- Email: [email protected]
Characterisation of tramadol, morphine and tapentadol in an acute pain model in Beagle dogs
Veterinary Anaesthesia and Analgesia, 2014, 41, 297–304
doi:10.1111/vaa.12140
RESEARCH PAPER
Characterisation of tramadol, morphine and tapentadol in an acute pain model in Beagle dogs Babette K€ ogel, Rolf Terlinden & Johannes Schneider Gr€ unenthal GmbH, Gr€ unenthal Innovation, Global Preclinical R&D, Aachen, Germany
Correspondence: Babette K€ ogel, Department of Pharmacology, Gr€ unenthal GmbH, Global Biomedical Sciences, Zieglerstrasse 6, 52078 Aachen, Germany. E-mail: [email protected]
Abstract Objective To evaluate the analgesic potential of the centrally acting analgesics tramadol, morphine and the novel analgesic tapentadol in a pre-clinical research model of acute nociceptive pain, the tailflick model in dogs. Study design Prospective part-randomized pre-clinical research trial. Animals Fifteen male Beagle dogs (HsdCpb:DOBE), aged 12–15 months. Methods On different occasions separated by at least 1 week, dogs received intravenous (IV) administrations of tramadol (6.81, 10.0 mg kg1), tapentadol (2.15, 4.64, 6.81 mg kg1) or morphine (0.464, 0.681, 1.0 mg kg1) with subsequent measurement of tail withdrawal latencies from a thermal stimulus (for each treatment n = 5). Blood samples were collected immediately after the pharmacodynamic measurements of tramadol to determine pharmacokinetics and the active metabolite O-demethyltramadol (M1). Results Tapentadol and morphine induced dosedependent antinociception with ED50-values of 4.3 mg kg1 and 0.71 mg kg1, respectively. In contrast, tramadol did not induce antinociception at any dose tested. Measurements of the serum levels of tramadol and the M1 metabolite revealed only marginal amounts of the M1 metabolite, which explains the absence of the antinociptive
effect of tramadol in this experimental pain model in dogs. Conclusions and clinical relevance Different breeds of dogs might not or only poorly respond to treatment with tramadol due to low metabolism of the drug. Tapentadol and morphine which act directly on l-opioid receptors without the need for metabolic activation are demonstrated to induce potent antinociception in the experimental model used and should also provide a reliable pain management in the clinical situation. The nonopioid mechanisms of tramadol do not provide antinociception in this experimental setting. This contrasts to many clinical situations described in the literature, where tramadol appears to provide useful analgesia in dogs for post-operative pain relief and in more chronically pain states. Keywords acute nociceptive pain, Beagle dog, morphine, tapentadol, tramadol.
Introduction Good pain management in dogs, allowing animals to recover from surgical procedures and return to physiological normality more rapidly, is not only important for veterinary medicinal practice but also when this animal species is used for research. In the pharmaceutical industry, Beagle dogs are used widely to characterise developmental compounds as potential drugs for human therapy. Analgesics used include nonsteroidal analgesics and opioids such as morphine (Johnston et al. 2008; Mathews 297
Tramadol, morphine and tapentadol in dogs B K€ogel et al.
2008) buprenorphine and methadone (Mathews 2008). Recently, tramadol has become popular for the therapy of moderate to severe pain in both humans and in animals. Tramadol is a centrally acting analgesic which has several modes of action. These include activation of opioid receptors (Hennies et al. 1988) preferentially of the l-subtype (Raffa et al. 1992) and enhancement of the extra-cellular concentrations of the monoamine neurotransmitters 5-hydroxytryptamine (serotonin) and noradrenaline by blocking the reuptake transporters and, in the case of serotonin, additional release mechanisms (Driessen & Reimann 1992; Raffa et al. 1992; Driessen et al. 1993). The opioid receptor affinity of tramadol is low, but the compound is metabolised extensively in rodents and humans by O- and N-demethylation, yielding metabolites with a higher l-opioid receptor affinity than that shown by the parent compound. O-demethyl tramadol (M1) is likely to be the main carrier of the opioid analgesic properties of tramadol (Hennies et al. 1988; Frink et al. 1996; Gillen et al. 2000). Tramadol is a racemic compound and the pharmacological activities of its (+)- and ()-enantiomers differ substantially. The l-opioid affinity is represented mainly in the (+)-enantiomer of tramadol and the (+)-M1 metabolite. The noradrenaline uptake inhibitory activity of tramadol resides in the ()-enantiomer, whereas the serotonin uptake and release are influenced by the (+)-enantiomer (Driessen & Reimann 1992; Driessen et al. 1993). The complex activities of both enantiomers of tramadol and the O-demethylmetabolite are complementary and necessary for the full analgesic efficacy (Raffa et al. 1992). There are little data in the literature concerning the use of tramadol in dogs, and dose range recommendations are vague or wide. Most of these studies are confined to pharmacokinetic data without referring to analgesic effects of tramadol in dogs (Wu et al. 2001; KuKanich & Papich 2004; McMillan et al. 2008; Giorgi et al. 2009; Saccomanni et al. 2010). There are only a few concrete reports on analgesic activity of tramadol in dogs either with or without concomitant pharmacokinetic parameters (Mastrocinque & Fantoni 2003; Rychel 2010; Kukanich & Papich 2011; Morgaz et al. 2013; Teixeira et al. 2013) The aim of this study was to conduct a pharmacodynamic investigation where direct measurement of nociception was combined with the measurement 298
of the serum levels of tramadol and the active metabolite O-demethyltramadol (M1). The opioid morphine and the novel centrally-acting analgesic tapentadol which in addition to its l-opioid agonistic activity also inhibits the neuronal uptake of noradrenaline were used as comparators (Tzschentke et al. 2006, 2007). Both act directly on l-opioid receptors without the need for metabolic activation and formation of analgesically relevant metabolites. The results of the analgesic test and the serum concentration measurements are discussed with respect to the species-specific (dog) metabolism and the pain type-dependent contribution of the various modes of analgesic action of tramadol. Material and methods Experiments were performed in accordance with the German Animal Welfare Law. Study protocols were approved by the internal animal welfare officer and the local Government Committee (District Council of Cologne 23.203.2-Gr, FG-PK-99-22) which is advised by an independent Ethics Committee. Animals The experiments were carried out in 15 male Beagle dogs, at the start of the study 12–15 month old and, weighing 11–13.5 kg. Their provenance and general keep are given in Appendix S1. In a first set of experiments tramadol and morphine were tested in a cohort of ten dogs. Each dose or vehicle (0.9% saline solvent of the drugs, used as placebo) group consisted of five dosing conditions (n = 5). Dogs were allocated partly randomly to the dose/treatments of tramadol, morphine and vehicle. It was ensured that dogs which received tramadol also received morphine. The dosing scheme is given in Appendix S2. In a second set of experiments, at a later date and in another cohort of dogs (n = 5), tapentadol was tested in a randomized order and among other research compounds. Tramadol and morphine was tested in the first set experiments, whereas tapentadol was tested in a second set of experiments. The vehicle was tested in each study. When any one dog received a maximum of 10 dosing conditions (over a period of 2 years), the dogs underwent a health-check that looked for physical and behavioural abnormalities. All dogs of this study were rated as healthy and thus suitable for rehoming as pets after the study.
© 2014 Association of Veterinary Anaesthetists and the American College of Veterinary Anesthesia and Analgesia, 41, 297–304
Tramadol, morphine and tapentadol in dogs B K€ogel et al.
