In Vivo Effect of Celecoxib and Tenoxicam on Oxidant/ Anti-Oxidant Status of Patients with Knee Osteoarthritis

Annals of Clinical & Laboratory Science, vol. 35, no. 2, 2005 137

In Vivo Effect of Celecoxib and Tenoxicam on Oxidant/ Anti-oxidant Status of Patients with Knee Osteoarthritis

Salih Ozgocmen,1 Ozge Ardicoglu,1 Hasan Erdogan,2 Ersin Fadillioglu,3 and Huseyin Gudul1 1Division of Rheumatology, Department of Physical Medicine and Rehabilitation, Firat University Faculty of Medicine, Elazig; 2Department of Physiology, Gaziosmanpasa University Faculty of Medicine, Tokat; 3Department of Physiology, Inonu University Faculty of Medicine, Malatya, Turkey

Abstract. The aim of this study was to compare the in vivo effects on free radical metabolism of 2 nonsteroidal anti-inflammatory drugs (NSAIDs): tenoxicam, an oxicam preferentially cyclooxygenase-1 (COX- 1) inhibitor, and celecoxib, a sulfonamide selective COX-2 inhibitor. The serum levels of oxidative stress- related enzymes (ie, xanthine oxidase (XO), superoxide dismutase (SOD), glutathione peroxidase (GSH- Px)), of a lipid peroxidation marker (malondialdehyde (MDA)), and of nitric oxide (NO) in patients with knee osteoarthritis were studied at baseline and after a 4-wk course of treatment with celecoxib (n = 11) and tenoxicam (n = 12). Celecoxib-treated patients had significant decrease in nitrite levels (p = 0.043), whereas SOD, XO, GSH-Px enzyme activities, and MDA levels did not change significantly compared to baseline. Tenoxicam-treated patients had significant decrease in nitrite levels (p = 0.036) and XO activity (p = 0.01), but their SOD, GSH-Px enzyme activities, and MDA levels were unchanged from baseline. There was significant correlation between the patients’ (n = 23) Western Ontario and McMaster Universities (WOMAC) LK3.0 Osteoarthritis Index, WOMAC-pain scores, and MDA levels (r = 0.50, p = 0.014) and the patients’ WOMAC-stiffness scores and XO enzyme activity (r = 0.46, p = 0.027) at baseline. Significant improvement was found in pain-VAS, patients’ global assessment, and WOMAC pain, stiffness, and physical function scores in celecoxib and tenoxicam-treated groups. In summary, our study revealed that tenoxicam may have antioxidant effects, and that celecoxib and tenoxicam may reduce nitrite levels, indicating an alteration of NO pathways. (received 23 November 2004; accepted 31 January 2005)

Keywords: osteoarthritis, celecoxib, tenoxicam, oxidative stress, free radicals, nitric oxide

Introduction responses where they participate in pathological processes including the arachidonic acid cascade and Formation of reactive oxygen species (ROS) has been phagocytosis [7]. Under normal homeostatic suggested to play important roles in various inflam- conditions, the production of endogenous ROS matory diseases including rheumatoid arthritis (RA) (generated by NADPH oxidase, xanthine oxidase and ankylosing spondylitis (AS) [1-3]. ROS are (XO), or cytochrome P450, etc) are balanced by the deleterious agents involved in cartilage degradation actions of cellular antioxidant defense systems, and chondrocyte survival [4-6]. ROS such as super- including enzymes (superoxide dismutase (SOD), -• • oxide anion (O2 ), hydroxyl radical ( OH), catalase (CAT), glutathione peroxidase (GSH-Px)), hydrogen peroxide (H2O2), and hypochlorous acid and non-enzymatic species (glutathione, ascorbate, (HOCl) are highly reactive chemical species that are tocopherol, retinol) to avoid oxidative stress. important in inflammatory and antibacterial Non-steroid anti-inflammatory drugs are widely used in the management of osteoarthritis (OA); their Address correspondence to Salih Ozgocmen, M.D., Firat Universitesi, Tip Fakultesi Fiziksel Tip ve Rehabilitasyon primary effects are on prostaglandin (PG) synthesis Anabilim Dali, TR-23119, Elazig, Turkey; tel 90 424 233 3555; through inhibition of the cyclooxygenase (COX) fax 90 424 248 0509; e-mail [email protected]. enzymes. Two isoforms of COX have been well-

