Aeg038 166.172

British Journal of Anaesthesia 90 (2): 166±72 (2003) DOI: 10.1093/bja/aeg038 Parecoxib sodium has opioid-sparing effects in patients undergoing total knee arthroplasty under spinal anaesthesia² R. C. Hubbard1*, T. M. Naumann2, L. Traylor1 and S. Dhadda1 1Pharmacia, 5200 Old Orchard Road, Skokie, IL 60077, USA. 2Hessing'sche Orthopedic Clinic, Hessingstr, Augsburg, Germany *Corresponding author. E-mail: [email protected] Background. This multicentre, double-blind, placebo-controlled study compared the opioid- sparing effectiveness and clinical safety of parecoxib sodium over 48 h, in 195 postoperative patients after routine total knee replacement surgery.
Methods. Elective total primary knee arthroplasty was performed under spinal anaesthesia, with a single dose of spinal bupivacaine 10±20 mg, and intraoperative sedation with midazolam 0.5±1.0 mg i.v., or propofol <6 mg kg±1 h±1. Patients were randomized to receive either parecoxib sodium 20 mg twice daily (bd) i.v. (n=65), parecoxib sodium 40 mg bd i.v. (n=67), or placebo (n=63) at the completion of surgery, and after 12, 24, and 36 h. Morphine (1±2 mg) was taken by patient-controlled analgesia or by bolus doses after 30 min.
Results. Patients receiving parecoxib sodium 20 mg bd and 40 mg bd consumed 15.6% and 27.8% less morphine at 24 h than patients taking placebo (both P<0.05). Both doses of parecoxib sodium administered with morphine provided signi®cantly greater pain relief than morphine alone from 6 h (P<0.05). A global evaluation of study medication demonstrated a greater level of satisfaction among patients taking parecoxib sodium than those taking placebo.
Parecoxib sodium administered in combination with morphine was well tolerated. However, a reduction in opioid-type side-effects was not demonstrated in the parecoxib sodium groups.
Conclusion. Parecoxib sodium provides opioid-sparing analgesic effects in postoperative patients.
Br J Anaesth 2003; 90: 166±72Keywords: analgesics, opioid; enzymes, cyclo-oxygenase-2, inhibition; pain, postoperativeAccepted for publication: October 2, 2002 Postsurgical pain has commonly been managed with Cyclo-oxygenase-2 (COX-2) selective inhibitors, which opioid analgesics alone. Although effective, opioids (e.g. only inhibit COX-2 and not COX-1 at therapeutic doses, morphine) are associated with adverse effects such as have been developed. Several studies have shown that they respiratory depression, sedation, nausea, vomiting, consti- provide ef®cacy similar to conventional NSAIDs, but with pation, and intestinal ileus.1±4 The availability of an improved safety and tolerability.11±14 However, until now, effective, but safer analgesic, that would be co-adminis- COX-2 selective inhibitors have only been available as oral tered and reduce the amount of opioids used, would formulations, which many patients may be unable to take, therefore be an advantage. Conventional non-steroidal especially in the perioperative period. Parecoxib sodium is anti-in¯ammatory drugs (NSAIDs) such as ketorolac have the ®rst parenteral form of a COX-2 selective inhibitor been shown to produce opioid-sparing effects when developed for acute pain. It is an amide pro-drug that is administered to postoperative surgical patients.5 6 rapidly hydrolysed in vivo to the active form, valdecoxib, a However, these agents are associated with reduced platelet COX-2 selective inhibitor with a COX-2:COX-1 selectivity function, leading to prolonged bleeding time, and gastric ratio of approximately 28 000:1. Clinical studies have ulceration.7±10 The clinical utility of conventional ² NSAIDs has, therefore, been considered to be limited in Declaration of interest. The study was sponsored by Pharmacia Corporation and parecoxib sodium was provided by Pharmacia the postoperative setting.
