牙周手术后使用依托考昔和塞来昔布预防疼痛的一个双盲，平行组，安慰剂对照，随机临床试验
The Use of Etoricoxib and Celecoxib for Pain Prevention After Periodontal Surgery A Double-Masked, Parallel-Group, Placebo-Controlled,Randomized Clinical Trial
Background:Postoperative pain is an adverse effect of peri-odontal surgeries and may therefore be prevented or mini-mized. This study was conducted to evaluate the clinicalefficacy of two selective cyclooxygenase-2 inhibitors, cele-coxib and etoricoxib, on pain prevention after periodontal sur-gery.
Methods:For this double-masked, parallel-group, placebo-controlled, and randomized clinical trial, 56 open-flap debride-ment surgeries were performed. The groups received threedifferent protocols 1 hour before surgery: 1) 200 mg celecoxib(and another 200 mg 12 hours after the first dose); 2) 120 mgetoricoxib; or 3) placebo. Pain intensity and discomfort wereassessed up to 2 days after surgery using the visual analogscale and the four-point verbal rating scale, respectively.Patients were instructed to take 750 mg acetaminophen asa rescue medication if necessary.
Results:Pain intensity levels in the etoricoxib group werelower than in the placebo group at the 2-, 3-, 4-, 5-, 6-, and7-hour periods after surgery (Kruskal-Wallis test;P<0.05).There was no statistically significant difference between cele-coxib and etoricoxib. Discomfort in the celecoxib group wassignificantly lower than in the placebo group only at the3-hour period (P=0.03). Rescue medication intake was signif-icantly less frequent in the etoricoxib group than in the placeboand celecoxib groups (analysis of variance;P=0.009).
Conclusion:It was concluded that a single etoricoxib dose isnot superior to two split doses of celecoxib when used for painprevention after open-flap debridement surgery.J Periodontol2011;82:1238-1244.
KEY WORDS Analgesia; celecoxib; etoricoxib; pain measurement; pain,postoperative; preanesthetic medication.
Periodontal therapy may cause a cer-tain degree of pain and discomfortto some patients.（见参考文献1）In two studies,>76% of the patients presented anyextent of pain after scaling and rootplaning, and the mean duration for mod-erate to intense pain was 6.1 hours.（见参考文献1,2）Periodontal surgery may also cause painand discomfort many times greater thanthat found after scaling and root planingalone.（见参考文献3）A total of 79% of the patientsrelated pain after open-flap debridementsurgery, 89% after gingivectomy, and93% after open-flap surgery with osseousresection. Such factors as age; surgicalmodality; surgery location, extent andduration; and anxiety may influence painperception.（见参考文献2
Non-steroidal anti-inflammatory drugs(NSAIDs) can be used for pain preventionand relief.（见参考文献4They act by suppressing cyclo-oxygenaseenzyme(COX),andasaconse-quence, the production of prostaglandins,prostacyclins, and thromboxanes is re-duced. By activating phagocyte cells, thenumber of proinflammatory cytokinesafter tissue incision is increased. Thecytokines and chemokines, extracellular to a reduction of the patient’s pain threshold.（见参考文献5At leasttwo isoforms of COX have been identified: COX-1 ispredominantly constitutive and expressed in differenttissues, such as stomach, intestine, kidneys, andplatelets, and also presents a protective gastrointesti-nal role; and COX-2, mainly produced after traumaand inductive for prostaglandins involved in inflam-matory response and pain mediation.（见参考文献4
In the 1990s, NSAIDs that selectively inhibitedCOX-2 were introduced in the market, with the pur-pose of minimizing the gastrointestinal adverseeffects related to conventional NSAIDs. These medi-cations brought some benefits, especially for thosepatients at risk for upper gastrointestinal bleeding orpeptic ulcer history.（见参考文献6In addition, these COX-2–selec-tive NSAIDs, generically known as “coxibs,” do notinhibit platelet aggregation like other conventionalNSAIDs, and can therefore be administered beforesurgery to prevent the initiation of the arachidonicacid cascade without the increased risks for periop-erative and postoperative bleeding.（见参考文献7Another advan-tage of the coxibs is their long action duration,avoiding frequent reingestion of the drug and facili-tating the patient’s compliance to treatment.（见参考文献8
