Pii: s0010-7824(01)00322-5

Anti-implantation effects of indomethacin and celecoxib in rats Nongluck Sookvanichsilp*, Pawitra Pulbutr Department of Pharmacology, Faculty of Pharmacy, Mahidol University, Bangkok, Thailand Abstract
Pregnant Wistar rats were used to investigate the anti-implantation effect of cyclooxygenase (COX) inhibitors, indomethacin (nonse- lective COX-1/COX-2 inhibitor) and celecoxib (specific COX-2 inhibitor). Indomethacin at doses of 2.5 and 5 mg/kg/day and celecoxib atdoses of 40, 80, and 160 mg/kg/day were orally administered once daily to each group (n ϭ 8) on Days 3–5 of pregnancy (Day 1 ϭ spermdetection). Indomethacin and celecoxib at anti-implantation dosages were further investigated for the effects on changes in endometrialvascular permeability in pregnant rats and uterine decidualization in pseudopregnant rats. The results demonstrated that indomethacin at adose of 5 mg/kg/day as well as celecoxib at doses of 80 and 160 mg/kg/day could significantly reduce the proportion of pregnant rats. Atthe anti-implantation dosages, they exhibited no significant effect on proportion of rats with blue dye sites in the endometrial vascularpermeability study, but they could significantly reduce the uterine decidualization. From these findings, the anti-implantation effect of thetwo COX inhibitors may principally be from decidualization defects, and COX inhibitors should, therefore, be used with caution inchildbearing age women. On the other hand, specific COX-2 inhibitors with their good gastric safety profile may have a potential role innonhormonal postcoital contraception. 2002 Elsevier Science Inc. All rights reserved.
Keywords: Anti-implantation; Celecoxib; COX inhibitors; Indomethacin; Postcoital contraception 1. Introduction
[4 –10]. Cyclooxygenase (COX) inhibitors, nonsteroidal an-tiinflammatory drugs (NSAIDs) used worldwide, inhibit Synchronized development of the embryo to the blasto- prostaglandin synthesis. COX exists in two isoforms, cyst stage by the time of arrival in the uterine cavity and COX-1 and COX-2. COX-1 is a constitutive enzyme differentiation of the uterus to the receptive state are essen- present in many tissues, including the stomach, kidneys, and tial to the implantation process [1]. The endometrial re- platelets. This isoform is responsible for the production of sponse to the embryo implantation varies in different spe- prostaglandins involved in general “housekeeping” activi- cies; however, one phenomenon appears to be constant: the ties, for example, maintenance of gastric mucosal integrity, implantation of the blastocyst is preceded by an increase in vascular hemostasis, and regulation of renal blood flow.
uterine vascular permeability leading to stromal edema [2].
COX-2 is an inducible enzyme and mainly produced at the The stromal cells, under the influence of progesterone, are sites of inflammation [11–13]. Indomethacin, a nonselective transformed into a complex tissue known as decidua (de- COX-1/COX-2 inhibitor, and celecoxib, a specific COX-2 cidualization) [1]. In rats, the induction of decidualization inhibitor with a good gastric safety profile, are used for by nontraumatic stimuli, such as intrauterine oil injection, anti-inflammatory and analgesic effects to treat various can occur on Day 5 of pseudopregnancy or after exposure of rheumatologic disorders. In addition, celecoxib is used to the uterus to the appropriate combination and concentrationof progesterone and estrogen [3].
reduce the number of polyps in patients with familial ad- Considerable evidence exists indicating that prostaglan- enomatous polyposis [13–15]. The adverse effect of the two dins are important in blastocyst implantation, endometrial COX inhibitors on blastocyst implantation was investigated vascular permeability, and uterine decidualization process in the present study. In addition, because both endometrialvascular permeability and uterine decidualization are crucialevents for implantation, the effects of the two COX inhib-itors on these changes were also the aims of the present * Corresponding author. Tel.: ϩ662-644-8677; fax: ϩ662-247-4696.