Analgesic test
Blood-sample collection and analyses
The tail-flick test was carried out in dogs using a modification of the method described by Adams et al. (1990). The tail flick latency (s), the time to withdraw the tail from a radiant heat source (bulb, 12V/50W) was measured. The apparatus used for nociceptive testing consisted of a hand-device which contained the bulb and was connected via a flexible wire to an apparatus which was synchronised with the activation of the bulb and measured the latency for the tail withdrawal (tail-flick analgesiometer, Labtec, Dr. Hess, Germany). The heat source was adjusted to produce a baseline tail flick latency of 3–5 seconds prior to any of the experiments and was left at a constant setting thereafter. A cut off time of 12 seconds was set to avoid tissue damage of the tail. The tails of the animals were shaved 1 day prior the tests and it was ensured that the skin of the test area was intact after shaving and when performing the trials. The hand-device positioned the bulb on the ventral surface of the tail. The increase in tail-flick latency was defined as antinociception and calculated as percent of the maximal possible effect (% MPE). The maximum possible antinociceptive effect was defined as the lack of a tail-flick during 12 seconds of exposure to the heat stimulus. The percentage of MPE was calculated according to the formula:
Thirty minutes before the start of the study, an 18-gauge soft cannula (Vasofix Braun€ ule, Luer Lock, BBraun Melsungen AG, 34209 Germany) was placed into the vena saphena, and fixed in place with a cohesive flex wrap bandage (Petflex, Medpets.de). Blood samples (2–3 mL, five times per dose) were obtained from the five dogs of the 10 mg kg1 tramadol treatment. Serum was obtained by centrifugation of the samples and they were stored frozen at 20 °C until assay. Samples were obtained at the following time points immediately after the tail-flick test: Pre-dose, 10, 20, 40 and 60 minutes after administration. Samples were analyzed for the concentrations of (+)- and ()-tramadol and (+)- and ()-M1, using a modification of a gas chromatographic method (Becker & Lintz 1986). Further details are given in Appendix S3.
½ðT1 T0 =T2 T0 Þ 100; where T0 and T1 were latencies before and after drug administration and T2 was the cut-off time which was set by the protocol. Drugs Drugs and doses tested were as follows; Tramadol HCl (Tramal, Gr€ unenthal GmbH) 6.81, 10.0 mg kg1, tapentadol HCl (Nucynta, Gr€ unenthal GmbH) 2.15, 4.64, 6.81 mg kg1 and morphine HCl (Merck AG, Germany) 0.464, 0.681, 1.0 mg kg1. The drugs were dissolved in 0.9% sodium chloride solution (Fresenius, Germany) and administered intravenously (IV) in an application volume of 5.0 mL kg1. Doses refer to the hydrochloride salts of the compounds. The vehicle 0.9% sodium chloride solution was used as placebo. All drugs or vehicle were administered slowly (over about 1 minute) into the right cephalic vein via a pre-placed 20-gauge cannula (Sterican, BBraun Melsungen AG, 34209 Germany).
Statistical analysis Tail-flick data were analyzed by a one-way repeated measure ANOVA (i.e. including both design factors, dose and time into the model). Post-hoc comparisons versus control were made with Bonferroni’s test. Effects were considered to be statistically significant if p ≤ 0.05. Statistical analysis was performed using SYSTAT, version 11, for Windows. ED50-values were calculated at the time of the maximal effect (i.e. 10 minutes after administration). Results In the tail-flick test, as measured by withdrawal latencies of the tail of the dogs, there was no antinociception measurable for tramadol at any time for both doses studied (6.81 and 10 mg kg1 IV) (Fig. 1). Tramadol was well tolerated up to a dose of 6.81 mg kg1. 10 mg kg1 was rated as the highest applicable dose due to side effects including timid behaviour, ataxia and, in one of the five dogs, a short lasting convulsion. Sedation was not observed. In contrast, morphine (Fig. 2) and tapentadol (Fig. 3) induced dose-dependent antinociception with ED50 values of 0.71 and 4.3 mg kg1, respectively. Peak effects were attained at 10 minutes after injection which was the time point of first measurement after drug administration. Following administration of morphine 0.681 mg kg1, mild sedation, salivation, ataxia occurred, and one dog vomited. Morphine at
© 2014 Association of Veterinary Anaesthetists and the American College of Veterinary Anesthesia and Analgesia, 41, 297–304
299
Tramadol, morphine and tapentadol in dogs B K€ogel et al.
Figure 1 Lack of effect after intravenous administration of tramadol in acute pain (tail-flick test) in Beagle dogs. Data are presented as % MPE (mean SD), MPE, maximum possible effect; n = 5 animals per group.
Figure 2 Dose-dependent effect after intravenous administration of morphine in acute pain (tail-flick test) in beagle dogs. Data are presented as % MPE (mean SD), *p < 0.05 versus placebo, MPE, maximum possible effect; n = 5 animals per group.
Figure 3 Dose-dependent effect after intravenous administration of tapentadol in acute pain (tail-flick test) in beagle dogs. Data are presented as % MPE (mean SD), *p < 0.05 versus placebo, MPE, maximum possible effect; n = 5 animals per group.
1.0 mg kg1 induced diarrhoea, ataxia, salivation and sedation. However, dogs were calm for most of the time during the study but remained responsive to touch or noise. Tapentadol at the two lower doses was well tolerated. However, at the highest dose (6.81 mg kg1) the dogs showed sedation 300
accompanied by reduced responsiveness, ataxia, salivation and diarrhoea. Tail-flick data were analysed by a one-way repeated measure ANOVA. Morphine: treatment F3,16 = 16.60, p < 0.001 time F4,46 = 18.40, p < 0.001, tapentadol: treatment F3,16 = 15.02, p < 0.001 time
© 2014 Association of Veterinary Anaesthetists and the American College of Veterinary Anesthesia and Analgesia, 41, 297–304
Tramadol, morphine and tapentadol in dogs B K€ogel et al. F4,46 = 28.46, p < 0.001 tramadol: treatment F2,12 = 0.860, p = 0.860 time F3,36 = 1.146, p = 0.344. Figure 4 shows the serum concentration profile of both tramadol enantiomers and that of (+)-M1. Low concentrations of the metabolite (+)-M1 of 11.9– 16.5 ng mL1 were measured, whereas ()-M1 concentrations were below the limit of quantification. On average, (+)-M1 reached only 0.8–1.8% of the (+)-tramadol concentrations. Discussion We intentionally used the tail-flick model to evaluate the opioid-like activity of test substances. We are aware that this model covers only a part of the broader range of clinical efficacy of tramadol. In this study, the l-opioid agonist morphine and the combined l-opioid agonist and noradrenaline uptake inhibitor tapentadol induced full antinociceptive efficacy. In contrast, the IV administration of tramadol, at a maximally tolerated IV dose, did not induce any antinociceptive effect in this canine model of acute pain. This finding is in contrast to the dose-dependent effects of tramadol in tail-flick tests in rodents (Friderichs et al. 1978; Raffa et al. 1992). In the tail-flick paradigm predominantly strong analgesics such as the l-opiate agonists and a2-adrenoceptor agonists are active (Pertovaara 1993; Fairbanks & Wilcox 1999; Negus et al. 2006). The lack of observed effect with tramadol in the dog has to be discussed on the basis of the complex pharmacological mode of action and the need for metabolic transformation, generating a potent l-opioid receptor activating metabolite. The binding affinity of the main metabolite M1 O-demethyl tramadol to l-opioid receptors in rat brain synaptosomes and to the human l-opioid receptor is more than two orders of magnitude
higher than that of the parent molecule (Hennies et al. 1988; Frink et al. 1996; Gillen et al. 2000). Thus, the M1 metabolite and especially its (+)-enantiomer represent the main l-opioid activity of tramadol. Tramadol is metabolized extensively in the Beagle breed of dog, as used in this study. Upon oral administration of tramadol (20 mg kg1) to dogs, 24 metabolites were found (Wu et al. 2001), the O-demethylated (M1) and N-demethylated (M2) metabolites and their conjugates being the major metabolites (Wu et al. 2001), similar to findings in humans. In dogs, the non-conjugated M2 was clearly the main metabolite in comparison to the M1 metabolite. Concomitant monitoring of the serum levels of tramadol and the M1 metabolite in the present study revealed a substantial systemic exposure to tramadol, but only marginal exposure to the (+)-M1 metabolite and even less for ()-M1. After an IV dose of 10 mg kg1 tramadol in our dogs, the highest measured plasma concentrations were 2081 ng mL1 for (+)-tramadol and 1907 ng mL1 for ()-tramadol, but only 17 ng mL1 for (+)-M1. These values suggest a slightly lower exposure when compared to the results of another study in male mixed breed dogs, where IV administration of 1, 2 and 4 mg kg1 tramadol resulted in maximum exposure levels between 820–4664 ng mL1 tramadol and 48– 135 ng mL1 M1 (McMillan et al. 2008). Following oral administration of 4 mg kg1 tramadol (as capsules) to Beagle dogs, a maximum serum concentration of 314 ng mL1 tramadol and 54 ng mL1 M1 were achieved (Negus et al. 2006). In a study by KuKanich & Papich (2004) higher oral dose of 10.2 mg kg1 tramadol resulted in maximum plasma concentrations of 1403 ng mL1 for tramadol and 449 ng mL1 for M1 in Beagle dogs. A clearly lower systemic expo-
Figure 4 Mean ( SD) serum concentrations of tramadol enantiomers and (+)- M1 in 5 Beagle dogs of the 10 mg kg1 IV dose group.
