/dl. Reference values of healthy subjects are between 250 and 300 CARR U; conditions of slight, medium, and high oxidative stress are defined, respectively, by values of 320 –360, 360– 400, and >400 CARR U.

314

S. Benedetti et al. / Clinical Biochemistry 37 (2004) 312–317

Table 1 Oxidative stress indices in patients at the 1st and 15th HBO exposure

Before HBO After HBO

Plasmatic ROM (CARR U)

Plasmatic MDA (Amol/l)

Erythrocyte MDA (Amol/grHb)

1st HBO

15th HBO

1st HBO

15th HBO

1st HBO

15th HBO

276 F 32 344 F 43

379 F 31* 393 F 30

10.91 F 1.28 12.12 F 0.89

15.97 F 2.44* 17.63 F 2.57

0.355 F 0.046 0.401 F 0.028

0.382 F 0.074 0.386 F 0.069

Reference group: ROM, 271 F 28 CARR U; plasmatic MDA, 9.89 F 0.85 Amol/l; erythrocyte MDA, 0.272 F 0.032 Amol/grHb. * P < 0.05 vs. 1st treatment (before HBO), t test for paired data.

Results obtained with d-ROM test have been validated by electronic spin resonance spectroscopy, the most fit experimental technique for revealing and identifying the presence of radicals in a biological sample [12].

bition monitored at 340 nm. One unit of activity is the amount of enzyme that determines a 50% inhibition of the NADH oxidation rate. CAT and GPx activities were determined in erythrocytes as previously described [14].

Determination of thiobarbituric reacting substances

Vitamin assay

Thiobarbituric reacting substances (TBARS) measurement, commonly known as MDA, was carried out in plasma and erythrocyte lysate at 535 nm according to the thiobarbituric acid test as described in Ref. [13].

Plasma levels of a-tocopherol and retinol were measured by reversed-phase HPLC following the method described in Ref. [16] with minor modifications. Briefly, 0.1 ml of plasma was added to 0.1 ml of ethanol (containing a-tocopherol acetate as internal standard, from Sigma, Milan, Italy) and to 0.5 ml of hexane. After centrifugation of the mixture, the hexane phase was drawn, concentrated to dryness under nitrogen current, and resuspended in 0.4 ml of methanol. The assay was performed using a Alltima C18 column (4.6
250 mm, 5 Am, from Alltech, Milan, Italy) equipped with a guard column Alltima C18 (4.6
7.5 mm, 5 Am). The eluent phase was constituted by methanol/water (98:2); the flow rate was 1.6 ml/min. UV detection was carried out at 292 nm for a-tocopherol and at 325 nm for retinol. All the organic solvents used were pure HPLC-grade from Carlo Erba, Milan, Italy. The HPLC instrumentation was from Jasco Corporation, Tokyo, Japan.

GSH determination GSH levels in hemolysate and plasma samples were measured spectrophotometrically at 412 nm by titration with 5,5V-dithiobis(2-nitrobenzoic acid) as described by Beutler [14]. Enzyme assay SOD activity was evaluated in erythrocytes following the method described in Ref. [15]. This method is based on the measurement of superoxide-driven NADH oxidation inhi-

Fig. 1. Effect of HBO on plasmatic ROM levels in patients exposed to 15 HBO treatments. Before the beginning of the HBO therapy, ROM levels were comparable in patients and reference subjects; increments of 24.6% and 37.3% were observed at the end of the 1st HBO treatment and before the 15th session, respectively (*P < 0.05 vs. 1st HBO, t test for paired data).

Fig. 2. Effect of HBO on plasma MDA levels in patients exposed to 15 HBO treatments. At the end of the 1st session, an increment of 11.1% in MDA levels was found; before the 15th treatment, the increment was of 46.4% (*P < 0.05 vs. 1st HBO, t test for paired data).

S. Benedetti et al. / Clinical Biochemistry 37 (2004) 312–317

315

Table 2 Non-enzymatic antioxidant levels in patients at the 1st and 15th HBO treatment

Before HBO After HBO

Plasmatic GSH (Amol/ml)

Erythrocyte GSH (Amol/grHb)

a-Tocopherol (Amol/l)

Retinol (Amol/l)

1st HBO

15th HBO

1st HBO

15th HBO

1st HBO

15th HBO

1st HBO

15th HBO

0.293 F 0.024 0.325 F 0.041

0.307 F 0.032 0.276 F 0.035

6.22 F 0.29 6.61 F 0.48

6.27 F 0.29 5.71 F 0.33

34.5 F 2.7 37.5 F 0.8

34.3 F 2.2 34.3 F 2.9

2.53 F 0.14 2.38 F 0.25

2.55 F 0.36 2.33 F 0.31

Reference group: plasmatic GSH, 0.362 F 0.029 Amol/ml; erythrocyte GSH, 6.70 F 0.24 Amol/grHb; a-tocopherol, 36.4 F 0.8 Amol/l; retinol, 2.82 F 0.15 Amol/l.

