PAIN Influence on platelet aggregation of i.v. parecoxib and acetaminophen in healthy volunteers

Transcription

1 British Journal of Anaesthesia 97 (2): (2006) doi: /bja/ael108 Advance Access publication May 12, 2006 PAIN Influence on platelet aggregation of i.v. parecoxib and acetaminophen in healthy volunteers E. Munsterhjelm 1 *, T. T. Niemi 1, O. Ylikorkala 2, P. J. Neuvonen 3 and P. H. Rosenberg 1 1 Department of Anaesthesiology and Intensive Care Medicine, 2 Department of Obstetrics and Gynaecology and 3 Department of Clinical Pharmacology, Helsinki University Hospital, Helsinki, Finland *Corresponding author: PO Box 340 (P-floor), FIN HUS, Finland. edward.munsterhjelm@hus.fi Background. Acetaminophen (paracetamol) alone or in combination with other analgesics is widely used for postoperative analgesia. While acetaminophen and non-steroidal antiinflammatory drugs inhibit platelet function, the cyclooxygenase-2 (COX-2) selectively inhibiting coxibs show no interference with platelet function. The authors studied the effect of a combination of i.v. parecoxib and acetaminophen on platelet function in healthy volunteers. Methods. Eighteen healthy, male volunteers (22 33 yr) received i.v. acetaminophen 1 g, parecoxib 40 mg+acetaminophen 1 g or placebo in a double-blind, crossover study. Platelet function was assessed by photometric aggregometry and by measuring the release of thromboxane B 2. Plasma acetaminophen concentrations were measured by high-performance liquid chromatography. Results. Platelet aggregation (median area under the curve) triggered with arachidonic acid 500 mm was 24.6, 3.9 and area units (P=0.02, all groups) after placebo, acetaminophen and parecoxib+acetaminophen, respectively. Inhibition of platelet aggregation showed no difference between acetaminophen alone and the combination (P=0.82). Aggregation triggered with arachidonic acid 750 or 1000 mm, adenosine diphosphate (ADP) 1.5 or 3 mm, or epinephrine 5 mm showed no differences between the groups. Release of thromboxane B 2 in response to ADP was inhibited similarly by both acetaminophen and the combination. Plasma acetaminophen concentrations were similar after acetaminophen and the combination. Conclusions. Acetaminophen and parecoxib showed no interaction in inhibiting platelet function. In combination they cause a mild degree of COX-1 inhibition corresponding to that of acetaminophen alone. Br J Anaesth 2006; 97: Keywords: analgesics non-opioid, acetaminophen; blood, platelets; pharmacology, drug interactions; thromboxane Accepted for publication: March 29, 2006 Postoperative analgesia often includes non-steroidal antiinflammatory drugs (NSAIDs). The desire to avoid NSAID side-effects launched the development of coxibs, cyclooxygenase-2 (COX-2)-selective inhibitors. 1 Increasing selectivity reduces risk of gastric toxicity 2 and platelet inhibition, 3 but may elevate the risk of cardiovascular adverse events. 4 Parecoxib, the pro-drug of valdecoxib, is the only coxib available for parenteral use, 5 and particularly useful after operation when nausea and vomiting prevent oral administration. Valdecoxib alone does not inhibit platelet aggregation. 6 Acetaminophen, available both for oral and parenteral use, differs from NSAIDs in that it inhibits prostaglandin synthesis mainly in the central nervous system. 7 However, acetaminophen has peripheral side-effects such as dosedependent inhibition of platelet aggregation 8 and gastric toxicity. 9 In postoperative analgesia, the combination of acetaminophen and a traditional NSAID is useful. 10 Therefore, the combination of acetaminophen and a coxib might be attractive. Theoretically, the coxib may reduce the risk of bleeding in comparison with non-selective NSAIDs, Ó The Board of Management and Trustees of the British Journal of Anaesthesia All rights reserved. For Permissions, please journals.permissions@oxfordjournals.org

