ANTI-INFLAMMATORY DRUGS FOR both acute and

Transcription

1 651 Influence of Nonsteroidal Anti-inflammatory Drug Treatment Duration and Time of Onset on Recovery From Exercise- Induced Muscle Damage in Rats Benoît M. Lapointe, PhD, PT, Pierre Frémont, MD, PhD, Claude H. Côté, PhD ABSTRACT. Lapointe BM, Frémont P, Côté CH. Influence of nonsteroidal anti-inflammatory drug treatment duration and time of onset on recovery from exercise-induced muscle damage in rats. Arch Phys Med Rehabil 2003;84: Objective: To determine if duration and time of onset of treatment with diclofenac sodium influence force recovery after exercise-induced muscle damage in rats. Design: Randomized placebo-controlled trial. Setting: Animal laboratory. Animals: A total of 217 female adult Wistar rats. Intervention: Rats were submitted to a protocol consisting of 450 eccentric contractions of the ankle dorsiflexors. Treatment by gavage with diclofenac sodium (1 mg/kg, twice daily) was started at different times pre- and postprotocol or for various treatment durations. Main Outcome Measures: In vitro contractile properties. Results: When treatment was initiated shortly postprotocol, force recovery was roughly proportional to treatment duration during the first 3 days but not at 7 and 28 days postprotocol. A 7-day treatment was no more effective than 1- or 2-day treatments when force was measured at 7 and 28 days; however, such prolonged treatment had no deleterious effect on muscle force at either time. A single-dose prophylactic treatment was as effective as a 2-day treatment initiated soon after the protocol when force was assessed 2 days postprotocol; on the other end, a treatment delayed for 3 days had no effect when force was measured at 7 days. Conclusions: Treatment with diclofenac sodium extending past the acute inflammatory phase was no more effective than short and timely treatment in this model of skeletal muscle damage. Key Words: Diclofenac; Exercise; Injuries; Muscles; Rehabilitation by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation ANTI-INFLAMMATORY DRUGS FOR both acute and chronic musculoskeletal problems are among the most widely prescribed medications in the field of physical medicine From the CHUL Research Center (Lapointe, Frémont, Côté) and Rehabilitation Department (Frémont, Côté), Laval University, Quebec City, QC, Canada. Supported by the Natural Sciences and Engineering Research Council of Canada (grant no ) and by fellowships from Réseau Provincial de Recherche en Adaptation et Réadaptation and Fonds de la Recherche en Santé du Québec. Presented in part at the REPAR Congress, June 4 6, 1998, Quebec City, QC. No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the author(s) or upon any organization with which the author(s) is/are associated. Correspondence to Claude H. Côté, PhD, CHUL Research Center, RC-9600, 2705 Blvd Laurier, Quebec City, QC G1V 4G2, Canada, Claude.H.Cote@ crchul.ulaval.ca. Reprints are not available /03/ $30.00/0 doi:s (03) and rehabilitation. The clinical evidence to support the use of nonsteroidal anti-inflammatory drugs (NSAIDs) is, however, limited, especially for acute conditions. 1-3 The difficulty in distinguishing the analgesic from the anti-inflammatory effects of NSAIDs on functional outcomes in humans greatly limits clinical studies that investigate their effectiveness in acute musculoskeletal conditions. Such difficulty can be overcome by using an animal model to further understand the influence of NSAIDs on soft tissue healing because more invasive techniques can be performed with animals. Results obtained with animals may guide future clinical research and possibly modify the use of NSAIDs until more clinical evidence of their efficacy is available. One experimental model for the study of muscle injury consists of submitting rat ankle dorsiflexors to eccentric (or lengthening) contractions. 4 Although the exact etiology is not entirely known, it is well accepted that unaccustomed eccentric contractions can cause exercise-induced muscle damage. 5,6 This syndrome is characterized not only by a decrease in force production and range of motion, but also by delayed-onset muscle soreness, stiffness, edema, leukocyte infiltration, and leakage of muscle proteins (eg, creatine kinase, troponin I, carbonic anhydrase III) into the plasma Although it is known that an inflammatory response is triggered by exerciseinduced muscle damage, its precise role in the degenerative and repair phases has yet to be clarified. The catabolic activity of leukocytes is well known, but several reports have shown that the inflammatory response is also implicated in different steps that lead to skeletal muscle repair Based on the observation that force production after exercise-induced muscle damage follows a biphasic curve, with a first decline immediately after the protocol and a second one 2 to 3 days postprotocol, it has been proposed that the inflammatory response was responsible for this second drop in force because the inflammatory response is allegedly fully developed at that time. 6,23 Accordingly, the use of flurbiprofen was shown to be beneficial in terms of force production in a rabbit model for the first few days after exercise-induced muscle damage. 