E-mail address: [email protected] (N. Sookvanichsilp).
0010-7824/02/$ – see front matter 2002 Elsevier Science Inc. All rights reserved.
PII: S 0 0 1 0 - 7 8 2 4 ( 0 1 ) 0 0 3 2 2 - 5 N. Sookvanichsilp, P. Pulbutr / Contraception 65 (2002) 373–378 Table 1Effect of indomethacin and celecoxib on implantation in rats a Day 1 ϭ the day of finding sperm in the vagina.
b % Implantation sites compared with the control group, calculated from the mean values.
* p Ͻ 0.05, ** p Ͻ 0.01, *** p Ͻ 0.001 compared with the control group.
2. Materials and methods
2.2. Effect of indomethacin and celecoxib on endometrialvascular permeability 2.1. Effect of indomethacin and celecoxib on implantation For the second experiment, vaginal smears were taken Female Wistar rats, weighing 160 –200 g, were obtained and examined daily in the same manner as mentioned from the National Animal Center, Mahidol University, Sa- above. The pregnant rats were divided into four groups, laya. They were housed in an air-conditioned room with free consisting of eight rats per group, as indicated in Table 2.
access to a commercial pellet diet and tap water. All rats Indomethacin at an oral dose of 5 mg/kg/day as well as were allowed to acclimatize for one week before any ex- celecoxib at oral doses of 80 and 160 mg/kg/day, doses that periments were started. The experiment protocol was ap- caused a significant reduction in the proportion of pregnant proved by the Animal Ethics Committee, Faculty of Phar- rats in the first experiment, were given to rats once daily on Days 3–5 of pregnancy. To determine endometrial vascular Vaginal smears were taken daily and examined micro- permeability, these rats were given an intravenous injection scopically. Female rats were caged individually with proven of 0.5 mL of 0.5% Evans blue dye in 0.9% sodium chloride fertile Wistar males in the afternoon of the proestrous stage.
solution via a tail vein [16] in the morning of Day 6 of Mating was confirmed by the presence of spermatozoa in pregnancy. Fifteen minutes after the Evans blue dye injec- the vaginal smear on the next morning (Day 1 of pregnan- tion, the rats were killed. The uteri were examined for the cy). The pregnant rats were divided into six groups, con- presence of blue dye sites, which is indicative of increased sisting of eight rats per group, as indicated in Table 1.
endometrial vascular permeability. The number of blue dye Indomethacin (Indocid 25 mg capsule, Merck Sharp & Dohme) at doses of 2.5 and 5 mg/kg/day and celecoxib(Celebrex 200 mg capsule, Searle) at doses of 40, 80, and 2.3. Effect of indomethacin and celecoxib on uterine 160 mg/kg/day were administered by oral intubation once daily to each group on Days 3–5 of pregnancy. The ratswere sacrificed on Day 8 of pregnancy, and the number of For this third experiment, pseudopregnant rats were used. Induction of decidualization was performed in a sim- Table 2Effect of indomethacin and celecoxib on Evans blue dye reaction in rats a Day 1 ϭ the day of finding sperm in the vagina.
b No significant differences in number of rats with blue dye sites between groups.
c % Blue dye sites compared with the control group, calculated from the mean values.
* p Ͻ 0.05 compared with the control group.
N. Sookvanichsilp, P. Pulbutr / Contraception 65 (2002) 373–378 Table 3Effect of indomethacin and celecoxib on uterine decidualization in rats Uterine wet weight (mg)/100 g body weight, mean Ϯ SEM n ϭ 8 a % Uterine weight compared with the control group, calculated from the mean values.
b No significant differences in mean weights of noninjected horns between groups.
* p Ͻ 0.01, ** p Ͻ 0.001 compared with the control group.