04 Ozgocmen 137-143 137 4/19/05, 7:21 PM 138 Annals of Clinical & Laboratory Science, vol. 35, no. 2, 2005

recognized (COX-1 and COX-2). COX-1 is a Exclusion criteria. Patients with the following constitutively expressed enzyme in many tissues, and conditions were excluded from the study: current COX-2 is an inducible enzyme predominantly drug usage (including supplementary vitamins) for expressed at the sites of inflammation. The activation uncontrolled concomitant disease or chronic of COX-1 promotes the release of eicosanoids that condition(s) that might interfere with the assessment are involved in physiologic processes including the of clinical findings of OA and serum oxidative stress production of thromboxane A2, prostacyclin, and markers; other prior disease or joint replacement at PGE2. Inhibition of COX-1 with non-selective the index joint; a history of GI ulcers and bleeding, NSAIDs can result in side effects such as platelet inflammatory arthritis, gout, pseudogout, or Paget’s dysfunction and gastrointestinal (GI) damage [8]. disease that might interfere with the assessments; Selective COX-2 inhibitors preferentially reduce history of excessive alcohol consumption, smoking, inflammation with fewer GI side effects, compared and regular aerobic exercise program; diagnosis of to traditional NSAIDs [9-11]. chronic pain syndrome (eg, fibromyalgia, chronic Some NSAIDs enhance ROS generation fatigue syndrome); intramuscular, intravenous, or whereas others attenuate ROS formation [12-20]. soft tissue corticosteroids within 1 mo prior to the Our purpose was to compare the in vivo effects on study; use of intra-articular corticosteroids in the ROS metabolism of celecoxib, a sulfonamide index knee joint within 2 mo prior to the study; selective COX-2 inhibitor, and tenoxicam, an intra-articular viscosupplementation in the index oxicam that preferentially inhibits COX-1. The knee joint during the past 6 mo, or intra-articular activities of oxidative stress-related enzymes (ie, XO, viscosupplementation in a non-index knee in the SOD, GSH-Px), malondialdehyde (MDA) level, past 3 mo; history of clinically significant intolerance and nitric oxide (NO) level in serum of OA patients to celecoxib and tenoxicam or known hyper- were studied at baseline and after a 4-wk course of sensitivity to sulfonamides. All of the patients treatment. voluntarily participated in the study and gave written informed consent. No supplementary Materials and Methods therapies, special diets, or aerobic exercise programs were allowed during the study period. Compliance Patients. Twenty-three patients with knee OA were was assessed by tablet counting; subjects with <70% administered either celecoxib or tenoxicam for 4 wk; compliance were excluded from the study. 11 patients were treated with celecoxib (200 mg po daily), and the rest with tenoxicam (20 mg po daily). Outcome measures. Pain was assessed using a visual Celecoxib-treated patients (2 men, 9 women) had a analog scale (VAS; 0-100 mm). Patient’s global mean age of 54.2 ± 9.6 yr (range 33-65) and a disease assessment (PGA) and Western Ontario and duration of 5.3 ± 5.6 yr (range 1-20). Tenoxicam- McMaster Universities (WOMAC) LK3.0 Osteo- treated patients (2 men, 10 women) had a mean age arthritis Index was performed at baseline and at the of 52.6 ± 6.0 yr (range 44-65) and a disease duration end of the treatment [22]. The WOMAC LK3.0 of 4.8 ± 3.5 yr (range 1-10). Osteoarthritis Index is a validated, multidimensional questionnaire of defined reliability, content, Inclusion criteria. The patients were 18-65 yr old, construct validity, and responsiveness. It consists of Functional Class I–III, with primary OA of the knee 24 questions (5 on pain, 17 on physical function, who met the ACR diagnostic criteria [21], defined and 2 on stiffness), each scored on a 5-point Likert by knee pain and recent radiographic evidence of scale (0 to 4, 0 representing none). The PGA was osteophytes. In addition, they had at least one of scored from 0 to 4 (0 representing very good). the following: age >50 yr, morning stiffness <30 min in duration, and/or crepitus, and a baseline visual Biochemical assays. Blood samples were obtained analog scale (VAS) pain intensity score of >30 mm at baseline and at the end of the clinical study period in the index joint. for analysis of oxidative stress-related analytes. The

