Share this article on:

A Randomized, Double-Blind Comparison Between Parecoxib Sodium and Propacetamol for Parenteral Postoperative Analgesia After Inguinal Hernia Repair in Adult Patients

Beaussier, M MD*; Weickmans, H MD*; Paugam, C MD*; Lavazais, S MD; Baechle, J P. MD; Goater, P MD§; Buffin, A MD; Loriferne, J F. MD; Perier, J F. MD#; Didelot, J P. MD**; Mosbah, A MD††; Said, R MD††; Lienhart, A MD, PhD*

doi: 10.1213/01.ANE.0000150972.88708.13
Ambulatory Anesthesia: Research Report

The newly injectable cyclooxygenase-2 selective nonsteroidal antiinflammatory drug, parecoxib, has never been compared with propacetamol, a parenteral formulation of acetaminophen. In this prospective, randomized, double-blind, double-dummy study, we randomly assigned 182 patients scheduled for initial inguinal hernia repair under general anesthesia to receive a single injection of 40 mg parecoxib or 2 injections of 2 g propacetamol within the first 12 h after surgery. The study variables were morphine consumption, pain at rest and while coughing, and patient satisfaction throughout the first 12 h postoperatively. For statistical analysis, we used the Student’s t-test, χ2, and covariance analysis. Total morphine consumption did not differ between the two groups. Pain was less intense in the parecoxib group at rest (P = 0.035) but did not differ for pain while coughing. The incidence of side effects was similar. Significantly more patients in the parecoxib group rated their pain management as good or excellent (87% versus 70% in the propacetamol group, P = 0.001). Within the first 12 h after inguinal hernia repair in adult patients, a single injection of parecoxib 40 mg compares favorably with 2 injections of propacetamol 2 g.

IMPLICATIONS: Within the first 12 h after open herniorrhaphy in adults, a single dose of IV parecoxib 40 mg compares favorably to propacetamol 2 g given twice. No difference was observed in side effects but patient satisfaction was better after parecoxib.

*CHU St. Antoine, Paris, France; †CHU Poitiers, France; ‡Clinique de la Sauvegarde, Lyon, France; §HIA Val de Grâce, Paris, France; ∥CHU Annecy, France; ¶Hôpital Ste Camille, Bry sur Marne, France; #CHU Tenon, Paris, France; **CHU Brabois, Vandoeuvre, les Nancy, France; and ††CHU Sahloul, Sousse, Tunisie

This study was support by a grant from Pfizer S.A.

Accepted for publication November 3, 2004.

Address correspondence and reprint requests to M. Beaussier, MD, Département d’Anesthésie Réanimation, Hôpital St-Antoine, 184 rue du Fbg St-Antoine, 75012 Paris, France. Address e-mail to

Non-opioid analgesics are often used as adjuvants to opiates in the postoperative period with the aim of improving analgesia and reducing morphine consumption. Nonsteroidal antiinflammatory drugs (NSAIDs) have become popular because of their powerful analgesic properties and lack of opioid-related side effects (1). Cyclooxygenase (COX)-2 selective inhibitor drugs (coxibs) are thought to have beneficial effects on inflammation and pain with concomitant preservation of homeostatic function and reduction in the incidence of side effects as compared with nonselective NSAIDs (2). Their benefits over placebo in improving pain and early recovery after inguinal herniorrhaphy have been recently demonstrated (3).

Parenteral formulations allow for intraoperative administration and may overcome the problem of bioavailability encountered with oral formulations within the perioperative setting (4). Parecoxib is the first injectable COX-2 selective inhibitor indicated for the treatment of acute postoperative pain (5). Once it is in the circulation, it is rapidly converted into valdecoxib which exhibits strong selective COX-2 inhibition (6,7). The analgesic efficacy of parecoxib has been considered superior to IV morphine 4 mg and as effective as the nonselective IV NSAID ketorolac, after gynecologic laparotomy (8) and orthopedic surgery (9). As compared with ketorolac, parecoxib exhibits a better safety profile. It especially does not impair platelet function (10) and has significantly less gastroduodenal toxicity (11).