2 Parecoxib, acetaminophen and platelets whereas the clinical significance of the weak platelet inhibition by acetaminophen is unclear. To our knowledge, the effect on platelet function of the combination of acetaminophen and a coxib has not yet been addressed experimentally. We hypothesize that platelet aggregation is equally inhibited by the combination and acetaminophen alone. Methods Eighteen healthy, non-smoking men between 22 and 33 yr of age volunteered in this double-blinded, randomized, placebo-controlled, crossover study, approved by the Ethics Committee for Studies in Healthy Subjects and Primary Care in the Hospital District of Helsinki and Uusimaa, and by the National Agency for Medicines in Finland. Written informed consent and normal plasma alanine transaminase and aspartate aminotransferase activities were a prerequisite for participation. The use of acetylsalicylic acid was forbidden for 10 days and that of other drugs affecting platelet function for 1 week before each experiment. Experimental procedures Every volunteer participated in three experiments with at least a 1 week interval between experiments. After 3 h of fasting, a venous blood sample (approximately 35 ml) was drawn from an antecubital vein through a 17 gauge cannula (VenflonÔ, Becton Dickinson, Franklin Lakes, NJ, USA). After the first sample, a dorsal vein of the contralateral hand was cannulated with a 20 gauge cannula and a slow infusion of 0.9% NaCl was started. Parecoxib 40 mg (Dynastat Ò, Pfizer, New York, USA) or placebo (NaCl 0.9%, Braun, Kronberg, Germany) was given as a 1 min bolus and the next blood sample was drawn after 50 min to allow time for biotransformation of parecoxib to valdecoxib. Approximately half an hour later acetaminophen 1 g (Perfalgan Ò, Bristol Myer Squibb, New York, USA) or placebo was given as a 10 min infusion, and the last blood sample was drawn 10 min after completion of the acetaminophen infusion. Drug combinations were parecoxib+acetaminophen, placebo+ acetaminophen and placebo+placebo. The infusions were blinded and administered in random order. Blood samples were collected into polypropylene tubes (Vacuette Ò, Greiner Bio-one, Austria) containing 3.2% buffered citrate, giving a volume ratio of 1:10. Samples for acetaminophen concentration measurement were collected into plastic EDTA-tubes (5.9 mg K2EDTA, VenoSafeÔ, Terumo Europe, Haasrode, Belgium). Laboratory tests Platelet aggregation Platelet aggregation was measured with a four channel photometric aggregometer (Packs-4, Helena Laboratories, USA) based on the method of Born. 11 Platelet-rich plasma (PRP) and platelet-poor plasma were prepared by centrifuging as described previously. 12 Aggregation was induced in 270 ml of PRP by adding 30 ml of one of the following triggers: adenosine diphosphate (ADP) at a final concentration of 1.5 or 3 mm, arachidonic acid at a final concentration of 500, 750 or 1000 mm, or epinephrine at a final concentration of 5 mm. These concentrations are known to cause platelet aggregation. Reagents were purchased from Sigma Aldrich (St Louis, USA) and Calbiochem (San Diego, USA). Aggregation was allowed to proceed for 300 s, the area under the curve of the aggregometry was recorded. Plasma for thromboxane B 2 (TxB 2 ) determination was prepared as described earlier. 12 Arachidonic acid EC 50 Part of the volunteers (n=9) were further investigated by determining the concentration of arachidonic acid causing 50% aggregation (EC 50 ) after the addition of acetaminophen 20 mg litre 1 to PRP in vitro. EC 50 determinations were performed in samples drawn before and 50 min after administration of parecoxib 40 mg. Aggregation was triggered with different concentrations of arachidonic acid ( mm) to achieve near half-maximal responses. Aggregation in plain PRP, triggered with arachidonic acid 1000 mm, was considered 100% and aggregation in PRP treated with acetylsalicylic acid 100 mm was considered 0%. The EC 50 values were determined with nonlinear regression based on the Hill equation. 