ϩ p Ͻ 0.05 compared with 160 mg/kg/day celecoxib-treated group.
ilar manner to those previously described [17,18]. Female of pregnant rats as well as the number of normal implanta- rats were ovariectomized and allowed one week to recover tion sites, when compared with controls (Table 1). Cele- from the operation. To obtain the state of pseudopregnancy, coxib at doses of 80 and 160 mg/kg/day, given on Days 3–5 the ovariectomized rats were treated by subcutaneous (sc) of pregnancy, also exhibited significant effect on implanta- injections with the hormone regimen of estradiol-17␤ (E ) tion. A complete inhibition of normal implantation, how- and progesterone (P ) at varying doses and duration. The ever, was seen only in 160 mg/kg/day celecoxib-treated rats.
regimen consisted of 0.2 ␮g E in the morning for three Both COX inhibitors at any dosages mentioned produced no days (Days Ϫ2, Ϫ1, and 0 of pseudopregnancy), 0.2 ␮g E gastrointestinal bleeding as evaluated by stool observation.
and 1 mg P in the afternoon of the equivalent of Day 0 of Indomethacin at a dose of 10 mg/kg/day, given once daily pseudopregnancy, 4 mg P in the afternoon of Days 2 and 3, on Days 3–5 of pregnancy, caused a severe gastrointestinal 0.3 ␮g E and 4 mg P in the afternoon of Day 4, and 0.1 ␮g E and 4 mg P in the morning of five following days (Days 5–9 of pseudopregnancy). Decidualization was induced on 3.2. Effect of indomethacin and celecoxib on endometrial Day 5 of pseudopregnancy by artificial deciduogenic stim- ulus, which was a unilateral intrauterine injection of 50 ␮Lof sesame oil at the ovarian end of the uterus. A tight At anti-implantation dosages (i.e., indomethacin at a ligature was tied just caudal to the site of injection to dose of 5 mg/kg/day and celecoxib at doses of 80 and 160 prevent leakage of oil from the uterus. The contralateral mg/kg/day, given on Days 3–5 of pregnancy), the two COX noninjected horn served as a non-pseudopregnant control.
inhibitors produced only a 25% reduction in the proportion On Days 3–5 of pseudopregnancy, the rats were admin- of rats with blue dye sites (Table 2). A slightly significant istered orally once daily indomethacin or celecoxib at anti- difference from controls (p ϭ 0.04) in the mean number of implantation doses as indicated in Table 3. The rats were blue dye sites found in the higher dose celecoxib-treated rats killed on Day 10 of pseudopregnancy, and uteri were then removed and blotted dry before weighing.
3.3. Effect of indomethacin and celecoxib on uterine Quantitative data were expressed as mean Ϯ SEM and Decidualization of the hormone-pretreated uterine horn analyzed by Student’s two-tailed, unpaired t-test by using induced by intrauterine oil injection exhibited about four the Microsoft Excel 97. Fisher’s exact test was used for times greater wet weight than the noninjected uterine horn.
