Antagonistic Effects of Ondansetron and Tramadol? A Randomized Placebo and Active Drug Controlled Study

The Journal of Pain, Vol, No 2 (December), 200: pp 274-28 Available online at www.sciencedirect.com Antagonistic Effects of Ondansetron and Tramadol? A Randomized Placebo and Active Drug Controlled Study Neele I. Rauers,* Frank Stüber, y Eun-Hae Lee,* Frank Musshoff, z Rolf Fimmers, x Martin Barann,* and Ulrike M. Stamer* * Department of Anesthesiology and Intensive Care Medicine, University of Bonn, Bonn, Germany. y Department of Anesthesiology and Pain Therapy, University Hospital of Bern, Inselspital, Bern, Switzerland. z Department of Forensic Medicine, University of Bonn, Bonn, Germany. x Department of Medical Biometry, Statistics, and Epidemiology, University of Bonn, Bonn, Germany. Abstract: Opposing effects of ondansetron and tramadol on the serotonin pathway have been suggested which possibly increase tramadol consumption and emesis when co-administered. In a randomized, double-blinded study, 79 patients received intravenous ondansetron, metoclopramide, or placebo for emesis prophylaxis. Analgesic regimen consisted of tramadol intraoperative loading and subsequent patient-controlled analgesia. Tramadol consumption and response to antiemetic treatment were compared. Additionally, plasma concentrations of ondansetron and ()O-demethyltramadol and CYP2D6 genetic variants were analyzed as possible confounders influencing analgesic and antiemetic efficacy. Tramadol consumption did not differ between the groups. Response rate to antiemetic prophylaxis was superior in patients receiving ondansetron (85.0%) compared with placebo (66.7%, P =.046), with no difference to metoclopramide (69.5%). Less vomiting was reported in the immediate postoperative hours in the verum groups (ondansetron 5.0%, metoclopramide 5.%) compared with placebo (8.6%; P =.0). Whereas plasma concentrations of ()O-demethyltramadol were significantly correlated to CYP2D6 genotype, no influence was detected for ondansetron. Co-administration of ondansetron neither increased tramadol consumption nor frequency of PONV in this postoperative setting. Perspective: Controversial findings were reported for efficacy of tramadol and ondansetron when co-administered due to their opposing serotonergic effects. Co-medication of these drugs neither increased postoperative analgesic consumption nor frequency of emesis in this study enrolling patients recovering from major surgery. ª 200 by the American Pain Society Key words: Analgesia, emesis, tramadol, ondansetron, serotonin. Nausea and vomiting as well as pain are major adverse events during postoperative anesthesia care, having significant impact on patients wellbeing and influencing nursing time, length of hospital stay, frequency of readmission, and costs. The weak opioid tramadol is widely used for postoperative analgesia as respiratory and sedative effects are minor compared Received October 3, 2009; Revised February 0, 200; Accepted March 4, 200. Supported by institutional resources and a grant of the R. Sackler Research Foundation. Address reprint requests to Dr Ulrike M. Stamer, Department of Anesthesiology and Intensive Care Medicine, Rheinische Friedrich-Wilhelms- Universität, Sigmund-Freud-Str 25, 5305 Bonn, Germany. E-mail: ulrike. stamer@ukb.uni-bonn.de 526-5900/$36.00 ª 200 by the American Pain Society doi:0.06/j.jpain.200.03.003 with WHO III opioids with comparable efficacy in analgesia. 27,3,40,43 However, an increased risk of nausea and vomiting is associated with tramadol, specifically if larger doses are intravenously injected quickly. 30 Selective 5-hydroxytryptamine type 3 (5-HT 3 ) receptor antagonists such as ondansetron are well established treatment options and thus were suggested to be of particular value in reducing tramadol associated emesis. 6 However, concerns were reported for the combination of ondansetron and tramadol because co-administration appeared to increase analgesic consumption with a subsequent increase in emetic sequelae and ondansetron failure. 4,8,9 The involvement of both drugs in the serotonin (5-HT) pathway was discussed as possible reason, with tramadol inhibiting 5-HT reuptake and thus, increasing the concentration of this neurotransmitter within the synaptic cleft and ondansetron blocking 5-HT 274