24 To our knowledge, no study has comprehensively examined how duration and time of initiation of NSAID treatment influence the extent of muscle damage, using force recovery after exercise-induced muscle damage as a marker. In this study, our first goal was to test the influence of diclofenac sodium treatment when it is initiated early postprotocol. Our underlying hypothesis was that treatment for 1 to 3 days could be beneficial because it would repress the catabolic phase of the inflammatory response only, but that, on the contrary, a 7-day treatment could be detrimental because it would interfere with the inflammatory response required for efficient healing. Our second goal was to evaluate how the time of initiation of treatment influences the outcome of exercise-induced muscle damage. To do so, a prophylactic and a delayed treatment, starting after the acute inflammatory phase, were studied. It was hypothesized that the prophylactic treatment would be more effective than

2 652 EFFECT OF COX REPRESSION IN EXERCISE-INDUCED MUSCLE DAMAGE, Lapointe Fig 1. Experimental design used to study the influence of diclofenac sodium treatment duration and time of onset on recovery from exercise-induced muscle damage in rats. Horizontal bars show the time of onset and the duration of diclofenac treatment for each experiment after the protocol of eccentric contractions. Appropriate placebo groups were run in parallel. Asterisks represent the time points of in vitro measurement of tetanic tension. the delayed treatment because it can interfere with the catabolic component of the inflammatory response. METHODS Experimental Design Experimental animals were submitted to the eccentric protocol and received diclofenac sodium (1mg/kg of body weight) administered by gastric gavage twice a day. Placebo animals underwent the same eccentric protocol but received water only, also by gavage. In all groups, the left hindlimb was shamoperated, with all manipulations performed except that no stimulation was given. Animals were randomly allocated to 1 of 3 series of experiments (fig 1). The series in which the effect of treatment duration with diclofenac sodium was studied consisted of 9 experimental groups. The duration of treatment ranged from 1 to 7 days, and force measurements were obtained at 1, 2, 3, and 7 days. In addition, to determine the long-term effect of a treatment extending past the acute inflammatory phase, a group was treated for 7 days, but force was measured only at 28 days. Data at each time point were analyzed by using a 1-way analysis of variance with the post hoc Fisher exact test. The series that examined the effect of treatment initiation time included 2 experimental groups. The first received only 1 dose of diclofenac sodium, 45 minutes before the protocol of contractions, and force was measured 2 days after the protocol. The second was treated for 4 days, but the treatment was initiated 3 days after the protocol. In this last series of experiments, Student t tests were performed to detect differences. In all cases, pairing was performed for comparison with the sham groups, and the level of statistical significance was set at P less than.05. All data were analyzed with the statistical software SPSS, version 4.0. a Eccentric Contractions Protocol The treatment and care of the animals were approved by, and followed the guidelines of the Laval University Hospital Research Center Animal Care and Use Committee. Rats were given a period of acclimatization and were maintained throughout the study on a 12-hour light-dark cycle in a controlled environment, with access to regular rat chow and water ad libitum. We used the same protocol as previously described to induce muscle damage of the ankle dorsiflexors. 13 Briefly, rats were first anesthetized with a cocktail of xylazine-ketamine (Rompun and Ketalar; intraperitoneal, 87.5 and 12.5mg/kg of body weight); additional ketamine injections were given as necessary. Before performing routine surgical preparation of the hindlimbs, Ringer s lactate solution was administered subcutaneously. The peroneal nerve of the right hindlimb was then exposed, and the animal was transferred to an apparatus designed to forcibly stretch the ankle dorsiflexor muscles in the last 40 of plantarflexion, with the foot tied to a circular motion motor. The protocol consisted of 450 contractions elicited every 2 seconds in 3 bouts of 5 minutes, with a 5-minute rest period between each bout. To ensure maximal stimulation, a frequency of 120Hz and a supramaximal voltage (3 6V) were used. When the protocol was completed, the muscle and fascia over the peroneal nerve were closed with synthetic absorbable 4-0 sutures (coated polyglactin [Vicryl]), and Michel s clips (11 mm) were used for the skin. Animals had a normal gait and no evident pain after recovering from the anesthesia. Measurement of Isometric Contractile Properties Measurements of isometric contractile properties were obtained in vitro as described earlier. 