© 2014 Association of Veterinary Anaesthetists and the American College of Veterinary Anesthesia and Analgesia, 41, 297–304
301
Tramadol, morphine and tapentadol in dogs B K€ogel et al.
sure of 216 ng mL1 for tramadol and 6 ng mL1 for M1 maximal concentrations were achieved following only a slightly lower dose of 9.9 mg kg1 (as oral tablets) in greyhound dogs (Kukanich & Papich 2011). The administration of tramadol to our Beagle dogs did not result in an antinociceptive effect in our study, albeit in these dogs morphine and tapentadol provided antinociception in the dog tail flick test, and showed full efficacy. Tapentadol possesses two mechanisms of action one involving l-opioid receptor agonism and the other uptake inhibition of noradrenaline; both these activities are relevant for its analgesic effect (Tzschentke et al. 2007, Schr€ oder et al. 2010). In contrast to tramadol, morphine and tapentadol activate l-opioid receptors without the need for metabolic transformation to an active metabolite. Inhibition of noradrenaline and serotonin uptake, the non-opioid mechanisms of tramadol, does not result in antinociception in experimental setting involving thermal nociceptive stimuli (Raffa & Friderichs 1996). There are at present no data showing the affinity of tramadol to canine l-opioid receptors. Opioid binding studies, however, indicate similar affinities across species from rodents to horses, albeit there are species differences in receptor densities (Yoburn et al. 1991; Thomasy et al. 2007). Further, single nucleotide polymorphisms in the canine MOR gene may account for the differences in opioid responses in various dog populations (Hawley & Wetmore 2010). Potentially, this may also influence pain perception. Thus, the data available on rat and human l-receptor affinities of tramadol and M1 and the information derived from the study in mice under inhibition of metabolic M1 formation, may help to explain the failure of tramadol to induce an antinociceptive activity in the dog tail flick test. In our study there was very little formation of the main tramadol metabolite M1 in dogs, as indicated by serum concentrations below 20 ng mL1 (Fig. 4). This concentration is well below the value of 147 ng mL1 determined at the threshold antinociceptive dose in mice (Friderichs & Becker 1991). Therefore by inference, the formation of the M1 metabolite in dogs may not to be sufficient to induce antinociception. However, there was a high level of tramadol, which by itself can contribute to an antinociceptive effect at least in mice as discussed above. A serum exposure of about 4000 ng mL1 tramadol (2081 ng mL1 (+)-tramadol and 1906 ng mL1 ()-tramadol) was measured 302
10 minutes after the IV administration of 10 mg kg1 in this study. In mice, according to the serum concentrations of 422 ng mL1 at the threshold dose level and 5953 ng mL1 at a full analgesic dose level with blockade of M1 formation might have been expected to be sufficient to induce an antinociceptive effect (Friderichs & Becker 1991). However, in our study, 30 minutes after administration, serum concentrations of 1312 ng mL1 for (+)-tramadol and 1185 ng mL1 for the ()-enantiomer were measured in the Beagle dogs. The total concentration of tramadol thus lies between the concentrations measured for both, threshold and maximum antinociceptive effects as derived from the mouse study (Friderichs & Becker 1991). Furthermore, the concentration of (+)-tramadol is close to its Ki value at the (rat) l-opioid receptor of 4.1 lM, which is equal to 1078 ng mL1 (Frink et al. 1996). Theoretically, from these extrapolations an antinociceptive effect might have been expected in the dog tail-flick test. However, the failure of tramadol to induce an antinociceptive effect shows that high concentrations of non-metabolized tramadol by itself could not compensate for the poor formation of the l-opioid active M1 metabolite, at least in these Beagle dogs. Known side effects which occur immediately after IV administration at 10 mg kg1 excluded an increase of dose in this test model. The formation of the metabolite M1 in humans is mediated by the hepatic enzyme CYP2D6. Poor metabolisers, due to their lack of this CYP2D6 enzyme, have impaired production of M1, so that tramadol has a reduced analgesic effect (Poulsen et al. 1996). There is no information for CYP-specific metabolism of tramadol in dogs. Differences in canine drug metabolism due to polymorphism are described for celecoxib (Paulson et al. 1999). Similarly, it is very likely that different dog populations may exist impacting upon tramadol metabolism. In humans, based on AUC calculations, the relative exposure to tramadol and to (+)-M1 was 7:1 in extensive metabolizers but was increased to 23:1 in poor metabolizers. This marked change is partly due to an increase in serum exposure of tramadol, but mostly to the large reduction of M1 formation in poor metabolizers (Poulsen et al. 1996). The relative exposure to tramadol and to M1 in extensive human metabolizers is similar to that seen in mice with unimpaired metabolism (Friderichs & Becker 1991). The tramadol to (+)-M1 concentration ratio of 122:1 in the present dog study is even lower than that calculated in human low metabolizers. The low
© 2014 Association of Veterinary Anaesthetists and the American College of Veterinary Anesthesia and Analgesia, 41, 297–304
Tramadol, morphine and tapentadol in dogs B K€ogel et al. formation of M1 from tramadol in dogs is accompanied by a lack of antinociception in this study. A residual effect was not detected. Autoreceptors on descending noradrenergic terminals are not necessarily involved in the a2-adrenoceptor mediated antinociception in the tail-flick test (Fairbanks & Wilcox 1999). This may explain the ineffectiveness of the non-opioid mechanisms of tramadol (spinal uptake inhibition of noradrenaline and serotonin) in this model despite the activity of a2-adrenoceptor agonists. As these agonists may act on targets other than spinal autoreceptors, their analgesic activity differs from that of uptake inhibitors. Due to the aforementioned considerations it is clear that a failure of tramadol in the dog tail flick paradigm does not necessarily relate to a lack of clinical efficacy and does not conflict with reports on the positive outcome of tramadol in canine studies using other pain conditions. In contrast, tramadol has been proved to be effective in clinical chronic pain conditions. Tramadol has been described as successful in the treatment of acute flares of arthritic pain and breakthrough pain in osteoarthritis in dogs (Rychel 2010). A clinical trial study in dogs of different breeds and ages measured analgesia following ovariohysterectomy using a subjective a pain score, and revealed that tramadol (2 mg kg1 IV) was as effective as morphine (0.2 mg kg1 IV) in postoperative pain management, and that these doses were well tolerated (Mastrocinque & Fantoni 2003). More recently published papers have also demonstrated efficacy of tramadol (2 or 3 mg kg IV) in adding to post-operative pain relief after a variety of clinical procedures (Morgaz et al. 2013; Teixeira et al. 2013). The efficacy of tramadol for post-operative analgesia dogs differs from the lack of antinociceptive effect shown in the present study. This difference might result from generation of M1. However, it is not known whether other breeds of dogs are capable of metabolising tramadol to form sufficient amounts of the M1 metabolite to provide antinociception and/or analgesia. In an experimental setting in healthy Greyhounds the oral administration (tablets, mean dose of 9.9 mg kg1) of tramadol increased the pain pressure threshold significantly despite low M1 concentrations (Kukanich & Papich 2011). In these clinical or experimental settings, inhibition of noradrenaline and serotonin uptake and release of serotonin by tramadol can contribute to analgesia (Driessen & Reimann 1992; Raffa et al. 1992; Driessen et al.