Rauers et al The Journal of Pain 275 effects at its specific receptors. This might partially antagonize tramadol induced analgesia mediated via serotonin. Conflicting results from published studies prompted this randomized controlled trial in patients recovering from major abdominal surgery. In contrast to an experimental setting enrolling healthy volunteers, most patients in the present clinical trial were on concomitant medication typical for patients with collective ages 50 to 70 years. Ondansetron treatment was compared with metoclopramide (MCP) and placebo. MCP is a potent dopamine receptor antagonist known for its antiemetic and prokinetic properties. It is a standard treatment in the perioperative setting, although serotonin receptor antagonists are more widely used to date. 32,40 Recent publications highlighted additional antagonistic effects of MCP on 5-HT 3 receptors. However, for these 5-HT 3 antagonistic effects MCP doses have to be considerably higher than in previously published protocols. 45,46 To date, efficacy of MCP in the treatment of postoperative nausea and vomiting is questioned in general. 2,6,9 As polymorphic cytochrome P450 (CYP)2D6 is involved in the metabolism of both drugs, a pharmacogenetic impact on their efficacy has to be considered. For m-opioid receptor mediated analgesia tramadol has to be metabolized by CYP2D6 to its active M-metabolite ()O-demethyltramadol. 33,35,4 In contrast, serotonergic and (nor-)adrenergic transmitted analgesia is predominantly mediated by the parent drug. 0,36 Previous studies indicated higher tramadol consumption in poor metabolizers and an increased risk of ondansetron failure in CYP2D6 ultra-rapid metabolizers. 7,24,33,35,39 As only a minority of Caucasian individuals carry these genotypes we were aware of the fact that the number of patients displaying these genotypes would be small and might only explain therapeutic failure in a few patients. Thus, this pharmacogenetic analysis can only be considered as exploratory. The hypothesis of this study was that postoperative tramadol consumption is increased when coadministered with ondansetron compared with placebo or MCP. In addition, antiemetic efficacy represented a secondary end point. CYP2D6 genetic variants were defined as possible confounders impacting ()O-demethyltramadol and ondansetron plasma concentrations as well as therapeutic failure of analgesic or antiemetic medication. Methods Patients Approval of the study design was obtained from the Institutional Review Board of the University of Bonn. After giving written informed consent, 87 patients (ASA classification I-III) scheduled for elective abdominal surgery were instructed in the details of the study, the use of the patient-controlled analgesia (PCA) device, and the numeric rating scale for pain intensities (NRS: 0 denotes no pain, 00 denotes worst pain imaginable). Exclusion criteria comprised alcoholism, drug dependence, clinically relevant compromised kidney or liver function, psychiatric diseases, epilepsy, history of vomiting from any organic etiology, contraindication for the use of study medications, known opioid intolerance, serious perioperative complications, and changes in anesthetic procedure. Preexisting medication was documented and discontinued only for the day of surgery with the exception of anti-hypertensive drugs and other drugs for treatment of cardiac disease. The patients individual risk for PONV was assessed preoperatively by a validated risk score (Apfel simplified risk score 3 ). Randomization and Blinding A double blinded, placebo-controlled parallel group study design was chosen. Assignment to study medications MCP, ondansetron or placebo was performed by a computer generated randomization list. Study drugs were prepared as 0-mL syringes containing either MCP 0 mg, ondansetron 4 mg, or placebo (.9% saline) by the hospital pharmacy and labelled as study medication. Physicians, nurses, patients, as well as laboratory staff were blinded to study medication. Clinical Study Protocol General anesthesia and postoperative analgesia were conducted using a standardized protocol: propofol 2 to 3 mg/kg, fentanyl.5 mg and cis-atracurium for induction and remifentanil, isoflurane and cis-atracurium for maintenance of anesthesia. About 40 minutes before termination of anesthesia the blinded study medication O, MCP, or P was given intravenously. After that, tramadol 3 mg/kg (maximum loading dose, 250 mg) and dipyrone g were infused intravenously. No anticholinergic drugs and no reversal of muscle relaxation were used. After emergence from anesthesia, patients were transferred to the recovery room. As soon as communication was possible, patients were asked about pain and nausea/ emesis. In case of nausea/emesis, antiemetic rescue medication (I) dimenhydrinate 62 mg, (II) droperidol.25 mg was administered intravenously. The analgesic regimen in the PACU consisted of further doses of tramadol 50 mg if pain scores were >40 at rest. For subsequent analgesic treatment on the general ward, patients could self-administer intravenous bolus doses of tramadol 20 mg via PCA with a lock-out time of 8 minutes. If analgesia remained insufficient or it turned out that emesis was associated with tramadol bolus administration and could not be controlled by antiemetic rescue medication the analgesic regime was changed to piritramide. Piritramide was administered in 2-mg increments if patients had unresolved pain in the recovery room. If tramadol PCA had to be terminated prematurely due to side effects or lacking efficacy, the analgesic regimen was changed to piritramide via PCA using a 2-mg bolus dose (conversion ratio :0). Opioid consumption was documented and the protocol of analgesic consumption via PCA device was printed out after discontinuation of therapy.