analysis of qualitative data. Statistical significance was set Mean weight of oil-injected (pseudopregnant) horns ob- tained from rats treated with indomethacin and celecoxib atanti-implantation doses (Table 3) given on Days 3–5 ofpseudopregnancy were very significantly lower than con- 3. Results
trols (p Ͻ 0.01 - Ͻ 0.001). Moreover, the weight of pseu-dopregnant horns from 160 mg/kg/day celecoxib group was 3.1. Effect of indomethacin and celecoxib on implantation significantly lower than that from 5 mg/kg/day indometha-cin group. No significant difference was found in the mean Indomethacin at a dose of 5 but not 2.5 mg/kg/day, given weight of the pseudopregnant horns when comparing the on Days 3–5 of pregnancy, significantly reduced the number N. Sookvanichsilp, P. Pulbutr / Contraception 65 (2002) 373–378 4. Discussion
Prostaglandins and COX-2 appeared to play pivotal role in decidualization [7–9,20,21]. Interestingly, in the present COX-derived prostaglandins are critical in female repro- study, indomethacin and celecoxib at anti-implantation duction. COX inhibitors have been reported to interfere with doses administered on Days 3–5 of pseudopregnancy ex- implantation [19 –21]. Significant reduction in the propor- hibited a highly significant decrease in uterine decidualiza- tion of pregnant rats treated with 5 mg/kg/day indometha- tion as assessed by uterine weight, when compared with cin, a nonselective COX-1/COX-2 inhibitor, on Days 3–5 of controls. Moreover, the highest degree of reduction of de- pregnancy in the present study was similar to the results cidualization was found with 160 mg/kg/day celecoxib, the reported by Phillips and Poyser [19] and Poyser [21]. In dose that exhibited a complete inhibition of blastocyst im- their studies, treatment of rats with sc indomethacin injec- plantation. Therefore, from these findings, the anti-implan- tion at a dose of 3 mg/kg twice daily on Days 3– 4 of tation effect of the two COX inhibitors may principally be pregnancy significantly reduced the number of implantation sites. However, there was no significant effect of indometh- Did the anti-implantation effect of indomethacin and acin on the proportion of rats with implantation or the mean celecoxib result from COX-2 inhibition or from COX-1/ number of implantation sites when indomethacin was given COX-2 inhibition? Many studies have indicated the role of sc at a dose of 1 mg/rat twice on Day 5 of pregnancy (the COX-2 in blastocyst implantation [20,30,31]. The role of day of blastocyst implantation); only the mean weight of COX-2 in implantation could also be indicated by the in- implantation sites was significantly less than in controls hibitory effect of DuP697 [20] and celecoxib (the present [16]. This could be due to too late an initiation of drug study). However, both COX-1 and COX-2 are expressed in the uterine epithelium at different times in early pregnancy.
Lim et al. [20] reported that treatment with DuP697, a COX-1 is expressed preferentially in uterine epithelium selective COX-2 inhibitor, on Days 3– 4 of pregnancy in prior to implantation. It is then down-regulated after embryo mice resulted in a dose-dependent inhibition of implanta- attachment, whereas COX-2 is expressed in the luminal tion. The interfering effect of celecoxib, which is a specific epithelium and subepithelial stromal cells at the time since COX-2 inhibitor, on implantation at doses of 80 and 160 attachment [12,32,33]. Overlapping pathways of the COX mg/kg/day on Days 3–5 of pregnancy in rats was also systems in female reproduction may exist so that COX-2 compensation occurs in the absence of COX-1 [34] and Administration of indomethacin and celecoxib at anti- COX-2 inhibition in COX-1(Ϫ/Ϫ) mice induced severe implantation dosages (5 mg/kg/day indomethacin and 80 reproductive failure [20], suggesting a lack of alternative and 160 mg/kg/day celecoxib on Days 3–5 of pregnancy) in sources of prostaglandin synthesis [10]. Whether celecoxib the present study produced only a 25% reduction in the at doses used in the present study (up to 160 mg/kg) could proportion of rats with blue dye sites (indicative of in- inhibit COX-1 was a question. Celecoxib demonstrated creased endometrial vascular permeability) when the inves- 375-fold selectivity for COX-2 over COX-1 [35]. In rats, tigation was performed in the morning of Day 6 of preg- celecoxib treatment decreased the stomach PGE nancy. Previous studies also revealed no significant effect of (COX-1) by approximately 40% at 10 mg/kg, but the de- indomethacin at a dose of 1 mg/rat twice on Day 5 of cline was not dose-related, and no gastrointestinal lesions pregnancy [16] or 3 mg/kg twice daily on Days 3– 4 of were observed at any doses (up to 200 mg/kg) [36,37], pregnancy [19] on the proportion of rats with uterine dye whereas indomethacin completely inhibited PGE synthesis sites or the mean number of dye sites when the investigation in the stomach at a dose of 3 mg/kg [36]. Therefore, at was done in the morning of Day 6 of pregnancy. Even anti-implantation doses in the present study, celecoxib though many authors reported the role of prostaglandins and COX-2 in endometrial vascular permeability [22–24], lack Although COX-1 expression is very low or undetectable of significant effect on endometrial vascular permeability of during the time of implantation [32], recently, Reese et al.