04 Ozgocmen 137-143 138 4/19/05, 7:21 PM Celecoxib, tenoxicam, and oxidative stress 139

blood was centrifuged and the supernatant serum from xanthine through the increase in absorbance was stored frozen until use for analysis of GSH-Px, at 293 nm [26]. A calibration curve was constructed XO, MDA, NO and SOD. by using 10-50 mU/ml concentrations of standard MDA levels were determined by reaction with XO solutions (Sigma X-1875). One unit of activity thiobarbituric acid (TBA) [23]. In the TBA test was defined as 1 µmol uric acid formed per min at reaction, MDA or MDA-like substances and TBA 37°C, pH 7.5. Results were expressed as U/mL. react to produce a pink pigment with an absorption GSH-Px (EC 1.6.4.2) activity was measured by maximum at 532 nm. The reaction was performed the method of Paglia and Valentine [27]. The at pH 2-3 and 90°C for 15 min. The sample was enzymatic reaction was initiated by the addition of then mixed with 2 volumes of cold 10% (w/v) H2O2 to the reaction mixture containing reduced trichloroacetic acid to precipitate proteins. The glutathione (GSH), reduced nicotinamide adenine precipitate was pelleted by centrifugation and an dinucleotide phosphate (NADPH), and glutathione aliquot of the supernatant was reacted with an equal reductase. The change in the absorbance at 340 nm volume of 0.67% (w/v) TBA in a boiling water bath was monitored by a spectrophotometer. Activity was for 10 min. After cooling, the absorbance was read expressed as U/L. at 532 nm. The results were expressed as µmol/L. Since NO is very labile, its direct measurement Statistical analysis. The Statistics Package for Social in the biological samples is difficult. In aqueous Sciences (SPSS Inc, Chicago, IL) was used for the solution, NO reacts with molecular oxygen and analyses. Results were expressed as mean ± SD. - accumulates in the plasma as nitrite (NO2 ) and Differences between the two groups at baseline were - nitrate (NO3 ) ions. These stable oxidation end assessed by independent sample t test. Changes products are readily measured in biological fluids observed before and after celecoxib or tenoxicam and have been used in vitro and in vivo as indicators were assessed by the paired sample t test. Spearman of NO production. The plasma total nitrite rank correlation and Pearson correlation coefficients concentration was accepted as an index of NO were used to assess the relationships between production. For total nitrite detection, serum was parameters. A two-tailed p value of <0.05 was treated with copperized cadmium granules to reduce considered statistically significant. - - NO3 to NO2 . Nitrite concentrations were quantified by a colorimetric assay based on the Griess Results reaction [24]. Briefly, a chromophore with a strong absorbance at 545 nm is formed by reaction of nitrite All patients in both groups completed the study. The with a mixture of N-naphthylethylenediamine and baseline oxidative stress-related measurements in sulphanilamide. A standard curve is established with celecoxib and tenoxicam groups did not have a set of serial dilutions (10-8 to 10-3 mol/L) of sodium significant difference when compared to each other nitrite and results are expressed as µmol/L of serum. (Table 1). Celecoxib-treated patients had a The principle of the total SOD (EC 1.15.1.1) significant decrease only in nitrite levels (p = 0.043), activity method is based on inhibition of nitroblue whereas SOD, XO, GSH-Px enzyme activities, or -• tetrazolium (NBT) reduction by O2 generated by MDA levels did not significantly change compared xanthine/xanthine oxidase [25]. Activity was assessed to baseline values (Table 1). Tenoxicam-treated in the ethanol phase of plasma after 1.0 ml of patients had a significant decrease in nitrite levels ethanol/chloroform mixture (5:3, v:v) was added to (p = 0.036) and XO activity (p = 0.01) and the same volume of plasma and centrifuged. One unchanged SOD, GSH-Px enzyme activities, and unit of SOD was defined as the enzyme amount MDA levels (p = 0.07, 0.61, and 0.34 respectively). causing 50% inhibition of NBT reduction rate. There were significant correlations between Activity was expressed as U/mL. patients’ (n = 23) WOMAC-pain scores and MDA Serum XO (EC 1.2.3.2) activity was measured levels (r = 0.50, p = 0.014) and between patients’ spectrophotometrically by the formation of uric acid WOMAC-stiffness scores and XO enzyme activity