Propacetamol is another injectable non-opioid analgesic. It is one of the most widely used drugs in the perioperative setting. After injection, it is rapidly and completely converted by plasma esterase into 50% paracetamol (acetaminophen) (12). Paracetamol may act through an inhibition of a COX-1 variant (13). It has clearly demonstrated analgesic properties and a good general tolerability (1,14).

The parenteral formulations of valdecoxib and acetaminophen have never been compared. The primary objective of this multicenter, prospective, double-blind randomized study was to compare the analgesic efficacy of parecoxib versus propacetamol while evaluating postoperative morphine consumption during the first 12 h after open inguinal hernia repair in adult patients. The a priori hypothesis was the noninferiority of parecoxib over propacetamol with respect to postoperative morphine consumption. Secondary objectives were to compare the pain intensity, tolerability, and treatment assessment by the patients.

Back to Top | Article Outline


This was a multicenter prospective, randomized, double-blind trial, to assess the noninferiority of parecoxib compared with propacetamol. An independent ethical committee approved the protocol design. Twenty investigation centers were recruited.

After written informed consent, patients ASA physical status I or II, aged 18–70 yr, scheduled for initial open inguinal hernia repair with Shouldice or apparented procedures under general anesthesia were randomly assigned to receive either parenteral parecoxib sodium or propacetamol in adjunction to morphine for postoperative pain relief.

The randomization was done centrally as blocks of 4 and with a 2:2 treatment ratio. In each center, treatment allocation was performed on the basis of one complete treatment block.

Exclusion criteria were recurrent herniorrhaphy, emergency conditions, laparoscopic procedure, pregnant or breast-feeding women, alteration in cognitive function, history of alcoholism or drug abuse, inguinal pain at inclusion or chronic pain, intake of narcotic analgesics or NSAIDs within 24 h before the administration of the study drugs (NSAIDs with a long half-life and aspirin at a dosage >325 mg had to be stopped at least 48 h before the procedure), and contraindications to the study drugs.

Back to Top | Article Outline

Anesthetic Procedure

Hydroxyzine (1.5–2 mg/kg) was given as oral premedication 1 h before surgery. In all patients, anesthetic induction was performed with propofol (2–3 mg/kg), sufentanil (0.2–0.3 μg/kg), or fentanyl (2–3 μg/kg) and, if needed, atracurium (0.5 mg/kg). Tracheas were intubated and patients’ lungs mechanically ventilated with a 50:50 mixture of oxygen and nitrous oxide. Petco2 was kept constant between 35 and 40 mm Hg. Maintenance of anesthesia was achieved with volatile anesthetics (isoflurane, desflurane, or sevoflurane) and boluses of sufentanil (0.075–0.15 μg/kg) or fentanyl (0.75–1.5 μg/kg) if needed. The last sufentanil or fentanyl injection had to be administered at least 30 min before the expected end of the surgery.

Back to Top | Article Outline

Analgesic Drug Delivery and Protocol Design

At the beginning of the wound closure (H0), patients received either 40 mg of parecoxib sodium IV or 2 g of propacetamol according to randomization number. Given that both treatments drugs have different presentations and mode of administration, a double-dummy design was applied. Propacetamol was administered by slow infusion over 15 min whereas parecoxib was injected by rapid bolus. Each patient received both an active product and the placebo of the other product. To preserve the double-blind assignment, treatments were prepared by a third person designated by the investigator. This third person did not have any contact with the patient or the investigator over the trial.

At the end of the procedure, the patients were transferred to the recovery room. At arrival, residual paralysis was antagonized with neostigmine and atropine if needed. After tracheal extubation, pain intensity was assessed with a 100-mm visual analog scale (VAS) and simple 4-point verbal rating scale (VRS) with no pain = 0, mild pain = 1, moderate pain = 2, and severe pain = 3. IV morphine boluses (2 mg) were administered and possibly repeated every 5 min as titration, until VRS <2.