Celecoxib was the first COX-2–selective NSAIDapproved by the Food and Drug Administration. It is in-dicated for the treatment of rheumatoid arthritis, osteo-arthritis, acute pain, and primary dysmenorrhea inadults.9This medication presents a specificity reasonfor COX-2 (COX-2/COX-1) equal to 30; eliminationhalf-life of 12 hours (which justifies its use twice a day);and a number needed to treat (NNT) of 4.2 for a 200-mgdose (i.e., 1 in 4.2 patients who take the drug mayobtain at least 50% pain relief after 4 to 6 hours).（见参考文献8,10
However, etoricoxib is considered to be a second-generation coxib. It is indicated for acute pain reliefin some countries, but its chronic use is not approvedin the United States.9It is >10 times more specific forCOX-2 than celecoxib, with a twice higher eliminationhalf-life, and its NNT is only 1.6 for dental proceduresin a 120-mg dose.（见参考文献8,11
We have previously demonstrated that two splitdoses of celecoxib and one preoperative dose of etor-icoxib, are superior to placebo for pain preventionafter periodontal surgery.12,13However, to the bestof our knowledge, the clinical comparison of thesetwo medication protocols for prevention of periodon-tal surgery pain is not available. Therefore, the pur-pose of this study is to compare the efficacy ofcelecoxib and etoricoxib on pain prevention afteropen-flap debridement surgery.
MATERIALS AND METHODS
Study Design
Sixty patients aged 18 to 56 years were selected at theDepartment of Periodontology, Ponta Grossa StateUniversity, Ponta Grossa, Parana′, Brazil, in 2008and 2009, to take part in this prospective, double-masked, parallel-group, placebo-controlled random-ized clinical trial. All patients had moderate or severechronic periodontitis（见参考文献14with clinical signs of inflam-mation (bleeding on probing or suppuration) afternon-surgical periodontal therapy and were scheduledfor open-flap debridement surgery on ≥1 quadrant(three to five teeth, maxilla or mandible). The follow-ing were excluded from the study: patients with a his-tory of systemic disease, such as diabetes mellitus,hypertension, or gastric ulcer; females who werepregnant or lactating; and patients allergic to anyof the formulations used in the study, using anal-gesics or anti-inflammatory drugs, or at risk for in-fective endocarditis. The nature of the study waspreviously explained to each patient, who signedan informed consent form approved by the Uni-versity’s Institutional Ethical Committee on HumanResearch.
Because stress and dental anxiety vary from pa-tient to patient and may influence pain perception,the State-Trait Anxiety Inventory（见参考文献15was applied atthe beginning of the study. Each patient was randomlyassigned to receive a different medication protocol:group 1 received celecoxib, 200 mg,1hourbe-fore surgery and another 200-mg dose 12 hoursafter the first dose; group 2 received etoricoxib,120 mg,§1 hour before surgery; and group 3 receivedplacebo 1 hour before surgery. For each procedure,a mucoperiosteal flap was raised under local anes-thesia (2% mepivacaine with 1:100,000 epinephrine)for complete scaling and root planing using Graceycurets and ultrasonic instrumentation. Patients wereinstructed to complete a pain diary every hour forthe first 8 hours after surgery, and three times a dayon the following day. The visual analog scale (VAS),which consists of a 100-mm line anchored by twoextremes (“no pain” and “pain as bad as it couldbe”), was used. Patients were asked to make a markon the line at the point that represents their levelof perceived pain.16Discomfort was assessed by afour-point verbal rating scale (VRS-4), which consistsof four options: 1) no discomfort, 2) some transitorydiscomfort, 3) persistent discomfort, or 4) pain.（见参考文献16For ethical reasons, all participants received rescuemedication (750 mg acetaminophen) and were in-structed to take one pill, if necessary, in case of pain,waiting?6 hours between intakes and writing in thediary each time the medication was used.