1993) In contrast, the present acute pain study differs from these settings in as far that only strong analgesic mechanisms (l-opioid and a2-adrenoceptor agonists) are effective in the tail-flick test. The directly acting l-opioid receptor agonist morphine and the combined l-opioid agonist and noradrenaline uptake inhibitor tapentadol do not need metabolic activation and thus are fully active in the dog tail-flick model (Tzschentke et al. 2006, Tzschentke et al. 2007). In the case of tapentadol, the analgesic effect in moderate to severe pain states is most probably mediated by its l-opioid agonistic activity. No analgesic effect of IV administered tramadol could be detected in this acute pain study in Beagle dogs. The lack of antinociception in this model is likely to be due to the low exposure to the active M1 metabolite which represents the main l-opioid activity of tramadol and the limited role that the other mechanisms contribute to analgesia in this acute pain model. The non-opioid mechanisms of tramadol do not work in this experimental setting, in contrast to many clinical conditions described in the literature, where tramadol is quite active and a useful analgesic drug in dogs especially in more chronic pain states. References Adams ML, Morris DL, Brase DA et al. (1990) Stereoselective effect of morphine on antinociception and endogenous opioid peptide levels in plasma but not cerebrospinal fluid of dogs. Life Sci 48, 917–924. Becker R, Lintz W (1986) Determination of tramadol in human serum by capillary gas chromatography with nitrogen-selective detection. J Chromatogr 377, 213– 220. Driessen B, Reimann W (1992) Interaction of the central analgesic, tramadol, with the uptake and release of 5hydroxytryptamine in the rat brain in vitro. Br J Pharmacol 105, 147–151. Driessen B, Reimann W, Giertz H (1993) Effects of the central analgesic tramadol on the uptake and release of noradrenaline and dopamine in vitro. Br J Pharmacol 108, 806–811. Fairbanks CA, Wilcox GL (1999) Moxonidine, a selective a2-adrenergic and imidazoline receptor agonist, produces spinal antinociception in mice. J Pharmacol Exp Ther 290, 403–412. Friderichs E, Becker R (1991) Correlation of tramadol and M1 serum levels with antinociceptive activity in mice. Abstract no. 36 of the 32nd Spring Meeting, Deutsche Gesellschaft f€ ur Pharmakologie und Toxikologie, Archives of Pharmacology, suppl. to Vol. 343.
© 2014 Association of Veterinary Anaesthetists and the American College of Veterinary Anesthesia and Analgesia, 41, 297–304
303
Tramadol, morphine and tapentadol in dogs B K€ogel et al.
Friderichs E, Felgenhauer F, Jongschaap P et al. (1978) Pharmacological studies on analgesia, dependence on and tolerance of tramadol, a potent analgetic drug. Arzneimittelforschung 28, 122–134. Frink MC, Hennies HH, Englberger W et al. (1996) Influence of tramadol on neurotransmitter systems of the rat brain. Arzneimittelforschung 46, 1029–1036. Gillen C, Haurand M, Kobelt DJ et al. (2000) Affinity, potency and efficacy of tramadol and its metabolites at the cloned human mu-opioid receptor. Naunyn Schmiedebergs Arch Pharmacol 362, 116–121. Giorgi M, Del Carlo S, Saccomanni G et al. (2009) Biopharmaceutical profile of tramadol in the dog. Vet Res Commun 33(Suppl. 1), 189–192. Hawley TA, Wetmore LA (2010) Identification of single nucleotide polymorphisms within exon 1 of the canine mu-opioid receptor gene. Vet Anaesth Analg 37, 79–82. Hennies HH, Friderichs E, Schneider J (1988) Receptor binding, analgesic and antitussive potency of tramadol and other selected opioids. Arzneimittelforschung 38, 877–880. Johnston SA, McLaughlin RM, Budsberg SC (2008) Nonsurgical management of osteoarthritis in dogs. Vet Clin North Am Small Anim Pract 38, 1449–1470. KuKanich B, Papich MG (2004) Pharmacokinetics of tramadol and the metabolite O-desmethyltramadol in dogs. J Vet Pharmacol Ther 27, 239–246. Kukanich B, Papich MG (2011) Pharmacokinetics and antinociceptive effects of oral tramadol hydrochloride administration in Greyhounds. Am J Vet Res 72, 256– 262. Mastrocinque S, Fantoni DT (2003) A comparison of preoperative tramadol and morphine for the control of early early postoperative pain in canine ovariohysterectomy. Vet Anaesth Analg 30, 220–228. Mathews KA (2008) Neuropathic pain in dogs and cats: if only they can tell us if they hurt. Vet Clin Small Anim 38, 1365–1414. McMillan CJ, Livingston A, Clark CR et al. (2008) Pharmacokinetics of intravenous tramadol in dogs. Can J Vet Res 72, 325–331. Morgaz J, Navarrete R, Mu~ noz-Rascon P et al. (2013) Postoperative analgesic effects of dexketoprofen, buprenorphine and tramadol in dogs undergoing ovariohysterectomy. Res Vet Sci 95, 278–282. Negus SS, Vanderah TW, Brandt MR et al. (2006) Preclinical assessment of candidate analgesic drugs: recent advances and future challenges. J Pharmacol Exp Ther 319, 507–514. Paulson SK, Engel L, Reitz B et al. (1999) Evidence for polymorphism in the canine metabolism of the cyclooxygenase 2 inhibitor, celecoxib. Drug Metab Dispos 27, 1133–1142. Pertovaara A (1993) Antinociception induced by alpha2-adrenoceptor agonists, with special emphasis on medetomidine studies. Prog Neurobiol 40, 691–709.
304
Poulsen L, Arendt-Nielsen L, Brøsen K et al. (1996) The hypoalgesic effect of tramadol in relation to CYP2D6. Clin Pharmacol Ther 60, 636–644. Raffa RB, Friderichs E (1996) The basic science aspect of tramadol hydrochloride. Pain Rev 3, 249–271. Raffa RB, Friderichs E, Reimann W et al. (1992) Opioid and nonopioid components independently contribute to the mechanism of action of tramadol, an ‘atypical’ opioid analgesic. J Pharmacol Exp Ther 260, 275–285. Rychel JK (2010) Diagnosis and treatment of osteoarthritis. Top Companion Anim Med 25, 20–25. Saccomanni G, Del Carlo S, Giorgi M et al. (2010) Determination of tramadol and metabolites by HPLC-FL and HPLC-MS/MS in urine of dogs. J Pharm Biomed Anal 53, 194–199. Schr€ oder W, Vry JD, Tzschentke TM et al. (2010) Differential contribution of opioid and noradrenergic mechanisms of tapentadol in rat models of nociceptive and neuropathic pain. Eur J Pain 14, 814–821. Teixeira RC, Monteiro ER, Campagnol D et al. (2013) Effects of tramadol alone, in combination with meloxicam or dipyrone, on postoperative pain and the analgesic requirement in dogs undergoing unilateral mastectomy with or without ovariohysterectomy. Vet Anaesth Analg 40, 641–649. Thomasy SM, Moeller BC, Stanley SD (2007) Comparison of opioid receptor binding in horse, guinea pig, and rat cerebral cortex and cerebellum. Vet Anaesth Analg 34, 351–358. Tzschentke TM, De Vry J, Terlinden R et al. (2006) Tapentadol hydrochloride. Drug. Future 31, 1053–1061. Tzschentke TM, Christoph T, Koegel B et al. (2007) (-)(1R,2R)-3-(3-dimethylamino-1-ethyl-2-methyl-propyl)phenol hydrochloride (tapentadol HCl): a novel mu-opioid receptor agonist/norepinephrine reuptake inhibitor with broad-spectrum analgesic properties. J Pharmacol Exp Ther 323, 265–276. Wu WN, McKown LA, Gauthier AD et al. (2001) Metabolism of the analgesic drug, tramadol hydrochloride, in rat and dog. Xenobiotica 31, 423–441. Yoburn BC, Lutfy K, Candido J (1991) Species differences in mu- and delta-opioid receptors. Eur J Pharmacol 193, 105–108. Received 4 February 2013; accepted 15 July 2013.
Supporting Information Additional Supporting Information may be found in the online version of this article: Appendix S1. Provenance and care of the dogs. Appendix S2. Table demonstrating dog number and dose conditions in first set of experiments. Appendix S3. Analytical method for tramadol and its metabolite M1.