After leaving the recovery room, patients could use immediate-release oral morphine tablets of 10 mg (Sevredol®; ASTA Medica, Merignac, France), upon request if VRS ≥2. In case of pain associated with nausea and/or vomiting symptoms, 5 mg of morphine was administered subcutaneously. Supplemental morphine (oral or subcutaneous) was allowed 1 h after the previous administration in case of persistent pain (VRS ≥2). Patients could use up to 4 tablets (or subcutaneous injections) within the first 12 h after the operation.

In the subgroup receiving propacetamol, the active molecule was reinjected at the 6th hour after H0, whereas at the same time, patients in subgroup parecoxib received a placebo in a double-blind manner.

Back to Top | Article Outline

Efficacy Assessments

The primary end-point of the study was the amount of morphine used by the patient during the first 12 h after the first injection of the study treatment. Secondary end-points were pain assessment at rest and while coughing, measured with a 4-point VRS and a 100-mm VAS at 30′, 45′, 1 h, 1.5 h, 2 h, then every hour until the 6th hour, and then every 2 h until the 12th hour after H0. Area under the curve (AUC) of VAS over the first 12 postoperative hours was used to study differences. AUC was calculated using the surface of trapeziums technique (15). Arterial blood pressure, cardiac and respiratory rates, and core temperature were recorded before each study drug administration. Ability to get up and related pain at the 6th hour after H0 were recorded. Pain assessment at the 6th hour was done before study drug reinjection.

Patients’ satisfaction with postoperative pain management was assessed at 6 and 12 h after H0 with a 4-point rating scale (poor = 0, fair = 1, good = 2, and excellent = 3).

Clinical tolerability and occurrence of adverse events were registered throughout the 24 postoperative hours.

According to European recommendations for noninferiority studies (available at, we planned to study morphine consumption in the “modified intent-to-treat” (M-ITT) population (i.e., all randomized patients having received at least one study medication dose) and in the “per protocol population” (PP). All other variables were studied on the M-ITT population.

The sample size calculation was based on morphine consumption in each group within the 12 h after study drug administration. No data concerning efficacy of propacetamol in this specific indication have been published. The morphine-sparing effect of propacetamol is generally considered to be close to 30% (16). Median morphine consumption by self-administration devices within the first 12 h after open inguinal hernia repair have been reported to be 22 mg (17). For sample size calculation, it has been estimated from prestudy pilot patients that the mean total consumption under propacetamol will be close to 15 mg (theoretical decrease by about 30%) with a standard deviation by 6 mg.

A priori hypothesis was the noninferiority of parecoxib regarding morphine consumption. The difference in morphine consumption (propacetamol − parecoxib) would be computed. If the lower bound of the 95% confidence interval of this difference exceeded −3 mg, then parecoxib would be considered noninferior to propacetamol, with an α risk of 5% and a power of 90%. Under this assumption, 70 patients in each group had to be included. Considering the likelihood of cancellation (major protocol deviation, patient withdrawal), it had been decided to include at least 80 patients in both groups.

Categorical data were compared with χ2 test (with Fisher’s exact correction when necessary) or Wilcoxon’s test and quantitative continuous variables with Student’s t-test. AUC of pain intensity within the first 12 h after H0 was compared between the groups using a covariance analysis. Results were presented as mean ± sd, median 95% confidence interval when needed. The threshold for statistical significance was set at 5%.

Back to Top | Article Outline


A total of 182 patients (92 patients in the parecoxib group and 90 in the propacetamol group) were included from March 2002 to April 2003 and received study medications (M-ITT population).

Demographic data and procedure duration are presented in Table 1. Intraoperative opioid consumption and time to tracheal extubation did not differ between groups (Table 1).

Table 1

Table 1

Back to Top | Article Outline

Postoperative Morphine Consumption

Total morphine consumption during the first 12 postoperative hours was computed as the sum of IV titration in the recovery room and oral morphine in the ward. The bioavailability of oral morphine was considered as 0.3. No patients experienced severe postoperative nausea and vomiting precluding the use of oral morphine administration.

