February 2011 Regular Article Biol. Pharm. Bull. 34(2) 233—237 (2011) 233
Influence of Nonsteroidal Anti-inflammatory Drugs on the Antiplatelet Effects of Aspirin in Rats
Yuuki AKAGI,* Yuta NIO, Shuji SHIMADA, and Takao AOYAMA Faculty of Pharmaceutical Sciences, Tokyo University of Science; 2641 Yamazaki, Noda, Chiba 278–8510, Japan. Received August 23, 2010; accepted November 16, 2010; published online November 24, 2010
Low-dose aspirin acts by irreversibly acetylating internal cyclooxygenase-1 (COX-1) on platelets, thereby suppressing platelet aggregation. Because nonsteroidal anti-inflammatory drugs (NSAIDs) also inhibit COX-1, the antiplatelet effects of aspirin may be suppressed when it is co-administered with NSAIDs. In this study, the influences of ibuprofen, loxoprofen sodium and etodolac on the antiplatelet effects of aspirin were investigated in male Sprague-Dawley (SD) rats. Aspirin and/or NSAIDs were administered orally at single or multiple daily
doses. Platelet aggregation (ADP and collagen were added as stimuli) and serum thromboxane B2 (TXB2) concen- trations were measured. The maximum inhibitions of aggregation in the aspirin before ibuprofen group were 41.0 7.8% for ADP and 38.7 5.4% for collagen at 6 h after administration; similar values were seen in the aspirin group; however, percent inhibitions in the aspirin before ibuprofen multiple administration group were lower than those in the aspirin group. Thus, the inhibitory effects of daily low-dose aspirin on platelets are
competitively inhibited by the prolonged use of multiple daily doses of ibuprofen. In contrast, serum TXB2 con- centrations in all groups were lower than those in the control group (drug-free). This suggests that the relation- ship between the inhibition of platelet COX-1 and the suppression of platelet aggregation is nonlinear. When aspirin was administered with loxoprofen sodium, similar effects were observed; however, with etodolac, the antiplatelet effects in all groups were equal to those in the aspirin group. Accordingly, if co-administration with NSAIDs is necessary with low-dose aspirin, a selective COX-2 inhibitor, such as etodolac, should be used. Key words aspirin; antiplatelet; nonsteroidal anti-inflammatory drug; drug interaction; rat; cyclooxygenase-1
Aspirin has long been used as an analgesic and antipyretic. with aspirin. Moreover, if inhibition of the antiplatelet effects Since the suppression of platelet aggregation by low-dose of aspirin is noted, it is necessary to consider alternatives in aspirin was noted in 1967,1) several large-scale clinical trials order to avoid such interactions. have demonstrated its antiplatelet efficacy.2) Aspirin was In this study, the influence of NSAIDs on the antiplatelet approved as a platelet aggregation inhibitor in Japan in 2000, effects of aspirin was investigated in rats. Ibuprofen, loxopro- and is now widely prescribed for the secondary prevention of fen sodium or etodolac (COX-2 selective inhibitor) were con- cardiovascular events.3) currently administrated with aspirin. First, we conducted a Low-dose aspirin inhibits cyclooxygenase-1 (COX-1) on study using a single daily dose of aspirin and/or NSAIDs, platelets, suppresses arachidonic acid metabolism, and pre- and measured platelet aggregation and serum thromboxane vents the synthesis of thromboxane A2 (TXA2), a compound B2 (TXB2). We then undertook a study of multiple daily that causes platelet aggregation. Aspirin acts on internal doses of aspirin (once a day) and/or NSAIDs (three times a COX-1 via a hydrophobic channel and irreversibly acetylates day) for 3 d, as this is a clinically used regimen. Finally, we Ser529.4) Access to the COX-1 active site is then impeded for devised a method to avoid the interactions that were the lifetime of the platelet. After COX-1 on platelets is acetyl- observed. ated by aspirin, the antiplatelet effects are thought to depend on platelet turnover (approx. 