© 2014 Association of Veterinary Anaesthetists and the American College of Veterinary Anesthesia and Analgesia, 41, 297–304
doi:10.1111/vaa.12140
RESEARCH PAPER
Characterisation of tramadol, morphine and tapentadol in an acute pain model in Beagle dogs Babette K€ ogel, Rolf Terlinden & Johannes Schneider Gr€ unenthal GmbH, Gr€ unenthal Innovation, Global Preclinical R&D, Aachen, Germany
Correspondence: Babette K€ ogel, Department of Pharmacology, Gr€ unenthal GmbH, Global Biomedical Sciences, Zieglerstrasse 6, 52078 Aachen, Germany. E-mail: [email protected]
Abstract Objective To evaluate the analgesic potential of the centrally acting analgesics tramadol, morphine and the novel analgesic tapentadol in a pre-clinical research model of acute nociceptive pain, the tailflick model in dogs. Study design Prospective part-randomized pre-clinical research trial. Animals Fifteen male Beagle dogs (HsdCpb:DOBE), aged 12–15 months. Methods On different occasions separated by at least 1 week, dogs received intravenous (IV) administrations of tramadol (6.81, 10.0 mg kg1), tapentadol (2.15, 4.64, 6.81 mg kg1) or morphine (0.464, 0.681, 1.0 mg kg1) with subsequent measurement of tail withdrawal latencies from a thermal stimulus (for each treatment n = 5). Blood samples were collected immediately after the pharmacodynamic measurements of tramadol to determine pharmacokinetics and the active metabolite O-demethyltramadol (M1). Results Tapentadol and morphine induced dosedependent antinociception with ED50-values of 4.3 mg kg1 and 0.71 mg kg1, respectively. In contrast, tramadol did not induce antinociception at any dose tested. Measurements of the serum levels of tramadol and the M1 metabolite revealed only marginal amounts of the M1 metabolite, which explains the absence of the antinociptive
effect of tramadol in this experimental pain model in dogs. Conclusions and clinical relevance Different breeds of dogs might not or only poorly respond to treatment with tramadol due to low metabolism of the drug. Tapentadol and morphine which act directly on l-opioid receptors without the need for metabolic activation are demonstrated to induce potent antinociception in the experimental model used and should also provide a reliable pain management in the clinical situation. The nonopioid mechanisms of tramadol do not provide antinociception in this experimental setting. This contrasts to many clinical situations described in the literature, where tramadol appears to provide useful analgesia in dogs for post-operative pain relief and in more chronically pain states. Keywords acute nociceptive pain, Beagle dog, morphine, tapentadol, tramadol.
Introduction Good pain management in dogs, allowing animals to recover from surgical procedures and return to physiological normality more rapidly, is not only important for veterinary medicinal practice but also when this animal species is used for research. In the pharmaceutical industry, Beagle dogs are used widely to characterise developmental compounds as potential drugs for human therapy. Analgesics used include nonsteroidal analgesics and opioids such as morphine (Johnston et al. 2008; Mathews 297
Tramadol, morphine and tapentadol in dogs B K€ogel et al.
2008) buprenorphine and methadone (Mathews 2008). Recently, tramadol has become popular for the therapy of moderate to severe pain in both humans and in animals. Tramadol is a centrally acting analgesic which has several modes of action. These include activation of opioid receptors (Hennies et al. 1988) preferentially of the l-subtype (Raffa et al. 1992) and enhancement of the extra-cellular concentrations of the monoamine neurotransmitters 5-hydroxytryptamine (serotonin) and noradrenaline by blocking the reuptake transporters and, in the case of serotonin, additional release mechanisms (Driessen & Reimann 1992; Raffa et al. 1992; Driessen et al. 1993). The opioid receptor affinity of tramadol is low, but the compound is metabolised extensively in rodents and humans by O- and N-demethylation, yielding metabolites with a higher l-opioid receptor affinity than that shown by the parent compound. O-demethyl tramadol (M1) is likely to be the main carrier of the opioid analgesic properties of tramadol (Hennies et al. 1988; Frink et al. 1996; Gillen et al. 2000). Tramadol is a racemic compound and the pharmacological activities of its (+)- and ()-enantiomers differ substantially. The l-opioid affinity is represented mainly in the (+)-enantiomer of tramadol and the (+)-M1 metabolite. The noradrenaline uptake inhibitory activity of tramadol resides in the ()-enantiomer, whereas the serotonin uptake and release are influenced by the (+)-enantiomer (Driessen & Reimann 1992; Driessen et al. 1993). The complex activities of both enantiomers of tramadol and the O-demethylmetabolite are complementary and necessary for the full analgesic efficacy (Raffa et al. 1992). There are little data in the literature concerning the use of tramadol in dogs, and dose range recommendations are vague or wide. Most of these studies are confined to pharmacokinetic data without referring to analgesic effects of tramadol in dogs (Wu et al. 2001; KuKanich & Papich 2004; McMillan et al. 2008; Giorgi et al. 2009; Saccomanni et al. 2010). There are only a few concrete reports on analgesic activity of tramadol in dogs either with or without concomitant pharmacokinetic parameters (Mastrocinque & Fantoni 2003; Rychel 2010; Kukanich & Papich 2011; Morgaz et al. 2013; Teixeira et al. 2013) The aim of this study was to conduct a pharmacodynamic investigation where direct measurement of nociception was combined with the measurement 298
of the serum levels of tramadol and the active metabolite O-demethyltramadol (M1). The opioid morphine and the novel centrally-acting analgesic tapentadol which in addition to its l-opioid agonistic activity also inhibits the neuronal uptake of noradrenaline were used as comparators (Tzschentke et al. 2006, 2007). Both act directly on l-opioid receptors without the need for metabolic activation and formation of analgesically relevant metabolites. The results of the analgesic test and the serum concentration measurements are discussed with respect to the species-specific (dog) metabolism and the pain type-dependent contribution of the various modes of analgesic action of tramadol. Material and methods Experiments were performed in accordance with the German Animal Welfare Law. Study protocols were approved by the internal animal welfare officer and the local Government Committee (District Council of Cologne 23.203.2-Gr, FG-PK-99-22) which is advised by an independent Ethics Committee. Animals The experiments were carried out in 15 male Beagle dogs, at the start of the study 12–15 month old and, weighing 11–13.5 kg. Their provenance and general keep are given in Appendix S1. In a first set of experiments tramadol and morphine were tested in a cohort of ten dogs. Each dose or vehicle (0.9% saline solvent of the drugs, used as placebo) group consisted of five dosing conditions (n = 5). Dogs were allocated partly randomly to the dose/treatments of tramadol, morphine and vehicle. It was ensured that dogs which received tramadol also received morphine. The dosing scheme is given in Appendix S2. In a second set of experiments, at a later date and in another cohort of dogs (n = 5), tapentadol was tested in a randomized order and among other research compounds. Tramadol and morphine was tested in the first set experiments, whereas tapentadol was tested in a second set of experiments. The vehicle was tested in each study. When any one dog received a maximum of 10 dosing conditions (over a period of 2 years), the dogs underwent a health-check that looked for physical and behavioural abnormalities. All dogs of this study were rated as healthy and thus suitable for rehoming as pets after the study.
© 2014 Association of Veterinary Anaesthetists and the American College of Veterinary Anesthesia and Analgesia, 41, 297–304
Tramadol, morphine and tapentadol in dogs B K€ogel et al.