04 Ozgocmen 137-143 139 4/19/05, 7:21 PM 140 Annals of Clinical & Laboratory Science, vol. 35, no. 2, 2005

Table 1. Oxidative stress-related parameters in serum of patients with knee ostoarthritis.

Parameter Celecoxib-treated group (n = 11) Tenoxicam-treated group (n = 12) baseline final p baseline final p mean ± SD mean ± SD mean ± SD mean ± SD

SOD (U/mL) 15.7 ± 4.6 18.9 ± 4.3 0.065 16.7 ± 6.6 21.3 ± 5.3 0.071

MDA (µmol/L) 1.6 ± 0.7 1.6 ± 0.4 0.789 1.4 ± 0.7 1.6 ± 0.6 0.345

GSH-Px (U/L) 29.9 ± 8.4 27.9 ± 7.1 0.509 27.3 ± 9.0 28.9 ± 7.1 0.615

XO (U/mL) 0.5 ± 0.4 0.4 ± 0.3 0.364 0.6 ± 0.3 0.2 ± 0.1 0.01

Nitrite (µmol/L) 45.9 ± 17.0 33.9 ± 6.3 0.043 42.5 ± 11.0 32.7 ± 10.0 0.036

SOD, superoxide dismutase; MDA, malondialdehyde; GSH-Px, glutathione peroxidase; XO, xanthine oxidase; p values represent baseline vs final measurements in each treatment group (paired t test).

Table 2. Outcome measures in patients with knee osteoarthritis.

Parameters Celecoxib-treated group (n = 11) Tenoxicam-treated group (n = 12)

WOMAC LK3.0 OA Index Pain Baseline, mean ± SD 12.3 ± 1.9 9.2 ± 4.1 Final, mean ± SD 7.0 ± 5.3 5.2 ± 3.7 Change, mean (95% CI) -5.3 (-8.7 to -1.8)a -4.0 (-6.2 to -1.8)c

Physical function Baseline, mean ± SD 36.4 ±10.4 27.4 ± 10.3 Final, mean ± SD 19.8 ± 12.5 17.2 ± 11.3 Change, mean (95% CI) -16.6 (-27.4 to -5.8)a -10.2 (-16.2 to -4.1)b

Stiffness Baseline, mean ± SD 2.1 ± 1.9 2.2 ± 1.1 Final, mean ± SD 1.4 ± 0.8 1.6 ± 1.0 Change, mean (95% CI) -0.7 (-2.3 to 0.8) -0.6 (-1.7 to 0.6)

Pain-VAS (mm) Baseline, mean ± SD 58.3 ±11.0 57.2 ± 14.5 Final, mean ± SD 28.1 ±22.4 22.7 ± 11.3 Change, mean (95% CI) -29.0 (-45.6 to -12.4)b -34.5 (-44.7 to -24.3)d