7 to 10 d), and to be maintained MATERIALS AND METHODS until new platelets, unacetylated by aspirin, are produced.5) Nonsteroidal anti-inflammatory drugs (NSAIDs), includ- Chemicals Aspirin (Lot No. TSH5399), ibuprofen (Lot ing ibuprofen, are also used as analgesics, antipyretics and No. TSG6469) and etodolac (Lot No. PEH6344) were pur- anti-inflammatory agents. Unlike aspirin, NSAIDs form a chased from Wako Pure Chemical Industries, and loxoprofen salt bridge with Arg120 of COX-1,6) and inhibit the enzy- sodium (Lot No. 6411) was kindly provided by Daiichi- matic activity of prostaglandin H2 synthesis. Suppression of Sankyo Co. (Tokyo, Japan). Platelet aggregation was induced platelet function is limited to a fixed time after taking with ADP (Wako Pure Chemical Industries, Ltd., Osaka, NSAIDs, because the COX-1 inhibition is reversible.7) Japan; Lot No. 105704) or collagen (Research Institute for The antiplatelet effects of aspirin may be decreased with the Functional Peptides, Yamagata, Japan; Lot No. 8HIA). co-administration of ibuprofen,4,8) and a warning is included All other chemicals were of analytical grade. in package inserts of both aspirin and ibuprofen. When Animals Male Sprague-Dawley (SD) rats (age, 8—9 ibuprofen binds with COX-1, it impedes aspirin acetylation weeks; weight, 250—300 g) were used in this study.11,12) of Ser529. Because numerous patients using low-dose aspirin They were housed under standard conditions (23 1 °C; rela- also take NSAIDs9) to relieve pain from conditions such as tive humidity, 55 5%) and maintained under a 12-h light/ arthritis and articular rheumatisms, the influence on the dark cycle. After a 12-h fast with free access to water, all rats antiplatelet effects of aspirin is of particular interest. Loxo- had an indwelling cannula (Polyethylene tube, SP31; Na- profen sodium is more frequently prescribed than ibuprofen tsume Seisakusho Co., Ltd., Tokyo, Japan) implanted in the in Japan,10) but no information is available on its interaction femoral artery for blood sampling. Implantation surgery was
∗ To whom correspondence should be addressed. e-mail: [email protected] © 2011 Pharmaceutical Society of Japan 234 Vol. 34, No. 2 performed under ether anesthesia. Rats were divided into each group (4 rats/group) at random. Study protocols were approved by the Institutional Animal Care and Use Commit- tee of the Tokyo University of Science. Preparation of Drug Solutions Aspirin (7.5 mg/ml), ibuprofen (10 mg/ml), loxoprofen sodium (2.5 mg/ml) and etodolac (2.5 mg/ml) were suspended in 0.5% carboxymethyl cellulose sodium (CMC-Na; Tokyo Chemical Industry, Co., Ltd., Tokyo, Japan; Lot No. C0045) solution. The final con- centration of ADP after addition to platelet-rich plasma (PRP) was 128 m M, and that of collagen was 8 and 25 mg/ml. Administration of Aspirin and/or NSAIDs The doses of aspirin, ibuprofen, loxoprofen sodium and etodolac were 15, 20, 5 and 5 mg/kg,13—18) respectively. The antiplatelet effects of aspirin13) and the analgesic and anti-inflammatory effects of NSAIDs have been confirmed at the present doses in rats.14—18) Drug solutions were orally administered at 2 ml/kg. The single daily dose groups included the aspirin, Fig. 1. Study Itinerary for Single Daily (A) and Multiple Daily Doses (B) NSAIDs, aspirin before NSAIDs, and NSAIDs before aspirin groups (Fig. 1A). In the study using multiple daily doses, at 2000 g for 15 min to remove platelets. Supernatants were NSAIDs were administered at 2, 6 and 10 h after aspirin for immediately stored at 30 °C until analysis. Serum TXB2 3 d (Fig. 1B). CMC-Na solution (drug-free) was adminis- was measured by enzyme-linked immunosorbent assay tered to control rats. (ELISA, Cayman Co., Ann Arbor, MI, U.S.A.) according to Blood Collection Approximately 2 ml of blood was col- the manufacturer’s instructions, and changes were analyzed lected slowly into a plastic syringe (volume: 2.5 ml; needle by Tukey’s t-test, with a value of p 