Analgesic test
Blood-sample collection and analyses
The tail-flick test was carried out in dogs using a modification of the method described by Adams et al. (1990). The tail flick latency (s), the time to withdraw the tail from a radiant heat source (bulb, 12V/50W) was measured. The apparatus used for nociceptive testing consisted of a hand-device which contained the bulb and was connected via a flexible wire to an apparatus which was synchronised with the activation of the bulb and measured the latency for the tail withdrawal (tail-flick analgesiometer, Labtec, Dr. Hess, Germany). The heat source was adjusted to produce a baseline tail flick latency of 3–5 seconds prior to any of the experiments and was left at a constant setting thereafter. A cut off time of 12 seconds was set to avoid tissue damage of the tail. The tails of the animals were shaved 1 day prior the tests and it was ensured that the skin of the test area was intact after shaving and when performing the trials. The hand-device positioned the bulb on the ventral surface of the tail. The increase in tail-flick latency was defined as antinociception and calculated as percent of the maximal possible effect (% MPE). The maximum possible antinociceptive effect was defined as the lack of a tail-flick during 12 seconds of exposure to the heat stimulus. The percentage of MPE was calculated according to the formula:
Thirty minutes before the start of the study, an 18-gauge soft cannula (Vasofix Braun€ ule, Luer Lock, BBraun Melsungen AG, 34209 Germany) was placed into the vena saphena, and fixed in place with a cohesive flex wrap bandage (Petflex, Medpets.de). Blood samples (2–3 mL, five times per dose) were obtained from the five dogs of the 10 mg kg1 tramadol treatment. Serum was obtained by centrifugation of the samples and they were stored frozen at 20 °C until assay. Samples were obtained at the following time points immediately after the tail-flick test: Pre-dose, 10, 20, 40 and 60 minutes after administration. Samples were analyzed for the concentrations of (+)- and ()-tramadol and (+)- and ()-M1, using a modification of a gas chromatographic method (Becker & Lintz 1986). Further details are given in Appendix S3.
½ðT1 T0 =T2 T0 Þ 100; where T0 and T1 were latencies before and after drug administration and T2 was the cut-off time which was set by the protocol. Drugs Drugs and doses tested were as follows; Tramadol HCl (Tramal, Gr€ unenthal GmbH) 6.81, 10.0 mg kg1, tapentadol HCl (Nucynta, Gr€ unenthal GmbH) 2.15, 4.64, 6.81 mg kg1 and morphine HCl (Merck AG, Germany) 0.464, 0.681, 1.0 mg kg1. The drugs were dissolved in 0.9% sodium chloride solution (Fresenius, Germany) and administered intravenously (IV) in an application volume of 5.0 mL kg1. Doses refer to the hydrochloride salts of the compounds. The vehicle 0.9% sodium chloride solution was used as placebo. All drugs or vehicle were administered slowly (over about 1 minute) into the right cephalic vein via a pre-placed 20-gauge cannula (Sterican, BBraun Melsungen AG, 34209 Germany).
Statistical analysis Tail-flick data were analyzed by a one-way repeated measure ANOVA (i.e. including both design factors, dose and time into the model). Post-hoc comparisons versus control were made with Bonferroni’s test. Effects were considered to be statistically significant if p ≤ 0.05. Statistical analysis was performed using SYSTAT, version 11, for Windows. ED50-values were calculated at the time of the maximal effect (i.e. 10 minutes after administration). Results In the tail-flick test, as measured by withdrawal latencies of the tail of the dogs, there was no antinociception measurable for tramadol at any time for both doses studied (6.81 and 10 mg kg1 IV) (Fig. 1). Tramadol was well tolerated up to a dose of 6.81 mg kg1. 10 mg kg1 was rated as the highest applicable dose due to side effects including timid behaviour, ataxia and, in one of the five dogs, a short lasting convulsion. Sedation was not observed. In contrast, morphine (Fig. 2) and tapentadol (Fig. 3) induced dose-dependent antinociception with ED50 values of 0.71 and 4.3 mg kg1, respectively. Peak effects were attained at 10 minutes after injection which was the time point of first measurement after drug administration. Following administration of morphine 0.681 mg kg1, mild sedation, salivation, ataxia occurred, and one dog vomited. Morphine at
© 2014 Association of Veterinary Anaesthetists and the American College of Veterinary Anesthesia and Analgesia, 41, 297–304
299
Tramadol, morphine and tapentadol in dogs B K€ogel et al.
Figure 1 Lack of effect after intravenous administration of tramadol in acute pain (tail-flick test) in Beagle dogs. Data are presented as % MPE (mean SD), MPE, maximum possible effect; n = 5 animals per group.
Figure 2 Dose-dependent effect after intravenous administration of morphine in acute pain (tail-flick test) in beagle dogs. Data are presented as % MPE (mean SD), *p < 0.05 versus placebo, MPE, maximum possible effect; n = 5 animals per group.
Figure 3 Dose-dependent effect after intravenous administration of tapentadol in acute pain (tail-flick test) in beagle dogs. Data are presented as % MPE (mean SD), *p < 0.05 versus placebo, MPE, maximum possible effect; n = 5 animals per group.
1.0 mg kg1 induced diarrhoea, ataxia, salivation and sedation. However, dogs were calm for most of the time during the study but remained responsive to touch or noise. Tapentadol at the two lower doses was well tolerated. However, at the highest dose (6.81 mg kg1) the dogs showed sedation 300
accompanied by reduced responsiveness, ataxia, salivation and diarrhoea. Tail-flick data were analysed by a one-way repeated measure ANOVA. Morphine: treatment F3,16 = 16.60, p < 0.001 time F4,46 = 18.40, p < 0.001, tapentadol: treatment F3,16 = 15.02, p < 0.001 time
© 2014 Association of Veterinary Anaesthetists and the American College of Veterinary Anesthesia and Analgesia, 41, 297–304
Tramadol, morphine and tapentadol in dogs B K€ogel et al. F4,46 = 28.46, p < 0.001 tramadol: treatment F2,12 = 0.860, p = 0.860 time F3,36 = 1.146, p = 0.344. Figure 4 shows the serum concentration profile of both tramadol enantiomers and that of (+)-M1. Low concentrations of the metabolite (+)-M1 of 11.9– 16.5 ng mL1 were measured, whereas ()-M1 concentrations were below the limit of quantification. On average, (+)-M1 reached only 0.8–1.8% of the (+)-tramadol concentrations. Discussion We intentionally used the tail-flick model to evaluate the opioid-like activity of test substances. We are aware that this model covers only a part of the broader range of clinical efficacy of tramadol. In this study, the l-opioid agonist morphine and the combined l-opioid agonist and noradrenaline uptake inhibitor tapentadol induced full antinociceptive efficacy. In contrast, the IV administration of tramadol, at a maximally tolerated IV dose, did not induce any antinociceptive effect in this canine model of acute pain. This finding is in contrast to the dose-dependent effects of tramadol in tail-flick tests in rodents (Friderichs et al. 1978; Raffa et al. 1992). In the tail-flick paradigm predominantly strong analgesics such as the l-opiate agonists and a2-adrenoceptor agonists are active (Pertovaara 1993; Fairbanks & Wilcox 1999; Negus et al. 2006). The lack of observed effect with tramadol in the dog has to be discussed on the basis of the complex pharmacological mode of action and the need for metabolic transformation, generating a potent l-opioid receptor activating metabolite. The binding affinity of the main metabolite M1 O-demethyl tramadol to l-opioid receptors in rat brain synaptosomes and to the human l-opioid receptor is more than two orders of magnitude
higher than that of the parent molecule (Hennies et al. 1988; Frink et al. 1996; Gillen et al. 2000). Thus, the M1 metabolite and especially its (+)-enantiomer represent the main l-opioid activity of tramadol. Tramadol is metabolized extensively in the Beagle breed of dog, as used in this study. Upon oral administration of tramadol (20 mg kg1) to dogs, 24 metabolites were found (Wu et al. 2001), the O-demethylated (M1) and N-demethylated (M2) metabolites and their conjugates being the major metabolites (Wu et al. 2001), similar to findings in humans. In dogs, the non-conjugated M2 was clearly the main metabolite in comparison to the M1 metabolite. Concomitant monitoring of the serum levels of tramadol and the M1 metabolite in the present study revealed a substantial systemic exposure to tramadol, but only marginal exposure to the (+)-M1 metabolite and even less for ()-M1. After an IV dose of 10 mg kg1 tramadol in our dogs, the highest measured plasma concentrations were 2081 ng mL1 for (+)-tramadol and 1907 ng mL1 for ()-tramadol, but only 17 ng mL1 for (+)-M1. These values suggest a slightly lower exposure when compared to the results of another study in male mixed breed dogs, where IV administration of 1, 2 and 4 mg kg1 tramadol resulted in maximum exposure levels between 820–4664 ng mL1 tramadol and 48– 135 ng mL1 M1 (McMillan et al. 2008). Following oral administration of 4 mg kg1 tramadol (as capsules) to Beagle dogs, a maximum serum concentration of 314 ng mL1 tramadol and 54 ng mL1 M1 were achieved (Negus et al. 2006). In a study by KuKanich & Papich (2004) higher oral dose of 10.2 mg kg1 tramadol resulted in maximum plasma concentrations of 1403 ng mL1 for tramadol and 449 ng mL1 for M1 in Beagle dogs. A clearly lower systemic expo-
Figure 4 Mean ( SD) serum concentrations of tramadol enantiomers and (+)- M1 in 5 Beagle dogs of the 10 mg kg1 IV dose group.