Statistics and data processing

Non-enzymatic antioxidant defenses

Results are expressed as means F standard error. The statistical analysis was carried out using the Student’s t test for paired data. Probability values of <0.05 were accepted. Statistics and graphs were obtained using the software Microcalk Origin 6.0 (Microcal Software, Inc., Northampton, MA, USA).

Table 2 shows the levels of the non-enzymatic antioxidants GSH, a-tocopherol, and retinol. After the 1st HBO exposure, an increment of GSH levels was observed; in fact, plasmatic GSH showed a 10.9% increase and erythrocyte GSH a 6.3% increase when compared to the starting levels. Before the 15th treatment, GSH values were comparable to those found before the 1st exposure, but at the end of the 15th HBO session, a decrease in GSH values was observed (10.1% and 8.9% for plasmatic and erythrocyte GSH, respectively). Regarding the liposoluble vitamins a-tocopherol and retinol, plasma levels did not show significant modifications during the repeated HBO exposures; a slight decrement of retinol plasma levels was observed at the end of the 1st and 15th HBO treatment compared to the starting values (5.9% and 8.6%, respectively).

Results Indices of oxidative stress Oxidative stress indices are shown in Table 1. Before the 1st HBO exposure, patients presented mean levels of plasmatic ROM within the reference range (250 – 300 CARR U) comparable to the reference group (276 F 32 vs. 271 F 28 CARR U). Following the 1st treatment, an increment of ROM levels (+24.6%) with values indicating slight oxidative stress (320 – 360 CARR U) was observed. Before the 15th HBO session, a condition of medium oxidative stress (360 – 400 CARR U) with values of 379 F 31 CARR U (+37.3%, P < 0.05) was reached (Fig. 1). Immediately after the 15th treatment, ROM levels of 393 F 30 CARR U were detected. The accumulation of ROM during the prolonged exposure to HBO was also confirmed by the increased levels of MDA; in fact, an increment of 11.1% in MDA plasmatic levels after the 1st HBO exposure and an increment of 46.4% ( P < 0.05) before the 15th session as compared to levels before the 1st treatment were observed (Fig. 2). Although less evident, an increment of erythrocyte MDA was also detected at the end of the 1st treatment (+13.0%) and before the 15th HBO session (+7.3%).

Erythrocyte enzymatic antioxidant defenses The response of the enzymatic antioxidant defense system to HBO is shown in Table 3. Preceding the HBO therapy, the activities of erythrocyte SOD, CAT, and GPx were comparable in patients and reference subjects. At the end of the 1st HBO session, SOD activity seemed to remain stable as compared to the starting value (2023 F 121 vs. 2016 F 128 U/grHb), while CAT and GPx activities increased, respectively, by 16.0% and 13.1% (12.3 F 1.4 vs. 10.6 F 0.6
104 U/grHB for CAT and 45.8 F 4.5 vs. 40.5 F 3.1 U/grHb for GPx). Immediately before the 15th HBO treatment, a significant decrease of SOD activity (21.1%) was observed compared to the value detected before the 1st HBO session (1590 F 190 vs. 2016 F 128 U/grHb, P < 0.05) (Fig. 3, panel A). The same course was found for CAT: its activity decreased by 20.8% before the 15th HBO session compared to the 1st treatment (8.4 F 1.4 vs. 10.6 F 0.6


Table 3 Erythrocyte enzymatic activities in patients at the 1st and 15th HBO session CAT (104 U/grHb)

SOD (U/grHb)

Before HBO After HBO

GPx (U/grHb)

1st HBO

15th HBO

1st HBO

15th HBO

1st HBO

15th HBO

2016 F 128 2023 F 121

1590 F 190* 1545 F 218

10.6 F 0.6 12.3 F 1.4

8.4 F 1.4* 9.5 F 1.1

40.5 F 3.1 45.8 F 4.5

43.8 F 2.6 47.8 F 3.7

Reference group: SOD, 1914 F 42 U/grHb; CAT, 9.7 F 0.6
104 U/grHb; GPx, 43.1 F 2.5 U/grHb. * P < 0.05 vs. 1st treatment (before HBO), t test for paired data.