Details of the pattern of morphine consumption over the first 12 postoperatively hours are given in Table 2. The 95% confidence interval for the treatment difference was computed at [−0.1; 3.5] with a mean value at 1.7 mg. Because the lower confidence limit exceeded −3 mg, the noninferiority of parecoxib over propacetamol was demonstrated.

Table 2

Table 2

To assess the statistical robustness of this result, we again calculated the difference in morphine consumption on the PP population. Fourteen patients (six in the parecoxib group and eight in the propacetamol group) had protocol deviations and were withdrawn from the M-ITT population by an external expert committee, blinded to the treatment group assignment. Exclusions were accounted for homogeneously between the two groups; they included: missing intake of study treatment, forbidden surgical procedures (tension-free techniques), implantation of prosthetic material, delay in treatment readministration, and forbidden concomitant medication. In the PP population, total morphine consumption was 6.6 ± 5.8 mg in the parecoxib group and 8.9 ± 6.5 mg in the propacetamol group. The 95% confidence interval for the treatment difference was computed at [0.4; 4.2] with a mean value at 2.3 mg. Because the lower confidence limit exceeded −3 mg, the noninferiority of parecoxib was again demonstrated in this population.

Back to Top | Article Outline

Pain Assessment

AUC of VAS score at rest was 199 ± 105 and 262 ± 168, respectively, in the parecoxib and propacetamol groups (P = 0.035) (Fig. 1). Pain induced by coughing did not differ significantly between the two groups, AUC being 372 ± 191 in the parecoxib group and 423 ± 222 in the propacetamol group (Fig. 2). VRS score of pain at rest and while coughing at the 1st, 6th, and 12th hours did not differ statistically between treatment groups. When asked, 74 patients in the parecoxib group and 69 in the propacetamol group were able to get up 6 h after the end of surgery (P = 0.54). Mean pain intensity with VAS score while getting up for the first time at the 6th hour was 33.5 ± 24 and 41.6 ± 25 mm, respectively, in the parecoxib and propacetamol groups (P = 0.16).

Figure 1

Figure 1

Figure 2

Figure 2

Back to Top | Article Outline

Safety Analysis

Two serious adverse events were reported during the first 12 h of evaluation. One patient in the parecoxib group complained of severe pruritus of the arm where the drug had been administered. There were no other signs of allergy. This event resolved after IV antihistaminic administration without any sequelae. Another patient in the propacetamol group experienced inguinoscrotal hematoma, probably related to surgery (no other condition, such as blood coagulation disorder, explained this complication). The most frequently reported minor adverse events were vomiting, urinary retention, nausea, malaise, and bradycardia. In the parecoxib and propacetamol groups, respectively, 27.1% and 25.5% of the patients experienced minor adverse events during the first 24 h. No difference was shown between the two groups. The most frequent minor adverse events are presented in Table 3.

Table 3

Table 3

Back to Top | Article Outline

Patients’ Assessment of the Pain Treatment

Significantly more patients in the parecoxib group than in the propacetamol group considered pain management as good or excellent at the 6th hour (80% versus 67%, P = 0.003) and at the 12th hour (87% versus 70%, P = 0.001) (Table 4).

Table 4

Table 4

Back to Top | Article Outline


We compared the analgesic efficacy and tolerability of the new parenteral COX-2 selective inhibitor, parecoxib, to propacetamol for pain relief after open herniorrhaphy. Overall morphine consumption did not differ between the two groups, but patients in the parecoxib group had better pain relief at rest, less oral morphine consumption in the ward, and better satisfaction concerning pain management.