© 2014 Association of Veterinary Anaesthetists and the American College of Veterinary Anesthesia and Analgesia, 41, 297–304
301
Tramadol, morphine and tapentadol in dogs B K€ogel et al.
sure of 216 ng mL1 for tramadol and 6 ng mL1 for M1 maximal concentrations were achieved following only a slightly lower dose of 9.9 mg kg1 (as oral tablets) in greyhound dogs (Kukanich & Papich 2011). The administration of tramadol to our Beagle dogs did not result in an antinociceptive effect in our study, albeit in these dogs morphine and tapentadol provided antinociception in the dog tail flick test, and showed full efficacy. Tapentadol possesses two mechanisms of action one involving l-opioid receptor agonism and the other uptake inhibition of noradrenaline; both these activities are relevant for its analgesic effect (Tzschentke et al. 2007, Schr€ oder et al. 2010). In contrast to tramadol, morphine and tapentadol activate l-opioid receptors without the need for metabolic transformation to an active metabolite. Inhibition of noradrenaline and serotonin uptake, the non-opioid mechanisms of tramadol, does not result in antinociception in experimental setting involving thermal nociceptive stimuli (Raffa & Friderichs 1996). There are at present no data showing the affinity of tramadol to canine l-opioid receptors. Opioid binding studies, however, indicate similar affinities across species from rodents to horses, albeit there are species differences in receptor densities (Yoburn et al. 1991; Thomasy et al. 2007). Further, single nucleotide polymorphisms in the canine MOR gene may account for the differences in opioid responses in various dog populations (Hawley & Wetmore 2010). Potentially, this may also influence pain perception. Thus, the data available on rat and human l-receptor affinities of tramadol and M1 and the information derived from the study in mice under inhibition of metabolic M1 formation, may help to explain the failure of tramadol to induce an antinociceptive activity in the dog tail flick test. In our study there was very little formation of the main tramadol metabolite M1 in dogs, as indicated by serum concentrations below 20 ng mL1 (Fig. 4). This concentration is well below the value of 147 ng mL1 determined at the threshold antinociceptive dose in mice (Friderichs & Becker 1991). Therefore by inference, the formation of the M1 metabolite in dogs may not to be sufficient to induce antinociception. However, there was a high level of tramadol, which by itself can contribute to an antinociceptive effect at least in mice as discussed above. A serum exposure of about 4000 ng mL1 tramadol (2081 ng mL1 (+)-tramadol and 1906 ng mL1 ()-tramadol) was measured 302
10 minutes after the IV administration of 10 mg kg1 in this study. In mice, according to the serum concentrations of 422 ng mL1 at the threshold dose level and 5953 ng mL1 at a full analgesic dose level with blockade of M1 formation might have been expected to be sufficient to induce an antinociceptive effect (Friderichs & Becker 1991). However, in our study, 30 minutes after administration, serum concentrations of 1312 ng mL1 for (+)-tramadol and 1185 ng mL1 for the ()-enantiomer were measured in the Beagle dogs. The total concentration of tramadol thus lies between the concentrations measured for both, threshold and maximum antinociceptive effects as derived from the mouse study (Friderichs & Becker 1991). Furthermore, the concentration of (+)-tramadol is close to its Ki value at the (rat) l-opioid receptor of 4.1 lM, which is equal to 1078 ng mL1 (Frink et al. 1996). Theoretically, from these extrapolations an antinociceptive effect might have been expected in the dog tail-flick test. However, the failure of tramadol to induce an antinociceptive effect shows that high concentrations of non-metabolized tramadol by itself could not compensate for the poor formation of the l-opioid active M1 metabolite, at least in these Beagle dogs. Known side effects which occur immediately after IV administration at 10 mg kg1 excluded an increase of dose in this test model. The formation of the metabolite M1 in humans is mediated by the hepatic enzyme CYP2D6. Poor metabolisers, due to their lack of this CYP2D6 enzyme, have impaired production of M1, so that tramadol has a reduced analgesic effect (Poulsen et al. 1996). There is no information for CYP-specific metabolism of tramadol in dogs. Differences in canine drug metabolism due to polymorphism are described for celecoxib (Paulson et al. 1999). Similarly, it is very likely that different dog populations may exist impacting upon tramadol metabolism. In humans, based on AUC calculations, the relative exposure to tramadol and to (+)-M1 was 7:1 in extensive metabolizers but was increased to 23:1 in poor metabolizers. This marked change is partly due to an increase in serum exposure of tramadol, but mostly to the large reduction of M1 formation in poor metabolizers (Poulsen et al. 1996). The relative exposure to tramadol and to M1 in extensive human metabolizers is similar to that seen in mice with unimpaired metabolism (Friderichs & Becker 1991). The tramadol to (+)-M1 concentration ratio of 122:1 in the present dog study is even lower than that calculated in human low metabolizers. The low
© 2014 Association of Veterinary Anaesthetists and the American College of Veterinary Anesthesia and Analgesia, 41, 297–304
Tramadol, morphine and tapentadol in dogs B K€ogel et al. formation of M1 from tramadol in dogs is accompanied by a lack of antinociception in this study. A residual effect was not detected. Autoreceptors on descending noradrenergic terminals are not necessarily involved in the a2-adrenoceptor mediated antinociception in the tail-flick test (Fairbanks & Wilcox 1999). This may explain the ineffectiveness of the non-opioid mechanisms of tramadol (spinal uptake inhibition of noradrenaline and serotonin) in this model despite the activity of a2-adrenoceptor agonists. As these agonists may act on targets other than spinal autoreceptors, their analgesic activity differs from that of uptake inhibitors. Due to the aforementioned considerations it is clear that a failure of tramadol in the dog tail flick paradigm does not necessarily relate to a lack of clinical efficacy and does not conflict with reports on the positive outcome of tramadol in canine studies using other pain conditions. In contrast, tramadol has been proved to be effective in clinical chronic pain conditions. Tramadol has been described as successful in the treatment of acute flares of arthritic pain and breakthrough pain in osteoarthritis in dogs (Rychel 2010). A clinical trial study in dogs of different breeds and ages measured analgesia following ovariohysterectomy using a subjective a pain score, and revealed that tramadol (2 mg kg1 IV) was as effective as morphine (0.2 mg kg1 IV) in postoperative pain management, and that these doses were well tolerated (Mastrocinque & Fantoni 2003). More recently published papers have also demonstrated efficacy of tramadol (2 or 3 mg kg IV) in adding to post-operative pain relief after a variety of clinical procedures (Morgaz et al. 2013; Teixeira et al. 2013). The efficacy of tramadol for post-operative analgesia dogs differs from the lack of antinociceptive effect shown in the present study. This difference might result from generation of M1. However, it is not known whether other breeds of dogs are capable of metabolising tramadol to form sufficient amounts of the M1 metabolite to provide antinociception and/or analgesia. In an experimental setting in healthy Greyhounds the oral administration (tablets, mean dose of 9.9 mg kg1) of tramadol increased the pain pressure threshold significantly despite low M1 concentrations (Kukanich & Papich 2011). In these clinical or experimental settings, inhibition of noradrenaline and serotonin uptake and release of serotonin by tramadol can contribute to analgesia (Driessen & Reimann 1992; Raffa et al. 1992; Driessen et al.