316

S. Benedetti et al. / Clinical Biochemistry 37 (2004) 312–317

Fig. 3. Erythrocyte SOD, CAT, and GPx activities evaluated immediately before the 1st and 15th HBO treatment. After the prolonged exposure to HBO, significant decrements were found for SOD and CAT activities (21.1% and 20.8%, respectively; *P < 0.05, t test for paired data); GPx was not affected by HBO.

104 U/grHb, P < 0.05; Fig. 3, panel B). Differently from SOD and CAT, no decrement in GPx activity was observed before the 15th HBO session (43.8 F 2.6 vs. 40.5 F 3.1 U/grHb; Fig. 3, panel C).

Discussion The presence of increased levels of reactive oxygen species in the blood of patients exposed to HBO has been previously demonstrated [5], but only a few studies have investigated the in vivo antioxidant response to oxidative stress in patients undergoing prolonged HBO therapy. In this clinical study, we originally evaluated the antioxidant status of 12 subjects exposed to 15 HBO treatments for pathological conditions related to hypoxia receiving no antioxidant supplementation. The possibility of accurately measuring the oxidative state in a biological system in a simple manner is of fundamental importance for clinical diagnostics. In this contest, the ‘‘d-ROM test’’ provides a simple, inexpensive, and practical method to identify subjects with a high level of oxidative stress [9,10]. Through the evaluation of hydroperoxides in plasma samples, no accumulation of ROM was found in patients before HBO exposure. However, immediately after the 1st HBO session, a condition of slight oxidative stress was observed, and after 15 HBO treatments, a situation of medium – high oxidative stress was reached. These data seem to indicate that when HBO exposure is prolonged, the antioxidant defense systems fail to maintain ROM levels within the reference range. The significant accumulation of ROM was also confirmed by the presence of increased levels of plasmatic MDA. Isoprostanes surely are more sensitive and specific markers of lipid peroxidation [17]; however, the plasmatic accumulation of MDA after the prolonged HBO exposure is so marked that we

think the interpretation of the data is not problematic in this contest. Regarding the in vivo antioxidant response to repeated HBO treatments, no significant modifications were noticed in the liposoluble vitamin levels; in fact, at the time of 15th HBO exposure, a-tocopherol and retinol plasma levels were almost comparable to those found before the 1st HBO session. One might ask why were there no changes in non-enzymatic antioxidants because one would expect these to be consumed first. However, it has been suggested that the depletion of antioxidants at the target sites due to oxidative stress activates signal transduction pathways with antioxidant mobilization from the body’s antioxidant stores through the plasma via a transport mechanism involving very low-density lipoproteins [18]. Our data seems to fit with this hypothesis. In accord with previous studies in animal models [19], an increment of plasmatic and erythrocyte GSH was detected immediately after the 1st HBO exposure, probably as an adaptative response to oxidative stress; however, this finding was not confirmed at the end of the 15th treatment when a decrement was observed. Significant modifications were found regarding the enzymatic antioxidant defense system. Generally, as response, an increased rate of radical production leads to an increment in the levels of antioxidant enzymes; in fact, CAT and GPx showed increased activities immediately after the 1-h HBO exposure, both after the 1st and 15th treatment. SOD activity apparently remained unchanged at the end of the HBO session, but preliminary experiments seem to indicate that the enzyme presents a maximum activity after 30 min of HBO exposure (unpublished observation). Interestingly, at the time of 15th treatment, a significant decrement in SOD and CAT activities (20% each) was observed compared to the 1st HBO exposure, while GPx activity did not undergo any appreciable modifications. The decreased antioxidant enzyme activity has been described in several in vivo situations [20,21], and it has been proposed that the most likely, species involved in SOD and CAT inactivation is singlet oxygen. In particular, Escobar et al. [22] demonstrated that SOD and CAT could be readily inactivated by singlet oxygen and peroxyl radicals, and when exposed to similar singlet oxygen concentrations, a parallel inactivation of both enzymes can be expected. This seems to correlate with our findings on the decrement of SOD and CAT activities in the presence of increased levels of plasmatic ROM. Further studies are already in progress investigating the possible relationship between the chemical modifications of the enzymes caused by oxygen reactive species and the consequent inactivation of the proteins. Moreover, a new clinical trial has just begun for the evaluation of the antioxidant response in patients similarly undergoing prolonged HBO therapy this time supplemented with oral antioxidants for the entire duration of the treatment.