25 Briefly, rats were anesthetized with sodium pentobarbital (Nembutal; 50mg/kg of body weight, intraperitoneal) and then received buprenorphine (Buprenex; 0.1 mg/kg of body weight, intraperitoneal) to ensure proper anesthesia. Extensor digitorum longus muscles were carefully dissected out and incubated at their optimal length at 25 C for 20 minutes in a buffered solution (Krebs- Ringer) that was supplemented with glucose and bubbled with carbogen (95% O 2,5%CO 2 ). A force-frequency relationship was determined by using 1 tetanic contraction of 400ms in duration every minute until maximum tetanic tension was obtained. On determination of tetanic tension, muscles were weighed to permit calculation of maximum specific tetanic tension, which is a normalization for the cross-sectional area. The value used for muscle density was 1.062g/cm 3, and the ratio of fiber length to muscle length was RESULTS Effect of Treatment Duration on Force Recovery In a preliminary experiment, we showed that 2-day or 7-day treatment with diclofenac sodium per se did not influence absolute and specific tetanic tension in control nonexercised rats (data not shown). We then examined the influence of various diclofenac treatment durations on absolute and specific maximum tetanic tension. One- and 2-day treatments led to similar significant protective influence on absolute and specific tetanic tensions compared with the placebo groups when stud-

3 EFFECT OF COX REPRESSION IN EXERCISE-INDUCED MUSCLE DAMAGE, Lapointe 653 Fig 2. Influence of diclofenac sodium treatment duration on maximum absolute tetanic tension measured in vitro. The NSAID group received 1mg/kg of diclofenac twice a day by gavage. Tension is expressed as a percentage of the value produced by the shamoperated limb. Values are mean standard error (SE) (n 6 8 muscles per group). Letters refer to statistically significant differences between groups (P<.05). Abbreviations: N1,2,3,7, treatment with NSAID for 1, 2, 3, and 7 days; P, placebo; S, sham. ied at day 2 (figs 2, 3). At day 3, only the rats treated for 3 days showed significantly higher absolute force production than the placebo group, but all groups were still significantly weaker than their sham counterparts. On the contrary, values for specific tetanic tension had a tendency to improve rather proportionally with treatment duration, as rats treated for 3 days showed significantly higher values than did those receiving the NSAIDs for just 1 day. Because calculation of specific tetanic tension involves muscle mass, which is influenced by edema, it Fig 3. Influence of diclofenac sodium treatment duration on maximum specific tetanic tension. Maximum specific tetanic tension was calculated as previously described. The NSAID group received 1 mg/kg of diclofenac twice a day. Tension is expressed as a percentage of the value produced by the sham-operated limb. Values are mean SE (n 6 8 muscles per group). Letters refer to statistically significant differences between groups (P<.05). See fig 2 for abbreviations. Fig 4. Long-term influence of a 7-day diclofenac sodium treatment on both maximum absolute and specific force production. Tetanic tension was measured in vitro at day 28 postprotocol. Rats received diclofenac sodium (1mg/kg, twice a day) for 7 days. Values are mean SE (n 5 7 muscles per group). No significant difference was observed between groups. Abbreviation: Po, tetanic tension. thus suggests that the time courses for the hemodynamic and catabolic influences of the inflammatory response may differ. At day 7, the values for absolute tetanic tension for all treatment durations were about midway between values for sham and placebo groups, which still differed significantly. The end result was that no significant difference was detected among the 3 groups. Specific tetanic tension was also globally the same in all NSAID groups, independent of treatment duration; surprisingly, only the 1-day treatment group had a value that did not differ from the sham group. We next tested whether diclofenac sodium, given for a period extending over the entire inflammatory period and during the early phase of the regeneration process, could impair long-term functional recovery. Rats were treated for 7 days with diclofenac, and tetanic tension was assessed 28 days after the protocol, at a time when force was back to normal values in the placebo group. Data indicate that, from a functional standpoint, the ability to generate isometric tension was not impaired by such a relatively long treatment (fig 4). Effects of Time of Initiation of Treatment on Force Recovery Because our results strongly suggested that inflammatory response associated events of exercise-induced muscle damage are concomitant with functional loss, we decided to verify whether a single dose of diclofenac sodium given before the eccentric contractions protocol could lead to a significant protective effect on contractile properties measured on day 2. In keeping with our working hypothesis, this prophylactic treatment had a strong protective influence on absolute tetanic tension, which did not differ significantly from the sham group value. However, the placebo group had a 32% deficit (fig 5). A significant protective effect could also be seen when results were expressed as specific tetanic tension. We then tested the hypothesis that initiating NSAID treatment, once the inflammatory response intensity was declining, would not have a beneficial effect. We treated the rats for 4 days beginning 3 days after the protocol. This delayed treatment showed no significant protective or detrimental effect, and values obtained at day 7 for both absolute and specific tetanic tension were within 5% of the placebo group s values (fig 6). However, both