Patient global assessment Baseline, mean ± SD 2.7 ± 0.8 2.7 ± 06 Final, mean ± SD 1.6 ± 1.0 1.3 ± 0.6 Change, mean (95% CI) 1.1 (0.4 to 1.8)a 1.4 (0.9 to 1.9)d

Baseline vs final measures by paired t test in celecoxib and tenoxicam treatment groups; a p = 0.007, b p = 0.003, c p = 0.002, d p = 0.0001

04 Ozgocmen 137-143 140 4/19/05, 7:21 PM Celecoxib, tenoxicam, and oxidative stress 141

(r = 0.46, p = 0.027) at baseline. Other oxidative formyl-methionyl-leucyl-phenylalanine (fMLP), stress-related measurements did not have significant calcium ionophore A23187, and serum-treated correlation with patients’ clinical measures (Table zymosan [28]. Furthermore, tenoxicam has been 2). shown to be a cofactor for the reduction of Significant improvement in the celecoxib- and peroxidases [29]. The in vitro effect of tenoxicam tenoxicam-treated groups was observed in the pain- enhancing erythrocyte CAT activity has been VAS, PGA, and WOMAC pain, stiffness, and demonstrated by Orhan and Sahin [30]. Similarly, physical function scores (Table 2). van Antwerpen and Neve [14] suggested that oxicams (ie, tenoxicam, piroxicam, meloxicam) were Discussion more reactive against ROS than nimesulide and ibuprofen. On the other hand, we must point out Our study revealed that serum nitrite levels decreased that these studies, including ours, represent short- with both celecoxib and tenoxicam treatments, but term effects of these drugs. neither drug caused significant changes in serum Xanthine oxidase is known to play a crucial role MDA levels or serum SOD and GSH-Px activities. in ischemia-reperfusion injury. During ischemia, A limited number of in vivo or in vitro studies have ATP is degraded to hypoxanthine and xanthine been reported regarding the effects of NSAIDs on dehydrogenase is converted to XO. During oxidant/antioxidant enzyme activities, nitric oxide reperfusion, XO catalyses the reaction of hypo- levels, and lipid peroxidation [8,12-20]. Xanthine xanthine or xanthine and molecular oxygen to oxidase enzyme activity significantly changed only superoxide radicals. These radicals rapidly react with in the tenoxicam-treated patients. To the best of our nitric oxide, peroxynitrite, and other reactive species knowledge, our study is the first to compare these 2 [31]. ROS attack polyunsaturated fatty acids drugs with particular reference to their effects on (PUFAs) in the membrane lipids to produce lipid oxidative stress metabolism using enzyme activities peroxidation. The assessments of serum MDA or 4- of XO, SOD, GSH-Px, and MDA and nitrite levels hydroxynonenal are the most commonly applied in OA patients. methods for the measurement of lipid peroxidation. In a recent study, Cimen et al [16] examined Related to partial oxygen pressure variations, the in vivo effects of celecoxib, ibuprofen, and mechanical stress, immunomodulatory or inflam- meloxicam on free radical metabolism of matory processes, chondrocytes mainly produce erythrocytes in patients with OA. The authors nitric oxide and superoxide [4]. The generation of concluded that these NSAIDs caused similar these radicals can lead to formation of peroxynitrite, impairment in enzymatic and non-enzymatic which is a more potent and long-lived oxidant and antioxidant defense systems in erythrocytes despite contributes to tissue injury mediated by their different mechanisms of action on COX. They inflammatory cells. There are divergent effects of found a significant decrease in erythrocyte SOD this interaction; first, superoxide limits some of the enzyme activity and antioxidant potentials in all effects of NO vasodilatation by converting it to groups treated with celecoxib, meloxicam, and ONOO-; secondly, scavenging superoxide anion ibuprofen [16]. In our study, both drugs increased while converting to ONOO- suggests that NO may the average serum SOD activities, but the difference provide a free radical scavenging effect [32,33]. was not statistically significant. Increased synovial fluid nitrite levels in OA Besides, we showed that tenoxicam reduced XO patients, reduced cartilage erosions with intra- activity and nitrite levels in vivo. The free radical articular injection of N-iminoethyl-L-lysine (L- scavenging activity of tenoxicam has been NIL), a selective inhibitor of inducible NO synthase demonstrated in experimental models of different (iNOS), incriminate NO as a potent mediator for cell-free systems by using different techniques [12]. cartilage destruction [34,35]. On the other hand, Tenoxicam has been shown to inhibit superoxide experimental data suggest that NO-mediated cell anion generation by neutrophils stimulated with N- death by apoptosis requires the generation of