the two COX inhibitors in the present study together with [10] showed that simultaneous inhibition of COX-1 and the results from those other two studies [16,19] may indicate COX-2 produced more effects on early pregnancy events that prostaglandins are not the only critical mediators in- than inhibition of either isoform alone. In the present study, volved in this event. This observation coincides with the indomethacin demonstrated a significant reduction in the previous findings that demonstrated that other mediators proportion of pregnant rats at a dose of 5 mg/kg, which was such as histamine [25,26] and platelet-activating factor [27– about 2–3 times higher than the anti-inflammatory doses 29] were able to produce increased endometrial vascular (approximately 2 mg/kg) [38], whereas a similar degree of permeability. Because the increased endometrial vascular reduction in the proportion of pregnant rats was found with permeability that leads to extravasation, uterine edema, and celecoxib at a dose of 80 mg/kg, which was about eight brings about blastocyst apposition appeared to be nonsig- times higher than doses used for anti-inflammation (approx- nificantly affected by COX inhibitor treatment in the present imately 10 mg/kg) [35,36]. Based on anti-inflammatory study, the critical mechanism(s) of the two COX inhibitors potency (COX-2 inhibition), a celecoxib dose required to should involve some of the other following events.
produce a significant effect on blastocyst implantation was N. Sookvanichsilp, P. Pulbutr / Contraception 65 (2002) 373–378 higher than an indomethacin dose. Poyser [21] also reported [9] Matsumoto H, Ma W, Smalley W, Trzaskos J, Breyer RM, Dey SK.
that NS-398, another COX-2 inhibitor, at a sc dose of 12 Diversification of cyclooxygenase-2-derived prostaglandins in ovula-tion and implantation. Biol Reprod 2001;64:1557– 65.
mg/kg/day on Days 3– 4 of pregnancy, failed to inhibit [10] Reese J, Zhao X, Ma WG, Brown N, Maziasz TJ, Dey SK. Compar- implantation in rats, although this compound produces anti- ative analysis of pharmacologic and/or genetic disruption of cyclo- inflammatory effect at the dose range (0.3–5 mg/kg) almost oxygenase-1 and cyclooxygenase-2 function in female reproduction as potent as indomethacin. It is possible that the dose of in mice. Endocrinology 2001;142:3198 –206.
NS-398 in that study was still too low to inhibit COX-2 [11] Lipsky PE. Specific COX-2 inhibitors in arthritis, oncology, and activity adequately in the rat uterus. Collectively, previous beyond: where is the science headed. J Rheumatol 1999;26(Suppl):25–30.
and present results suggest that inhibition of COX-2 alone [12] Mitchell JA, Warner TD. Cyclo-oxygenase-2: pharmacology, physi- could produce an inhibitory effect on blastocyst implanta- ology, biochemistry, and relevance to NSAID therapy. Br J Pharma- tion, but inhibition of both COX-1 and COX-2 isoforms probably produce greater adverse effect on implantation.
[13] Dannhardt G, Kiefer W. Cyclooxygenase inhibitors— current status In conclusion, the present results support the role of and future prospects. Eur J Med Chem 2001;36:109 –26.
[14] Robert LJ, Morrow JD. Analgesic-antipyretic and anti-inflammatory prostaglandins on blastocyst implantation as demonstrated agents and drugs employed in the treatment of gout. In: Hardman JG, by both COX inhibitors indomethacin and celecoxib inter- Limbird LE, Gillman AG, editors. The pharmacological basis of fering with normal blastocyst implantation. The anti-im- therapeutics. 10th ed. New York: McGraw-Hill, 2001. p. 687–731.
plantation effect of COX inhibitors could principally be the [15] McEvoy GK, editor. AHFS (American Hospital Formulary Service).
result of decidualization defects. Therefore, any NSAIDs, Drug information. Bethesda: American Society of Health-System i.e., conventional NSAIDs that are nonselective COX-1/ Pharmacists 2001:1924 –34,1976 – 83.