0.05 being considered size: 21 G) containing a 1/10 volume of 3.13% citrate significant. sodium, before the first administration (0 h; Figs. 1A-I) and at 6 and 12 h after administration. For multiple daily doses, RESULTS blood was taken before aspirin administration (0 h), and at 6 and 12 h after administration on day 3. PRP was obtained In the aspirin (Asp) group, the maximum inhibition of from the upper layer after centrifuging blood at 200 g at platelet aggregation was 44.0 3.9% for ADP (128 m M) and room temperature for 10 min. Platelet-poor plasma (PPP) 37.7 4.3% for collagen (25 mg/ml) at 6 h after administra- was obtained after centrifuging the residue at 1700 g at tion (p 0.01). In contrast, the percent inhibitions with room temperature for 15 min. ibuprofen (Ibu), loxoprofen sodium (Lox) and etodolac (Eto) Measurement of Platelet Aggregation Platelet aggre- were 0.7—18.5% (Figs. 2A, B), which were nearly identical gation was measured using a light transmission aggregome- to the values in the control group, and suppression of platelet ter, PRP313M (IMI Co., Ltd., Saitama, Japan), according to aggregation was not observed. These inhibition rates were the Born and Cross Method.19) PRP (100 ml) was pre-incu- similar at 12 h. bated in a cuvette at 37 0.5 °C for 15 min in the PRP313M. In the study using a single daily dose of aspirin and Platelet aggregation stimulus (ADP or collagen) was added NSAIDs, the maximum inhibitions of aggregation in the to PRP and the aggregation rate was recorded for 10 min. As- aspirin before ibuprofen (Asp→Ibu) group were 41.0 7.8% suming light transmission of PPP to be 100%, that of the for ADP and 38.7 5.4% for collagen at 6 h, which was simi- PRP sample was recorded as the aggregation rate. The ratios lar to values in the Asp group. Percent inhibition in the of maximum aggregation rate after 6 and 12 h to that before ibuprofen before aspirin (Ibu→Asp, ADP: 4.5 2.5%, colla- administration were then calculated. For multiple daily doses, gen: 2.2 1.0%, p 0.01) group was significantly lower than the ratios of maximum aggregation rate after 6 and 12 h to that in the Asp group (Figs. 3A1, A2). These inhibition rates that of 0 h on day 3 in the control group were calculated. Re- were similar at 12 h. For loxoprofen sodium, the maximum sults were expressed as maximum inhibition of platelet ag- inhibition of platelet aggregation was equal to that of ibupro- gregation (mean S.E.), and Tukey’s test was used for statis- fen (Figs. 3B1, B2); however, the levels were similar to those tical analysis to determine significance among the groups. A in the Asp group, regardless of the order of co-administration value of p 0.05 was considered to be significant. As there with etodolac, and no suppression of antiplatelet effects was are many variables that affect platelet aggregation,20,21) we observed (Figs. 3C1, C2). carefully standardized the conditions for blood collection and In contrast, the maximum inhibitions of aggregation in the time until measurement. aspirin before ibuprofen multiple administration (Asp→Ibu, Measurement of Serum TXB2 Platelet COX-1 activity multiple) group were 14.3 14.0% for ADP and 1.8 6.0% was determined by measuring serum levels of TXB2, the for collagen at 0 h on day 3, and these values were lower than major stable metabolite of TXA2, in platelets. To minimize those in the aspirin multiple administration (Asp, multiple) the influence of TXB2 derived from hematopoietic cells, group (ADP: p 0.05, collagen: p 0.01). Similar values TXB2 concentrations of the supernatant from collagen-stimu- were seen at 6 and 12 h (Figs. 4A1, A2). For loxoprofen lated PRP were determined.22) PRP (100 ml) was stimulated sodium (Asp→Lox, multiple), percent inhibition was equal with 8 mg/ml collagen for 10 min, followed by centrifugation to ibuprofen (Figs. 4B1, B2); however, almost no significant February 2011 235
