1 Department of Clinical Pharmacy and Toxicology of the Leiden University Medical Center, Leiden, The Netherlands. 2 Juliana's Children Hospital, The Hague, The Netherlands. 3 Centre for Human Drug Research, Leiden, The Netherlands
* Corresponding author. E-mail: j.zwaveling{at}lumc.nl
Accepted for publication February 24, 2004.
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Methods. The pharmacokinetics of rectal tramadol and its active metabolite were studied in 12 young children (age: 16 yr) postoperatively. On the basis of these data, a population model was constructed. Using this model, the pharmacokinetics of different doses of tramadol were calculated.
Results. The pharmacokinetics of rectal tramadol could be adequately described by a one-compartment model. The pharmacokinetic parameters derived from the model indicate that a low variability was present. Elimination half-life was 4.3 (0.2) h (SEM) and the apparent clearance was 16.4 (1.5) litre h1 (SEM).
Conclusions. The study showed that after rectal administration, tramadol is absorbed at a reasonable rate and with a low inter-individual variability in small children. The data also suggested that a rectal dose of tramadol 1.52.0 mg kg1 is therapeutic.
Keywords: analgesia, paediatric ; analgesic administration, rectal ; analgesics opioid, tramadol ; pain
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
It thus seems that tramadol may be suitable to treat postoperative pain in children. However, after i.v. and oral administration, peak concentrations are reached rapidly and this has been associated with postoperative nausea and vomiting.7 This limits the use of tramadol as a postoperative analgesic especially in day surgery. Rectal administration of tramadol may be an alternative in this situation. It is convenient to use, and is established treatment for postoperative pain in adults.8 However, both the pharmacokinetics and effects of rectal tramadol in children are unknown.
We developed a tramadol suppository and investigated its pharmacokinetics. Particular interest was paid to the variability in absorption and the concentrationtime profiles. These data were also used to estimate the therapeutic dose. For this last objective, the requirement was that plateau tramadol concentrations were at least 100 ng ml1, which is the lower end of the effective concentration range in adults.9
![]() |
Patients and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Patients
Patients were included in the study if they were admitted to the hospital for minor surgery and it was planned that they had to stay for postoperative care in the hospital for at least 24 h. Their body weight had to be between 12.5 and 25 kg. Postoperative analgesia would otherwise have included acetaminophen and/or a non-steroidal anti-inflammatory drug. Patients were excluded if they had to undergo abdominal surgery or surgery for which postoperative treatment with opioids was planned. Induction of general anaesthesia was carried out using propofol and alfentanil, sufentanil or remifentanil. In some patients, vecuronium (for intubation) and midazolam (premedication) were used. Maintenance of anaesthesia was with isoflurane or sevoflurane, and if necessary an opiate. Regional block or infiltration was carried out with bupivacaine or lidocaine.
Patients receiving enzyme inhibitors or inducers (CYP450 2D6) and/or drugs acting on the CNS, other than those necessary for anaesthesia, were excluded.
Experimental medication
In the study, suppositories consisting of tramadol 25 mg in hard fat (Witepsol H15) were used. The suppositories were prepared in the LUMC hospital pharmacy and analysed for tramadol content and content uniformity by the quality assurance laboratory of the LUMC hospital pharmacy.
Study days
The suppository was inserted immediately after induction of anaesthesia. Venous blood samples (1 ml; plain tubes) were taken at 0, 0.45, 2, 4, 6, 9 and 24 h after drug administration through an i.v. cannula. The samples were centrifuged for 5 min (speed 4000 rounds min1), and the serum was stored at 20°C until analysis.