04 Ozgocmen 137-143 141 4/19/05, 7:21 PM 142 Annals of Clinical & Laboratory Science, vol. 35, no. 2, 2005

additional ROS [6]. A combination of COX-2 2004;24:80-83. inhibitors and hyaluronic acid has been shown to 4. Henrotin YE, Bruckner P, Pujol JPL. The role of reactive oxygen species in homeostasis and degradation of cartilage. reduce NO production by highly degenerated Osteoarthritis Cartilage 2003;11:747-755. chondrocytes [36] and COX-2 has been suggested 5. Del Carlo M, Loeser RF. Increased oxidative stress with to have a proapoptotic function on synovial fibro- aging reduces chondrocyte survival: Correlation with blasts from human OA synovium [37]. Nevertheless, intracellular glutathione levels. Arthritis Rheum 2003; NSAIDs were shown to inhibit NO-induced 48:3419-3430. 6. Del Carlo M, Loeser RF. Nitric oxide-mediated chondro- apoptosis independent of their effects on COX cyte cell death requires the generation of additional activity [38]. These data are in agreement with reactive oxygen species. Arthritis Rheum 2002;46:394- previous experimental studies indicating that 403. inhibition of PGE2 by COX-2 inhibitors, or 7. Halliwell B, Gutteridge JMC. Free Radicals in Biology and Medicine, 3rd ed. Clarendon Press, Oxford, 1999. addition of exogenous PGE2 to the media, do not 8. Fosslien E. Adverse effects of nonsteroidal anti-inflam- affect spontaneous NO production significantly matory drugs on the gastrointestinal system. Ann Clin [33]. Our findings–similar nitrite lowering effect Lab Sci 1998;28:67-81. of tenoxicam, a traditional nonselective NSAID, and 9. Vane JR, Botting RM. Mechanism of action of nosteroidal celecoxib, a COX-2 selective NSAID–are not anti-inflammatory drugs. Am J Med 1998;104;2S-8S. 10. Lane NE. Pain management in osteoarthritis: The role of surprising. Our data warrant further studies to assess COX-2 inhibitors. J Rheumatol 1997;24(Suppl 49);20- the interaction between the protective effects of 24. nitric oxide in the microvasculature (ie, myocardial 11. Hinz B, Brune K. Pain in osteoarthritis: new drug and ischemia-reperfusion injury, adult respiratory distress mechanisms. Curr Opin Rheumatol 2004;16:628-633. 