Ó The Board of Management and Trustees of the British Journal of Anaesthesia 2003 Opioid-sparing with parecoxib sodium after surgery demonstrated that parecoxib sodium is effective in several consent and medical history were obtained, and patients models of postoperative pain.15±17 In additional studies, underwent physical examination, and testing of vital signs parecoxib sodium has been shown to have no effect on and clinical laboratory values. Elective total primary knee platelet function or gastric mucosa at doses up to 40 mg arthroplasty was performed under spinal anaesthesia.
twice daily (bd).18±20 Duration of analgesia was de®ned as the time from initiation The present study was designed to compare the opioid- of spinal analgesia until the ®rst toe movements after sparing effectiveness and clinical safety of parecoxib operation. Patients were eligible for enrolment at the end of sodium 20 mg bd i.v. and 40 mg bd i.v. over a 48-h period surgery.
in postoperative patients after total knee replacement, a Patients were randomized, according to a computer- surgical model known to cause severe pain.
generated randomization schedule, to receive either parecoxib sodium 20 mg bd i.v., parecoxib sodium 40 mg bd i.v., or placebo i.v. at the completion of surgery, and at 12, 24, and 36 h after the administration of the ®rst dose of This was a multicentre, randomized, double-blind, placebo- study medication. Morphine doses (1±2 mg) could be taken controlled, multiple-dose study conducted at 10 sites in by patient-controlled analgesia (PCA) or by bolus, and were Belgium, France, Germany, and Sweden. Enrolment ranged permitted at any time from 30 min after the end of surgery.
from 13 to 33 patients per participating investigator, and Patients remained at the study site until after the ®nal was conducted between December 1999 and August 2000. assessment at 48 h. Those requiring further analgesia at any The protocol and amendments were reviewed by the time point in addition to morphine and the study medication appropriate independent ethics committees, and the study were withdrawn from the study after completing pain was conducted in accordance with the ethical principles that assessments and a global evaluation of the study have their origins in the Declaration of Helsinki.
Patients' assessments Patients aged >18 yr, who had undergone routine total Patients assessed their pain intensity as none', mild', primary replacement of one knee, performed under a moderate' or severe' at 2, 4, 6, 9, 12, 18, 24, 36 and 48 h standardized regimen of spinal anaesthesia, were included after administration of the ®rst dose of study medication. If in the study. All patients received a single dose of spinal patients were asleep during the scheduled time for pain bupivacaine 10±20 mg, and intraoperative sedation with assessment, they were not awakened, except at the 12-, 24-, midazolam 0.5±1.0 mg i.v., or propofol <6 mg kg±1 h±1. and 36-h assessments, which were performed immediately Eligible patients were in satisfactory health, as determined before the receipt of study medication. Ambulation was by medical history and physical examination. Women were restricted for 15 min before each pain assessment.
eligible for the study only if they were not lactating and Patients also provided a global evaluation of study veri®ed as not being pregnant.
medication at 24 h (before the administration of the third Patients were excluded from the study if they had dose of study medication), and at 48 h after the adminis- undergone a revision to a previous knee replacement, tration of the ®rst dose of study medication, using a four- emergency knee replacement, or knee replacement as a point scale (1=poor, 2=fair, 3=good, 4=excellent).
result of a trauma. Patients with asthma or bronchospasm, requiring treatment with glucocorticoids, were excluded, as were those with in¯ammatory bowel disease, a chronic or acute renal or hepatic disorder, or a signi®cant coagulation Ef®cacy and safety endpoints defect. Patients were not considered eligible if they had Ef®cacy measures included the cumulative amount of upper gastrointestinal ulceration or bleeding up to 60 days morphine consumed at 2, 4, 6, 9, 12, 18, 24, 36, and 48 h, before receiving study medication. In addition, patients and the amount of morphine consumed in the periods 0±2, were excluded if they had used long-acting NSAIDs in the 4 2±4, 4±6, 6±9, 9±12, 12±18, 18±24, 24±36 and 36±48 h after days before the ®rst dose of study medication, or if they had the ®rst dose of study medication. Other ef®cacy measures taken antidepressants, narcotic analgesics, antihistamines, were the proportion of patients requiring morphine between anxiolytics, hypnotics, sedatives, NSAIDs, or cortico- the time points; time to the ®rst dose of morphine; time to steroids up to 24 h before receipt of the study medication the last dose of morphine; and patients' global evaluation of (except routine preoperative medication).
the study medication.
