| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
From the *Department of Anesthesiology and Intensive Care Medicine, University of Saarland, Homburg/Saar, Germany; and Department of Anesthesiology and Intensive Care Medicine, St.-Marien-Hospital Lünen, Lünen, Germany.
Address correspondence and reprint requests to Ulrich Grundmann, MD, Department of Anesthesiology and Intensive Care Medicine, University of Saarland, D-66421 Homburg/Saar, Germany. Address e-mail to grundmann.ulrich{at}web.de.
 |
Abstract |
|---|
 Top Abstract IntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
METHODS |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
After informed consent, patients were assigned randomly to one of four study groups. Randomization was performed using a sealed opaque envelope with a computer-generated block random allocation (http://www.randomizer.org/). The four study groups were 1) parecoxib 40 mg, 2) paracetamol 1 g, 3) metamizol 1 g, and 4) placebo. The patients received the drugs dissolved in 100 mL normal saline via IV infusion over 15 min. Patients of the placebo group received 100 mL of normal saline as a control. The study solutions were prepared by one of the researchers (WW) who was not involved in the intraoperative and postoperative treatment of these patients, whereas postoperative data were collected by another anesthesiologist (CW) who was blinded as to the drugs used. The group assignment code was retained until the conclusion of the study. To ensure patient safety, however, a sealed opaque envelope containing the name of the drug was kept with the patient in the operating room (OR), the postanesthesia care unit (PACU), and on the ward to permit immediate unmasking if necessary.
All patients had fasted overnight and were premedicated with 10 mg diazepam orally both the night before surgery and 90 min before induction of anesthesia. On arrival in the OR an 18-gauge cannula was inserted into a forearm vein and standard monitors were applied. Heart rate, oxygen saturation, and the bispectral index of the electroencephalogram (BIS A-2000 monitor, software version XP; Aspect Medical Systems Inc., Newton, MA) were monitored continuously and noninvasive arterial blood pressure was recorded every 5 min throughout anesthesia. Before induction all patients received 5 mL/kg of IV fluid over 20 min. Oxygen was administered via an anesthetic breathing circuit and facemask. After 5 min of administration of oxygen anesthesia was induced in all patients with 2 mg/kg propofol and 2 µg/kg fentanyl followed by 0.5 mg/kg atracurium to facilitate tracheal intubation and to maintain neuromuscular blockade, monitored by the train-of-four stimulation method using a peripheral nerve stimulator. After tracheal intubation, patients received a total IV anesthesia with 0.25 µg · kg–1 · min–1 remifentanil and 3–6 mg · kg–1 · h–1 propofol. The infusion rate of propofol was adjusted to maintain the Bispectral index values between 40 and 45. The lungs were ventilated with an air/oxygen mixture (Fio2 0.35–0.45) at a flow rate of 1 L/min in a semiclosed system (Cato; Dräger AG, Lübeck, Germany). The inspired oxygen and end-tidal concentrations of carbon dioxide (CO2) were measured continuously at the proximal end of the endotracheal tube using a calibrated infrared gas analyzer (Dräger pm 8050, Dräger). Ventilation was adjusted to maintain normocapnia as measured by end-tidal CO2. Forty-five minutes before the expected end of surgery the non-opioid analgesics were infused over a period of 15 min. At the completion of surgery the administration of maintenance anesthetics was discontinued without tapering. The lungs of each patient were ventilated with 100% oxygen at a flow rate of 5 L/min. Spontaneous recovery of neuromuscular function was confirmed by train-of-four monitoring. The trachea was extubated when adequate spontaneous ventilation (tidal volume >5 mL/kg) and response to verbal commands were established. The last surgical stitch was defined as the end of surgery. During emergence from anesthesia the following recovery times were recorded: end of surgery until spontaneous opening of eyes, tracheal extubation, stating name and date of birth.
Thereafter, the patients were directly transferred to the PACU, where further clinical observations were done by an independent, blinded observer who was unaware of the administered study drugs. During a 120-min observation period hemodynamic variables were recorded. The severity of postoperative side effects (nausea, vomiting, shivering, and pruritus) was quantified using different scales. The scale for nausea, pruritus, and shivering was 0 = none, 1 = mild, 2 = severe. The scale for postoperative vomiting was 0 = none, 1 = 1 or 2 episodes, and 3 = more than 2 episodes. In the PACU patients were asked every 10 min to quantify their pain experience at rest on a visual analog scale (VAS) between 0 and 100, with 0 representing no pain and 100 the worst imaginable pain. Postoperative pain was treated by self-administration of small doses of IV piritramide using a patient-controlled analgesia (PCA) pump (Injectomat®, PCA-PACOM; Fresenius AG, Germany). The PCA was programmed to administer a bolus dose of 3 mg piritramide on patient demand, with a 2 h limit of 30 mg and a lockout interval of 5 min. The cumulative piritramide consumption at discharge from the PACU and side effects were recorded. During the preoperative assessment patients were instructed how to use the VAS and the PCA pump. Furthermore any intraoperative and postoperative adverse events (e.g., alterations of heart rate and arterial blood pressure) after the administration of the study drugs were assessed and recorded.