12. Maffei Facino RM, Carini M, Saibene L. Scavenging of syndrome) and interaction of nitric oxide pathways free radicals by tenoxicam: a participating mechanism in with side effects or adverse events related to NSAIDs. the antirheumatic/anti-inflammatory efficacy of the drug. In summary, our study showed that tenoxicam Arch Pharm 1996:329;457-463. may have antioxidant effects, and that celecoxib and 13. Parij N, Nagy AM, Fondu P, Neve J. Effects of non- tenoxicam may reduce nitrite levels, indicating an steroidal anti-inflammatory drugs on the luminol and lucigenin amplified chemiluminescence of human alteration in NO pathways. The relationship neutrophils. Eur J Pharmacol 1998;352:299-305. between patients’ pain and stiffness scores and 14. Van Antwerpen P, Neve J. In vitro comparative assessment oxidative stress-related measurements warrants of the scavenging activity against three reactive oxygen further research to assess the potential role of free species of non-steroidal anti-inflammatory drugs from oxicam and sulfonilide families. Eur J Pharm 2004;496; radicals and antioxidants in the pathogenesis and 55-61. clinical features of OA. Knowledge of the underlying 15. Facino RM, Carini M, Aldini G. Antioxidant activity of mechanisms might offer new therapeutic approaches nimesulide and its main metabolites. Drugs 1993;46 for joint diseases and prevention of cartilage damage. (Suppl 1):15-21. 16. Cimen MYB, Bolgen-Cimen O, Eskandari G, Sahin G, References Erdogan C, Atik U. In vivo effects of meloxicam, celecoxib and ibuprofen on free radical metabolism in human erythrocytes. Drug Chem Toxicol 2003;26:169-176. 1. Lunec J, Halloran SP, White AG, Dormandy TL. Free- 17. Zheng SX, Mouithys-Mickalad A, Deby-Dupont GP, radical oxidation (peroxidation) products in serum and Deby CM, Maroulis AP, Labasse AH, Lamy ML, Crielaard synovial fluid in rheumatoid arthritis. J Rheumatol JM, Reginster JY, Henrotin YE. In vitro study of the anti- 1981;8:233-245. oxidant properties of nimesulide and 4-OH nimesulide: 2. Akyol O, Isci N, Temel I, Ozgocmen S, Uz E, Murat M, effects on HRP- and luminol-dependent chemilumin- Buyukberber S. The relationships between plasma and escence produced by human chondrocytes. Osteoarthritis erythrocyte antioxidant enzymes and lipid peroxidation Cartilage 2000;8:419-425. in patients with rheumatoid arthritis. Joint Bone Spine 18. Mouithys-Mickalad AM, Zheng SX, Deby-Dupont GP, 2001;68:311-317. Deby CM, Lamy MM, Reginster JY, Henrotin YE. In 3. Ozgocmen S, Sogut S, Ardicoglu O, Fadillioglu E, vitro study of the antioxidant properties of non steroidal Pekkutucu I, Akyol O. Serum nitric oxide, catalase, anti-inflammatory drugs by chemiluminescence and superoxide dismutase, and malondialdehyde status in electron spin resonance (ESR). Free Radic Res 2000;33: patients with ankylosing spondylitis. Rheumatol Int 607-621.