Failure in pain relief after inguinal hernia repair remains a matter of concern (18). Multimodal analgesia is highly recommended with the aim of improving pain relief while reducing morphine consumption. In this perspective, early parenteral administration of non-opioid analgesics within the intraoperative period could help to control pain within the initial postoperative period. Parenteral formulations have the advantage of allowing for intraoperative administration and providing complete drug bioavailability. In the postoperative period, parenteral formulation of acetaminophen provides a significantly better and longer analgesic effect than the same dosage in oral form (4).

Parecoxib and propacetamol have a short onset of action and complete analgesic action occurs within the first hour after the injection of both drugs (4,8). However, unlike parecoxib, which can be administrated as a bolus, propacetamol has to be administered as a slow infusion over 15 minutes (12). That is why we applied a double-dummy parallel group study. Analgesic efficacy after 2 g of propacetamol, corresponding to 1 g of acetaminophen, lasts approximately 6 hours (19,20) and it is common clinical practice to repeat its administration after this time interval (21). Because analgesic efficacy does not seem to differ between 20 mg and 40 mg of parecoxib (22), the larger dose is longer-lasting and a single injection has been considered to be efficient within the 12-hour period of evaluation (23).

Many studies have found nonselective NSAIDs to be superior to paracetamol (acetaminophen) to relieve pain, especially after minor surgical procedures, but the magnitude of the difference is small (24). Concerning parenteral administration, two previous studies found that propacetamol exhibited a similar analgesic effect compared with the nonselective NSAID ketorolac (25,26). COX-2 selective NSAIDs rofecoxib or celecoxib have been found to provide greater analgesic efficacy than paracetamol when given orally (27,28). In the present study, parecoxib and propacetamol exhibited similar analgesic properties, although it is noteworthy that parecoxib was superior in alleviating pain relief at rest and in providing better patient satisfaction.

The main criteria of analgesic efficacy was total morphine consumption, including IV titration and oral doses. After leaving the recovery room, patients could take oral morphine tablets when VRS was >1. The validity and safety of such patient-controlled oral analgesia have been demonstrated (29). Concerns raised about the use of oral morphine administration in the postoperative period have been mostly related to the pharmacokinetics of sustained-release formulations or to the intestinal ileus that occurs after digestive surgery (30,31). After inguinal hernia repair, gastrointestinal function is not impaired by the surgery and patients are permitted to drink very early in the immediate postoperative period. In the present study, we did not observe nausea and vomiting precluding the use of the oral route.

For the calculation of total morphine consumption, the bioavailability of oral morphine has been assumed to be 0.3, as previously shown in healthy volunteers or cancer patients (32). However, morphine’s bioavailability is probably reduced by 20% in the immediate postoperative period (29,30). In the present study, whatever the unknown morphine systemic concentrations, consumption of oral morphine tablets was significantly less in the parecoxib group than in the propacetamol group. Significantly fewer patients in the parecoxib group needed morphine pills in the ward. This difference was significant only during the first six hours after surgery. After this period, pain intensity was very low and oral morphine demand was small in both groups.

Pain scores remained low with minor differences between groups. This was probably the result of study design, allowing patients to take the morphine they needed for pain relief. Because a single injection of parecoxib compared favorably to propacetamol given twice during the 12-hour evaluation period, the longer duration of action of parecoxib can be confirmed and considered as a notable advantage.

In two recent placebo-controlled studies, COX-2 inhibitors, when started preoperatively, hastened and improved the quality of recovery (3,33). In the current study, the time course of recovery was not measured and the incidence of side effects did not differ after parecoxib or propacetamol administration when given at the completion of surgery. However, administration of parecoxib was accompanied by a significant enhancement of patients’ satisfaction concerning their pain management as compared with propacetamol. This result was unexpected and out of proportion with the weak analgesic benefit of parecoxib and the similar occurrence of opioid-related side effects between groups. Because COXs are involved in the regulation of numerous central nervous system processes (34), it is possible that this satisfaction improvement may have been partly related to a direct psychologically beneficial action of NSAIDs. It is noteworthy that previous comparisons between coxibs and acetaminophen always led to the same conclusion (27,28).