Statistical Analysis
The power calculation was performed using datapreviously published by the authors.12When the sample size in each of the three groups was 20,a two-sided test would have 86% power at an effectsize of 0.20 and a 0.05 level. These calculationswere made using software.
Once the pain scores did not show normal distribu-tion and homogeneity of variances, the Kruskal-Wallis non-parametric test was used to determinethe difference among the groups within each hourfor the VAS. The Kruskal-Wallis test was also usedto evaluate the ordinal categoric data of VRS-4.In case of statistical differences, the Mann-WhitneyU(Wilcoxon rank-sum test) test was used for pairwisecomparisons. One-way analysis of variance was usedto compare the State-Anxiety, Trait-Anxiety, andnumber of rescue medication pills required by eachgroup. Thex2test was used to compare the numberof patients requiring rescue medication in each group.For all tests, thePvalue was set at 0.05 and software was used.
RESULTS
Fifty-six patients (26 males and 30 females; 20 to 56years old; mean age: 38–8 years) completed thestudy. Two patients did not return after surgery. An-other two patients did not fulfill the pain diary properlyand their data were discarded. The mean number ofteeth involved in the surgeries was 3.8 teeth and sur-gery duration mean was 54 minutes.
The group treated with etoricoxib showed signifi-cantly less pain than the placebo group at 2-, 3-, 4-,5-, 6-, and 7-hour periods. However, there were nostatistically significant differences between etoricoxiband celecoxib. Celecoxib showed significantly lowerpain levels than placebo only at the 3-hour period.No statistically significant differences were foundon the second day. The mean, standard deviation(SD), and median of pain intensity for the celecoxib,etoricoxib, and placebo groups at each time periodare shown in Table 1.
However, discomfort was not different between theetoricoxib and placebo groups. The only statisticallysignificant difference in discomfort was observedbetween the celecoxib and placebo groups at hour 3(P = 0.03). The distribution (percent) of patientswith “no discomfort,” “some transitory discomfort,”“persistent discomfort,” or “pain” for the celecoxib,etoricoxib, and placebo groups at each time periodon the first postoperative day is shown in Figure 1.
Seventy-five percent of the patients in the placebogroup required rescue medication. In the test groups,55% of the patients who used celecoxib and 12% ofthose who received etoricoxib needed rescue medi-cation. This difference was statistically significant(P=0.0005). Furthermore, rescue medication intakewas significantly less frequent for the etoricoxibgroup than for the placebo and celecoxib groups (P=0.009) (Fig. 2). There was no statistically significant difference among the groups concerning the psycho-metric instruments (State-Trait Anxiety Inventory)(Table 2).
No adverse side effects were reported for any med-ication. The patients’ most frequent postoperativecomplaint was related to dentin hypersensitivity,which was treated using a 2% potassium nitrate and2% sodium fluoride desensitizing agent after sutureremoval.