S. Benedetti et al. / Clinical Biochemistry 37 (2004) 312–317

References [1] Tibbles PM, Edelsberg JS. Hyperbaric-oxygen therapy. N Engl J Med 1996;334:1642 – 8. [2] Leach RM, Rees PJ, Wilmshurst P. Hyperbaric oxygen therapy. BMJ 1998;317:1140 – 3. [3] Bakker DJ. Hyperbaric oxygen therapy: past, present and future indications. Adv Exp Med Biol 1992;317:95 – 105. [4] Jamieson D, Chance B, Cadenas E, Boveris A. The relation of free radical production to hyperoxia. Annu Rev Physiol 1986;48: 703 – 19. [5] Narkowicz CH, Vial JH, McCartney P. Hyperbaric oxygen therapy increases free radical levels in the blood of humans. Free Radic Res Commun 1993;19:71 – 80. [6] Harabin AL, Braisted JC, Flynn ET. Response of antioxidant enzymes to intermittent and continuous hyperbaric oxygen. J Appl Physiol 1990;69:328 – 35. [7] Etlik O, Tomur A, Dundar K, Erdem A, Gundogan NU. The effect of antioxidant vitamins E and C on lipoperoxidation of erythrocyte membranes during hyperbaric oxygenation. J Basic Clin Physiol Pharmacol 1997;8:269 – 77. [8] Dennog C, Radermacher P, Barnett YA, Speit G. Antioxidant status in humans after exposure to hyperbaric oxygen. Mutat Res 1999;428: 83 – 9. [9] Cesarone MR, Belcaro G, Carratelli M, et al. A simple test to monitor oxidative stress. Int Angiol 1999;18:127 – 30. [10] Trotti R, Carratelli M, Barbieri M. Performance and clinical application of a new, fast method for the detection of hydroperoxides in serum. Panminerva Med 2002;44:37 – 40. [11] Trotti R, Carratelli M, Barbieri M, et al. Oxidative stress and a thrombophilic condition in alcoholics without severe liver disease. Haematologica 2001;86:85 – 91. [12] Alberti A, Bolognini L, Macciantelli D, Carratelli M. The radical

[13] [14] [15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

317

cation of N,N-diethyl-para-phenylenediamine: a possible indicator of oxidative stress in biological samples. Res Chem Intermed 2000; 26:253 – 67. Buege JA, Aust SD. Microsomal lipid peroxidation. Methods Enzymol 1978;52:302 – 10. Beutler E. Red cell metabolism. A manual of biochemical methods. New York: Grune and Stratton; 1984. Paoletti F, Aldinucci D, Mocali A, Caparrini A. A sensitive spectrophotometric method for the determination of superoxide dismutase activity in tissue extracts. Anal Biochem 1986;154:536 – 41. Catignani GL. An HPLC method for the simultaneous determination of retinol and alpha-tocopherol in plasma or serum. Methods Enzymol 1986;123:215 – 9. Roberts II LJ, Morrow JD. Products of the isoprostane pathway: unique bioactive compounds and markers of lipid peroxidation. Cell Mol Life Sci 2002;59:808 – 20. Elsayed NM. Antioxidant mobilization in response to oxidative stress: a dynamic environmental – nutritional interaction. Nutrition 2001;17: 828 – 34. Shang F, Gong X, Egtesadi S, et al. Vitamin C prevents hyperbaric oxygen-induced growth retardation and lipid peroxidation and attenuates the oxidation-induced up-regulation of glutathione in guinea pigs. J Nutr Biochem 2002;13:307 – 13. Fernandez V, Llesuy S, Solari L, Kipreos K, Videla L, Boveris A. Chemiluminescent and respiratory responses related to thyroid hormone-induced liver oxidative stress. Free Radic Res Commun 1988;5: 77 – 84. Llesuy S, Evelson P, Gonzalez-Flecha B, et al. Oxidative stress in muscle and liver of rats with septic syndrome. Free Radic Biol Med 1994;16:445 – 51. Escobar JA, Rubio MA, Lissi EA. Sod and catalase inactivation by singlet oxygen and peroxyl radicals. Free Radic Biol Med 1996;20: 285 – 90.