4 654 EFFECT OF COX REPRESSION IN EXERCISE-INDUCED MUSCLE DAMAGE, Lapointe Fig 5. Influence of a single prophylactic diclofenac sodium dose on both maximum absolute and specific maximum force production. Tetanic tension was measured in vitro 2 days after the eccentric protocol. Rats were given a single dose of diclofenac (1 mg/kg) 45 minutes before the eccentric protocol. Values are mean SE (n 4 7 muscles per group). *Significantly different from the sham group (P<.05). Significantly different from the NSAID group (P<.05). groups produced significantly lower tensions than did the sham group. DISCUSSION In agreement with our first hypothesis, we showed that early and short treatments with diclofenac sodium during the acute inflammatory phase had a beneficial effect on the muscle s ability to generate maximal force in this animal model of skeletal muscle damage. As expected, treatment with diclofenac did not provide complete protection because tetanic tension values of the treated groups were still significantly lower than the value for the sham group. This may be explained by the mechanical nature of the primary muscle damage, on which diclofenac has no influence. Furthermore, partial establishment of the inflammatory response obviously occurred as treatment was initiated 6 hours after the protocol, a delay that was necessary to allow recovery of the gag reflex before gavage of the animals. Comparison of the time course for absolute and specific tetanic tension recovery produced an interesting issue. Although recovery of absolute tetanic tension at day 7 was unrelated to treatment duration during the acute inflammatory phase, there was a strong tendency for better protection regarding specific tetanic tension when the treatment covered this entire phase. It therefore suggests that a short and early treatment may be appropriate to control the catabolic component of the inflammatory response but insufficient to give satisfactory results in terms of control of tissue edema, because specific tetanic tension decreased when compared with absolute tetanic tension. Although we have no evidence that edema per se can decrease force production in vitro, in humans it can lead to pain that is associated with muscle swelling and increased compartment pressure. 27,28 This can subsequently have a negative influence on force production. 29,30 Hence, controlling edema in clinical settings may be very desirable. Although this study suggests that it may be beneficial to use diclofenac sodium or other NSAIDs after muscle damage, the efficiency is related to the rapidity of onset of treatment, as stated in our second hypothesis. The observation that a singledose prophylactic treatment was as effective as a 1- or 2-day treatment that was initiated after exercise-induced muscle damage is in line with findings from Hasson et al. 31 When such an observation is coupled with the fact that a delayed treatment initiated after the acute inflammatory phase did not have an effect on tetanic tension, it supports the conclusion that rapidity of treatment onset is an important and significant variable in exercise-induced muscle damage outcome. Theoretically, this could be expected because the sooner one begins to interfere with prostaglandin E 2 (PGE 2 ) production, the better the inflammatory response repression will be. In line with this view, it is hard to understand the rationale of a treatment aimed at inhibiting production of the proinflammatory PGE 2 when the acute inflammatory phase and cyclooxygenase-2 (COX-2) activity are phasing out or are over (ie, in the regeneration phase). In this study, we found that extending treatment beyond the acute inflammatory phase into the repair phase was not more effective in helping muscle recovery, as assessed through measurement of tetanic tension. However, contrary to what we stated in our hypothesis, these longer treatments did not appear to harm muscle repair because in vitro isometric force returned to a normal value 28 days postprotocol after a 7-day treatment, just as in the placebo group. On the contrary, Mishra et al, 24 in a rabbit model of exercise-induced muscle damage, observed that a 7-day treatment with flurbiprofen was deleterious for recovery of muscle force at 28 days postprotocol. This discrepancy may be attributed to the use of a different NSAID, to the protocol of lengthening contractions, or to the animals used in the experiment. Furthermore, note that the assessment of functional recovery based on tetanic tension indicates the capacity of a muscle to produce its maximum isometric contractile performance, but it does not necessarily give an accurate estimation of the muscle s structural integrity or mechanical resistance. In line with this, we 13 recently showed that the use of diclofenac sodium for 7 days interferes with processes that are implicated in the adaptive response to eccentric contractions. Hence, at day 28 postprotocol, complete recovery of tetanic tension in both the NSAID and placebo groups implies that the intrinsic capacity of the muscle to contract had completely recovered. Nonetheless, it does not mean that the mechanical resistance to daily occurring eccentric contractions had recov- Fig 6. Influence of a delayed treatment with diclofenac sodium on both maximum absolute and specific force production. Tetanic tension was measured in vitro 7 days after the eccentric protocol. Rats received diclofenac (1mg/kg, twice a day) starting 3 days after the eccentric protocol until they were killed. Values are mean SE (n 4 7 muscles per group). *Significantly different from the sham group (P<.05).