04 Ozgocmen 137-143 142 4/19/05, 7:21 PM Celecoxib, tenoxicam, and oxidative stress 143

19. Basivireddy J, Jacob M, Pulimood AB, Balasubramanian 29. Ichihara S, Tomisawa H, Fukuzawa H, Tateishi M, Joly KA. Indomethacin-induced renal damage: role of oxygen R, Heintz R. Oxidation of tenoxicam by leukocyte free radicals. Biochem Pharmacol 2004;67:587-599. peroxidases and H2O2 produces novel products. Drug 20. Nakamura Y, Kozuka M, Naniwa K, Takabayashi S, Metab Dispos 1989;17:463-468. Torikai K, Hayashi R, Sato T, Ohigashi H, Osawa T. 30. Orhan H, Sahin G. In vitro effects of NSAIDS and Arachidonic acid cascade inhibitors modulate phorbol paracetamol on oxidative stress-related parameters of ester-induced oxidative stress in female ICR mouse skin: human erythrocytes. Exp Toxicol Pathol 2001;53:133- differential roles of 5-lipoxygenase and cyclooxygenase-2 140. in leukocyte infiltration and activation. Free Radic Biol 31. Akyol O, Herken H, Uz E, Fadillioglu E, Unal S, Sogut Med 2003;35:997-1007. S, Ozyurt H, Savas HA. The indices of endogenous 21. Altman RD, Asch E, Bloch D, Bole G, et al. Development oxidative and antioxidative processes in plasma from of criteria for the classification and reporting of schizophrenic patients. The possible role of oxidant/ osteoarthritis: Classification of osteoarthritis of the knee. antioxidant imbalance. Prog Neuropsychopharmacol Biol Arthritis Rheum 1986;29:1039-1049. Psychiatr 2002,26:995-1005. 22. Bellamy N, Buchanan WW, Goldsmith CH, Campbell 32. Clancy RM, Amin AR, Abramson SB. The role of nitric J, Stitt LW. Validation study of WOMAC: a health status oxide in inflammation and immunity. Arthritis Rheum instrument for measuring clinically important patient 1998;41:1141-1151. relevant outcomes to antirheumatic drug therapy in 33. Lotz M. The role of nitric oxide in articular cartilage patients with osteoarthritis of the hip or knee. J Rheumatol damage. Rheum Dis Clin North Am 1999;25:269-282. 1988;15:1833-1840. 34. Karan A, Karan MA, Vural P, Erten N, Tascioglu C, Aksoy 23. Esterbauer H, Cheeseman KH. Determination of alde- C, Canbaz M, Oncel A. Synovial fluid nitric oxide levels hydic lipid peroxidation products: Malonaldehyde and in patients with knee osteoarthritis. Clin Rheumatol 4-hydroxynonenal. In: Oxygen Radicals in Biological 2003;22:397-399. Systems (Packer L, Glazer AN, Eds), Method Enzymol 35. Pelletier JP, Lascou-Coman V, Jovanovic D, Fernandes 1990;186:407-421. JC, Manning P, Connor JR, et al. Selective inhibition of 24. Cortas NK, Wakid NW. Determination of inorganic inducible nitric oxide synthase in experimental osteo- nitrate in serum and urine by a kinetic cadmium– arthritis is associated with reduction in tissue levels of reduction method. Clin Chem 1990;36:1440-1443. catabolic factors. J Rheumatol 1999;26:2002-2014. 25. Sun Y, Oberley LW, Li Y. A simple method for clinical 36. Takabashi T, Uemura Y, Tagushi H, Ogawa Y, Yashida S, assay of superoxide dismutase. Clin Chem 1988;34:497- Toda M, Kobayashi T, Seguchi H, Tani T. Cross talk 500. between COX-2 inhibitor and hyaluronic acid in osteo- 26. Prajda N, Weber G. Malign transformation–linked arthritis chondrocytes. Int J Mol Med 2004;14:139-144. imbalance: decreased XO activity in hepatomas. FEBS 37. Jovanovic DV, Mineau F, Notoya K, Reboul P, Martel- Lett 1975;59:245-249. Pelletier J, Pelletier JP. Nitric oxide induced cell death in 27. Paglia DE, Valentine WN. Studies on the quantitative human osteoarthritic synoviocytes is mediated by tyrosine and qualitative characterization of erythrocyte glutathione kinase activation and hydrogen peroxide and/or super- peroxidase. J Lab Clin Med 1967;70:158-170. oxide formation. J Rheumatol 2002;29:2165-2175. 28. Colli S, Colombo S, Tremoli E, Stragliotto E, Nicosia S. 38. Yoon YB, Kim SJ, Hwang SG, et al. Non-steroidal anti- Effects of tenoxicam on superoxide anion formation, beta- inflammatory drugs inhibit nitric oxide induced apoptosis glucuronidase release and fMLP binding in human and differentiation of articular chondrocytes independent neutrophils: comparison with other NSAIDs. Pharmacol of cyclooxygenase activity. J Biol Chem 2003;278:15319- Res 1991;23:367-379. 15325.

04 Ozgocmen 137-143 143 4/19/05, 7:21 PM