DISCUSSION
Some patients do not respond well to non-surgical peri-odontal treatment and may therefore be referred to sur-gical procedures, such as open-flap debridementsurgeries. However, pain and discomfort are expectedafter these procedures, and an efficient medicationprotocol may minimize these effects.（见参考文献2,17-21
The open-flap debridement surgery is a pain modeladopted and accepted in the literature because of itsfrequent use in periodontal practice, ease to recruitpatients, and the possibility of a standardized surgicalprocedure.（见参考文献13,21
As we have previously discussed elsewhere,（见参考文献13preemptive analgesia is believed to promote improved clinical results forpain prevention better thantreatment initiated after sur-gery. It is defined as a treat-ment that prevents theestablishment of sensitiza-tion caused by incisionaland inflammatory injuries; itstarts before incision andcovers both the period of sur-gery and the initial postop-erative period.（见参考文献22However,clinical trials and systematicreviews have failed to demon-strate this superiority.（见参考文献20,23-25The use of short-durationdrugs, such as ibuprofen anddiclofenac, may be an expla-nation. These drugs coverthe period of surgery, but notthe immediate postoperativeperiod, when generation ofnociceptive stimuli may beintense for 12 to 48 hours.（见参考文献22Postoperative pain generallylasts for 24 hours, withgreater intensity at 6 to 8hours.（见参考文献26Hence, an efficientanalgesia protocol shouldcover at least the first day af-ter surgery. Waiting for paininitiation after surgery to med-icate produces unnecessary discomfort and may re-duce the efficacy of any posterior treatment.（见参考文献27Theadvantages of COX-2 selective drugs versus traditionalNSAIDs include less adverse reactions related to gas-trointestinal problems; absence of platelet aggre-gation inhibition, which may cause perioperativebleeding complications when a preemptive medicationis used; long action duration and elimination half-life;and greater and long-lasting pain relief. In addition, ad-verse effects related to the use of COX-2–selectivedrugs, such as kidney or cardiovascular problems,were only associated with their chronic use.（见参考文献11
Etoricoxib is a novel NSAID, highly selective forCOX-2. It is rapidly absorbed and great plasmaticlevels are reached after 1 hour. Its elimination half-life is 25 hours.（见参考文献10Malmstrom et al.（见参考文献28stated that120 mg should be the minimum dose for the bestanalgesic action of this drug, justifying the doseused in this trial. However, many countries have re-cently withdrawn the 120-mg dose presentationbecause of increased cardiovascular risks. It is im-portant to note that there is no current evidence thata single 120-mg dose may lead to these adverseeffects.
However, celecoxib may also be used in higherdoses than those administered in this study, whichmay enhance its analgesic effect. The 400-mg doseof celecoxib NNT is 2.5, almost half the NNT fora 200-mg dose (NNT=4.2).（见参考文献8However, its indicationfor 12/12 hours reingestion may be difficult for pa-tient compliance to treatment.10This same medica-tion protocol was used in one previous study, wherewe demonstrated that celecoxib was superior to pla-cebo for pain prevention after open-flap debridementsurgery on the first postoperative day.（见参考文献12Interest-ingly, in this study there was no significant differencebetween celecoxib and placebo or between celecoxiband etoricoxib groups at any time period evaluated.Celecoxib’s mechanism of action to prevent or con-trol pain, although similar to that of etoricoxib (they belong to the same family),seems to be less effec-tive.（见参考文献8,11The reason why etoricoxib, but not cele-coxib, is demonstrated to be statistically superiorto placebo may be because of its pharmacologicproperties.（见参考文献10
Several instruments have been proposed to evalu-ate clinical pain intensity. The VAS, used in this study,is frequently used. It has a positive correlation to thenumerical rate scale-101 (pain level rated by 0 to100 basis), and it is equally sensitive in assessingacute pain after surgery. The VRS-4 scale can alsobe used, but it is considered to be less effective thanthe numeric scales.（见参考文献29
Postoperative pain after periodontal surgery gen-erally lasts for 24 hours,（见参考文献17,26which influenced thedecision to compare two different clinically testedand validated protocols that could offer a 24-hourpain coverage for the patient. It was not the purposeof this study to compare different drug dosages, butrather the protocols. In the case of celecoxib, but notfor etoricoxib, another postoperative dose is neces-sary for effective 24-hour pain control. Because thiswas a parallel-design and not a crossover study, theabsence of a placebo pill 12 hours after the preoper-ative etoricoxib intake definitively did not influencethe results of this study. Besides, the patients andthe surgeon were masked to the treatment, and therewas no contact or communication among the pa-tients enrolled in the study.