13 Thromboxane B 2 concentration Thromboxane B 2 (TxB 2 ), the stable metabolite of thromboxane A 2 (TxA 2 ), is formed during aggregation. TxB 2 concentrations in PRP triggered with ADP 3mM or arachidonic acid 1000 mm were determined with a radioimmunoassay as described earlier. 14 The inter-assay coefficient of variation was 26% (n=10), but to avoid the impact of this variation the samples compared were run in the same batch of the assay (intra-assay coefficient of variation n=17% 12 ). Acetaminophen concentration Blood samples drawn after administration of the second drug were centrifuged at 3000 g for 10 min and plasma was stored at 20 C. Acetaminophen concentration was determined using high-performance liquid chromatography. 15 The limit of quantification was 0.1 mg litre 1, and the day-to-day coefficients of variation were 4.2% at 2.5 mg litre 1 and 4.1% at 14.4 mg litre 1 (n=4). Statistical analysis We tested the hypothesis of a difference in platelet aggregation between acetaminophen and parecoxib+acetaminophen greater than 1 SD (a-error=5%). We considered a difference smaller than 1 SD to be of minor clinical significance. As no difference was anticipated, we increased the power of the 227

3 Munsterhjelm et al. study to 95% in order to control the b-error. The sample-size needed under these circumstances was n= Data were tested for normality with the Kolmogorow Smirnov test. Non-normally distributed data were analysed with Friedman s test (repeated measures ANOVA on ranks); three groups, and the Wilcoxon matched pairs signed rank sum test; acetaminophen vs combination. From normally distributed data, confidence intervals were calculated using the appropriate t-distribution. Statistical testing was with SigmaStat for Windows Version 2.03 (SPSS Inc. Chicago, IL, USA). Results All 18 volunteers completed the study according to the protocol. Two volunteers showed an abnormally long lag phase before secondary aggregation in response to arachidonic acid in pre-infusion aggregation, and these data were excluded from further statistical analysis. Parecoxib showed no inhibition (P=0.21), whereas acetaminophen showed a clear inhibition of platelet aggregation triggered with arachidonic acid 500 mm (P=0.019, Fig. 1). No statistical difference occurred between the combination and acetaminophen alone (P=0.82), although aggregation varied in five volunteers. Mean plasma acetaminophen concentrations were similar in both groups; 12.7 mg litre 1 (95% CI ) in the acetaminophen group and 12.6 mg litre 1 (95% CI ) in the combination group. Those three volunteers showing more inhibition by the combination had plasma acetaminophen concentrations of 11.9, 9.7 and 14.2 mg litre 1 after the combination and 16.8, 11.5 and 12.8 mg litre 1 after acetaminophen alone, respectively, whereas those two showing a clear inhibition only by acetaminophen alone had plasma acetaminophen concentrations of 14.9 and 10.9 mg litre 1 after the combination and 14.6 and 13.8 mg litre 1 after acetaminophen alone. To further elucidate any possible interaction between parecoxib and acetaminophen, we determined arachidonic acid EC 50 -values in vitro, after adding acetaminophen 20 mg litre 1 to samples drawn before and after administration of parecoxib 40 mg. The selected nine volunteers, marked with closed circles in Figure 1, included those with varying inhibition by acetaminophen and the combination. Mean EC 50 was 843 mm (95% CI ) before and 742 mm (95% CI ) after parecoxib administration. Mean change after parecoxib was 10.8% (95% CI 23.7 to 2.3). When triggered with arachidonic acid 750 or 1000 mm, ADP (1.5 or 3 mm), or epinephrine (5 mm), we detected no inhibition of platelet aggregation by either acetaminophen or the combination (Table 1). Release of TxB 2 during aggregation triggered with ADP was inhibited similarly by acetaminophen and by the combination (Table 2). When triggered with arachidonic acid, we observed no differences in TxB 2 release