Data were processed in Microsoft® Excel 2002, SigmaStat® 3.0.1 and SigmaPlot® 8.02 (SPSS Science Software GmbH, Erkrath, Germany). To calculate the number of patients needed for the study, we evaluated the anesthetic and postoperative charts of patients in our hospital that had been operated on before the start of this study. Based on the assumptions that 1) 1 g metamizol would produce a 55% decrease in the VAS score (sd, 14) and 2) a type I error 
= 0.05 and Type II error ß = 0.2, an a priori power analysis suggested a sample size of 18 patients for each group. To compensate for dropouts, e.g., withdrawal of consent or technical problems with the PCA pump, 20 patients per group were included. Mean values and sd were calculated for all variables. Differences among the groups were calculated by one-way analysis of variance followed by pairwise comparisons using a post hoc Student-Newman-Keuls-test. Categorical data were analyzed using 
2 or Fisher’s exact test as appropriate. Differences were judged significant at P < 0.05.
|
RESULTS |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The four groups were similar with respect to gender, age, height, body mass index, ASA physical status, and duration of surgery. Despite random assignment, patients of the parecoxib group had a slightly higher body weight than those in the placebo group (Table 1). None of the study drugs had any major effects on the hemodynamic variables, and there were no significant between-group differences. Emergence from anesthesia was comparable for all groups studied (Table 2). There was no significant difference in the time required for emergence from anesthesia as defined by eye opening, time to tracheal extubation, and orientation time (time at which each patient was able to state his or her name and date of birth).
|
|
Data of the PACU evaluation are presented in Table 3. At arrival in the PACU postoperative VAS pain scores were significantly lower in the metamizol group compared with the paracetamol, parecoxib, and placebo groups. In addition in the metamizol group, significantly fewer patients required additional PCA compared with the other groups. However, in patients requiring additional pain therapy there was no significant difference in time to first request for piritramide and cumulative consumption of piritramide as assessed by the PCA data in the PACU. At discharge from the PACU 2 h after admission, VAS pain scores at rest were similar in all groups. Overall, we observed infrequent nausea and vomiting although the patients had not received an antiemetic prophylaxis. One patient in each treatment group suffered from mild nausea, and there was one patient receiving parecoxib with one episode of vomiting immediately after arrival in the PACU. Shivering was rare and mostly mild in all groups except for placebo, but this difference did not reach statistical significance. None of the patients complained of pruritus and no other complications were observed. The incidence of agranulocytosis was not addressed in any special way in the study design, but routine postoperative blood cell counts revealed no evidence of this adverse effect and all the patients were discharged from the hospital in good condition.
|
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Compared with previous studies addressing pain relief after lumbar disk surgery (4,6) the magnitude of the VAS scores in patients receiving placebo were minor in our trial. This low pain intensity might be explained, in part, by the microsurgical technique resulting in a small tissue trauma and by successful removing of herniated nucleus pulposus material causing radicular symptoms. Because the microsurgical technique was used in the previous studies as well, the reason for this difference remains unclear.
Although the introduction of microsurgical techniques has reduced analgesic requirements, postoperative pain caused by the operation itself or by the wound is unavoidable. Opioid analgesics are the traditional first-line medication in this setting but may induce unwanted side effects, such as nausea and vomiting, sedation, and respiratory depression. Several studies have shown that non-opioid analgesics provide effective pain relief in patients with acute postoperative pain after lumbar disk surgery, either as a substitute for or as an adjunct to opioid analgesia (4,7,8). However, this is the first prospective, randomized, double-blind, placebo-controlled study that compares IV administered parecoxib, paracetamol, and metamizol for pain relief in the early postoperative period after lumbar disk surgery.
In the present study the injectable form of analgesics was chosen, as in the perioperative setting many patients cannot tolerate oral medication or may have variable gastrointestinal absorptive function. Parecoxib, the only parenterally administered coxib available, is a prodrug that is converted in the liver to its active metabolite valdecoxib (9). In the paracetamol group we administered a new, ready-to-use IV solution of paracetamol (PerfalganTM 10 mg/mL; Bristol-Myers Squibb GmbH, München, Germany). Results of a bioequivalence study comparing this new paracetamol solution and the IV prodrug propacetamol indicate that 1 g of paracetamol administered as PerfalganTM is equivalent to 2 g of propacetamol with minor application side effects (10). The third analgesic used in this study, metamizol, a pyrazolone derivate, provides additional antipyretic, antispasmodic, and antiinflammatory effects. It is a very popular non-opioid analgesic in Germany, Spain, and South America whereas in other countries it has been banned because of its disputed association with potentially life-threatening agranulocytosis (11).