276 The Journal of Pain Ondansetron Co-administered With Tramadol During the 48-hour study period, nausea and vomiting were assessed regularly for absence or presence and treated with antiemetic rescue medication if needed. In addition, patient pain scores under rest and movement/ coughing were recorded using the NRS. Observation time points for emesis and pain intensities were hourly up to the 8th hour, at 2 hours, and then every 6 hours up to the 48th hour. As individual experience and subjective estimation of nausea/vomiting episodes might differ considerably from investigators assessment an additional questionnaire was filled in by each patient after 48 hours. Questions aimed at nausea as well as vomiting during 3 time intervals: the first hours after surgery, during the first postoperative night, and on the first postoperative day. Laboratory Analysis Blood samples were drawn 30, 90, and 80 minutes after drug administration, centrifuged, and frozen at 80 C. All laboratory analyses were performed after enrollment of the last patient. For CYP2D6 genotyping, a PCR and real-time PCR-based (LightCycler, Roche Diagnostics, Mannheim, Germany) melting point analysis for detection of *3,*4,*5,*6,*7,*8,*0, and *4 alleles and the gene duplication/multiduplication was used. 39,4 According to the number of functionally active CYP2D6 alleles patients were assigned to group PM (poor metabolizers, no functionally active CYP2D6 gene), group HZ/IM (heterozygous individuals carrying only functionally active allele or intermediate metabolizers with decreased, but not absent enzymatic activity), group EM (extensive metabolizers with 2 active alleles and normal enzyme activity), or group UM (ultra-rapid metabolizers carrying a duplication/ multiduplication of the CYP2D6 gene resulting in increased enzyme production). Plasma concentrations of ()O-demethyltramadol were measured by a liquid chromatographic-tandem mass spectrometric procedure with atmospheric pressure chemical ionization (LC-APCI- MS/MS). 26 Ondansetron concentrations were determined in a similar way using a Chiralcel OD-R column with a mobile phase of water/acetonitrile 5 mm ammonium formiate in a gradient program with the following transitions: ondansetron: m/z/ 293.8 -> 70.0, 84.0; propanolol (internal standard) m/z/ 259.95 ->6., 82.9. The limit of detection (signal-to-noise ratio, :3) was.2 ng/ml, the limit of quantification (3 times the LOD) was.6 ng/ml, respectively. Statistical Analysis According to previous data, cumulative tramadol consumption was estimated to amount to 220 6 95 mg. Power analysis showed that 54 subjects per group were required to detect a -mg difference in tramadol consumption at a power of.8 and a 2-sided level of significance of.05. For comparison of cumulative tramadol consumption (mean 6 SEM), ANOVA followed by pairwise comparison of the 3 groups by t tests was used. For this closed test procedure, resulting P values were corrected by replacing them by the ANOVA P value if this was larger. Complete response rates to antiemetic prophylaxis (no nausea, no vomiting, no antiemetic rescue medication up to the 8th hour) as a secondary end point and individual patient assessment of nausea and vomiting were compared using Fisher s exact test for the comparison of all 3 groups as well as for subsequent pair wise comparisons. Mean AUC (area under the time-concentration-curve) for plasma concentrations of ondansetron and ()O-demethyltramadol were calculated and compared between CYP2D6 genotypes (ANOVA). Results Demographic Data and Genotypes Complete data of all 79 patients receiving study medication could be analyzed (Fig ). Demographic and surgery related data as well as preoperative Apfel scores did not differ between the study groups (Table ). Due to enrollment of a large number of patients with prostatic cancer, 64.8% of the participants were male. With the exception of 0 individuals, patients were of Caucasian (middle European) origin. Tramadol Consumption and Analgesic Efficacy There was no difference in tramadol consumption during the first 8 postoperative hours, with 267.0 6 9.8, 298. 