[16] Kennedy TG. Evidence for a role of prostaglandins in the initiation of COX-2 inhibitors (such as diclofenac, ibuprofen, naproxen, blastocyst implantation in the rat. Biol Reprod 1977;16:286 –91.
and piroxicam) and newer NSAIDs that are either selective [17] Kennedy TG, Lukash LA. Induction of decidualization in rats by COX-2 inhibitors (such as etodolac, nimesulide, nabum- intrauterine infusion of prostaglandins. Biol Reprod 1982;27:253– 60.
etone, and meloxicam) or specific COX-2 inhibitors (such [18] Orlando-Mathur CE, Kennedy TG. An investigation into the role of as rofecoxib and parecoxib) should be used with caution in neutrophils in decidualization and early pregnancy in the rat. BiolReprod 1993;48:1258 – 65.
childbearing age women. In fact, there are some case reports [19] Phillips CA, Poyser NL. Studies on the involvement of prostaglandins of NSAID-induced reversible infertility in female long-term on implantation in the rat. J Reprod Fertil 1981;62:73– 81.
NSAID users [39,40]. On the other hand, because of the [20] Lim H, Paria BC, Das SK, et al. Multiple female reproductive failure anti-implantation effect of specific COX-2 inhibitors to- in cyclooxygenase-2 deficient mice. Cell 1997;91:197–208.
gether with their good gastric safety profile, a potential role [21] Poyser NL. A comparison of the effects of indomethacin and NS-398 of these compounds in nonhormonal postcoital contracep- (a selective prostaglandin H synthase-2 inhibitor) on implantation inthe rat. Prostaglan Leukot Essent Fatty Acids 1999;61:297–301.
[22] Kennedy TG. Prostaglandins and increased endometrial vascular per- meability resulting from application of an artificial stimulus to theuterus of the rat sensitized for decidual cell reaction. Biol Reprod References
[23] Kennedy TG. Prostaglandin E , adenosine 3Ј:5Ј-cyclic monophos- [1] Goodman HM. Hormonal control of reproduction in the female: phate and changes in endometrial vascular permeability in rat uteri pregnancy and lactation. In: Johnson LR, editor. Essential medical sensitized for decidual cell reaction. Biol Reprod 1983;29:1069 –76.
physiology. 2nd ed. Philadelphia: Lippincott-Raven, 1998. p. 661–73.
[24] Hamilton GS, Kennedy TG. Uterine vascular changes after unilateral [2] Borini A, Asch RH. Factors that affect implantation in the human. In: intrauterine infusion of indomethacin and prostaglandin E to rats Asch RH, Studd JWW, editors. Annual progress in reproductive sensitized for the decidual cell reaction. Biol Reprod 1994;50:757– medicine, 1993. Lancaster: The Parthenon Publishing Group, 1993. p.
[25] Hoos PC, Hoffman LH. Effect of histamine receptor antagonists and [3] Kennedy TG, Ross HE. Temporal- and hormone-dependent changes indomethacin on implantation in the rabbit. Biol Reprod 1983;29: in uterine sensitization for the decidual cell reaction and decidualiza- tion in vitro of rat endometrial stromal cells. J Reprod Fertil 1997; [26] Brandon JM, Wallis RM. Effect of mepyramine, a histamine H -, and burimamide, a histamine H -receptor antagonist, on ovum implanta- [4] Frank GR, Brar AK, Cedars MI, Handwerger S. Prostaglandin E tion in the rat. J Reprod Fertil 1977;50:251– 4.
enhances human endometrial stromal cell differentiation. Endocrinol- [27] Harper MJ. Platelet-activating factor: a paracrine factor in preimplan- tation stages of reproduction? Biol Reprod 1989;40:907–13.