Although there is no doubt that metamizol may cause agranulocytosis, reports on the risk of metamizol-associated agranulocytosis suggest widely varying estimates. Although Hedenmalm and Spigset (12) reported an incidence of 1 case per 1431 prescriptions in Sweden, Ibanez et al. (13) concluded that in Spain the absolute risk of metamizol-associated agranulocytosis at usual doses and for short treatment periods is very small, with a calculated incidence of 0.56 cases per million inhabitants per year. Thus, considering these data, uncertainty remains and the only way to clarify the real incidence would be to do a large prospective study in countries that are routine users of metamizol. Until then, patients should probably be monitored for blood dyscrasias, especially if long-term use is intended. If agranulocytosis occurs, therapy with broad-spectrum antibiotics and hematopoietic growth factors will reduce the mortality in those patients. However, in the discussion of metamizol-induced agranulocytosis the risk of this side effect should be considered in comparison with other potentially life-threatening adverse effects of alternative analgesics. Although NSAIDs account for a substantial risk of gastrointestinal bleeding, renal failure, or severe skin reaction, metamizol is relatively safe concerning these side effects. Andrade et al. (14), evaluating epidemiological studies of non-narcotic analgesic safety published from 1970 to 1995, estimated the excess mortality as a result of agranulocytosis, aplastic anemia, anaphylaxis, and serious upper gastrointestinal complications as 25/100 million for metamizol, 20/100 million for paracetamol, 185/100 million for aspirin, and 592/100 million for diclofenac; most of these complications were related to gastrointestinal side effects. The authors concluded that a relative risk estimate of 300 or more for the association of metamizol with agranulocytosis would have been necessary for the excess mortality of metamizol to be comparable to that of aspirin or diclofenac.
Although all the drugs studied belong to the group of non-opioid analgesics, they act by different mechanisms. The analgesic action of parecoxib results from the inhibition of the COX-2 isoenzyme that plays an important role in the synthesis of prostaglandin E2 in the traumatized area by increasing the threshold of activation of the nociceptors. In contrast, despite the long use of metamizol and paracetamol their mode of action is still not fully understood. Generally considered as belonging to the NSAIDs there is only a weak inhibition of prostaglandin synthesis and a lack of other typical actions of NSAIDs, such as antiplatelet activity and gastrotoxicity, suggesting a distinct mode of action. A third COX isoenzyme, COX 3, was identified, and the inhibition of this isoform could represent a primary central mechanism by which these drugs decrease pain (15).
As information indicating equipotent doses of the non-opioid analgesics are still missing for lumbar disk surgery, we decided to choose the maximum doses recommended by the manufacturers for initial IV use in adults, i.e., parecoxib 40 mg, paracetamol 1 g, and metamizol 1 g, respectively. Thus, even if it is assumed that in the present study the dosing of paracetamol and parecoxib was too small for adequate pain relief when compared to metamizol, simply increasing the dose of paracetamol or parecoxib cannot be recommended.
A comparison of our results concerning the analgesic efficacy with the results of previous studies is difficult because of different application times, routes, and dosages used for the respective non-opioids. In a placebo-controlled study designed to evaluate the concept of balanced analgesia after vertebral disk surgery Fletcher et al. (16) compared IV propacetamol and ketoprofen given alone or in combination postoperatively. Similar to the present study with paracetamol, pain scores and cumulative morphine consumption did not differ in the group receiving propacetamol compared with placebo. Hans et al. (17) also failed to demonstrate an analgesic efficacy of propacetamol hydrochloride in the early postoperative period after lumbar disk surgery when administered at the end of the procedure. Because data for parecoxib and metamizol are missing for lumbar spine surgery, we have to consider the results of trials in different types of surgery. Avellaneda et al. (18), using a dosage of 2 g metamizol and a dosage of 1 g propacetamol, demonstrated comparable analgesic effects of metamizol and propacetamol after heart surgery. Consistent with the results of our study, metamizol provided better pain relief than paracetamol after oral administration of 1 g every 6 h of the respective analgesic after ambulatory hand surgery (19). The effect of parecoxib on postoperative pain relief was studied in minor and major surgical procedures. In contrast to our findings, parecoxib usually demonstrated analgesic efficacy superior to placebo and similar to ketorolac (20–22).