6 32.2, and 267.2 6 22. mg for groups O, MCP, and P, respectively. Initial tramadol titration dose in the recovery room after emergence of anesthesia amounted to 55.7 6 7.4, 54.4 6 6.4, and 37.6 6 6.3 mg in groups O, MCP, and P, respectively (P =.06). Additional analgesic rescue medication piritramide was necessary in 38 patients, 2 on ondansetron (8.8 6 2.2 mg), 2 on MCP (3.2 6 8.0 mg), and 4 on placebo (7.2 6.4 mg, P =.6) in the recovery room. During the subsequent PCA period group O patients did not use more tramadol compared with the other groups. In patient of group MCP, tramadol was discontinued due to lack of analgesic efficacy. Genotyping revealed a PM genotype. All group O patients had sufficient pain control by tramadol. Neither pain scores at rest nor at movement differed between the groups. Antiemetic Efficacy of Study Medication Complete response to antiemetic prophylaxis without any nausea and vomiting during the first 8 postoperative hours was achieved in 85.0%, 69.5%, and 66.7% of groups O, MCP, and P patients, respectively, with an improved symptom control in group O compared with group P (P =.03) (Fig 2). In the recovery room, 30 patients (O:9/MCP:0/P:) were given antiemetic rescue medication; 24 (7/8/9) had symptom control after dose and 6 needed 2 doses. After referral to a general ward, 32 (8/3/) individuals were treated with rescue medication, with 20 (4/9/7) patients having sufficient symptom control after one dose of rescue medication. In 3 individuals of group P, tramadol was discontinued

Rauers et al The Journal of Pain 277 Assessed for eligibility n=2 Met inclusion criteria + gave written informed consent 87 Excluded Not meeting inclusion criteria 8 Refused to participate 2 Other reasons 4 Randomization Ondansetron Did not receive ondansetron: Intraoperative complications Postop. mechanical ventilation Protocol violation Received ondansetron 63 MCP Did not receive MCP: Intraoperative complications Postop. mechanical ventilation Received MCP 62 2 59 Placebo Did not receive placebo: Intraoperative complications Postop. mechanical ventilation Received placebo 62 Additional piritramide in the PACU No 48 Yes 2 due to lack of analgesic efficacy 0 Additional piritramide in the PACU No 47 Yes 2 due to lack of analgesic efficacy Additional piritramide in the PACU No 46 Yes 4 due to lack of analgesic efficacy 0 Analgesia Antiemetic rescue medication No 5 Yes due to emesis 9 0 Antiemetic rescue medication No 4 Yes due to emesis 9 0 Antiemetic rescue medication No 40 Yes 20 due to emesis 3 Emesis Analgesic consumption analyzed Antiemetic efficacy analyzed Analgesic consumption analyzed Antiemetic efficacy analyzed 59 59 Analgesic consumption analyzed Antiemetic efficacy analyzed Analysis Figure. Flow chart of patients through each stage of the randomized trial. PACU, postanesthesia care unit; T, tramadol. due to continuous emesis. Two of these patients were switched to piritramide rescue medication. The patients questionnaires revealed no statistically significant difference in the frequency of nausea for the 3 treatment groups (20.0/ 22.0/ 28.8%) for the first postoperative hours; however, a significant difference in the number of patients with vomiting (5.0/ 5./ 8.6%; P =.0). The 2 group O patients carrying a CYP2D6 duplication were both male, with an Apfel score of 2, and did not have nausea or vomiting during the study period. CYP2D6 Genotypes and Drug Concentrations Frequencies of genotype groups PM, HZ/IM, EM, and UM amounted to 0.%, 3.3%, 53.6%, and 5.0%, respectively (Table ). Genotype frequencies did not deviate from Hardy-Weinberg equilibrium. Plasma concentrations of ondansetron did not differ between the genotype groups (Fig 3A), with an AUC of 68.8 6 22.7, 95.6 6 20.2, 93.4 6 2.6, and 74.5 6 39.4 ng h/ml in PM, HZ/IM, EM, and UM, respectively. In contrast, concentrations of ()O-demethyltramadol were significantly elevated in UMs (AUC: 279. 6.7 ng h/ml) compared with the other genotypes and thus were consistent with the CYP2D6 duplication / multiduplication (P <.00, Fig 3B). AUCs amounted to 47.3 6.0 in group EM, 38.0 6 2.5 in group HZ/IM, and 22.2 6 2.4 ng h/ml in group PM. Discussion This randomized, controlled, double-blinded study showed that tramadol s analgesic efficacy and ondansetron s antiemetic effects were maintained if both drugs were co-administered. Thus, some previous controversial findings could not be confirmed. Considering the concomitant administration of tramadol and ondansetron, these results for the primary end point (analgesic efficacy of tramadol) were unexpected, since antagonistic actions between ondansetron and tramadol have been described previously. 4 For the secondary end point (ondansetron s antiemetic efficacy), the results were not surprising, as ondansetron is generally found to be more effective than placebo or MCP 0 mg. However, for the secondary end point, only those patients at risk for emesis are relevant. Thus, this analysis needs larger patient numbers to draw firm conclusions considering antiemetic efficacy of ondansetron when co-administered with tramadol. Although MCP demonstrated some benefit considering vomiting during the first postoperative hours compared with placebo, it failed to improve overall complete response. However, the dose might have been too low, and from a retrospective view it could be considered to be an active placebo only. 2,9 As antiemetic rescue medication was immediately available this single dose antiemetic / placebo approach was reasonable, although in clinical practice, high-risk patients should receive multiple interventions in