Etoricoxib was superior to placebo in preventingpain at the 2-, 3-, 4-, 5-, 6-, and 7-hour periods. Inaddition, the use of etoricoxib promoted significantlyless rescue medication intake (P=0.009). The reasonsome of the data in Table 1 seem to be different (butthere is no statistical difference among them) is thatthe Kruskal-Wallis test does not compare parameters(mean or SD). It assigns ranks for each value and thencompares the groups. However, it had to be used be-cause of the absence of normal distribution andhomogeneity of variances.
Anxiety seems to play an important role among thefactors that may influence postoperative pain and dis-comfort.（见参考文献30Some psychometric instruments aim to as-sess the patient’s level of anxiety at the moment (stateanxiety), or on the daily routine (trait anxiety).（见参考文献15How-ever, the anxiety evaluated in this study did not signif-icantly vary among the three groups (Table 2).
CONCLUSIONS
The adoption of a preoperative medication protocolusing 120-mg etoricoxib may be considered as ef-fective as using two split doses of 200-mg celecoxib for pain prevention after open-flap debridement sur-geries. However, limitations of this study include thesurgical approach adopted and the parallel-groupmethod used. Open-flap debridement surgeries areexpected to generate less pain intensity than othersurgical procedures, such as mucogingival surgeryor bone ressective surgery.（见参考文献18Even the control grouppresented low pain intensity values in this study,which may mask the real influence of the medicationon pain prevention. In addition, pain is highly sub-jective and may vary among patients. Therefore,crossover design studies seem to be ideal for thatevaluation.（见参考文献13
ACKNOWLEDGMENTS
The authors thank Upper Level Personnel Improve-ment Coordination (CAPES), Brasilia, Brazil, for thescholarship (JPS) that partially funded the study.The authors report no conflicts of interest related tothis study.
REFERENCES
1. Pihlstrom BL, Hargreaves KM, Bouwsma OJ, MyersWR, Goodale MB, Doyle MJ. Pain after periodontalscaling and root planing.J Am Dent Assoc1999;130:801-807.
2. Canakcxi CF, Canakc xi V. Pain experienced by patientsundergoing different periodontal therapies.J Am DentAssoc2007;138:1563-1573.
3. Matthews DC, McCulloch CA. Evaluating patient per-ceptions as short-term outcomes of periodontal treat-ment: A comparison of surgical and non-surgicaltherapy. J Periodontol1993;64:990-997.
4. Jeske AH. COX-2 inhibitors and dental pain control.JGt Houst Dent Soc1999;71:39-40.
5. Czock D, Keller F, Rasche FM, Ha¨ussler U. Pharmaco-kinetics and pharmacodynamics of systemically ad-ministered glucocorticoids.Clin Pharmacokinet2005;44:61-98.
6. Hawkey CJ, Jones JI. Gastrointestinal safety of COX-2specific inhibitors.Gastroenterol Clin North Am2001;30:921-936.
7. Straube S, Derry S, McQuay HJ, Moore RA. Effect ofpreoperative Cox-II-selective NSAIDs (coxibs) onpostoperative outcomes: A systematic review of ran-domized studies.Acta Anaesthesiol Scand2005;49:601-613.
8. Derry S, Barden J, McQuay HJ, Moore RA. Single doseoral celecoxib for acute postoperative pain in adults.Cochrane Database Syst Rev2008;8:CD004233.
9. Khan AA, Dionne RA. The COX-2 inhibitors: Newanalgesic and anti-inflammatory drugs. Dent ClinNorth Am2002;46:679-690.
10. Shi S, Klotz U. Clinical use and pharmacologicalproperties of selective COX-2 inhibitors.Eur J ClinPharmacol2008;64:233-252.
11. Clarke R, Derry S, Moore RA, McQuay HJ. Singledose oral etoricoxib for acute postoperative pain inadults. Cochrane Database Syst Rev 2009;15:CD004309.
12. Pilatti GL, Andre′ dos Santos F, Bianchi A, CavassimR, Tozetto CW. The use of celecoxib and dexameth-asone for the prevention and control of postoperativepain after periodontal surgery.J Periodontol2006;77:1809-1814.