Safety was assessed for the 48 h after the ®rst dose of study medication by the incidence of treatment-emergent adverse events, results from physical examinations, and During the pretreatment period, de®ned as 14 days before changes from baseline in vital signs and clinical laboratory the administration of study medication, written informed values (biochemistry, haematology and urinalysis).
Hubbard et al.
Table 1 Patient characteristics. Data are actual values, or mean (range or %) Parecoxib sodium 20 mg bd i.v.
Parecoxib sodium 40 mg bd i.v.
166.08 (144.0±186.0) 166.42 (148.0±192.0) 163.58 (142.0±184.0) 80.85 (46.0±112.0) 79.19 (44.0±109.0) 80.77 (48.0±115.0) Surgical procedure Duration of surgery (h) Duration of anaesthesia (h) Statistical analyses one to two missing values, and the last-observation-carried- The sample size calculation was based on the amount of forward method was used if there were three or more morphine expected to be consumed within 24 h, and consecutive missing values, or if there were no evaluations determined that 60 patients per group would be suf®cient to after a certain time point.
detect a difference of >20% decrease in the total amount of morphine used over the 24-h postoperative dose period between parecoxib sodium 20 mg bd, parecoxib sodium Results 40 mg bd, and placebo groups.21 All ef®cacy analyses were performed on the modi®ed Patients intent-to-treat (ITT) cohort. This included all patients who A total of 195 patients were randomized to placebo (n=63), were randomized, received at least one dose of study parecoxib sodium 20 mg bd (n=65), or parecoxib sodium medication, underwent a surgical procedure of no more than 40 mg bd (n=67). The physical characteristics of the 4 h duration, did not require additional analgesic medication treatment groups were comparable; most patients were within the ®rst 30 min, and were connected to PCA within female (68±78%) and Caucasian (97±100%). In addition, 140 min of surgery.
there were no differences between treatment groups in the The cumulative amount of morphine administered at each type of surgical procedure (surgery on right or left knee), time point, and the amount of morphine consumed during duration of surgery, or duration of anaesthesia (Table 1).
each ®xed time interval, was analysed using analysis of A total of 19 (10%) patients were withdrawn from the variance (ANOVA), with treatment and centre as factors. study (Table 2), thus 176 (90%) patients completed it. The The median time to the ®rst and last dose of morphine was analyses of ef®cacy measures were performed on the ITT calculated using the Kaplan±Meier product limit estimator cohort, which included all patients who were randomized, with Miller's adjustment. For patients who withdrew from received at least the ®rst dose of study medication, whose the study before 48 h, the time to the last dose of morphine operative time was <4 h, did not require analgesia within was considered censored at the time of withdrawal.22 30 min of the end of surgery, and whose PCA was instituted Ninety-®ve per cent con®dence intervals for the median within 140 min of wound closure. The ITT cohort included time to the last dose of morphine were calculated using the 189 patients: 63 patients in the placebo group; 61 patients in Simon and Lee method.23 Overall and pair-wise log-rank the parecoxib sodium 20 mg bd group; and 65 patients in the tests were used to determine the statistical signi®cance of parecoxib sodium 40 mg bd group.
the treatment group differences in the distribution of time to the ®rst dose of morphine and to the last dose of morphine.
The proportion of patients requiring morphine during each Morphine use time period, and patients' global evaluation were analysed Analysis of the modi®ed ITT cohort demonstrated that using the Cochran±Mantel±Haenszel test, adjusted by patients receiving parecoxib sodium 20 mg bd or 40 mg bd centre. Effect of sex on morphine consumption was consumed signi®cantly less morphine at 24 h than patients analysed using Fisher's exact test.
taking placebo (both P<0.05; Fig. 1). Mean cumulative Time-speci®c pain intensity (categorical) was analysed amounts of morphine consumed over 24 h were 43.5 mg in using ANOVA. Linear interpolation was used to estimate the placebo group, and 36.7 mg and 31.4 mg in the Opioid-sparing with parecoxib sodium after surgery Table 2 Patient numbers during the study Placebo Parecoxib sodium Parecoxib sodium 20 mg bd i.v.
40 mg bd i.v.
Reason for withdrawal Treatment failure Pre-existing violation Prohibited medication Other non-compliance Fig 2 Mean PCA morphine consumed during ®xed time intervals after surgery in patients taking parecoxib sodium 20 mg bd i.v., parecoxib sodium 40 mg bd i.v. and placebo. *P<0.05, signi®cant difference in morphine consumption between parecoxib sodium groups and placebo.
Fig 1 Amount of PCA morphine (mg) consumed within 24 h of surgery in the parecoxib sodium and placebo groups. *P<0.05, signi®cant difference between parecoxib sodium groups and placebo.
parecoxib sodium 20 mg bd and 40 mg bd groups, Patients in the parecoxib sodium groups consumed less morphine than patients in the placebo group throughout the study. The reductions in morphine consumption with parecoxib sodium compared with placebo were signi®cant Fig 3 The cumulative amount of mean PCA morphine consumed over during the 6±9 and 18±24 h time intervals for the 20 mg bd 48 h after surgery in patients taking parecoxib sodium 20 mg bd i.v., dose, and during the 4±6, 6±9 and 18±24 h time intervals for parecoxib sodium 40 mg bd i.v., and placebo. Patients in the parecoxib the 40 mg bd dose (Fig. 2). At each time point after surgery, sodium groups consumed less morphine than patients in the placebo group throughout the study. The differences were signi®cant (*P<0.004) both doses of parecoxib sodium reduced the cumulative at every time point from 6 h for parecoxib sodium 40 mg bd i.v., and at morphine consumption compared with placebo (Fig. 3). 9 and 24 h for parecoxib sodium 20 mg bd i.v.
These reductions were signi®cant for patients taking parecoxib sodium 40 mg bd at every time point from 6 h (P<0.004). In patients taking parecoxib sodium 20 mg bd, Pain intensity signi®cant reductions in morphine consumption compared Parecoxib sodium administered with morphine provided with placebo were obtained at 9 and 24 h.
greater pain relief than morphine alone at every time point In each time period, the proportion of patients requiring after surgery compared with placebo. The reductions in pain morphine was lower in the parecoxib sodium groups than in intensity with both doses of parecoxib sodium were the placebo group (except the parecoxib sodium 40 mg bd statistically signi®cantly different compared with placebo group at 2±4 h). At 9±12 h, the parecoxib sodium 40 mg bd at 6 h, 24 h, and 36 h (P<0.05; Fig. 4). There were no group had a signi®cantly (P<0.011) lower percentage of signi®cant differences in pain intensity between the two patients requiring morphine than the placebo group. The parecoxib sodium treatment groups at any time point.
three treatment groups were comparable with respect to both median time to the ®rst dose of morphine (range 1.53±2.02 h) and median time to the last dose of morphine Patients' global evaluation of study medication (range 44.02±46.30). The use of morphine was not affected The patients' global evaluation of study medication showed by the patient's sex (P=0.19).
that patients treated with parecoxib sodium had greater Hubbard et al.
Fig 5 Global evaluation of study medication at 24 h in patients taking PCA morphine and either parecoxib sodium 20 mg bd i.v., parecoxib sodium 40 mg bd i.v., or placebo. Both doses of parecoxib sodium were associated with better scores than placebo; these differences were signi®cant for parecoxib sodium 40 mg bd i.v. (P=0.004).
Fig 4 Mean pain intensity scores over the 48 h after surgery in patients using PCA morphine with either parecoxib sodium 20 mg bd i.v., parecoxib sodium 40 mg bd i.v. or placebo. Pain intensity was scored as none (0), mild (1), moderate (2), or severe (3). *P<0.05, signi®cantly Table 3 Number (%) of adverse events experienced by >5% of patients in different from placebo.
any treatment group. *P<0.001 vs placebo Parecoxib sodium Parecoxib sodium 20 mg bd i.v.
40 mg bd i.v.
patient satisfaction scores than those taking placebo treat- ment. At 24 h after the ®rst dose of study medication, both Hypertension doses of parecoxib sodium were associated with signi®- Hypotension cantly better scores than placebo (P<0.018). The percentage Fever of patients who rated their study medication as good' or Nausea 22 (34.9) 18 (27.7) 12 (19.0) 14 (21.5) excellent' was 71% and 79% in the parecoxib sodium Postoperative anaemia 3 (4.8) 20 mg bd and 40 mg bd groups, compared with 53% in the Urinary retention placebo group (Fig. 5). At 48 h, parecoxib sodium 40 mg bd was associated with statistically signi®cantly better scores than placebo (P=0.004). A total of 92% of patients rated their study medication as good' or excellent', compared enced by >5% of patients in any treatment group are with 70% in the placebo group. After 48 h of treatment, summarized in Table 3.
parecoxib sodium 20 mg bd i.v. had numerically, but not Only one patient in each parecoxib sodium group statistically signi®cantly, better scores than placebo discontinued treatment as a result of adverse events: pruritus (P=0.083). A total of 84% of parecoxib sodium 20 mg bd- and rash in one patient taking parecoxib sodium 20 mg bd; treated patients rated their study medication as good' or and vomiting in one patient taking parecoxib sodium 40 mg excellent'.
bd. In the placebo group, three patients discontinued treatment because of adverse events: they were nausea and vomiting; nausea and hypertension; and nausea. One placebo-treated patient died during the study from an acute pulmonary embolism.
All 195 patients were included in the analysis of safety. The overall incidence of adverse events among the three treatment groups was similar. Adverse events were reported Discussion for 68.3% (43/63) of patients in the placebo group, 66.2% This study shows that parecoxib sodium 20 mg and 40 mg (43/65) of patients in the parecoxib sodium 20 mg bd group, bd i.v. produces signi®cant opioid-sparing effects compared and 70.1% (47/67) of patients in the parecoxib sodium 40 mg with placebo in postoperative patients after knee replace- bd group. Patients receiving parecoxib sodium 40 mg bd ment surgery. Mean morphine consumption over 24 h was experienced signi®cantly less fever than placebo patients reduced by 15.6% and 27.8% in the parecoxib sodium 20 mg (1.5% vs 19.0%; P<0.001; Table 3). The incidence of bd and 40 mg bd groups, respectively. At each time point opioid-type side-effects (nausea, vomiting, constipation, over the 48-h postoperative period, both doses of parecoxib intestinal ileus, central nervous system effects) was similar. sodium reduced morphine consumption in a dose-dependent The use of antiemetic medication was not different between manner compared with placebo. The combination of better groups. Most (>85%) of the adverse events in each group pain control and tolerability in the presence of parecoxib were mild to moderate in severity. Adverse events experi- sodium affected the patients' global evaluation of the study Opioid-sparing with parecoxib sodium after surgery medication as, at 24 and 48 h, there was a greater level of required. Consequently, additional clinical studies, using satisfaction among patients taking parecoxib sodium than different clinical designs or different patient populations, those taking placebo.
will be helpful to de®ne more fully any clinical bene®t of the The use of analgesic combination therapy to achieve opioid-sparing effects of parecoxib sodium.
reductions in opioid use and improved pain control, as COX-2 selective inhibitors that provide opioid-sparing recommended by The Royal College of Anaesthetists and effects, without disrupting platelet function, are ideal in the others, is supported in this study.24±26 Patients in the management of postoperative patients. Although oral COX-2 parecoxib sodium treatment groups not only required less selective inhibitors have demonstrated opioid-sparing morphine, but experienced greater pain relief than patients effects when administered preoperatively to patients under- receiving morphine alone at every time point after surgery going spinal fusion surgery,35 parecoxib sodium is the only compared with placebo. These results demonstrate one injectable form of a COX-2 selective inhibitor in develop- advantage of multi-model therapy, that patients experience ment. This is particularly important in the acutely painful less pain even though they require less narcotic.
postoperative setting because many surgical patients cannot Parecoxib sodium is a COX-2 selective inhibitor, which tolerate oral medication or may have variable perioperative reduces the number of adverse events associated with non- gastrointestinal absorptive function.
selective COX-1 inhibition. These include upper gastro- Parecoxib sodium has demonstrated analgesic ef®cacy in intestinal ulceration and bleeding, renal dysfunction, and several postoperative pain models, and has shown a rapid bleeding related to platelet inhibition. In this study, onset of action and long duration.16 17 36 The results of the parecoxib sodium 20 mg bd i.v. and 40 mg bd i.v. were present study indicate that parecoxib sodium also provides a safe and well tolerated, as most of the adverse events in each signi®cant opioid-sparing effect when administered over a group were mild to moderate in intensity. There were no 48-h period in postoperative patients after total knee clinically important adverse gastrointestinal, platelet- replacement, a surgical model known to cause severe pain.
related, or renal side-effects observed in this trial, although the numbers studied were relatively small.
A reduction in opioid-type side-effects was not demon- Acknowledgements strated in this study, however. Some postoperative side- The authors would like to extend their thanks to the following investigators effects, such as nausea and vomiting, which had a higher who participated in the trial: Dr Kai Bernsmann (St Josef Hospital, Bochum Germany), Dr Frederic Camu (Academisch Ziekenhuis, Brussels, incidence in patients taking parecoxib sodium 40 mg bd Belgium), Dr Christian Conseiller (Hospital Cochin, Paris, France), Dr than in patients taking parecoxib sodium 20 mg bd or Erna Van Droogenbroeck (Aalsters Stededijke Ziekenhuis, Aalst, placebo, are affected by multiple factors in the surgical Belgium), Dr Kristine Fonck (Univ Ziekenhuis Gent, Gent, Belgium), Dr Rene Heylen (Sol Campus St Jan, Gent, Belgium), Dr Vincent L. Hoffman environment. For example, the type of surgery, the anaes- (Univ Ziekenhuis Antwerpen, Edegem, Belgium), Dr Klaus Ruhnau thetic and associated medication, and the pain itself.27 It (Marien Hospital Buer gGmbH, Gelsenkirchen, Germany) and Dr Soren may not be a reasonable expectation that the occurrence of Toksvig-Larsen (Landskrona Hospital, Landskrona, Sweden).
such multifactorial events in the immediate postoperative period would be modi®ed by postoperative administration References of parecoxib sodium. There are several other potential explanations for the absence of any reduction in opioid-type 1 Bovill JG, Sebel PS, Stanley TH. Opioid analgesics in anesthesia: side-effects. First, opioid symptoms and physiological with special reference to their use in cardiovascular anesthesia.
Anesthesiology 1984; 61: 731±55 effects were not prospectively identi®ed for speci®c meas- 2 Guignard B, Bossard AE, Coste C. Acute opioid tolerance: urement in this trial. The use of speci®c data collection intraoperative remifentanil increases postoperative pain and instruments (such as the Opiate Symptom Distress morphine requirement. Anesthesiology 2000; 93: 409±17 Questionnaire; On ®le, Pharmacia), physical examination 3 Holmes B, Ward A. Meptazinol. A review of its pharmaco- (such as listening for bowel sounds), or physiological dynamic and pharmacokinetic properties and therapeutic monitoring (oxygen saturation) are methods that may be ef®cacy. Drugs 1985; 30: 285±312 able to evaluate more clearly the effects of opioid reduction.
4 Lewis KS, Han NH. Tramadol: a new centrally acting analgesic.
Am J Health Syst Pharm 1997; 54: 643±52 Second, the degree of reduction in opioid use may not have 5 Burns JW, Aitken HA, Bullingham RE, McArdle CS, Kenny GN.
been suf®cient to result in clinically meaningful reductions Double-blind comparison of the morphine sparing effect of in their side-effects, or the trial may have been too small to continuous and intermittent i.m. administration of ketorolac. Br J identify differences between treatment groups. A trial of this Anaesth 1991; 67: 235±8 size may also be underpowered to identify reductions in 6 Gillies GW, Kenny GN, Bullingham RE, McArdle CS. The uncommon symptoms, such as respiratory depression. A morphine sparing effect of ketorolac tromethamine. A study of a review of published opioid-sparing trials with ketorolac new, parenteral non-steroidal anti-in¯ammatory agent after demonstrates that opioid-sparing effects are variable and not abdominal surgery. Anaesthesia 1987; 42: 727±31 7 Borda IT, Koff R, eds. NSAIDs: a Pro®le of Adverse Effects.
always signi®cantly demonstrated.28±34 The use of a patient Philadelphia: Hanley and Belfus Inc.; 1995 population more susceptible to opioid side-effects than the 8 Brooks PM, Day RO. Nonsteroidal antiin¯ammatory drugsÐ carefully selected population in this clinical trial may be differences and similarities. N Engl J Med 1991; 324: 1716±25 Hubbard et al.
9 Gabriel SE, Jaakkimainen L, Bombardier C. Risk for serious 23 Simon R, Lee YJ. Nonparametric con®dence limits for survival gastrointestinal complications related to use of nonsteroidal probabilities and median survival time. Cancer Treat Rep 1982; 66: anti-in¯ammatory drugs. A meta-analysis. Ann Intern Med 1991; 24 Acute Pain Management Guidelines Panel. Acute Pain 10 Garcia Rodriguez LA, Jick H. Risk of upper gastrointestinal Management: Operative or Medical Procedures and Trauma.
bleeding and perforation associated with individual non-steroidal Clinical Practice Guidelines. AHCPR 02-0032. Rockville: Agency anti-in¯ammatory drugs. Lancet 1994; 343: 769±72 for Health Care Policy Research, Public Health Service, US 11 Bensen WG, Fiechtner JJ, McMillen JI, et al. Treatment of Department of Health and Human Services, 1992 osteoarthritis with celecoxib, a cyclooxygenase-2 inhibitor: a 25 Guidelines for the Use of Non-steroidal Anti-in¯ammatory randomized controlled trial. Mayo Clin Proc 1999; 74: 1095±105 Drugs in the Perioperative Period. The Royal College of 12 Geis GS. Update on clinical developments with celecoxib, a new Anaesthetists, March 1998 speci®c COX-2 inhibitor: what can we expect? J Rheumatol 1999; 26 Carr DB, Goudas LC. Acute pain. Lancet 1999; 353: 2051±8 27 Moote C. Ef®cacy of nonsteroidal anti-in¯ammatory drugs in the 13 Gierse JK, McDonald JJ, Hauser SD, Rangwala SH, Koboldt CM, management of postoperative pain. Drugs 1992; 44: 14±29 Seibert K. A single amino acid difference between cyclo- 28 Fredman B, Olsfanger D, Flor P, Jedeikin R. Ketorolac does not oxygenase-1 (COX-1) and -2 (COX-2) reverses the selectivity decrease postoperative pain in elderly men after transvesical of COX-2 speci®c inhibitors. J Biol Chem 1996; 271: 15810±14 prostatectomy. Can J Anaesth 1996; 43: 438±41 14 Simon LS, Weaver AL, Graham DY, et al. Anti-in¯ammatory and 29 Vanlersberghe C, Lauwers MH, Camu F. Preoperative ketorolac upper gastrointestinal effects of celecoxib in rheumatoid administration has no preemptive analgesic effect for minor arthritis: a randomized controlled trial. JAMA 1999; 282: 1921±8 orthopaedic surgery. Acta Anaesthesiol Scand 1996; 40: 948±52 15 Desjardins PJ, Grossman EH, Kuss ME, et al. The injectable 30 Parker RK, Holtmann B, Smith I, White PF. Use of ketorolac cyclooxygenase-2-speci®c inhibitor parecoxib sodium has after lower abdominal surgery. Effect on analgesic requirement analgesic ef®cacy when administered preoperatively. Anesth Analg 2001; 93: 721±7 and surgical outcome. Anesthesiology 1994; 80: 6±12 16 Rasmussen L, Steckner K, Hogue C, Kuss ME, Torri S, Hubbard 31 Green CR, Pandit SK, Levy L, Kothary SP, Tait AR, Schork MA.