278 The Journal of Pain Ondansetron Co-administered With Tramadol Table. Demographic and Perioperative Data of Patients Assigned to the 3 Treatment Groups ONDANSETRON MCP PLACEBO Age (y) 57.2 6 2.3 58. 63.7 57.6 6 2.7 Sex: male / female 38 / 22 39 / 20 39 / 2 Body weight (kg) 74.2 6 2.8 78.0 6 6.8 78.2 6 4.9 Height (cm) 73.0 6 8.4 72.5 6 9.4 74.26 8.5 Duration of surgery (min) 69.5 6 7.8 75.8 6.4 75.2 6 67.8 Kind of surgery (No. of patients) Nephrectomy/enucleation of kidney tumor 6 4 7 Prostatectomy 9 7 4 Major abdominal 8 23 2 Reinsertion of ureter 4 4 6 Lateral thoracotomy 3 2 ASA physical status I / II / III (No. of patients) 8 / 42 / 0 8 / 36 / 5 8 / 36 / 6 PONV Score / 2 / 3 / 4* (No. of patients) 2 / 32 / 2 / 4 3 / 27 / 2 / 7 0 / 3 / 6 / 3 Intraoperative tramadol loading dose (mg) 25. 6 46. 222.9 6 4.5 224.5 6 4.9 Pain scores (NRS) after awakening 49.8 6 23.5 50.8. 6 24.8 5.6 6 22.0 2 h at rest 33.3. 6 5.7 34.8 6 7. 34. 6 6.3 6 h at rest 2.9 6 5.2 23.9 6 3.0 26.6 6 5.5 2 h at movement / coughing 57.9 6 20.4 58.4 6 6.9. 6 2.5 6 h at movement / coughing 45.4 6 4. 44.3 6 3.6 48. 6 5.4 CYP2D6 genotype group PM / HZ/IM / EM / UM (No. of patients) 3 / 20 / 35 / 2 7 / 2 / 28 / 3 8 / 5 / 33 / 4 Abbreviation: ASA, American Society of Anesthesiologists. NOTE. Data are presented as mean 6 SD or number of patients. *PONV Score 3,23 : Validated simplified risk score for PONV calculated by the number of risk factors: female sex, point; nonsmoking, point; history of PONV / motion sickness, point; postoperative opioids, point. parallel. 3 Intentionally, not only patients with Apfel scores $3 were enrolled but also patients with lower risks because our aim was not to investigate treatment of PONV in general but treatment of tramadol associated nausea and vomiting. Ondansetron has a mean plasma half-life of up to 5.5 hours in healthy volunteers.,38 Because it was given for prophylaxis, comparable to most previous studies, the antiemetic effect was only evaluated within the first 8 postoperative hours. During the later course, additional stimuli such as first mobilization or intake of oral fluids can also cause emetic sequelae. Analgesia and Tramadol Consumption Serotonin is one of the pivotal neurotransmitters of the descending pathways that down-modulates spinal nociception. Tramadol does not only produce analgesia via opioid receptors but also via inhibition of the reuptake of 5-HT and norepinephrine probably by additional facilitation of 5-HT release at the spinal cord. 0,28,37 Different sites of serotonin receptors are discussed for possible involvement; for example, spinal presynaptic 5-HT 3 subtypes have been implicated in antinociception in animal models. 25 Furthermore, a dopaminergic component has been postulated as well. 42 Trials investigating tramadol s analgesic efficacy if coadministered with ondansetron were not unequivocal up to date. 4,6,8,9,8,29 However, studies considerably differed in their design, patients enrolled, kind and duration of surgical procedure, drugs used for anesthesia, as well as dosing and timing of drugs. In 3 of these studies, tramadol requirements by PCA were increased when ondansetron was co-administered for antiemetic prophylaxis. 4,8,9 In contrast, 3 other studies and the present investigation showed no difference for tramadol consumption. 6,8,29 One possible explanation for this is the higher intraoperative tramadol loading dose which may have covered possible lower analgesic needs in some patients. Tramadol loading doses ranged between 50 mg up to 4 times to 2 mg/kg in previous trials. 4,6,8,9,8 However, type of surgery was not comparable to our major procedures. Animal trials revealed conflicting results. Erhan et al 4 did not find an ondansetron related decrease of tramadol s analgesic effectiveness in mice. In contrast, Dürsteler et al 3 demonstrated the presence of a functional antagonism. Interestingly, ondansetron as well as tramadol alone were able to induce antinociception and furthermore, analgesic response depended on the type of the nociceptive stimulus. Diverse mechanism of tramadol action and increasing 5-HT as well as norepinephrine levels in the CNS and probably in the periphery (gut) were discussed. 