aspirin-induced inhibition of platelet COX-1 in rats. In addi- tion to ibuprofen, loxoprofen sodium (well-prescribed in Japan) and etodolac (a COX-2 selective inhibitor) were used in this experiment. Platelet aggregation in the Asp→Ibu group was equal to that in the Asp group, thus suggesting that the influence of ibuprofen could be eliminated by administering aspirin 2 h before ibuprofen. In contrast, the percent inhibition observed in the Asp and Asp→Ibu groups was not observed in the Ibu→Asp group. These results show that the antiplatelet effects of aspirin are suppressed by ibuprofen when adminis- tered before aspirin (Figs. 3A1, A2). It has been reported that there is no decrease in antiplatelet effects when giving healthy subjects aspirin before ibuprofen; however, such effects were observed when giving ibuprofen before aspirin.4) The results of our experiment in rats showed similar trends, and we thus considered our animal model to reflect the inter- action between aspirin and NSAIDs. Co-administration of loxoprofen sodium with aspirin had a similar influence to that of ibuprofen (Figs. 3B1, B2). This interaction is consis- tent with the report that NSAIDs competitively inhibit the binding of aspirin at the acetylating site in platelet COX-1.24) Therefore, our findings suggest that there would be almost no interaction when aspirin is given before ibuprofen or loxo- profen sodium. By the way, ibuprofen and loxoprofen sodium
significantly decreased serum TXB2 levels (i.e., COX-1 is inhibited), but these NSAIDs did not suppress platelet aggre- gation (Figs. 2A—C). The mechanism is unclear; however, it has been reported that antiplatelet effect of aspirin is seen 25) Fig. 2. ADP-Induced (A) and Collagen-Induced (B) Platelet Aggregation when over 90% of TXA2 production is suppressed. It has and Serum TXB2 Levels (C) in Rats at 6 h after Oral Administration of also been suggested that the interaction between meloxicam Ibuprofen ( , Ibu), Loxoprofen Sodium ( , Lox), Etodolac ( , Eto) Alone, and COX-1 may be weaker than that between aspirin and or CMC-Na ( , Cont.) COX-1.26) Considering them, antiplatelet effect might not be Data points for A and B are means S.E., and the line in C indicates mean value seen without quite a lot of and continuous suppression of (n 4; ∗ p 0.05; † p 0.01). TXA2 production. The maximum inhibition of aggregation in both the aspirin differences between the Asp multiple group and aspirin be- before etodolac (Asp→Eto) and etodolac before aspirin fore etodolac multiple administration (Asp→Eto, multiple) groups (Eto→Asp) was equal to that in the Asp group (Figs. group were observed at 6 and 12 h (Figs. 4C1, C2). 3C1, C2). Serum TXB2 levels in the etodolac group at 6 h In the Asp, Asp→Ibu and Ibu→Asp groups, serum TXB2 were lower than those in the control group; however, they concentrations were 1.44 0.54, 0.45 0.17 and 2.81 0.78 were significantly higher than that in the ibuprofen and loxo- ng/ml at 6 h after administration (p 0.01), which were lower profen sodium groups (Fig. 2C). This suggests that etodolac than those in the control group (44.77 8.91 ng/ml, Fig. 5A). slightly inhibits COX-1, but ibuprofen and loxoprofen
Moreover, serum TXB2 levels in the Asp→Ibu multiple sodium exhibit greater COX-1 inhibitory activity, and the in- groups (1.91 0.86 ng/ml) were also lower than those in the teraction between aspirin and COX-1 may be stronger than multiple control group (36.49 12.71 ng/ml, Fig. 5B). In the that between NSAID and COX-1. There is also a side pocket 27) case of loxoprofen sodium and etodolac, serum TXB2 con- in the hydrophobic channel of COX-2, to which etodolac is centrations in all groups were lower than those in the control thought to primarily bind, but because of its weak inhibition group, regardless of administration order or the use of multi- of COX-1, it has very little influence on the antiplatelet ple doses (Figs. 5A, B). effects of aspirin. In our single-dose study, inhibition by NSAIDs could be DISCUSSION avoided