Several limitations of the study deserve discussion. The lack of a placebo group does not permit conclusions about the interest in using non-opioid analgesics after inguinal hernia repair under general anesthesia. Even if the analgesic benefit of both drugs over placebo has been demonstrated in other surgical models, it is difficult to know whether propacetamol was less effective than parecoxib or simply ineffective in our chosen population. The limited period of evaluation was not long enough to determine the potential differences between the two drugs for complication rates and in clinically relevant outcomes, such as resumption of physical activities. In this study, we did not measure the time course of postoperative recovery. It should be noted that the marginal differences between the two drugs are unlikely to translate into clinically relevant variables of recovery.

In conclusion, within the first 12 hours after inguinal hernia repair in adult patients, a single parenteral injection of parecoxib 40 mg compares favorably with 2 injections of propacetamol 2 g.

The authors thank Michel Bunodière, MD for helpful suggestions in the study design and Maëva Deniaud (Soladis Inc.) for assistance in statistical analysis.

Back to Top | Article Outline


1. White PF. The role of non-opioid analgesic techniques in the management of pain after ambulatory surgery. Anesth Analg 2002;94:577–85.
2. Fitzgerald GA, Patrono C. The coxibs, selective inhibitors of cyclooxygenase-2. N Engl J Med 2001;345:433–42.
3. Ma H, Tang J, White PF, et al. Perioperative rofecoxib improves early recovery after outpatient herniorrhaphy. Anesth Analg 2004;98:970–5.
4. Jarde O, Boccard E. Parenteral versus oral route increases paracetamol efficacy. Clin Drug Invest 1997;14:474–81.
5. Cheer S, Goa K. Parecoxib (parecoxib sodium). Drugs 2001;61:1133–41.
6. Talley JJ, Bertenshaw SR, Carter JS, et al. N-[[(5-methyl-3-phenylisocazol-4-Yl)-phenyl]sulfonyl] propanamide, sodium salt, parecoxib sodium: a potent selective inhibitor of Cox-2 for parenteral administration. J Med Chem 2000;43:1661–3.
7. Fricke J, Varkalis J, Zwillich S, et al. Valdecoxib is more efficacious than rofecoxib in relieving pain associated with oral surgery. Am J Ther 2002;9:89–97.
8. Barton S, Langeland F, Snabes M, et al. Efficacy and safety of intravenous parecoxib sodium in relieving acute postoperative pain following gynecologic laparotomy surgery. Anesthesiology 2002;96:306–14.
9. Rasmusen G, Steckner K, Hogue C, et al. Intravenous parecoxib sodium for acute pain after orthopedic knee surgery. Am J Orthop 2002;31:336–43.
10. Noveck R, Laurent A, Kuss M, et al. Parecoxib sodium does not impair platelet function in healthy elderly and non-elderly individuals: two randomised, controlled trials. Clin Drug Invest 2001;21:465–76.
11. Stoltz R, Harris S, Kuss M, et al. Upper GI mucosal effects of parecoxib sodium in healthy elderly subjects. Am J Gastroenterol 2002;97:65–71.
12. Depre M, van Hecken A, Verbesselt R, et al. Tolerance and pharmacokinetic of propacetamol, a paracetamol formulation for intravenous use. Fundam Clin Pharmacol 1992;6:259–62.
13. Chandrasekharan NV, Dai H, Roos KL, et al. COX 3, a cyclooxygenase variant inhibited by acetaminophen and other analgesic/antipyretic drugs: cloning, structure and expression. Proc Natl Acad Sci USA 2002;99:13926–31.
14. Dahl V, Raeder JC. Non-opioid postoperative analgesia. Acta Anaesthesiol Scand 2000;44:1191–203.
15. Press WH, Flannery BP, Teukolsky SA, Vetterling WT. Numerical recipes in Fortran 77: the art of scientific computing. 2nd ed. Cambridge: Cambridge University Press, 1993.