3 Nausea and Vomiting Two previous studies reported either an increase of vomiting scores in ondansetron treated patients compared with placebo with no difference in nausea score or higher nausea scores compared with MCP treatment. 4,6 As some authors reported incidences, others calculated scores, a comparison of study results is difficult. Higher vomiting scores were supposed to be correlated to the intensity of tramadol use in the first 8 hours induced by ondansetron co-administration. 4 In contrast, de Witte et al 9 did not report any differences in the incidence of PONV between ondansetron

Rauers et al The Journal of Pain 279 A (+)O-demethyltramadol ng/ml 40 20 00 80 40 20 0 PM EM HZ / IM UM 0 30 90 20 50 80 Minutes B 80 Figure 2. Percentage of patients with 5%/95% confidence interval in group O (ondansetron), MCP (metoclopramide), and P (placebo) with complete response to antiemetic prophylaxis (neither nausea, nor vomiting, nor antiemetic rescue medication). Fisher s exact test 8-hour interval: P =.046 (corrected); paired Fisher s exact test group O versus group P: P =.03. Ondansetron ng/ml 75 70 65 55 50 0 30 90 20 50 80 Minutes (24-hour cumulative incidence 40%) and placebotreated groups (50%), although tramadol consumption was also increased. In clinical practice, patients generally reduce PCA demands if they experience a clear-cut relationship between PCA bolus doses and emesis. They intend to discontinue therapy, which also occurred in patient of the present trial. Overall, number of patients investigated up to date was small; thus, some of the studies might have been underpowered. Furthermore, there was no stratification for genotype, which could explain some of the therapeutic failures, with high tramadol consumption on the one hand (eg, CYP2D6 PM) and lacking antiemetic efficacy of ondansetron (UM) on the other hand. To date, the underlying mechanisms of tramadol induced emesis are not fully understood. In part, opioid effects contribute; though this does not explain the increased incidence during the initial titration phase. Increased 5-HT concentrations appear to be a reasonable explanation. 5 Another consideration aims at an increase of (nor-)adrenaline (nor-epinephrine) levels by inhibition of reuptake, thus producing emesis. 7,2 CYP2D6-Dependent Metabolism Frequency of genotypes was comparable to previous reports enrolling predominantly Caucasian individuals. 39,4 The impact of CYP2D6 polymorphisms on tramadol analgesia and antiemetic efficacy of ondansetron as well as tropisetron are well described. 7,5,24,33,39,4 Due to the limited number of individuals receiving ondansetron, pharmacogenetic data and drug concentrations were only analyzed exploratively. In contrast to the expected high concentrations of ()O-demethyltramadol in UMs, Figure 3. (A) Plasma concentrations (mean 6 SEM) of ondansetron, and (B) ()O-demethyltramadol 30, 90, and 80 minutes after intravenous administration of ondansetron and tramadol allocated to CYP2D6 genotype group. As patients were allowed to receive additional tramadol in the recovery room between time point 90 and 80 minutes, plasma concentration of ()Odemethyltramadol increased in several subjects. PM, poor metabolizers; HZ / IM, heterozygous carriers or intermediate metabolizers; EM, extensive metabolizers; UM, ultra-rapid metabolizers. the previously discussed low ondansetron concentrations could only be confirmed in one of the UMs. 7,24 These findings might be explained by further polymorphic genes involved in drug uptake and enzymatic cleavage,5,20,23 or co-administration of drugs involved in the serotonergic pathways. 2,22,34,44 Conclusions In this postoperative setting after major abdominal surgery, tramadol s analgesic effect was not decreased by co-administration of ondansetron compared with placebo or MCP, nor was the frequency of PONV increased. However, this does not exclude any possible opposing actions of these drugs within the serotonergic pathway. Acknowledgments The authors thank all colleagues, the staff of the postanesthesia care unit, and the general wards involved in this study. Furthermore, we acknowledge the excellent technical assistance of Makbule Kobilay (Department of Anesthesiology and Intensive Care Medicine, University of Bonn) and Jörg Bayer (Department of Forensic Medicine, University of Bonn).