by administering aspirin 2 h before the single NSAID dose. However, NSAIDs taken the day before may The antiplatelet effects of low-dose aspirin are suppressed inhibit the antiplatelet effects of aspirin4) during the actual by ibuprofen, and this interaction is thought to be due to clinical regimen. In our second study with multiple daily ibuprofen interfering with the inhibition of platelet COX- doses under a clinically relevant regimen (aspirin: once a day, 1.4,9) This interaction has not been reported when using NSAIDs: three times a day), the Asp→Ibu multiple and COX-2 selective inhibitors, such as celecoxib and diclofenac Asp→Lox multiple groups did not show reduced platelet sodium.4,23) However, the influence of other NSAIDs on the aggregation (Figs. 4A1, A2, B1, B2). This was in contrast to antiplatelet effects of aspirin remains unclear. We therefore the Asp multiple group, and showed that the inhibitory conducted a study to assess whether NSAIDs interfere with effects of daily low-dose aspirin on platelets are competi- 236 Vol. 34, No. 2
Fig. 3. ADP-Induced (A1, B1, C1) or Collagen-Induced (A2, B2, C2) Platelet Aggregation in Rats after Oral Administration of Aspirin ( ), Aspirin be- fore NSAIDs ( ), or NSAIDs before Aspirin ( ) (n 4; Means S.E.; ∗ p 0.05 versus Aspirin; † p 0.01 versus Aspirin)
Fig. 4. ADP-Induced (A1, B1, C1) or Collagen-Induced (A2, B2, C2) Platelet Aggregation in Rats after Oral Administration of Multiple Daily Doses of Aspirin ( ), Aspirin before NSAIDs ( ), or CMC-Na ( , Cont.) (n 4; Means S.E.; ∗ p 0.05 versus Aspirin; † p 0.01 versus Aspirin) tively inhibited by the prolonged use of multiple daily doses suggest that co-administering aspirin with these NSAIDs on of ibuprofen or loxoprofen sodium, even when aspirin was a daily basis might suppress the antiplatelet effects of aspirin. administered before the first dose of NSAIDs. Our results Serum TXB2 levels in the Asp, Asp→Ibu, Ibu→Asp, and February 2011 237
no influence on platelet aggregation was observed when tak- ing a single dose of aspirin before NSAIDs; however, multi- ple doses did not eliminate this interaction, irrespective of when they were administered. In contrast, etodolac did not interfere with the irreversible inhibition so much. Accord- ingly, if co-administration of NSAIDs with low-dose aspirin is necessary, a COX-2 selective inhibitor, such as etodolac, is recommended.
Acknowledgment We would like to thank Daiichi-Sankyo Co. for kindly providing loxoprofen sodium.
REFERENCES
1) Weiss H. J., Aledort L. M., Lancet, 2, 495—497 (1967). 2) Yasue H., Ogawa H., Tanaka H., Miyazaki S., Hattori R., Saito M., Ishikawa K., Masuda Y., Yamaguchi T., Motomiya T., Tamura Y., Am. J. Cardiol., 83, 1308—1313 (1999). 3) ISIS-2 (Second International Study of Infarct Survival) Collaborative Group, Lancet, 2, 349—360 (1988). 4) Catella-Lawson F., Reilly M. P., Kapoor S. C., Cucchiara A. J., De- marco S., Tournier B., Vyas S. N., Fitzgerald G. A., N. Engl. J. Med., 345, 1809—1817 (2001). 5) Burch J. W., Stanford N., Majerus P. W., J. Clin. Invest., 61, 314—319 Fig. 5. Serum TXB2 Concentrations in Rats at 6 h after Oral Administra- (1978). tion of Aspirin (Asp), Aspirin before NSAIDs, or NSAIDs before Aspirin 6) Greig G. M., Francis D. A., Falgueyret J. P., Ouellet M., Percival M. (A), and Multiple Daily Doses of Aspirin and Aspirin before NSAIDs (B) D., Roy P., Bayly C., Mancini J. A., O’Neill G. P., Mol. Pharmacol., The line indicates mean value (n 4; ∗ p 0.05 versus control; † p 0.01 versus con- 52, 829—838 (1997). trol; ibuprofen: Ibu; loxoprofen sodium: Lox; etodolac: Eto; control: Cont.) 7) Cronberg S., Wallmark E., Soderberg I., Scand. J. Haematol., 33, 155—159 (1984). 8) Sohn S., Kwon K., Pediatr. Cardiol., 29, 153—156 (2008). Asp→Ibu multiple groups were significantly lower than 9) Pathmakanthan S., O’Donovan D. G., Sheehan K. M., Murray F. E., Ir. those in the control group (Figs. 5A, B) For loxoprofen Med. J., 91, 58—60 (1998). 