16. Delbos A, Boccard E. Morphine sparing effect of propacetamol after ligamentoplasty of the knee. J Pain Symptom Manage 1995;10:279–86.
17. Harrison C, Morris S, Harvey J. Effect of ilioinguinal and iliohypogastric nerve block and wound infiltration with 0.5% bupivacaine on postoperative pain after hernia repair. Br J Anaesth 1994;72:691–3.
18. Rawal N, Hylander J, Nydahl PA, et al. Survey of postoperative analgesia following ambulatory surgery. Acta Anaesthesiol Scand 1997;41:1017–22.
19. Delacroix P, Kenesi C, Stoppa R. Evaluation of analgesic efficacy of propacetamol versus placebo in patients with postoperative pain. Semin Hop Paris 1985;61:2739–42.
20. Ang ET, Goldfarb G, Boccard E. Analgesic efficacy of propacetamol hydrochlorate 2 g versus pentazocine 30 mg after orthopedic surgery. Eur J Pain 1990;11:137–42.
21. Romsig J, Moiniche S, Dahl JB. Rectal and parenteral paracetamol, and paracetamol in combination with NSAIDs, for postoperative analgesia. Br J Anaesth 2002;88:215–26.
22. Tang J, Li S, White P, et al. Effect of parecoxib, a novel intravenous cyclooxygenase type-2 inhibitor, on the postoperative opioid requirement and quality of pain control. Anesthesiology 2002;96:1305–9.
23. Daniels S, Grossman E, Kuss M, et al. A double-blind, randomized comparison of intramuscularly and intravenously administered parecoxib sodium versus ketorolac and placebo in a post-oral surgery pain model. Clin Ther 2001;23:1018–31.
24. Hyllested M, Jones S, Pedersen JL, Kehlet H. Comparative effect of paracetamol, NSAIDs or their combination in postoperative pain management: a qualitative review. Br J Anaesth 2002;88:199–214.
25. Varassi G, Marinangeli F, Agro F, et al. A double-blinded evaluation of propacetamol versus ketorolac in combination with patient-controlled analgesia morphine: analgesic efficacy and tolerability after gynecologic surgery. Anesth Analg 1999;88:611–6.
26. Zhou T, Tang J, White P. Propacetamol versus ketorolac for treatment of acute postoperative pain after total hip or knee replacement. Anesth Analg 2001;92:1569–75.
27. Issioui T, Klein K, White P, et al. Cost-efficacy of rofecoxib versus acetaminophen for preventing pain after ambulatory surgery. Anesthesiology 2002;97:931–7.
28. Watcha M, Issioui T, Klein K, White PF. Costs and effectiveness of rofecoxib, celecoxib and acetaminophen for preventing pain after ambulatory otolaryngologic surgery. Anesth Analg 2003;96:987–94.
29. Striebel HW, Scheitza W, Philippi W, et al. Quantifying oral analgesic consumption using a novel method and comparison with patient-controlled intravenous analgesic consumption. Anesth Analg 1998;86:1051–3.
30. Pinnock CA, Derbyshire DR, Achola KJ, Smith G. Absorption of controlled release morphine sulphate in the immediate postoperative period. Br J Anaesth 1986;58:868–71.
31. McCormack JP, Warriner B, Levine M, Glick N. A comparison of regularly dosed oral morphine and on-demand intramuscular morphine in the treatment of postsurgical pain. Can J Anaesth 1993;40:819–24.
32. Kalso E, Vainio A. Morphine and oxycodone hydrochloride in the management of cancer pain. Clin Pharmacol Ther 1990;47:639–46.
33. Gan T, Joshi G, Viscusi E, et al. Preoperative parenteral parecoxib and follow-up oral valdecoxib reduce length of stay and improve quality of patient recovery after laparoscopic cholecystectomy surgery. Anesth Analg 2004;98:1665–73.
34. Yermakova A, O’Banion MK. Cyclooxygenases in the central nervous system: implication for treatment of neurological disorders. Curr Pharm Des 2000;6:1755–76.
© 2005 International Anesthesia Research Society