280 The Journal of Pain Ondansetron Co-administered With Tramadol References. Aapro M: 5-HT(3)-receptor antagonists in the management of nausea and vomiting in cancer and cancer treatment. Oncology 69:97-09, 2005 2. Apfel CC, Kranke P, Katz MH, Goepfert C, Papenfuss T, Rauch S, Heineck R, Greim CA, Roewer N: Volatile anaesthetics may be the main cause of early but not delayed postoperative vomiting: A randomized controlled trial of factorial design. Br J Anaesth 88:659-668, 2002 3. Apfel CC, Läärä E, Koivuranta M, Greim CA, Roewer N: A simplified risk score for predicting postoperative nausea and vomiting: Conclusions from cross-validations between two centers. Anesthesiology 9:693-700, 999 4. Arcioni R, della Rocca M, Romanò S, Romano R, Pietropaoli P, Gasparetto A: Ondansetron inhibits the analgesic effects of tramadol: A possible 5-HT(3) spinal receptor involvement in acute pain in humans. Anesth Analg 94: 553-557, 2002 5. Barann M, Urban B, Stamer U, Dorner Z, Bönisch H, Brüss M: Effects of tramadol and O-demethyl-tramadol on human 5-HT reuptake carriers and human 5-HT3A receptors: A possible mechanism for tramadol-induced early emesis. Eur J Pharmacol 53:54-58, 2006 6. Broome IJ, Robb HM, Raj N, Girgis Y, Wardall GJ: The use of tramadol following day-case oral surgery. Anaesthesia 54: 289-292, 999 7. Candiotti KA, Birnbach DJ, Lubarsky DA, Nhuch F, Kamat A, Koch WH, Nikoloff M, Wu L, Andrews D: The impact of pharmacogenomics on postoperative nausea and vomiting: Do CYP2D6 allele copy number and polymorphisms affect the success or failure of ondansetron prophylaxis? Anesthesiology 02:543-549, 2005 8. Cubukçu Z, Ozbek H, Günesx Y, Gündüz M, Ozcengiz D, Isxik G: Effect of ondansetron in lower extremity bone surgery on morphine and tramadol consumption using patient controlled analgesia. Agri 9:36-4, 2007 9. De Witte JL, Schoenmaekers B, Sessler DI, Deloof T: The analgesic efficacy of tramadol is impaired by concurrent administration of ondansetron. Anesth Analg 92:39-32, 200 0. Desmeules JA, Piguet V, Collart L, Dayer P: Contribution of monoaminergic modulation to analgesic effect of tramadol. Br J Clin Pharmacol 48:7-2, 996. Dixon CM, Colthup PV, Serabjit-Singh CJ, Kerr BM, Boehlert CC, Park GR, Tarbit MH: Multiple forms of cytochrome P450 are involved in the metabolism of ondansetron in humans. Drug Metab Dispos 23:225-230, 995 2. Duman EN, Kesim M, Kadioglu M, Yaris E, Kalyoncu NI, Erciyes N: Possible involvement of opioidergic and serotonergic mechanisms in antinociceptive effect of paroxetine in acute pain. J Pharmacol Sci 94:6-65, 2004 3. Dürsteler C, Mases A, Fernandez V, Pol O, Puig MM: Interaction between tramadol and two anti-emetics on nociception and gastrointestinal transit in mice. Eur J Pain 0:629-638, 2006 4. Erhan E, Onal A, Kocabas S, Parlar A, Yegul I, Kosay S: Ondansetron does not block tramadol-induced analgesia in mice. Methods Find Exp Clin Pharmacol 27:629-632, 2005 5. Fischer V, Vickers AE, Heitz F, Mahadevan S, Baldeck JP, Minery P, Tynes R: The polymorphic cytochrome P-4502D6 is involved in the metabolism of both 5-hydroxytryptamine antagonists, tropisetron and ondansetron. Drug Metab Dispos 22:269-274, 994 6. Gan TJ, Meyer TA, Apfel CC, Chung F, Davis PJ, Habib AS, Hooper VD, Kovac AL, Kranke P, Myles P, Philip BK, Samsa G, Sessler DI, Temo J, Tramèr MR, Vander Kolk C, Watcha M: Society for Ambulatory Anesthesia: Society for Ambulatory Anesthesia guidelines for the management of postoperative nausea and vomiting. Anesth Analg 05:65-628, 2007 7. García-Quetglas E, Azanza JR, Sádaba B, Muñoz MJ, Gil I, Campanero MA: Pharmacokinetics of tramadol enantiomers and their respective phase I metabolites in relation to CYP2D6 phenotype. Pharmacol Res 55:22-30, 2007 8. Helmy SAK: Prophylactic anti-emetic efficacy of ondansetron in laparoscopic cholecystectomy under total intravenous anaesthesia: A randomised, double-blind comparison with droperidol, metoclopramide and placebo. Anaesthesia 54:266-27, 999 9. Henzi I, Walder B, Tramèr MR: Metoclopramide in the prevention of postoperative nausea and vomiting: A quantitative systematic review of randomized, placebocontrolled studies. Br J Anaesth 83:76-77, 999 20. Ho KY, Gan TJ: Pharmacology, pharmacogenetics, and clinical efficacy of 5-hydroxytryptamine type 3 receptor antagonists for postoperative nausea and vomiting. Curr Opin Anaesthesiol 9:6-6, 2006 2. Högger P: Adverse effects of opioid analgesic treatment are correlated with a significant elevation in plasma epinephrine in healthy humans. Eur J Clin Pharmacol 56: 463-468, 2000 22. Hunt TE, Wu W, Zbuzek VK: Ondansetron blocks nifedipine-induced analgesia in rats. Anesth