between groups. Discussion Our findings show that platelet function, assessed with photometric aggregometry and by measuring release of TxB 2,is equally inhibited by acetaminophen alone and the combination of acetaminophen and parecoxib. Parecoxib exhibits Aggregation AUC Baseline aggregation Placebo Pare Placebo Acet Comb Fig 1 Platelet aggregation triggered with arachidonic acid 500 mm. Each volunteer (n=16) received placebo, acetaminophen 1 g (Acet), or the combination of parecoxib 40 mg and acetaminophen 1 g (Comb). Parecoxib only sampling (Pare) was before acetaminophen administration in the combination group. Closed circles indicate those nine volunteers selected for arachidonic acid EC 50 determinations in vitro. *P=0.019; all groups (Friedman s test), and P=0.82; acetaminophen vs combination (Wilcoxon matched pairs signed rank sum test). 228

4 Parecoxib, acetaminophen and platelets Table 1 Platelet aggregation triggered with arachidonic acid, ADP or epinephrine. Data are aggregation (AUC, 10 3 area units) reported as medians (25th/75th percentiles). Each volunteer (n=18, *n=16) received placebo, acetaminophen 1 g, or the combination of parecoxib 40 mg and acetaminophen 1 g. Aggregation triggered with arachidonic acid 500 mm is shown in Figure 1. Statistical tests are Friedman s test (all groups) and Wilcoxon matched pairs signed rank sum test (acetaminophen vs combination). Aggregation trigger Sampling Drug combination P-value (all groups) Parecoxib+ placebo acetaminophen acetaminophen P-value acetaminophen vs combination Arachidonic acid (750 mm)* Arachidonic acid (1000 mm)* Baseline aggregation After first drug After second drug Baseline aggregation After first drug After second drug 25.5 (24.8/26.1) 25.6 (24.6/26.2) 25.5 (24.5/26.1) 25.9 (25.5/26.3) 25.4 (25.1/26.5) 25.6 (25.1/26.1) 25.4 (24.7/26.3) 25.8 (25.1/26.4) 24.2 (21.9/26.2) 25.2 (24.3/26.2) 26.1 (25.1/26.5) 24.6 (22.8/26.5) 25.5 (25.1/26.2) 25.7 (25.4/26.3) 24.8 (24.0/25.9) 25.4 (24.6/26.3) 25.6 (24.6/26.6) 25.5 (25.1/26.6) ADP (1.5 mm) Baseline aggregation 12.2 (8.2/18.2) 10.9 (8.4/21.4) 11.2 (8.1/16.0) 0.41 After first drug 11.3 (7.4/19.3) 13.4 (7.0/19.8) 10.9 (8.6/20.7) 0.61 After second drug 12.1 (8.2/21.3) 9.2 (6.9/19.4) 11.0 (7.1/20.0) ADP (3.0 mm) Baseline aggregation 23.0 (19.6/24.8) 21.1 (14.6/24.8) 22.0 (17.1/23.7) 0.35 After first drug 23.2 (15.9/24.3) 23.1 (16.3/24.9) 21.5 (14.9/25.0) 0.68 After second drug 23.0 (18.9/25.3) 17.9 (15.4/24.8) 20.3 (16.9/24.5) Epinephrine (5 mm) Baseline aggregation 20.3 (14.5/22.2) 18.9 (16.2/22.2) 19.7 (15.1/21.0) 0.95 After first drug 18.7 (13.6/21.0) 18.3 (15.0/22.1) 18.7 (13.2/20.7) 0.68 After second drug 18.7 (12.8/22.3) 15.9 (11.8/19.8) 17.9 (10.5/20.7) Table 2 TxB 2 release from activated platelets. Data are concentrations (mg litre 1 ) reported as medians (25th/75th percentiles). Each volunteer (n=18) received placebo, acetaminophen 1 g, or the combination of parecoxib 40 mg and acetaminophen 1 g. Statistical tests are Friedman s test (all groups) and Wilcoxon matched pairs signed rank sum test (acetaminophen vs combination). Aggregation trigger Sampling Drug combination P-value (all groups) Parecoxib+ placebo acetaminophen acetaminophen P-value acetaminophen vs combination Arachidonic acid (1000 mm) Baseline After first drug After second drug 1334 (1145/1501) 1433 (1118/1846) 1478 (1200/1661) 1530 (1319/1761) 1460 (1337/1630) 1252 (1057/1518) 1518 (1284/1696) 1393 (1263/1809) 1372 (983/1559) ADP (3 mm) Baseline 24.9 (19.2/38.9) 30.4 (11.8/39.9) 27.5 (13.6/42.3) After first drug 25.7 (12.2/37.8) 28.4 (16.1/44.4) 27.3 (11.8/36.0) After second drug 28.7 (20.1/42.7) 16.8 (8.69/30.0) 19.7 (11.5/31.6) no inhibition of platelet function, 3 and our results confirm these observations. Acetaminophen, on the other hand, dose-dependently inhibits platelet aggregation. 8 The ability of NSAIDs to inhibit platelet function is directly related to their ability to inhibit COX-1. Valdecoxib, to which parecoxib is hydrolysed by a cytochrome P450-independent mechanism in the liver, has low affinity for COX-1. 