13. Steffens JP, Santos FA, Sartori R, Pilatti GL. Pre-emptive dexamethasone and etoricoxib for pain anddiscomfort prevention after periodontal surgery: Adouble-masked, crossover, controlled clinical trial.JPeriodontol2010;81:1153-1160.
14. Armitage GC. Development of a classification systemfor periodontal diseases and conditions. Ann Peri-odontol1999;4:1-6.
15. Spielberger CD, Gorsuch RL, Lushene R.Manual forthe State-Trait Anxiety Inventory. Palo Alto: Consul-ting Psychologists Press; 1970:3-14.
16. Jensen MP, Karoly P, Braver S. The measurement ofclinical pain intensity: A comparison of six methods.Pain1986;27:117-126.
17. Gallardo F, Rossi E. Analgesic efficacy of flurbipro-fen as compared to acetaminophen and placeboafter periodontal surgery. JPeriodontol1990;61:224-227.
18. Curtis JW Jr., McLain JB, Hutchinson RA. The in-cidence and severity of complications and pain fol-lowing periodontal surgery. J Periodontol 1985;56:597-601.
19. Minutello JS, Newell DH, Thrash WJ, Terezhalmy GT.Evaluation of preoperative diflunisal for postoperativepain following periodontal surgery.J Periodontol1988;59:390-393.
20. Vogel RI, Desjardins PJ, Major KV. Comparison ofpresurgical and immediate postsurgical ibuprofen onpostoperative periodontal pain.J Periodontol1992;63:914-918.
21. Rashwan WA. The efficacy of acetaminophen-caffeine compared to Ibuprofen in the control ofpostoperative pain after periodontal surgery: Acrossover pilot study.JPeriodontol2009;80:945-952.
22. Kissin I. Preemptive analgesia.Anesthesiology2000;93:1138-1143.
23. Jung YS, Kim MK, Um YJ, Park HS, Lee EW, KangJW. The effects on postoperative oral surgery painby varying NSAID administration times: Compari-son on effect of preemptive analgesia. Oral SurgOral MedOralPatholOralRadiolEndod2005;100:559-563.
24. Moiniche S, Kehlet H, Dahl JB. A qualitative andquantitative systematic review of preemptive anal-gesia for postoperative pain relief: The role oftiming of analgesia. Anesthesiology 2002;96:725-741.
25. Ong CK, Lirk P, Seymour RA, Jenkins BJ. The efficacyof preemptive analgesia for acute postoperative painmanagement: A meta-analysis.Anesth Analg2005;100:757-773.
26. Seymour RA, Meechan JG, Blair GS. An investigationinto post-operative pain after third molar surgeryunder local analgesia.Br J Oral Maxillofac Surg1985;23:410-418.
27. Desjardins PJ, Shu VS, Recker DP, Verburg KM, WoolfCJ. A single preoperative oral dose of valdecoxib,a new cyclooxygenase-2 specific inhibitor, relievespost-oral surgery or bunionectomy pain.Anesthesiol-ogy2002;97:565-573.
28. Malmstrom K, Sapre A, Couglin H, Agrawal NG,Mazenko RS, Fricke JR Jr. Etoricoxib in acutepain associated with dental surgery: A randomized,double-blind, placebo- and active comparator-con-trolled dose-ranging study. Clin Ther2004;26:667-679.
29. Breivik H, Borchgrevink PC, Allen SM, et al. Assess-ment of pain.Br J Anaesth2008;101:17-24.
30. Scott DS, Hirschman R. Psychological aspects ofdental anxiety in adults.J Am Dent Assoc1982;104:27-31.Correspondence: Professor Gibson Luiz Pilatti, PontaGrossa State University, Department of Periodontology,4748 Carlos Cavalcanti Av., 84030-900, Ponta Grossa,PR, Brazil. Fax: 55-42-32203741; e-mail gibsonpilatti@Gmail.com.
Submitted November 12, 2010; accepted for publication
December 27, 2010.