5 EFFECT OF COX REPRESSION IN EXERCISE-INDUCED MUSCLE DAMAGE, Lapointe 655 ered to the same extent in both groups of animals. This may be particularly relevant in work and sports settings, where individuals are rapidly exposed to these same damaging contractile patterns on a daily basis. Our hypothesis that treatment with NSAIDs beyond the acute inflammatory phase would be deleterious is supported by recent data. Using a model of pleurisy, Gilroy et al 32 observed that the usual and rapidly occurring COX-2 expression leading to PGE 2 production lasted about 24 hours. The novelty in their study was an unexpected second peak of COX-2 expression at the end of the acute inflammatory phase. This second peak was concomitant with the production of prostaglandin J 2 (PGJ 2 ) instead of PGE 2. This is noteworthy, because PGJ 2 has anti-inflammatory effects partly through interference with important transcription factors for COX-2 expression. Hence, repressing COX-2 after the acute inflammatory phase may not be appropriate. 33,34 CONCLUSION Given the scarcity of clinical data and the well-known side effects of NSAIDs in clinical practice, this study may offer some rationale for the treatment of skeletal muscle damage. Although the physiopathology of exercise-induced muscle damage likely differs from what is seen with contusion or strain injuries, this study provides evidence to support the notion that use of NSAIDs should be restricted to the acute inflammatory phase. Furthermore, it should be noted that these concepts are applicable when the inflammatory response is not unduly reactivated after the initial trauma. Acknowledgment: We thank Sébastien Richard for technical assistance. References 1. Almekinders LC. Anti-inflammatory treatment of muscular injuries in sport. An update of recent studies. Sports Med 1999;28: Lecomte JM, Lacroix VJ, Montgomery DL. A randomized controlled trial of the effect of naproxen on delayed onset muscle soreness and muscle strength. Clin J Sport Med 1998;8: Weiler JM. Medical modifiers of sports injury. The use of nonsteroidal antiinflammatory drugs (NSAIDs) in sports soft-tissue injury. Clin Sports Med 1992;11: McCully KK, Faulkner JA. Injury to skeletal muscle fibers of mice following lengthening contractions. J Appl Physiol 1985;59: Armstrong RB, Warren GL, Warren JA. Mechanisms of exerciseinduced muscle fibre injury. Sports Med 1991;12: Faulkner JA, Brooks SV, Opiteck JA. Injury to skeletal muscle fibers during contractions: conditions of occurrence and prevention. Phys Ther 1993;73: MacIntyre DL, Reid WD, McKenzie DC. Delayed muscle soreness. The inflammatory response to muscle injury and its clinical implications. Sports Med 1995;20: Smith LL. Acute inflammation: the underlying mechanism in delayed onset muscle soreness? Med Sci Sports Exerc 1991;23: Sorichter S, Puschendorf B, Mair J. Skeletal muscle injury induced by eccentric muscle action: muscle proteins as markers of muscle fiber injury. Exerc Immunol Rev 1999;5: Stauber WT, Clarkson PM, Fritz VK, Evans WJ. Extracellular matrix disruption and pain after eccentric muscle action. J Appl Physiol 1990;69: Child R, Brown S, Day S, Donnelly A, Roper H, Saxton J. Changes in indices of antioxidant status, lipid peroxidation and inflammation in human skeletal muscle after eccentric muscle actions. Clin Sci (Colch) 1999;96: Hellsten Y, Frandsen U, Orthenblad N, Sjodin B, Richter EA. Xanthine oxidase in human skeletal muscle following eccentric exercise: a role in inflammation. J Physiol (Lond) 1997;498: Lapointe BM, Fremont P, Cote