[5] Majerus PW. Prostaglandins: critical roles in pregnancy and colon [28] Ahmed A, Dearn S. The role of platelet-activating factor and its receptor in endometrial receptivity. Adv Exp Med Biol 1997;416: [6] Marions L, Danielsson KG. Expression of cyclo-oxygenase in human endometrium during the implantation period. Mol Hum Reprod 1999; [29] Ahmed A, Dearn S, Shams M, et al. Localization, quantification, and activation of platelet-activating factor receptor in human endome- [7] Lim H, Dey SK. PPAR␦ functions as prostacyclin receptor in blas- trium during the menstrual cycle: PAF stimulates NO, VEGF, and tocyst implantation. Trend Endocrinol Metab 2000;11:137– 42.
FAKpp125. Fed Am Soc Exper Biol J 1998;12:831– 43.
[8] Scherle PA, Ma W, Lim H, Dey SK, Trzaskos JM. Regulation of [30] Jacobs AL, Hwang D, Julian J, Carson DD. Regulated expression of cyclooxygenase-2 induction in the mouse uterus during decidualiza- prostaglandin endoperoxide synthase-2 by uterine stroma. Endocri- tion. An event of early pregnancy. Biol Chem 2000;275:37086 –92.
N. Sookvanichsilp, P. Pulbutr / Contraception 65 (2002) 373–378 [31] FitzGerald GA, Patrono C. The human pharmacology of the coxibs: [36] In vivo selectivity of SC-58635: pharmacological inhibition of pros- inhibitors of cyclooxygenase-2. In: Vane JR, Botting RM, editors.
taglandins produced in inflammed pouch (COX-2) vs stomach London: William Harvey Press, 2001. p. 442– 60.
(COX-1) in rats. G.D. Searle & Co., Document No. BRD97D1834, [32] Chakraborty I, Das SK, Wang J, Dey SK. Development expression of the cyclo-oxygenase-1 and cyclo-oxygenase-2 in the peri-implanta- [37] Penning TD, Talley JJ, Bertenshaw SR, et al. Synthesis and biological tion mouse uterus and their differential regulation by the blastocyst evaluation of the 1,5-diarylpyrazole class of cyclooxygenase-2 inhib- and ovarian steroids. J Mol Endocrinol 1996;16:107–22.
itors: identification of 4-[5-(4-methylphenyl)-3-(trifluoromethyl)-1H- [33] Tsuboi K, Sugimoto Y, Iwane A, Yamamoto K, Yamamoto S, pyrazol-1-yl]benze nesulfonamide (SC-58635, celecoxib). J Med Ichikawa A. Uterine expression of prostaglandin H synthase in late pregnancy and during parturition in prostaglandin F receptor-deficient [38] Sayar K, Melli M. Effect of combination of misoprostol and indo- mice. Endocrinology 2000;141:315–24.
methacin on eicosanoid production in carrageenan-induced air pouch [34] Reese J, Brown N, Paria BC, Morrow J, Dey SK. COX-2 compen- inflammation in rats. Eur J Pharmacol 1999;369:365–71.
sation in the uterus of COX-1 deficient mice during the pre-implan- [39] Calmels C, Dubost JJ, Jasmin-Lebrun C, Sauvezie B. A new case of tation period. Mol Cell Endocrinol 1999;150:23–31.
NSAID-induced infertility. Rev Rheum Engl Ed 1999;66:167– 8.
[35] Muscara MN, McKnight W, Asfaha S, Wallace JL. Wound collagen [40] Mendonca LL, Khamashta MA, Nelson-Piercy C, Hunt BJ, Hughes deposition in rats: effects of an NO-NSAID and a selective COX-2 GR. Non-steroidal anti-inflammatory drugs as a possible cause for inhibitor. Br J Pharmacol 2000;129:681– 6.
reversible infertility. Rheumatology 2000;39:880 –2.

Source: http://research.msu.ac.th/rds/pis_pdf/24-09-2013_04-09.pdf