Mu-opioid receptor agonists have been considered the cornerstone of systemic postsurgical analgesia for over a century; however, emerging evidence supports the routine postoperative co-administration of various nonopioid analgesics1 for the purpose of improving analgesic efficacy, more effectively reducing movement-evoked pain2 and reducing opioid-related adverse effects.3 Although several recent reviews advocating “multimodal” analgesia discuss the use of up to eight different drug classes (e.g., opioids, local anesthetics, acetaminophen, nonsteroidal antiinflammatory drugs [NSAIDs], anticonvulsants, N-methyl-d-aspartate antagonists, α-2-adrenergic agonists and corticosteroids), the current evidence base of rigorous comparative trials to support the use of specific drug combinations is relatively limited to combinations of an opioid with various nonopioid analgesics and the combination of a NSAID with local anesthetics or acetaminophen.4 Several trials have evaluated the postoperative combination of acetaminophen with NSAIDs5; however, fewer trials are available to guide the rational use of other nonopioid drug combinations. Recent trials of inpatients after hysterectomy2,6 and spinal fusion7 suggest that combining cyclooxygenase-2 inhibitors (COX-2I) with α-2-Δ calcium channel subunit antagonists (i.e., gabapentin and pregabalin) may provide (particularly on postoperative days 1–2) superior reduction of movement-evoked pain and accelerated pulmonary recovery without substantial increases in side effect burden.
Since the discovery of diverse COX enzymes and the potential for better gastrointestinal safety with COX-2 selective inhibition, a growing number of postoperative pain trials are being conducted involving COX-2Is.8 However, very few investigations have evaluated the COX-2I, meloxicam, and for this reason we chose to study it in the present combination trial. Meloxicam is a COX-2I which, at doses as low as 7.5 mg, has demonstrated efficacy for the short-term treatment of postoperative pain9 and, in some trials, with comparable efficacy10 and less blood loss11 compared with nonspecific NSAIDs. Although not readily apparent, it should be noted that cardiovascular problems seen with rofecoxib and other COX-2Is could potentially also occur with meloxicam. Gabapentin is a 3-alkylated analog of γ-amino butyric acid whose foremost putative analgesic mechanism is the modulation of α-2-Δ calcium channel subunits.12 Many small trials have demonstrated the safety and efficacy of gabapentin, at doses as low as 300 mg, for pain after various surgical procedures (including laparoscopic cholecystectomy) with observed benefits including reduction of movement-evoked pain, opioid sparing, and preoperative anxiolysis.13–15 Given a lack of adverse drug interactions, their distinct analgesic mechanisms and preclinical evidence of analgesic additivity or synergism between these two drug classes,16–18 meloxicam and gabapentin appear to have distinct analgesic mechanisms which may interact favorably in combination. Thus, the purpose of this trial was to test the hypothesis that a meloxicam-gabapentin combination has superior efficacy versus either drug alone in ambulatory patients after laparoscopic cholecystectomy. Given previous evidence that gabapentin2 and meloxicam9 reduce both spontaneous and evoked pain, this trial was designed to comprehensively evaluate both these effects for a combination of gabapentin and meloxicam.
This trial received institutional approval from the Queen's University Research Ethics Board and government regulatory approval from Health Canada. Eligible patients aged 18 yr or older with a body mass index <36 kg/m2 that fit an ASA I or II classification before elective laparoscopic cholecystectomy were enrolled between November 2004 and January 2008. Trial exclusions were: 1) hypersensitivity to any drugs to be used in the study, 2) serious organ disease/dysfunction, 3) persistent preoperative pain, 4) daily intake or intake within 48 h before surgery, of any glucocorticoid drugs, NSAIDs, or other analgesics, not including daily administration of ≤325 mg of aspirin for cardiovascular prophylaxis, 5) evidence of substance or alcohol abuse, 6) major psychiatric disorder, 7) bleeding disorder, 8) peptic ulcer disease, 9) moderate to severe asthma, 10) seizure disorder requiring treatment with an anticonvulsant, and 11) language barrier to communicating with research staff.
This was a single-center, parallel, randomized, double-blind trial with three treatment groups: (a) meloxicam 15 mg PO daily for 3 days from the morning of surgery 1 h preoperatively up to and including the morning of postoperative day 2, (b) 1600 mg PO daily on the day of surgery (1200 mg PO before surgery and 400 mg PO on the evening of surgery) and gabapentin 1200 mg PO daily on postoperative days 1 and 2 (400 mg PO TID on postoperative days 1 and 2) for 3 days, and (c) meloxicam 15 mg PO daily for 3 days plus gabapentin 1600 mg daily on the day of surgery and gabapentin 1200 mg daily on postoperative days 1 and 2. On the day of surgery, a larger preoperative dose of gabapentin (or matching placebo) was administered to exert a potential preemptive effect and also to partially compensate for the midday gabapentin dose precluded by recovery from anesthesia. Study medications were encapsulated in identically appearing red (rofecoxib or “rofecoxib” placebo) and gray (gabapentin or “gabapentin” placebo) gelatin capsules by the Kingston General Hospital Investigational Pharmacy (KGH-IP) to maintain double-blind conditions as per a double-dummy design. At study commencement, a pharmacist from the KGH-IP prepared a concealed, computer-generated random treatment allocation schedule, which randomized these three treatments, in blocks of three, to a series of patient numbers. After informed consent, enrolled patients were assigned the next consecutive patient number and the corresponding set of study medications was dispensed by the KGH-IP. No one other than the KGH-IP pharmacists was aware of any patient's treatment allocation during the trial. While in hospital, nursing staff administered and observed study medication consumption. Patients received red capsules and gray capsules, which were identical in appearance across treatment groups as per a double dummy design. Two red capsules (each containing either meloxicam 7.5 mg or placebo) were administered once daily in the morning starting preoperatively on the day of surgery (1 h preop) and continuing through postoperative days 1 and 2. One gray capsule (containing either gabapentin 400 mg or placebo) was administered three times daily on postoperative days 1 and 2. On the day of surgery, three gray capsules (containing either gabapentin 400 mg or placebo) were administered 1 h before surgery and 1 gray capsule (containing either gabapentin 400 mg or placebo) was administered on the evening of surgery (i.e., the midday dose of gray capsules was omitted on the day of surgery).
Perioperative Study Protocol
Intraoperatively, patients received a balanced anesthetic at the discretion of the attending anesthesiologist who was blinded to treatment. Although not frequently used in this setting at our institution, the use of ketamine was specifically forbidden in this study because of its analgesic effects and the potential that unbalanced use across treatment arms might skew study results. IV fentanyl (2–5 μg/kg) was the only opioid administered intraoperatively and any other analgesic drugs (including glucocorticoids such as dexamethasone) were forbidden. Throughout laparoscopic surgery, carbon dioxide insufflation pressures were limited to no higher than 15 mm Hg. Upon completion of surgery, all trocar insertion sites were infiltrated by the surgeon with up to 15 mL of 0.25% bupivicaine without epinephrine. All patients received ondansetron 4 mg IV, 30 min before anticipated completion of surgery. After surgery, patients were evaluated by study personnel for all the day- of-surgery outcome measures (see below) upon arrival in the postanesthetic care unit (PACU) and then every 30 min thereafter until discharge from hospital. All patients received clinician-administered analgesia with IV fentanyl 12.5–25 μg every 3 min as needed and no other analgesic drugs were allowed. Upon discharge from hospital, patients were prescribed either codeine 30–60 mg or morphine 5–10 mg (as per the blinded nonstudy clinician prescriber's preference) PO every 3 h as needed and no other analgesic drugs were allowed until the end of postoperative day 3. A nurse in the PACU evaluated each patient every 15 min and instituted oxygen therapy for an oxygen saturation of <92%.
The primary outcome was pain intensity (0–10 numerical rating scale), at rest and with movement on the day of surgery in PACU. Shoulder pain was also specifically rated as a secondary outcome on the day of surgery. Pain at rest and with movement on postoperative days 1 and 2 were included as secondary outcomes based on telephone calls made at approximately 11 am on each of those days. Other secondary outcome measures included pain intensity 30 days postoperatively and opioid consumption, peak expiratory flow (PEF), forced expiratory volume in 1 s (FEV1), forced vital capacity (FVC), pain-related interference,19 patient satisfaction, frequency of moderate or severe adverse effects, oxygen saturation, frequency of supplemental oxygen requirement, time to hospital discharge, and time to return to work.
Baseline PEF, FEV1, and FVC were measured preoperatively on the day of surgery. At each assessment after surgery in the PACU, pain (0–10 numerical rating scale) was recorded as per the following progression: 1) pain at rest, 2) pain evoked by sitting, in a standardized fashion, from the supine position, followed by a 60-s rest period, 3) pain evoked by peak expiration using a peak flowmeter (Spirolab II, Roxon Medi-tech Ltd., Montreal, Quebec, Canada) followed by a 60-s rest period, and finally 4) pain evoked by cough. Also at each assessment, interval and cumulative fentanyl consumption, oxygen saturation (%) and current oxygen administration, side effects (elicited with the question: “Are you having any symptoms other than pain? If yes, rate them as: 0—none, 1—mild, 2—moderate, 3—severe.”), and the interval from PACU admission to the time at which patients met a postanesthesia discharge score of 920 were recorded. On postoperative days 1, 2, and 30, patients were contacted by telephone and asked to rate their pain at rest, then upon sitting or standing and then upon coughing. Patients also reported on opioid consumption and the presence and severity of side effects and rated how much pain interferes (0–10) with general activity, mood, walking, relations with others, and sleep. On postoperative day 3, patients were contacted by telephone to record their total opioid consumption since hospital discharge and also asked to rate their satisfaction (0, completely dissatisfied; 10, completely satisfied) with the postoperative pain management they received.
Statistical analyses were conducted by the trial's statistician (DT). Sample size calculation was based on the null hypothesis of no pain difference across all groups. Assuming a baseline pain intensity of 5 on a 0–10 numerical rating scale, a pain variance of 2.5 from previous estimates, and using the Bonferroni method to adjust for two pairwise comparisons (i.e., combination versus meloxicam and combination versus gabapentin), we calculated that a sample of 30 patients per treatment group would provide an 80% chance (at an α level of 0.05) of detecting a mean difference among treatment groups of 1.5 on a 0–10 numerical rating scale.
Repeated measures, such as pain, morphine consumption, and PEF/FEV1/FVC were analyzed by the following method: A linear mixed model with terms of treatment, time, and interaction between treatment and time were first fitted with the data.21 If the interaction term was significant, the treatment groups were then compared at each time point in cross-sectional analyses using the method of nonparametric analysis of variance (ANOVA) with the Wilcoxon's test as a special case when there are two groups.22 Otherwise, a model with treatment and time variables only was refitted. Exploratory cross-sectional analyses comparing treatment groups at each time point were also performed using nonparametric ANOVA method regardless of whether or not a treatment by time interaction was significant. Continuous data measured at one single point, such as pain on day 30, pain interference, and patient satisfaction, or obtained as a summary measure for repeated measures such as time to hospital discharge, were analyzed using the nonparametric ANOVA mentioned above. Proportion data, such as frequency of adverse effects and patients requiring oxygen, were analyzed using Fisher's exact method.23 For all analyses, the three pairwise comparisons at 0.05 level were only made if the overall test for the difference among all groups was significant at the 0.05 level. This Fisher's least significant difference method provides protection for multiple comparisons.24 All statistical analyses were conducted using SAS (statistical analysis system) software version 9.
Figure 1 describes patient flow through the trial including hospital admissions and study withdrawals. Patients were recruited between January 2005 and December 2007. No changes were made to surgical, anesthetic or postoperative analgesic techniques over this time period. Regardless of the timing of trial exit, all patients receiving study drug were tracked over at least 30 days for adverse effects and major adverse reactions. Outcomes were included for analysis from all patients until their withdrawal from or completion of the trial. Table 1 describes the demographic and baseline characteristics of patients included in the efficacy analysis.
Figure 2 shows mean spontaneous and evoked pain scores on the day of surgery. Because more than 50% of study data were missing at 0 and 30 min (because of postanesthetic sedation) and at 150 min and later (because of patient discharges from the PACU), day of surgery pain data are reported and analyzed only for 60, 90, and 120 min. The percentage of pain data available for analysis at 60, 90, and 120 min varied from 67% to 100%, depending on the specific pain measure (e.g., rest versus sitting) and time point. Primary repeated measures analyses showed a significant effect of treatment by time interaction for rest pain only (P = 0.01). Subsequent cross-sectional analysis showed that rest pain at 60 min in the gabapentin group was lower than in the meloxicam group (P = 0.003). A significant effect of time was found for all other pain measures (except shoulder pain, P = 0.7) such that pain decreased from 60 to 120 min (P = 0.0005, 0.005, <0.0001 for pain evoked by peak expiration, sitting, and cough, respectively), but no significant effect of treatment was found for any other pain measures (P = 0.7, 0.3, 0.3, and 0.08 for shoulder pain, pain evoked by peak expiration, sitting, and cough pain, respectively). Only one exploratory cross-sectional analysis was statistically significant and suggested that cough pain at 60 min in the gabapentin group was lower than in the meloxicam group (P = 0.01). On postoperative days 1, 2, and 30, there were no significant effects of treatment group on spontaneous or movement-evoked pain measures (Table 2).
Opioid Consumption, Oxygen Requirement, and Spirometric Measures
There were no statistically significant differences among treatment groups in opioid consumption either on the day of surgery or during the interval from hospital discharge to postoperative day 3 (Fig. 2). In the PACU, oxygen administration was required by 53.4%, 53.3%, and 51.7% of patients in the meloxicam, gabapentin, and combination groups, respectively. Table 3 describes PEF, FEV1, and FVC from 60 to 120 min after surgery. Primary repeated measures analyses showed no significant effect of treatment by time interaction for any of the spirometric measures (P = 0.44, 0.07, and 0.35 for PEF, FEV1, and FVC, respectively). A significant effect of time was found for all spirometric measures such that these measures improved from 60 to 120 min (P < 0.0003 for all measures), but no significant effect of treatment was found for any of these measures (P = 0.3, 0.9, 0.9 for PEF, FEV1, and FVC, respectively). The only statistically significant exploratory cross-sectional analysis suggests that PEF at 120 min in the combination group was higher than in the meloxicam group (P = 0.02).
Adverse Effects, Overall Patient Satisfaction, and Pain-Related Interference
No major adverse events were encountered during the trial follow-up period other than those described in the “Subjects” section above. Table 4 describes the frequency of treatment-emergent, moderate, or severe adverse effects with at least a 5% incidence in one of the treatment groups. For individual side effects, only nausea on the day of surgery was significantly different among treatment groups (P = 0.03) and pairwise comparisons showed nausea was significantly less frequent in the combination group than in the meloxicam group (P = 0.016) but not the gabapentin group (P = 0.77). No significant differences among treatment groups were observed for overall patient satisfaction or pain-related interference (data not shown).
Time to Hospital Discharge and Return to Work
Median time (hours) from PACU admission to meeting PACU discharge criteria was 2.92, 2.83, and 2.75 for meloxicam, gabapentin, and combination groups, respectively (not significant). Mean time (days) from surgery to return to work was 13.6, 11.7, and 10.6 for meloxicam, gabapentin, and combination groups, respectively (not significant).
These data suggest little or no benefit of combining meloxicam and gabapentin for pain relief after laparoscopic cholecystectomy. It is unlikely that these apparently negative results were due to chance, insensitivity of measurement scales, or bias. Observed temporal changes in pain on the day of surgery were internally consistent across all pain measures and levels of reported pain intensity were externally consistent with those of other laparoscopic cholecystectomy trials.25,26 Our observed difference in 60-min rest pain between the gabapentin and meloxicam groups suggest that this trial had adequate statistical power and assay sensitivity to detect significant treatment differences. Also consistent with these results, data from our previous hysterectomy trial similarly showed that pain during rofecoxib-gabapentin combination treatment was either greater or minimally different than during the single-drug gabapentin treatment on the day of surgery.2
The apparent lack of superiority of a gabapentin-meloxicam versus single-drug gabapentin might be explained by a limited role of COX inhibition immediately postoperatively on the day of surgery.27 Because the release of cytokines and inflammatory mediators is not readily appreciated until about 4 h after surgery,28–30 it is possible that NSAIDs/COX-2Is have lower efficacy on the day of surgery when compared with subsequent postoperative days which may, in part, explain the trends favoring gabapentin and meloxicam-gabapentin over meloxicam alone. Although differences in pain between combination and single-drug therapy on postoperative days 1 and 2 were not statistically significant, we cannot exclude an additive effect on these days because trends favoring the combination were observed consistently across all pain measures and opioid consumption. Also, pain scores on postoperative days 1 and 2 were substantially lower than on the day of surgery leading to a possible “floor” effect thus making it more difficult to demonstrate significant differences within the constraints of the trial's statistical power.
Combination therapy was associated with less-frequent nausea versus single-drug meloxicam treatment. Because trends toward less-opioid use in the combination group on postoperative days 1–3 were not statistically significant, the possibility of opioid sparing as an explanation for less nausea cannot be confirmed. It is, however, possible that gabapentin exerts a specific antiemetic effect as suggested by one previous postoperative trial.31 Our observations of different frequencies of nausea across treatment groups should be interpreted with caution; however, because they were elicited with open-ended side effect questioning as opposed to a specific instrument for evaluating nausea and vomiting.
Exploratory analyses suggest that PEF 120 min after surgery was significantly higher during combination versus meloxicam treatment. This might be partly explained by a nonsignificant trend toward lower intensity of pain evoked by peak expiration given previous observations of an inverse relationship between evoked pain and pulmonary performance32 and of improved lung function with analgesic treatment.2,33,34 However, despite this difference in pulmonary performance, no significant differences in supplemental oxygen administration were observed.
In conclusion, data from this trial provide no support for the combined use of meloxicam and gabapentin for pain relief after laparoscopic cholecystectomy. These results suggest that perioperative analgesic polypharmacy may not always be necessary or appropriate and further emphasize the need for continued rigorous evaluation of different analgesic combinations on a procedure-specific, and combination-specific, basis to better guide the rational polypharmacy of postoperative pain.
The authors thank Deborah Shore, RN, and Chris Gray for their valuable assistance with several aspects of this study.
1. Kehlet H, Dahl JB. The value of “multimodal” or “balanced analgesia” in postoperative pain treatment. Anesth Analg 1993;77:1048–56
2. Gilron I, Orr E, Tu D, O'Neill JP, Zamora JE, Bell AC. A placebo-controlled randomized clinical trial of perioperative administration of gabapentin, rofecoxib and their combination for spontaneous and movement-evoked pain after abdominal hysterectomy. Pain 2005;113:191–200
3. Marret E, Kurdi O, Zufferey P, Bonnet F. Effects of nonsteroidal antiinflammatory drugs on patient-controlled analgesia morphine side effects: meta-analysis of randomized controlled trials. Anesthesiology 2005;102:1249–60
4. White PF, Kehlet H, Neal JM, Schricker T, Carr DB, Carli F. Fast-Track Surgery Study Group. The role of the anesthesiologist in fast-track surgery: from multimodal analgesia to perioperative medical care. Anesth Analg 2007;104:1380–96
5. 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
6. Turan A, White PF, Karamanlioglu B, Memis D, Tasdogan M, Pamukçu Z, Yavuz E. Gabapentin: an alternative to the cyclooxygenase-2 inhibitors for perioperative pain management. Anesth Analg 2006;102:175–81
7. Reuben SS, Buvanendran A, Kroin JS, Raghunathan K. The analgesic efficacy of celecoxib, pregabalin, and their combination for spinal fusion surgery. Anesth Analg 2006;103:1271–7
8. Gilron I, Milne B, Hong M. Cyclooxygenase-2 inhibitors in postoperative pain management: current evidence and future directions. Anesthesiology 2003;99:1198–208
9. Thompson JP, Sharpe P, Kiani S, Owen-Smith O. Effect of meloxicam on postoperative pain after abdominal hysterectomy. Br J Anaesth 2000;84:151–4
10. Nekoofar MH, Sadeghipanah M, Dehpour AR. Evaluation of meloxicam (A cox-2 inhibitor) for management of postoperative endodontic pain: a double-blind placebo-controlled study. J Endod 2003;29:634–7
11. Weber EW, Slappendel R, Durieux ME, Dirksen R, van der Heide H, Spruit M. COX 2 selectivity of non-steroidal anti-inflammatory drugs and perioperative blood loss in hip surgery. A randomized comparison of indomethacin and meloxicam. Eur J Anaesthesiol 2003;20:963–6
12. Taylor CP, Gee NS, Su TZ, Kocsis JD, Welty DF, Brown JP, Dooley DJ, Boden P, Singh L. A summary of mechanistic hypotheses of gabapentin pharmacology. Epilepsy Res 1998;29: 233–49
13. Ho KY, Gan TJ, Habib AS. Gabapentin and postoperative pain—a systematic review of randomized controlled trials. Pain 2006;126:91–101
14. Gilron I. Gabapentin and pregabalin for chronic neuropathic and early postsurgical pain: current evidence and future directions. Curr Opin Anaesthesiol 2007;20:456–72
15. Tiippana EM, Hamunen K, Kontinen VK, Kalso E. Do surgical patients benefit from perioperative gabapentin/pregabalin? A systematic review of efficacy and safety. Anesth Analg 2007; 104:1545–56
16. Hurley RW, Chatterjea D, Rose Feng M, Taylor CP, Hammond DL. Gabapentin and pregabalin can interact synergistically with naproxen to produce antihyperalgesia. Anesthesiology 2002;97:1263–73
17. Picazo A, Castañeda-Hernández G, Ortiz MI. Examination of the interaction between peripheral diclofenac and gabapentin on the 5% formalin test in rats. Life Sci 2006;79:2283–7
18. Yoon MH, Yaksh TL. Evaluation of interaction between gabapentin and ibuprofen on the formalin test in rats. Anesthesiology 1999;91:1006–13
19. Mendoza TR, Chen C, Brugger A, Hubbard R, Snabes M, Palmer SN, Zhang Q, Cleeland CS. The utility and validity of the modified brief pain inventory in a multiple-dose postoperative analgesic trial. Clin J Pain 2004;20:357–62
20. Chung F. Are discharge criteria changing? J Clin Anesth 1993; 5:64S–68S
21. Verbeke G, Molenberghs G, eds. Linear mixed models in practice: a SAS-oriented approach. New York: Springer, 1997
22. Lehmann EL. Nonparametrics: statistical methods based on ranks. San Francisco: Holden-Day, 1975
23. Agresti A. Categorical data analysis. New York: Wiley, 1990
24. Snedecor GW, Cochran WG. Statistical methods. 7th ed. Ames, IA: The Iowa State University Press, 1980
25. Pandey CK, Priye S, Singh S, Singh U, Singh RB, Singh PK. Preemptive use of gabapentin significantly decreases postoperative pain and rescue analgesic requirements in laparoscopic cholecystectomy. Can J Anaesth 2004;51:358–63
26. Gan TJ, Joshi GP, Viscusi E, Cheung RY, Dodge W, Fort JG, Chen C. 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
27. Puolakka PA, Puura AI, Pirhonen RA, Ranta AU, Autio V, Lindgren L, Rorarius MG. Lack of analgesic effect of parecoxib following laparoscopic cholecystectomy. Acta Anaesthesiol Scand 2006;50:1027–32
28. Buvanendran A, Kroin JS, Berger RA, Hallab NJ, Saha C, Negrescu C, Moric M, Caicedo MS, Tuman KJ. Upregulation of prostaglandin E2 and interleukins in the central nervous system and peripheral tissue during and after surgery in humans. Anesthesiology 2006;104:403–10
29. Colley CM, Fleck A, Goode AW, Muller BR, Myers MA. Early time course of the acute phase protein response in man. J Clin Pathol 1983;36:203–7
30. Wallace AM, Sattar N, Mcmillan DC. The co-ordinated cytokine/hormone response to acute injury incorporates leptin. Cytokine 2000;12:1042–5
31. Pandey CK, Priye S, Ambesh SP, Singh S, Singh U, Singh PK. Prophylactic gabapentin for prevention of postoperative nausea and vomiting in patients undergoing laparoscopic cholecystectomy: a randomized, double-blind, placebo-controlled study. J Postgrad Med 2006;52:97–100
32. Gilron I, Tod D, Goldstein DH, Parlow JL, Orr E. The relationship between movement-evoked versus spontaneous pain and peak expiratory flow after abdominal hysterectomy. Anesth Analg 2002;95:1702–7
33. Liu J, Ding Y, White PF, Feinstein R, Shear JM. Effects of ketorolac on postoperative analgesia and ventilatory function after laparoscopic cholecystectomy. Anesth Analg 1993;76: 1061–6
© 2009 International Anesthesia Research Society
34. Sinatra RS, Shen QJ, Halaszynski T, Luther MA, Shaheen Y. Preoperative rofecoxib oral suspension as an analgesic adjunct after lower abdominal surgery: the effects on effort-dependent pain and pulmonary function. Anesth Analg 2004;98:135–40