1 Acetaminophen is a weak inhibitor of COX COX-1 is one link in the chain of events leading to platelet aggregation. Platelet activation is initiated by exposed subendothelial matrix and the production of thrombin at the site of vascular injury. 18 Activated platelets secrete ADP, stored in intracellular granula, and release arachidonic acid from their cellular membrane. Arachidonic acid is transformed to prostaglandin H 2 by COX-1, and by thromboxane synthase further to TxA 2. Both TxA 2 and ADP bind to specific G-protein-coupled receptors on the surface of the platelet. 19 The activated receptors initiate intracellular signalling leading to activation of the GP IIb IIIa (also known as integrin a IIb b 3 ) complex responsible for aggregation. We triggered platelet aggregation with arachidonic acid, ADP or epinephrine. The most sensitive trigger in detecting inhibition of platelet COX is a low concentration of arachidonic acid, as it competes with the drug for binding to the enzyme. 20 Arachidonic acid 1000 and 750 mm triggered aggregation insensitive to acetaminophen, which agrees with our previous results in healthy volunteers, 8 and the addition of parecoxib showed no additional effect. With arachidonic acid 500 mm, on the other hand, acetaminophen 1 g, and the combination of acetaminophen 1 g and parecoxib 40 mg were clearly inhibitory in most volunteers. Lowering the concentration of arachidonic acid in the assay, however, not only increases the sensitivity to the drugs, but also increases the amount of random variation as the individual aggregation threshold is approached. 229

5 Munsterhjelm et al. A high degree of both inter- and intrasubject variability in aggregation triggered with arachidonic acid has been previously reported. 21 Part of the variation in our results can be explained by random variation in acetaminophen concentrations between experiments; two volunteers with a different degree of aggregation after acetaminophen and combination showed a plasma acetaminophen concentration more than 1 SD higher in the non-aggregating sample. Our interpretation that the variation is by random rather than an effect of parecoxib is further supported by the lack of difference between arachidonic acid EC 50 -values before and after parecoxib administration. With ADP or epinephrine as aggregation triggers, we detected no inhibitory effect. As these triggers bind directly to receptors on the surface of the platelet, they trigger aggregation relatively insensitive to COX inhibition. This is also in line with our previous results with increasing doses of acetaminophen. 8 Another approach to evaluate COX inhibition is to measure the release of the stable end-product of the pathway of prostaglandin synthesis in platelets, that is TxB 2. We detected no difference in TxB 2 release, further supporting the lack of interaction between acetaminophen and parecoxib. The release of TxB 2 also allows a comparison with non-selective NSAIDs. In a previous study, median TxB 2 release in response to ADP after an i.v. infusion of diclofenac (1.1 mg kg 1 ) was only 3.3% of preinfusion release, 12 as compared with 55% after acetaminophen and 72% after the combination in this study. Platelet aggregation is tightly regulated. Endothelial cells synthesize prostacyclin that binds to G-protein-coupled receptors on the surface of platelets. 22 Activation of these receptors opposes the transduction of the pro-aggregatory signal. In contrast to TxA 2 formation in platelets, prostacyclin in endothelial cells is synthesized from arachidonic acid by COX-2. The COX-2 selective coxibs therefore inhibit the production of prostacyclin without affecting TxA 2 production, as shown by McAdam and colleagues 23 already in The authors of that paper raised concern about possible cardiovascular toxicity. Only in recent years has this important question been addressed in large randomized clinical trials As a result, rofecoxib was withdrawn from the market. Clinical implications Parecoxib is useful in postoperative analgesia after, for instance, gynaecological laparotomy and total hip and knee arthroplasty The analgesic effect of acetaminophen is also well documented. 