10) Ise Y., Murota Y., Takayama K., Narita M., Suzuki T., Soh S., sodium, a similar trend was seen (Figs. 5A, B). It is thought Katayama S., Hirano M., Yakugaku Zasshi, 123, 613—618 (2003). that ibuprofen and loxoprofen sodium themselves inhibit 11) Tanaka T., Takei M., Fukuta Y., Higashino R., Fukuda Y., Nomura Y., COX-1 activity (Fig. 2C), and they interfered with aspirin Ito S., Kurimoto T., Suzuki Y., Biol. Pharm. Bull., 22, 257—260 acetylation of platelet COX-1 at 6 h after administration. The (1999). antiplatelet effects of aspirin also decreased with pre-admin- 12) Sudo T., Kessen Shiketsushi, 14, 265—269 (2003). 13) Zoja C., Benigni A., Livio M., Bergamelli A., Orisio S., Abbate M., istration and multiple administration of ibuprofen or loxopro- Bertani T., Remuzzi G., Am. J. Pathol., 134, 1027—1038 (1989). fen sodium. In either case, ibuprofen or loxoprofen sodium 14) Adams S. S., Mccullough K. F., Nicholson J. S., Arch. Int. Pharmaco- inhibits considerable platelet COX-1 reversibly, and the bind- dyn., 178, 115—129 (1969). ing of NSAID with COX-1 could be influenced by aspirin.26) 15) Noguchi M., Kimoto A., Gierse J. K., Walker M. C., Zweifel B. S., However, not all bindings of NSAID might be displaced; Nozaki K., Sasamata M., Biol. Pharm. Bull., 28, 2075—2079 (2005). 16) Shi J. M., Lai S.G., Xu C. J., Duan G. L., Duan L. I., Acta Pharmacol. antiplatelet effects could have been suppressed, compared to Sin., 25, 996—999 (2004). aspirin. Our results support the hypothesis that inhibition of 17) Suyama H., Kawamoto M., Gaus S., Yuge O., Brain Res., 1010, 144— COX-1 by NSAIDs, including ibuprofen and loxoprofen 150 (2004). sodium, does not persist throughout the dosing interval. The 18) Inoue K., Fujisawa H., Motonaga A., Inoue Y., Kyoi T., Ueda F., relationship between the inhibition of platelet COX-1 and the Kimura K., Biol. Pharm. Bull., 17, 1577—1583 (1994). 19) Born G. V., Nature (London), 194, 927—929 (1962). suppression of platelet aggregation is not linear. There are 20) Satoh K., Ozaki Y., Thrombosis and Circulation, 12, 345—348 (2004). some routes to activate platelets, besides arachidonic acid- 21) Cattaneo M., Arterioscl. Thromb. Vasc. Biol., 24, 1980—1987 (2004). induced signal pathways. Therefore, platelets can be activat- 22) Ohmori T., Yatomi Y., Nonaka T., Kobayashi Y., Madoiwa S., Mimuro ed by some agonists leading to activation and aggregation.26) J., Ozaki Y., Sakata Y., J. Thromb. Haemost., 4, 1271—1278 (2006). 23) Renda G., Tacconelli S., Capone M. L., Sacchetta D., Santarelli F., Sci- In contrast to ibuprofen and loxoprofen sodium, the maxi- ulli M. G., Zimarino M., Grana M., D’Amelio E., Zurro M., Price T. mum inhibition of aggregation in the aspirin before etodolac S., Patrono C., Caterina R. De., Patrignani P., Clin. Pharmacol. Ther., multiple administration group (Asp→Eto, multiple) was 80, 264—274 (2006). almost equal to that in the Asp multiple group (Figs. 4C1, C2). 24) Loll P. J., Picot D., Garavito R. M., Nat. Struct. Biol., 2, 637—643 As the COX-1 inhibitory activity of etodolac is weak, little (1995). 25) Nishikawa M., Tamaru S., Jpn. J. Clin. Pharmacol. Ther., 40, 253— difference was seen between the single- and multiple-dose 260 (2009). studies. Therefore, when low-dose aspirin needs to be admin- 26) Ryn J. V., Kink Eiband M., Kuritsch I., Feifel U., Hanft G., Wallenstein istered with NSAID, a COX-2 selective inhibitor, such as G., Trummlitz G., Pairet M., J. Clin. Pharmacol., 44, 777—784 etodolac,28) seems to be the best selection. (2004). In conclusion, our findings suggest that the inhibitory 27) Luong C., Miller A., Barnett J., Chow J., Ramesha C., Browner M. F., Nat. Struct. Biol., 3, 927—943 (1995). effects of aspirin on platelet aggregation are competitively 28) Kawai S., Nishida S., Kato M., Furumaya Y., Okamoto R., Koshino T., inhibited by both ibuprofen and loxoprofen sodium. Almost Mizushima Y., Eur. J. Pharmacol., 347, 87—94 (1998).