Analg 82: 498-500, 996 23. Janicki PK: Cytochrome P450 2D6 metabolism and 5-hydroxytryptamine type 3 receptor antagonists for postoperative nausea and vomiting. Med Sci Monit :RA322-RA328, 2005 24. Kaiser R, Sezer O, Papies A, Bauer S, Schelenz C, Tremblay PB, Possinger K, Roots I, Brockmöller J: Patient-tailored antiemetic treatment with 5-hydroxytryptamine type 3 receptor antagonists according to cytochrome P-450 2D6 genotypes. J Clin Oncol 20: 2805-28, 2002 25. Millan MJ: Serotonin and pain: Evidence that activation of 5-HTA receptors does not elicit antinociception against noxious thermal, mechanical and chemical stimuli in mice. Pain 58:45-6, 994 26. Musshoff F, Madea B, Stüber F, Stamer UM: Enantiomeric determination of tramadol and O-desmehyltramadol by liquid chromatography mass spectrometry and application to postoperative patients receiving tramadol. J Anal Toxicol 30:463-467, 2006 27. Ng KF, Yuen TS, Ng VM: A comparison of postoperative cognitive function and pain relief with fentanyl or tramadol patient-controlled analgesia. J Clin Anesth 8:205-20, 2006 28. Oliva P, Aurilio C, Massimo F, Grella A, Maione S, Grella E, Scafuro M, Rossi F, Berrino L: The antinociceptive effect of tramadol in the formalin test is mediated by the serotonergic component. Eur J Pharmacol 445:70-85, 2002

Rauers et al The Journal of Pain 28 29. Öztekin S, Ozzeybek D, Tasxdöğen A, Kilercik H, Kara HC: Comparison of the antiemetic efficacy of tropisetron and droperidol with patient-given tramadol. J Int Med Res 3: 267-27, 2003 30. Pang WW, Mok MS, Huang S, Hung CP, Huang MH: Intraoperative loading attenuates nausea and vomiting of tramadol patient-controlled analgesia. Can J Anaesth 47: 968-973, 2000 3. Pang WW, Mok MS, Lin CH, Yang TF, Huang MH: Comparison of patient-controlled analgesia (PCA) with tramadol or morphine. Can J Anaesth 46:030-035, 999 32. Pang WW, Wu HS, Lin CH, Chang DP, Huang MH: Metoclopramide decreases emesis but increases sedation in tramadol patient-controlled analgesia. Can J Anaesth 49: 029-033, 2002 33. Pedersen RS, Damkier P, Brøsen K: Enantioselective pharmacokinetics of tramadol in CYP2D6 extensive and poor metabolizers. Eur J Clin Pharmacol 62:53-52, 2006 34. Pickering G, Loriot MA, Libert F, Eschalier A, Beaune P, Dubray C: Analgesic effect of acetaminophen in humans: First evidence of a central serotonergic mechanism. Clin Pharmacol Ther 79:37-378, 2006 35. Poulsen L, Arendt-Nielsen L, Brøsen K, Sindrup SH: The hypoalgesic effect of tramadol in relation to CYP2D6. Clin Pharmacol Ther :636-644, 996 36. Raffa RB, Friderichs E, Reimann W, Shank RP, Codd EE, Vaught JL: Opioid and non-opioid components independently contribute to the mechanism of action of tramadol, an atypical opioid analgesic. J Pharmacol Exp Ther 2:275-285, 992 37. Raffa RB, Friderichs E, Reimann W, Shank RP, Codd EE, Vaught JL, Jacoby HI, Selve N: Complimentary and synergistic antinociceptive interaction between the enantiomers of tramadol. J Pharmacol Exp Ther 267:33-340, 993 38. Simpson KH, Hicks FM: Clinical pharmacokinetics of ondansetron: A review. J Pharm Pharmacol 48:774-78, 996 39. Stamer UM, Lehnen K, Höthker F, Bayerer B, Wolf S, Hoeft A, Stüber F: Impact of CYP2D6 genotype on postoperative tramadol analgesia. Pain 05:23-238, 2003 40. Stamer UM, Maier C, Grond S, Veh-Schmidt B, Klaschik E, Lehmann KA: Tramadol in the management of postoperative pain: A double-blind, placebo- and drug-controlled study. Eur J Anaesthesiol 4:646-654, 997 4. Stamer UM, Musshoff F, Kobilay M, Hoeft A, Madea B, Stüber F: Concentrations of tramadol and O-desmethyltramadol in different CYP2D6 genotypes. Clin Pharmacol Ther 82:4-47, 2007 42. Taylor BK, Joshi C, Uppal H: Stimulation of dopamine D2 receptors in the nucleus accumbens inhibits inflammatory pain. Brain Res 987:35-43, 2003 43. Vickers MD, O Flaherty D, Szekely SM, Read M, Yoshizumi J: Tramadol: Pain relief by an opioid without depression of respiration. Anaesthesia 47:29-296, 992 44. Villikka K, Kivistö KT, Neuvonen PJ: The effect of rifampin on the pharmacokinetics of oral and intravenous ondansetron. Clin Pharmacol Ther 65:377-38, 999 45. Walkembach J, Brüss M, Urban BW, Barann M: Interactions of metoclopramide and ergotamine with human 5-HT(3A) receptors and human 5-HT reuptake carriers. Br J Pharmacol 46:543-552, 2005 46. Wallenborn J, Gelbrich G, Bulst D, Behrends K, Wallenborn H, Rohrbach A, Krause U, Kühnast T, Wiegel M, Olthoff D: Prevention of postoperative nausea and vomiting by metoclopramide combined with dexamethasone: Randomised double blind multicentre trial. BMJ 333:324, 2006