10 As far as we know, no studies have explored the analgesic effect of parecoxib and acetaminophen in combination. Acetaminophen has been combined with other coxibs, but data are sparse and still inconclusive. The combination of celecoxib and acetaminophen was more effective than either drug alone after otolaryngological surgery, 30 but rofecoxib plus acetaminophen equalled acetaminophen alone after tonsillectomy. 31 The pro-aggregatory effect of valdecoxib may have disadvantages after certain types of surgery. Patients undergoing cardiovascular surgery are at high risk of experiencing thrombotic events, and valdecoxib increases this risk. 32 Coxibs are thus contraindicated after this type of surgery. In other types of surgery, however, the risk of intra- and postoperative bleeding may prevail. After tonsillectomy, for example, non-selective NSAIDs increase the risk of reoperation because of haemorrhage. 33 Under such circumstances coxibs may be justified. Our results indicate that the combination of parecoxib and acetaminophen causes a mild degree of COX-1 inhibition, but any clinical conclusion should be drawn with care. Our study population comprised young healthy males not necessarily representative of the general population. Furthermore, our methodology included no tests of clinical bleeding or thrombosis. We still believe that the combination of acetaminophen and coxibs may prove useful in patients with a low risk of thrombotic events, who are susceptible to haemorrhagic complications of non-selective NSAIDs. Acknowledgements The authors thank Professor Hannele Yki-Järvinen for the laboratory facilities; Anna Becker and Mia Biström for skilled technical assistance; Kristiina Koivisto, Maija-Liisa Mäkinen and Marja-Leena Ylinen for taking good care of the volunteers; and our volunteers for smooth cooperation. This work was financed by departmental funding and a grant from the Medical Society of Finland, Helsinki, Finland. References 1 Tacconelli S, Capone ML, Sciulli MG, Ricciotti E, Patrignani P. The biochemical selectivity of novel COX-2 inhibitors in whole blood assays of COX-isozyme activity. Curr Med Res Opin 2002; 18: Silverstein FE, Faich G, Goldstein JL, et al. Gastrointestinal toxicity with celecoxib vs nonsteroidal anti-inflammatory drugs for osteoarthritis and rheumatoid arthritis: the CLASS study: a randomized controlled trial. JAMA 2000; 284: Noveck RJ, Laurent A, Kuss M, Talwalker S, Hubbard RC. Parecoxib sodium does not impair platelet function in healthy elderly and non-elderly individuals. Clin Drug Invest 2001; 21: Bresalier RS, Sandler RS, Quan H, et al. Cardiovascular events associated with rofecoxib in a colorectal adenoma chemoprevention trial. N Engl J Med 2005; 352: Cheer SM, Goa KL. Parecoxib (parecoxib sodium). Drugs 2001; 61: Leese PT, Recker DP, Kent JD. The COX-2 selective inhibitor, valdecoxib, does not impair platelet function in the elderly: results of a randomized controlled trial. J Clin Pharmacol 2003; 43: Flower RJ, Vane JR. Inhibition of prostaglandin synthetase in brain explains the anti-pyretic activity of paracetamol (4-acetamidophenol). Nature 1972; 240: Munsterhjelm E, Munsterhjelm NM, Niemi TT, Ylikorkala O, Neuvonen PJ, Rosenberg PH. Dose-dependent inhibition of platelet function by acetaminophen in healthy volunteers. Anesthesiology 2005; 103: Garcia Rodriguez LA, Hernandez-Diaz S. Risk of uncomplicated peptic ulcer among users of aspirin and nonaspirin nonsteroidal antiinflammatory drugs. Am J Epidemiol 2004; 159:

6 Parecoxib, acetaminophen and platelets 10 Rømsing J, Møiniche S, Dahl JB. Rectal and parenteral paracetamol, and paracetamol in combination with NSAIDs, for postoperative analgesia. Br J Anaesth 2002; 88: Born GVR. Quantitative investigations into the aggregation of blood platelets. J Physiol 1962; 162: 67 8P 12 Munsterhjelm E, Niemi TT, Syrjälä MT, Ylikorkala O, Rosenberg PH. Propacetamol augments inhibition of platelet function by diclofenac in volunteers. Br J Anaesth 2003; 91: Bowen WP, Jerman JC. Nonlinear regression using spreadsheets. Trends Pharmacol Sci 1995; 16: Viinikka L, Ylikorkala O. Measurement of thromboxane B 2 in human plasma or serum by radioimmunoassay. Prostaglandins 1980; 20: Pufal E, Sykutera M, Rochholz G, Schutz HW, Sliwka K, Kaatsch HJ. Determination of paracetamol (acetaminophen) in different body fluids and organ samples after solid-phase extraction using HPLC and an immunological method. Fresenius J Anal Chem 2000; 367: Armitage P, Berry G, Matthews JNS. Statistical Methods in Medical Research, 4th Edn. Bodmin, UK: Blackwell Science, 2002: Mitchell JA, Akarasereenont P, Thiemermann C, Flower RJ, Vane JR. Selectivity of nonsteroidal antiinflammatory drugs as inhibitors of constitutive and inducible cyclooxygenase. Proc Natl Acad Sci USA 1993; 90: Andrews RK, Berndt MC. Platelet physiology and thrombosis. Thromb Res 2004; 114: Murugappan S, Shankar H, Kunapuli SP. Platelet receptors for adenine nucleotides and thromboxane A2. Semin Thromb Hemost 2004; 30: Munsterhjelm E, Niemi TT, Ylikorkala O, Silvanto M, Rosenberg PH. Characterisation of inhibition of platelet function by paracetamol and its interaction with diclofenac in vitro. Acta Anaesthesiol Scand 2005; 49: Burke J, Kraft WK, Greenberg HE, et al. Relationship of arachidonic acid concentration to cyclooxygenase-dependent human platelet aggregation. J Clin Pharmacol 2003; 43: Armstrong RA. Platelet prostanoid receptors. Pharmacol Ther 1996; 72: McAdam BF, Catella-Lawson F, Mardini IA, Kapoor S, Lawson JA, FitzGerald GA. Systemic biosynthesis of prostacyclin by cyclooxygenase (COX)-2: the human pharmacology of a selective inhibitor of COX-2. Proc Natl Acad Sci USA 1999; 96: Bresalier RS, Sandler RS, Quan H, et al. Cardiovascular events associated with rofecoxib in a colorectal adenoma chemoprevention trial. N Engl J Med 2005; 352: Solomon SD, McMurray JJ, Pfeffer MA, et al. Cardiovascular risk associated with celecoxib in a clinical trial for colorectal adenoma prevention. N Engl J Med 2005; 352: Malan TP, Jr, Gordon S, Hubbard R, Snabes M. The cyclooxygenase-2-specific inhibitor parecoxib sodium is as effective as 12 mg of morphine administered intramuscularly for treating pain after gynecologic laparotomy surgery. Anesth Analg 2005; 100: Barton SF, Langeland FF, Snabes MC, et al. Efficacy and safety of intravenous parecoxib sodium in relieving acute postoperative pain following gynecologic laparotomy surgery. Anesthesiology 2002; 97: Malan TP, Jr, Marsh G, Hakki SI, Grossman E, Traylor L, Hubbard RC. Parecoxib sodium, a parenteral cyclooxygenase 2 selective inhibitor, improves morphine analgesia and is opioidsparing following total hip arthroplasty. Anesthesiology 2003; 98: Hubbard RC, Naumann TM, Traylor L, Dhadda S. Parecoxib sodium has opioid-sparing effects in patients undergoing total knee arthroplasty under spinal anaesthesia. Br J Anaesth 2003; 90: Issioui T, Klein KW, White PF, et al. The efficacy of premedication with celecoxib and acetaminophen in preventing pain after otolaryngologic surgery. Anesth Analg 2002; 94: Pickering AE, Bridge HS, Nolan J, Stoddart PA. Double-blind, placebo-controlled analgesic study of ibuprofen or rofecoxib in combination with paracetamol for tonsillectomy in children. Br J Anaesth 2002; 88: Nussmeier NA, Whelton AA, Brown MT, et al. Complications of the COX-2 inhibitors parecoxib and valdecoxib after cardiac surgery. N Engl J Med 2005; 352: Moiniche S, Romsing J, Dahl JB, Tramer MR. Nonsteroidal antiinflammatory drugs and the risk of operative site bleeding after tonsillectomy: a quantitative systematic review. Anesth Analg 2003; 96: