Intravenous Ketoprofen in Postoperative Pain Treatment After Major Abdominal Surgery

World J. Surg. 29, 446–449 (2005) DOI: 10.1007/s00268-004-7612-0

Intravenous Ketoprofen in Postoperative Pain Treatment after Major Abdominal Surgery

Dagmar Oberhofer, M.D., M.S. Jasna Skok, M.D., M.S. Visnja Nesek-Adam, M.D., M.S.

Department of Anaesthesiology and Intensive Care, Sveti Duh General Hospital, Sveti Duh 64, 10000 Zagreb, Croatia

Published Online: March 23, 2005

Abstract. In recent years considerable attention has been paid to the mended in the management of postoperative pain and thus to treatment of postoperative pain, with regard to the favorable effect of improve analgesia (additive or synergistic effect) and to reduce adequate analgesia on patient outcome. Multimodal analgesia (e.g., opioids and nonsteroidal anti-inflammatory drugs [NSAIDs] or local anes- their side effects[3–5]. Systemic use of opioids (intramuscular, thetics) is recommended for effective postoperative pain relief. There are intravenous, continuous infusion, patient-controlled analge- few data on the use of NSAIDs in postoperative pain treatment after sia—PCA) remains the most widely employed method of abdominal surgery. We conducted a randomized, double-blind, placebo- treatment for moderate to severe postoperative pain. Recent controlled study to assess the analgesic efficacy and safety of ketoprofen studies have shown the addition of non-steroidal anti-inflam- after major abdominal surgery. One and nine hours postoperatively patients received 100 mg of ketoprofen i.v. (n = 21) or placebo (n = 22) in matory drugs (NSAIDs) to opioid therapy to decrease the need addition to a pain-treatment protocol consisting of continuous infusion of of the latter by 20% to 30%, to improve analgesia, and to pos- tramadol 200 mg and metamizol 5 g over 24 hours with additional 25 mg sibly reduce the rate of opioid-related side effects, primarily i.v. tramadol in case of inadequate analgesia. Pain was assessed by nu- respiratory depression and nausea and/or vomiting [1]. Rela- meric rating scale at rest and at deep breath 3, 6, 12, and 24 hours postoperatively and the total dose of tramadol used in the first 24 hours tively few studies have tackled the issue of postoperative anal- was recorded. Patients in the ketoprofen group had significantly lower gesia by combined administration of opioids and NSAIDs after pain scores both at rest and at deep breath, at 3 (p < 0.01), 6, and 12 major abdominal surgery [4, 6–12], which is associated with the hours (p < 0.05) postoperatively. The 24-hour use of tramadol was sig- most severe pain after thoracic procedures. p nificantly lower in the ketoprofen group ( < 0.01), with less nausea and Ketoprofen is a NSAID, a phenylpropionic acid derivative, vomiting. There were no bleeding complications or other adverse events related to ketoprofen therapy. The study showed the value of short-term that has been in clinical use since 1973, and is available as an use of ketoprofen to improve the quality of analgesia after major ab- intravenous (i.v.) formulation [13]. The aim of the present dominal surgery without significant adverse effects. study was to assess the analgesic efficacy of i.v. ketoprofen administered with continuous infusion of the weak synthetic opioid tramadol [14] and metamizol, a pyrazolone derived analgoantipyretic [13] during the first 24 hours after major abdominal surgery. We also recorded possible complications Postoperative pain control frequently proves inadequate in spite associated with ketoprofen therapy, excessive bleeding and re- of considerable advancement in the understanding of pain nal dysfunction. pathophysiology. The fear of pain is a factor for most patients scheduled for an operative procedure. In addition to its favorable psychological effect, efficient pain relief is also considered Patients and Methods to reduce the stress response (endocrine-metabolic and inflam- The study protocol was approved by the hospital Ethics Com- matory) to surgical trauma and to stimulate early patient mittee, and a written informed consent was obtained from all mobilization, thus improving the procedure outcome [1, 2]. The patients before the study. The study included 44 adult patients use of a single analgesic to treat moderate to severe postoper- undergoing major abdominal surgery (upper or upper and lower ative pain has proved inadequate to ensure optimal analgesia. medial laparotomy) and treated postoperatively in the surgical Therefore, multimodal analgesia—i.e., administration of agents intensive care unit (ICU). Exclusion criteria were as follows: and methods of different sites and modes of action—is recom- preoperative use of analgesics, allergy to NSAIDs, history of peptic ulcer disease, coagulopathy, or renal dysfunction. Patients Presented in part at the 9th Symposium on Intensive Care Medicine, June 24–27, 2002, Brijuni Islands, Croatia, and at the Euroanaesthsia 2004 submitted to abdominoperineal resection of rectum for carcinoma Meeting, June 5–8, 2004, Lisbon, Portugal. were also considered ineligible because of extensive surgical trauma and an increased risk of postoperative hemorrhage. The Correspondence to: Visnja Nesek-Adam, M.D., M.S., e-mail: most common indications for the operation were colon and gas- [email protected] tric neoplasms, obstructive icterus of malignant etiology, liver Oberhofer et al.: Ketoprofen for Postoperative Analgesia 447

Table 1. Patient and surgery characteristics. Table 2. Pain assessment at rest using numeric rating scale (NRS) at different times after surgery. Control (n = 22) Ketoprofen (n = 21) Time (hours) Control (n = 22) Ketoprofen (n = 21) p Value Age (years) 66 (8.3) 63 (11.6) Sex (M:F) 13:9 12:9 3 4.8 (1.96) 3.0 (1.74) 0.002 Weight (kg) 69 (12.1) 73 (13.4) 6 3.7 (2.14) 2.5 (1.81) 0.046 ASA grade 12 2.9 (2.61) 1.5 (1.24) 0.030 2 13 (59.1%) 12 (57.1%) 24 0.75 (1.04) 0.4 (0.82) n.s. 3 7 (31.8%) 9 (42.9%) 4 2 (9.1%) Values are expressed as mean (S.D.); n.s. = non-significant Surgical procedure Colon surgery 12 (54.5%) 9 (42.9%) (resections, reanastomoses) Gastric surgery 5 (22.7 %) 5 (23.8 %) Statistics Liver resection 1 (4.8 %) Biliary-digestive anastomosis 3 (13.6 %) 2 (9.5 %) Sample size was calculated on the assumption that the analgesic for bile obstruction effect of ketoprofen manifested by a 30% NRS score and/or total Miscellaneous 2 (9.1%) 4 (19.0 %) tramadol dose reduction would be clinically relevant. With a Values are mean (S.D.) or number (proportion). mean NRS score of 4.9 (1.7) 2–4 hours after abdominal surgery recorded for the in-house pain treatment protocol (tramadol + metamizol infusion), the level of significance of 5% (a = 0.05) and study power of 80% (1 ) b = 0.80), each group should have metastases, retroperitoneal tumor, and other intraabdominal included 21 patients [15]. Data are presented as mean and stan- pathology. dard deviation (S.D.), unless stated otherwise. Patient data were Forty-four patients were randomized to the control (n = 23) or not statistically analyzed, because patients were randomly as- ketoprofen (n = 21) group. All patients received general anes- signed to the control or ketoprofen group. Within-group Kol- thesia. The agents used for induction of anesthesia were as con- mogorov-Smirnov testing yielded normal data distribution. sidered appropriate by the anesthesiologist in charge of the StudentÕs t-test was used to compare NRS score and 24-hour patient. Balanced anesthesia with 40% oxygen in nitrous oxide, tramadol use between the groups. A p value <0.05 was con- isoflurane, fentanyl, and a neuromuscular blocking drug chosen sidered statistically significant. by the anesthesiologist was used for maintenance. At the end of the surgical procedure patients were extubated and transferred to Results the surgical ICU. Upon ICU admission and vital sign assessment, pain severity was evaluated by use of the numeric rating scale Forty-four patients entered the study. One patient from the (NRS), previously explained to the patient On NRS, 0 denotes no control group was excluded 10 hours after the initiation of pain and 10 denotes the most severe pain. Patients with A NRS tramadol and metamizol infusion for the occurrence of general- score at rest ‡4 received additional fentanyl (Jannsen, Belgium) in ized urticarial rash. The patient received no antibiotic or other a dose of 0.05 mg for NRS £ 3. This was followed by continuous medication that may have induced the allergic reaction either at infusion of 200 mg of tramadol (Bayer Pharma, Germany) and 5 g the time of or immediately before the reaction. The control and of metamizol (Pliva, Croatia) in 500 ml of saline over 24 hours in ketoprofen groups were comparable with respect to age, gender, all patients, with an added 25 mg i.v. bolus of tramadol in case of body weight, ASA (American Association of Anesthesislogists) inadequate analgesia. A NRS score ‡4 at rest or patientÕs request status, and type of operation (Table 1). for analgesia was considered indicative of inadequate analgesia. Pain assessment by NRS at rest and at deep breath at dif- In addition to tramadol and metamizol, the ketoprofen group ferent postoperative time intervals is presented in Tables 2 and patients were administered 100 mg of ketoprofen (Lek, Slovenia) 3, respectively. The NRS score was significantly lower in the in 100 ml of saline over 30 minutes at 1 and 9 hours postopera- ketoprofen group at 3, 6, and 12 hours postoperatively, both at tively. The control group patients received placebo—i.e., 100 ml rest and at deep breath. The difference reached a higher level of of saline—at the same time intervals. The bottles containing sal- significance at 3 hours (p < 0.01) than at 6 and 12 h (p < 0.05) ine with ketoprofen or placebo were protected by aluminium foil postoperatively, both at rest and at deep breath. The 24-hour for ketoprofen photosensitivity. All patients received ranitidine use of tramadol was significantly lower (p < 0.01) in the keto- i.v. postoperatively. The NRS assessment of pain severity was profen group, 193 (31) mg than in the control group, 235 (40) done by the nurse on duty at 3, 6, 12, and 24 hours postoperatively mg (Table 4). Respiratory depression was not observed in any of at rest and at deep breath. The 24-hour use of tramadol, side the ketoprofen or control group patients. The incidence of effects in terms of nausea and/or vomiting, and the possible nausea and/or vomiting was higher in the controls (n = 7) than occurrence of excessive postoperative bleeding were recorded in in the ketoprofen (n = 4) group. During the postoperative each patient Platelets, prothrombin time, activated partial period, none of the ketoprofen group patients developed thromboplastin time, fibrinogen, urea, and creatinine were excessive bleeding; coagulation tests and platelet count were determined on postoperative days 1 and 2 and compared with within the normal limits on postoperative day 1. A slight urea preoperative values for the possible ketoprofen effect on platelet increase was recorded in the early postoperative period in one and renal function. The patients operated on for carcinoma and patient from the ketoprofen and control group each; the latter those at an increased risk of thromboembolic complications re- also showed a creatinine increase. However, urea and creatinine ceived low molecular weight heparin prophylaxis 12 hours post- normalized in both patients on postoperative day 2 after volume operatively. replacement therapy. 448 World J. Surg. Vol. 29, No. 4, April 2005

Table 3. Pain assessment at deep breath using numerical rating scale Table 4. Twenty-four hour tramadol consumption. (NRS) at different times after surgery. Control (n = 22) Ketoprofen (n = 21) p Value Time (hours) Control (n = 22) Ketoprofen (n = 21) p Value Tramadol 235 (40.1) 193 (31.0) < 0.001 3 5.9 (1.74) 4.3 (1.77) 0.005 consumption (mg) 6 4.9 (1.70) 3.6 (1.83) 0.019 12 4.2 (2.08) 2.8 (1.36) 0.014 Values are expressed as mean (S.D.). 24 2.2 (1.27) 1.6 (1.53) n.s.

Values are expressed as mean (S.D.); n.s. = non-significant. arachidonic acid to prostaglandins and thromboxanes. Some of these prostaglandins have a protective action on the gastric mu- Discussion cosa and renal hemodynamics, and thromboxane A2 is responsi- ble for platelet aggregation. The major concern in the Study results showed the systemic use of the NSAID ketoprofen in perioperative use of NSAIDs is their potential to increase combination with tramadol and metamizol improved the quality of bleeding. Only in one study [10] did the authors find an increased analgesia after major abdominal surgery, manifested by a signifi- incidence of hemorrhagic complications in patients who received cantly lower NRS score in the early postoperative period. The indomethacin, as compared to placebo, after major abdominal higher efficacy of analgesia was associated with a lower total dose of surgery. In their study of the analgesic efficacy of diclofenac after tramadol administered over the first 24 hours after the operation. major abdominal surgery, Hodsman et al. [7] report on two dic- This finding is consistent with the results reported from other lofenac group patients submitted to abdominoperineal resection studies investigating the analgesic efficacy of various NSAIDs in of the rectum with extensive pelvic dissection. Both patients combination with opioids, most commonly morphine, in this cate- developed postoperative bleeding and required reoperation. We gory of surgical patients [4, 6–12]. In the studies of postoperative did not include patients undergoing abdominoperineal resection analgesia after gynecological, orthopedic, or abdominal surgery, of the rectum in our series because of extensive surgical trauma various NSAIDs were used, usually ketorolac, diclofenac, and ke- and a large raw area, which increases the risk of postoperative toprofen, all of which are available in parenteral formulation. Al- hemorrhage. In our study, there was no case of excessive bleeding though these agents were generally administered by i.m. injection, in the ketoprofen group during the postoperative period, and we considered this route of administration inappropriate for the none of the patients from either group required reoperation. It is high level of patient discomfort. The more so, in case of repeated worthwile to note that most of our patients received thrombo- NSAID doses, i.m. injections would have been contraindicated by prophylaxis with low molecular weight heparin postoperatively. the low molecular weight heparin thromboprophylaxis initiated 12 None of the patients developed gastrointestinal side effects or hours after the operation. Total morphine dosage administered i.v. renal dysfunction attributable to the use of ketoprofen. This is by a PCA system over a particular period of time was in many in keeping with other studies where short-term perioperative use studies used as a measure of postoperative analgesia efficacy. This (< 1 week) of NSAIDs did not increase the incidence of side method of postoperative analgesia has not yet been routinely used effects when patients at risk were excluded [3, 4]. in our department. Instead, we use continuous infusion of 200–300 In conclusion, i. v. ketoprofen used as an adjuvant to an mg tramadol and 5 g metamizol in 500 ml of saline over 24 hours, opioid-based postoperative pain management protocol signifi- with the addition of 50 mg tramadol i.v. in case of inadequate cantly improves the quality of postoperative analgesia after analgesia. The combination of tramadol and metamizol has an major abdominal surgery, as demonstrated by reduced postop- additive or synergistic action [16], and according to our experience, erative pain scores and 24-hour opioid use. There was no in- the protocol mentioned above allows for good postoperative anal- crease in the risk of bleeding, and use of ketoprofen may reduce gesia with a minimal risk of significant side effects such as respi- the incidence of nausea and vomiting. Therefore, we suggest ratory depression. A similar protocol of continuous tramadol and that NSAIDs be more often included, along with systemic opi- ketorolac infusion over 48 hours in 585 patients submitted to major oids, in the multimodal treatment of postoperative pain after abdominal surgery [17] showed a high quality of postoperative pain major surgery. relief without any serious side effects at a limited cost. The PCA method enables more precise opioid titration tailored to individual References patient needs than continuous infusion with intermittent i.v. bo- luses. That is why in many studies using the PCA method of opioid 1. Kehlet K, Holte K. Effect of postoperative analgesia on surgical administration in the postoperative period, the improved quality of outcome. Br. J. Anaesth. 2001;87:62–72 2. Gottschalk A, Wu CL, Ochroch EA. Current treatment options for pain relief attributable to the use of an NSAID manifested pri- acute pain. Expert Opin. Pharmacother. 2002;3:1599–1611 marily as a total opioid dose reduction. In the present study, the 3. Kehlet H, Dahl JB. The value of ‘‘multimodal’’ or ‘‘balanced anal- main effect was subjective pain relief, as manifested by a lower NRS gesia’’ in postoperative pain treatment. Anest. Analg. 1993;77:1048– score at 3, 6, and 12 hours postoperatively. The NRS was chosen for 1056 pain assessment, as it was found to be easier to use than the visual 4. Jin F, Chung F. Multimodal analgesia for postoperative pain control. J. Clin. Anesth. 2001;13:524–539 analogue scale in elderly patients, who frequently suffer visual and 5. Dahl JB, Kehlet H. Non-steroidal anti-inflammatory drugs: rationale coordination difficulties. for use in severe postoperative pain. Br. J. Anaesth. 1991;66:703–712 Nonsteroidal anti-inflammatory drugs achieve their analgesic 6. Gillies GWA, Kenny GNC, Bullingham RES, et al. The morphine effect mainly in the periphery by reducing the concentration of sparing effect of ketorolac tromethamine. Anaesthesia 1987;42:727– 731 prostaglandins, which have an important role in the initiation of 7. Hodsman NBA, Burns J, Blyth A, et al. The morphine sparing effects pain signals. They inhibit cyclooxygenases (COX-1 and COX-2), of diclofenac sodium following abdominal surgery. Anaesthesia the enzymes that are involved in the oxidative metabolism of 1987;42:1005–1008 Oberhofer et al.: Ketoprofen for Postoperative Analgesia 449

8. Burns JW, Aitken HA, Bullingham RES, et al. Double-blind com- 13. Roberts, LJ, II, morrow, JD (2001) "Analgesic-antipyretic and anti- parison of the morphine sparing effect of continuous and intermit- inﬂammatory agents and drugs employed in the treatment of gout" In: tent I.M. administration of ketorolac. Br. J. Anaesth. 1991;67:235– Hardman, JG, Limbird, LE, Goodman, A (editors), McGraw-Hill, 238 New York, pp 687-731 9. Ready LB, Brown CR, Stahlgren LH, et al. Evaluation of intravenous 14. Bamigbade TA, Langford RM. The clinical use of tramadol hydro- ketorolac administered by bolus or infusion for treatment of post- chloride. Pain Rev. 1998;5:155–182 operative pain. Anesthesiology 1994;80:1277–1286 15. Campbell MJ, Julious SA, Altaian DG. Estimating sample sizes for 10. Reasbeck PG, Rice ML, Reasbeck JC. Double-blind controlled trial binary, ordered categorical, and continuous outcomes in two group of indomethacin as an adjunct to narcotic analgesia after major comparisons. Br. Med. J. 1995;311:1145–1148 abdominal surgery. Lancet 1982;2:115–118 16. Montes A, Warner W, Puig MM. Use of intravenous patient-con- 11. Rao AS, Cardosa M, Inbasegaran K. Morphine-sparing effect of trolled analgesia for the documentation of synergy between tramadol ketoprofen after abdominal surgery. Anaesth. Intens. Care and metamizol. Br. J. Anaesth. 2000;85:217–223 2000;28:22–26 17. Pieri M, Meacci L, Santini L, et al. Control of acute pain after major 12. Turner GA, Gorringe J. Indomethacin as adjunct analgesia following abdominal surgery in 585 patients given tramadol and ketorolac by open cholecystectomy. Anaesth. Intens. Care 1994;22:25–29 intravenous infusion. Drugs Exp. Clin. Res. 2002;28:113–118