CH. Adaptation to lengthening contractions is independent of voluntary muscle recruitment but relies on inflammation. Am J Physiol Regul Integr Comp Physiol 2002;282:R MacIntyre DL, Sorichter S, Mair J, Berg A, McKenzie DC. Markers of inflammation and myofibrillar proteins following eccentric exercise in humans. Eur J Appl Physiol 2001;84: Robertson TA, Maley MA, Grounds MD, Papadimitriou JM. The role of macrophages in skeletal muscle regeneration with particular reference to chemotaxis. Exp Cell Res 1993;207: Round JM, Jones DA, Cambridge G. Cellular infiltrates in human skeletal muscle: exercise induced damage as a model for inflammatory muscle disease? J Neurol Sci 1987;82: Cantini M, Massimino ML, Bruson A, Catani C, Dalla Libera L, Carraro U. Macrophages regulate proliferation and differentiation of satellite cells. Biochem Biophys Res Commun 1994;202: Massimino ML, Rapizzi E, Cantini M, et al. ED2 macrophages increase selectively myoblast proliferation in muscle cultures. Biochem Biophys Res Commun 1997;235: McLennan IS. Resident macrophages (ED2- and ED3-positive) do not phagocytose degenerating rat skeletal muscle fibres. Cell Tissue Res 1993;272: Pimorady-Esfahani A, Grounds MD, McMenamin PG. Macrophages and dendritic cells in normal and regenerating murine skeletal muscle. Muscle Nerve 1997;20: St Pierre BA, Tidball JG. Differential response of macrophage subpopulations to soleus muscle reloading after rat hindlimb suspension. J Appl Physiol 1994;77: Tidball JG. Inflammatory cell response to acute muscle injury. Med Sci Sports Exerc 1995;27: MacIntyre DL, Reid WD, Lyster DM, Szasz IJ, McKenzie DC. Presence of WBC, decreased strength, and delayed soreness in muscle after eccentric exercise. J Appl Physiol 1996;80: Mishra DK, Friden J, Schmitz MC, Lieber RL. Anti-inflammatory medication after muscle injury. A treatment resulting in short-term improvement but subsequent loss of muscle function. J Bone Joint Surg Am 1995;77: Lapointe BM, Cote CH. Anesthetics can alter subsequent in vitro assessment of contractility in slow and fast skeletal muscles of rat. Am J Physiol 1999;277:R Segal SS, Faulkner JA. Temperature-dependent physiological stability of rat skeletal muscle in vitro. Am J Physiol 1985;248:C Chleboun GS, Howell JN, Conatser RR, Giesey JJ. Relationship between muscle swelling and stiffness after eccentric exercise. Med Sci Sports Exerc 1998;30: Friden J, Sfakianos PN, Hargens AR, Akeson WH. Residual muscular swelling after repetitive eccentric contractions. J Orthop Res 1988;6: Crenshaw AG, Thornell LE, Friden J. Intramuscular pressure, torque and swelling for the exercise-induced sore vastus lateralis muscle. Acta Physiol Scand 1994;152: Howell JN, Chleboun G, Conatser R. Muscle stiffness, strength loss, swelling and soreness following exercise-induced injury in humans. J Physiol (Lond) 1993;464: Hasson SM, Daniels JC, Divine JG, et al. Effect of ibuprofen use on muscle soreness, damage, and performance: a preliminary investigation. Med Sci Sports Exerc 1993;25: Gilroy DW, Colville-Nash PR, Willis D, Chivers J, Paul-Clark MJ, Willoughby DA. Inducible cyclooxygenase may have antiinflammatory properties. Nat Med 1999;5: Colville-Nash PR, Gilroy DW. COX-2 and the cyclopentenone prostaglandins a new chapter in the book of inflammation? Prostaglandins Other Lipid Mediat 2000;62: Colville-Nash PR, Gilroy DW. Potential adverse effects of cyclooxygenase-2 inhibition: evidence from animal models of inflammation. BioDrugs 2001;15:1-9. Supplier a. SPSS Inc, 233 S Wacker Dr, 11th Fl, Chicago, IL