R. Intravenous parecoxib sodium for acute pain after Intraoperative ketorolac has an opioid-sparing effect in women postorthopedic knee surgery. Am J Orthop 2002; 31: 334±43 after diagnostic laparoscopy but not after laparoscopic tubal 17 Barton SF, Langeland FF, Snabes MC, et al. Ef®cacy and safety of ligation. Anesth Analg 1996; 82: 732±7 intravenous parecoxib sodium in relieving acute post-operative 32 Blackburn A, Stevens JD, Wheatley RG, Madej TH, Hunter D.
pain following gynecologic laparotomy surgery. Anesthesiology Balanced analgesia with intravenous ketorolac and patient- 2002; 97: 306±14 controlled morphine following lower abdominal surgery. J Clin 18 Noveck RJ, Laurent A, Kuss ME, Talwalker S, Hubbard RC. The Anesth 1995; 7: 103±8 COX-2 speci®c inhibitor, parecoxib sodium, does not impair 33 Reuben SS, Connelly NR, Lurie S, Klatt M, Gibson CS. Dose- platelet function in healthy elderly and non-elderly individuals.
response of ketorolac as an adjunct to patient-controlled Clin Drug Invest 2001; 21: 465±76 analgesia morphine in patients after spinal fusion surgery.
19 Harris SI, Kuss M, Hubbard RC, Goldstein JL. Upper Anesth Analg 1998; 87: 98±102 gastrointestinal safety evaluation of parecoxib sodium, a new 34 Rogers JE, Fleming BG, Macintosh KC, Johnston B, Morgan- parenteral cyclooxygenase-2-speci®c inhibitor compared with Hughes JO. Effect of timing of ketorolac administration on ketorolac, naproxen and placebo. Clin Ther 2001; 23: 1422±8 patient-controlled opioid use. Br J Anaesth 1995; 75: 15±8 20 Stoltz RR, Harris SI, Kuss ME, et al. Upper GI mucosal effects of 35 Reuben SS, Connelly NR. Postoperative analgesic effects of parecoxib sodium in healthy elderly subjects. Am J Gastroenterol celecoxib or rofecoxib after spinal fusion surgery. Anesth Analg 2000; 91: 1221±5 21 Delbos A, Boccard E. The morphine-sparing effect of 36 Daniels SE, Grossman EH, Kuss ME, Talwalker S, Hubbard RC. A propacetamol in orthopedic postoperative pain. J Pain Symptom double-blind, randomized comparison of intramuscularly and Manage 1995; 10: 279±86 intravenously administered parecoxib sodium versus ketorolac 22 Miller RG. Survival Analysis. Somerset: John Wiley and Sons, 1998; and placebo in a post-oral surgery pain model. Clin Ther 2001; 23:

Source: http://www.eao.chups.jussieu.fr/polys/certifopt/saule_coxib/ainsanticox2/1hubbardetal.pdf