1993) In contrast, the present acute pain study differs from these settings in as far that only strong analgesic mechanisms (l-opioid and a2-adrenoceptor agonists) are effective in the tail-flick test. The directly acting l-opioid receptor agonist morphine and the combined l-opioid agonist and noradrenaline uptake inhibitor tapentadol do not need metabolic activation and thus are fully active in the dog tail-flick model (Tzschentke et al. 2006, Tzschentke et al. 2007). In the case of tapentadol, the analgesic effect in moderate to severe pain states is most probably mediated by its l-opioid agonistic activity. No analgesic effect of IV administered tramadol could be detected in this acute pain study in Beagle dogs. The lack of antinociception in this model is likely to be due to the low exposure to the active M1 metabolite which represents the main l-opioid activity of tramadol and the limited role that the other mechanisms contribute to analgesia in this acute pain model. The non-opioid mechanisms of tramadol do not work in this experimental setting, in contrast to many clinical conditions described in the literature, where tramadol is quite active and a useful analgesic drug in dogs especially in more chronic pain states. References Adams ML, Morris DL, Brase DA et al. (1990) Stereoselective effect of morphine on antinociception and endogenous opioid peptide levels in plasma but not cerebrospinal fluid of dogs. Life Sci 48, 917–924. Becker R, Lintz W (1986) Determination of tramadol in human serum by capillary gas chromatography with nitrogen-selective detection. J Chromatogr 377, 213– 220. Driessen B, Reimann W (1992) Interaction of the central analgesic, tramadol, with the uptake and release of 5hydroxytryptamine in the rat brain in vitro. Br J Pharmacol 105, 147–151. Driessen B, Reimann W, Giertz H (1993) Effects of the central analgesic tramadol on the uptake and release of noradrenaline and dopamine in vitro. Br J Pharmacol 108, 806–811. Fairbanks CA, Wilcox GL (1999) Moxonidine, a selective a2-adrenergic and imidazoline receptor agonist, produces spinal antinociception in mice. J Pharmacol Exp Ther 290, 403–412. Friderichs E, Becker R (1991) Correlation of tramadol and M1 serum levels with antinociceptive activity in mice. Abstract no. 36 of the 32nd Spring Meeting, Deutsche Gesellschaft f€ ur Pharmakologie und Toxikologie, Archives of Pharmacology, suppl. to Vol. 343.
© 2014 Association of Veterinary Anaesthetists and the American College of Veterinary Anesthesia and Analgesia, 41, 297–304
303
Tramadol, morphine and tapentadol in dogs B K€ogel et al.
Friderichs E, Felgenhauer F, Jongschaap P et al. (1978) Pharmacological studies on analgesia, dependence on and tolerance of tramadol, a potent analgetic drug. Arzneimittelforschung 28, 122–134. Frink MC, Hennies HH, Englberger W et al. (1996) Influence of tramadol on neurotransmitter systems of the rat brain. Arzneimittelforschung 46, 1029–1036. Gillen C, Haurand M, Kobelt DJ et al. (2000) Affinity, potency and efficacy of tramadol and its metabolites at the cloned human mu-opioid receptor. Naunyn Schmiedebergs Arch Pharmacol 362, 116–121. Giorgi M, Del Carlo S, Saccomanni G et al. (2009) Biopharmaceutical profile of tramadol in the dog. Vet Res Commun 33(Suppl. 1), 189–192. Hawley TA, Wetmore LA (2010) Identification of single nucleotide polymorphisms within exon 1 of the canine mu-opioid receptor gene. Vet Anaesth Analg 37, 79–82. Hennies HH, Friderichs E, Schneider J (1988) Receptor binding, analgesic and antitussive potency of tramadol and other selected opioids. Arzneimittelforschung 38, 877–880. Johnston SA, McLaughlin RM, Budsberg SC (2008) Nonsurgical management of osteoarthritis in dogs. Vet Clin North Am Small Anim Pract 38, 1449–1470. KuKanich B, Papich MG (2004) Pharmacokinetics of tramadol and the metabolite O-desmethyltramadol in dogs. J Vet Pharmacol Ther 27, 239–246. Kukanich B, Papich MG (2011) Pharmacokinetics and antinociceptive effects of oral tramadol hydrochloride administration in Greyhounds. Am J Vet Res 72, 256– 262. Mastrocinque S, Fantoni DT (2003) A comparison of preoperative tramadol and morphine for the control of early early postoperative pain in canine ovariohysterectomy. Vet Anaesth Analg 30, 220–228. Mathews KA (2008) Neuropathic pain in dogs and cats: if only they can tell us if they hurt. Vet Clin Small Anim 38, 1365–1414. McMillan CJ, Livingston A, Clark CR et al. (2008) Pharmacokinetics of intravenous tramadol in dogs. Can J Vet Res 72, 325–331. Morgaz J, Navarrete R, Mu~ noz-Rascon P et al. (2013) Postoperative analgesic effects of dexketoprofen, buprenorphine and tramadol in dogs undergoing ovariohysterectomy. Res Vet Sci 95, 278–282. Negus SS, Vanderah TW, Brandt MR et al. (2006) Preclinical assessment of candidate analgesic drugs: recent advances and future challenges. J Pharmacol Exp Ther 319, 507–514. Paulson SK, Engel L, Reitz B et al. (1999) Evidence for polymorphism in the canine metabolism of the cyclooxygenase 2 inhibitor, celecoxib. Drug Metab Dispos 27, 1133–1142. Pertovaara A (1993) Antinociception induced by alpha2-adrenoceptor agonists, with special emphasis on medetomidine studies. Prog Neurobiol 40, 691–709.
304
Poulsen L, Arendt-Nielsen L, Brøsen K et al. (1996) The hypoalgesic effect of tramadol in relation to CYP2D6. Clin Pharmacol Ther 60, 636–644. Raffa RB, Friderichs E (1996) The basic science aspect of tramadol hydrochloride. Pain Rev 3, 249–271. Raffa RB, Friderichs E, Reimann W et al. (1992) Opioid and nonopioid components independently contribute to the mechanism of action of tramadol, an ‘atypical’ opioid analgesic. J Pharmacol Exp Ther 260, 275–285. Rychel JK (2010) Diagnosis and treatment of osteoarthritis. Top Companion Anim Med 25, 20–25. Saccomanni G, Del Carlo S, Giorgi M et al. (2010) Determination of tramadol and metabolites by HPLC-FL and HPLC-MS/MS in urine of dogs. J Pharm Biomed Anal 53, 194–199. Schr€ oder W, Vry JD, Tzschentke TM et al. (2010) Differential contribution of opioid and noradrenergic mechanisms of tapentadol in rat models of nociceptive and neuropathic pain. Eur J Pain 14, 814–821. Teixeira RC, Monteiro ER, Campagnol D et al. (2013) Effects of tramadol alone, in combination with meloxicam or dipyrone, on postoperative pain and the analgesic requirement in dogs undergoing unilateral mastectomy with or without ovariohysterectomy. Vet Anaesth Analg 40, 641–649. Thomasy SM, Moeller BC, Stanley SD (2007) Comparison of opioid receptor binding in horse, guinea pig, and rat cerebral cortex and cerebellum. Vet Anaesth Analg 34, 351–358. Tzschentke TM, De Vry J, Terlinden R et al. (2006) Tapentadol hydrochloride. Drug. Future 31, 1053–1061. Tzschentke TM, Christoph T, Koegel B et al. (2007) (-)(1R,2R)-3-(3-dimethylamino-1-ethyl-2-methyl-propyl)phenol hydrochloride (tapentadol HCl): a novel mu-opioid receptor agonist/norepinephrine reuptake inhibitor with broad-spectrum analgesic properties. J Pharmacol Exp Ther 323, 265–276. Wu WN, McKown LA, Gauthier AD et al. (2001) Metabolism of the analgesic drug, tramadol hydrochloride, in rat and dog. Xenobiotica 31, 423–441. Yoburn BC, Lutfy K, Candido J (1991) Species differences in mu- and delta-opioid receptors. Eur J Pharmacol 193, 105–108. Received 4 February 2013; accepted 15 July 2013.
Supporting Information Additional Supporting Information may be found in the online version of this article: Appendix S1. Provenance and care of the dogs. Appendix S2. Table demonstrating dog number and dose conditions in first set of experiments. Appendix S3. Analytical method for tramadol and its metabolite M1.
© 2014 Association of Veterinary Anaesthetists and the American College of Veterinary Anesthesia and Analgesia, 41, 297–304