There are possible concerns regarding the side effects of the analgesics used in this study. Both metamizol and paracetamol are reported to cause hypotension that may be poorly tolerated by critically ill patients (18,23) or by patients in a prone position, which is usually used for lumbar spine surgery. In contrast, based on the results obtained in the present study, the three study drugs were well tolerated in this relatively healthy population with no difference from placebo. Of note, the analgesics showed no adverse hemodynamic effects, probably because they were continuously infused over 15 minutes rather than injected as a bolus. In the present study there were no adverse cardiovascular events such as myocardial infarction, thromboembolism, or deep vein thrombosis. It must be emphasized that none of the studies evaluating COX-2-inhibitors for postoperative analgesia reported an increased risk of cardiovascular events, except for two patient studies after coronary artery bypass graft surgery comparing IV parecoxib/oral valdecoxib versus placebo (24,25). Based on these data, valdecoxib and parecoxib are considered to be contraindicated in patients undergoing coronary artery bypass graft surgery. The precise mechanism responsible for the increased cardiovascular risk is unknown, but there is a mechanistic hypothesis that COX-2 inhibitors might increase the risk of thromboembolic events because of the inhibition of prostacyclin in the vascular endothelium without the concomitant inhibition of platelet thromboxane A2 generation, causing an imbalance of procoagulatory and anticoagulatory factors. The relevance of the increased cardiovascular risk caused by coxibs after long-term treatment for patients on short-term postoperative pain therapy is unclear. Nevertheless, the German Drug Safety Board ("Arzneimittelkommission der Deutschen Ärzteschaft") recommends not using coxibs in the perioperative setting (26); however, this recommendation was published several weeks after the present study had been finished. It is noteworthy that in postmarketing experience serious skin reactions have been described with parecoxib and valdecoxib, and the reported rate appears to be more than with other COX-2-inhibitors. Obviously patients appear at highest risk for these events early in the course of therapy and also patients without a history of sulfonamide allergy are at risk. Based on these new data, the Food and Drug Administration and the European Agency for the Evaluation of Medicinal Products now recommend discontinuing the use of parecoxib/valdecoxib at the first appearance of skin rash, mucosal lesions, or any other signs of hypersensitivity.
In this study piritramide, a synthetic opioid structurally related to meperidine, was chosen for PCA. Its relative analgesic potency compared with morphine is approximately 0.7. Because piritramide is distributed extensively and eliminated slowly, the pharmacokinetic profile of the drug allows for intermittent bolus administration even when relatively constant effect compartment concentrations are desirable, e.g., for PCA (27).
Despite avoiding antiemetic prophylaxis, the incidence of postoperative nausea and vomiting in all groups was less than previously reported (6). Besides patient-related factors, the use of total IV anesthesia with propofol may have contributed to these results, as previous studies indicated that propofol appears to have an inherent antiemetic effect (28,29).
In terms of pharmacoeconomics, acquisition costs at our hospital are 0.18
for 1 g of metamizol, 6.86 for 40 mg of parecoxib, and 2.12 for 1 g of paracetamol. Thus, postoperative pain treatment with metamizol is not only the most effective but also the least expensive method. This cost-saving effect will be much more pronounced if the analgesics are continued to the maximal dose for daily use of metamizol 5 g, parecoxib 80 mg, and paracetamol 4 g respectively.
The study also has certain limitations. The lack of significant differences among the analgesic effects of paracetamol, parecoxib, and placebo must be interpreted with some caution, as they could be real or could be related to the methodology of the study evaluating patients undergoing lumbar microdiscectomy, which might have lacked the appropriate assay upside sensitivity related to the moderate level of pain (VAS < 30), probably as a result of the microsurgical technique and successful removing of herniated nucleus pulposus material. In addition, pain intensity was evaluated only at rest; thus it remains unclear if there would be identical results for movement-related pain relief. Consequently, further studies in more painful surgical models are needed for clarification. Furthermore, the limited period of evaluation (2 hours in the PACU) was not long enough to give adequate long-term outcome information. However, with respect to the investigated drugs and side effects, our data sufficiently demonstrate that non-opioid treatment is not associated with an increase of adverse effects compared with placebo.
In conclusion, in patients undergoing lumbar microdiscectomy, the IV administration of a single dose of metamizol 1 g provides significantly better pain control in the early postoperative period compared with other non-opioids without increasing adverse side effects. Parecoxib and paracetamol failed to improve postoperative pain relief when compared with placebo, but this lack of differences must be interpreted with some caution because of the low VAS scores observed in this study.
|
Footnotes |
|---|
|
REFERENCES |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  G. Riest, J. Peters, M. Weiss, S. Dreyer, P. D. Klassen, B. Stegen, A. Bello, and M. Eikermann Preventive effects of perioperative parecoxib on post-discectomy pain Br. J. Anaesth., February 1, 2008; 100(2): 256 - 262. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