Analytical procedures
Serum samples were analysed for tramadol and O-desmethyltramadol using a modified high pressure liquid chromatography method.10 11 To 200 µl homogenized serum, 10 µl verapamil (internal standard, 5 µg ml1), 100 µl ammonium hydroxide (25%), and 2 ml methyl tertbutylether were added. After vortex mixing for 15 s and centrifugation (5 min, 4000 rounds min1), the organic layer was removed and evaporated under nitrogen (40°C). The dry extract was reconstituted in 100 µl of methanol and transferred into the microvial of the autosampler (Gynkotek). Subsequently, 40 µl was injected onto the analytical column (Hypersil silica 3 µm, 125 x 4.6 mm ID, Shandon). The signal was quantified using fluorescence detection (excitation wavelength: 275 nm; emission wavelength of 310 nm; Waters). The mobile phase consisted of perchloric acid (70%, 25 µl) in methanol (500 ml), and the flow-rate was 1 ml min1. Calibration curves were obtained by adding tramadol and O-desmethyltramadol to drug-free human serum in the range 25500 ng ml1 with a fixed amount of internal standard. Precision and accuracy studies in plasma showed the coefficient of variation to be less than 10%, with a high accuracy for both within-day (n=6) and day-to-day (n=7) assays (<8%). Total recovery of verapamil, tramadol, and O-desmethyl-tramadol was 100, 95, and 89%, respectively. The lower limit of quantification for tramadol and O-desmethyltramadol was 5 and 3 ng ml1, respectively.
Data analysis
The serum concentrationtime profiles of tramadol and O-desmethyltramadol were analysed using a population pharmacokinetic approach with non-linear mixed effect modelling (Version V; NONMEM Project Group, UCSF, San Francisco, CA). First order conditional estimation was used with the interaction option, assuming multiplicative inter- and intra-subject error distributions. The parameters that were estimated were the absorption half-life, the elimination half-life, and the apparent clearance (Cl/F), in which F is the fractional bioavailability. The resulting empirical Bayes estimates were used in a simulation to predict maximal concentrations, and the time interval during which tramadol concentrations would exceed 100 ng ml1 after administration of doses of 1 or 2 mg kg1 in the population were included in this study.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
One patient expelled the suppository shortly after administration. The data of this patient were not analysed and replaced by data from a newly recruited subject. Three subjects had minor gastrointestinal complaints during the postoperative course, resulting in vomiting in two patients (at 56 h after administration). Depression of respiration was not observed.
Tramadol pharmacokinetics
The average concentrationtime curves of tramadol and O-desmethyltramadol are shown in Figure 1. Mean (SD) maximal plasma concentrations of tramadol and its metabolite were 200 (60) and 35 (15) ng ml1. The mean time of the maximal serum concentrations of tramadol and its metabolite was 2.4 (1.0) and 3.9 (1.1) h after administration.
|
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
This study showed that after rectal administration, tramadol was well absorbed and showed a relatively low variability in absorption and clearance. It may have been preferable to perform a study with both i.v. and rectal administration of tramadol to fully characterize the kinetics (especially the fractional bioavailability). This would however, have also required us to study the pharmacokinetics of an i.v. dose of tramadol, in the same children. Because of the additional burden on the children for research purposes only, we decided against this for ethical reasons. Notwithstanding, some important observations could be made.
The study showed again that it is not justified to extrapolate data from adults to (young) children. Both primary pharmacokinetic parameters, clearance and volume of distribution, were different from those reported in adults. The apparent clearance (CL/F) in children (0.93 litre kg1 h1) was twice as high as in adults (0.49 litre kg1 h1), and the apparent volume of distribution (V/F) in children (6.1 litre kg1) was slightly higher than in adults (4.0 litre kg1).12 The latter may be attributed to the higher water to fat ratio in children. Although comparisons between studies in different groups of patients should be made with care, these findings indicate that the shorter elimination half-life in children is likely as a result of a higher tramadol clearance. This is in agreement with the observation of a relationship between liver size and the rate of metabolism of hepatically metabolized drugs, which cannot be explained by an increased activity of cytochrome P450.13 The difference in tramadol clearance between children and adults after rectal administration is in contrast to the finding that after i.v. administration no difference in clearance between the two age groups was found.14 Apparently, there are differences in the degree of absorption after rectal administration between children and adults. One likely explanation for this is a more alkaline pH of the rectal mucosa in children.15 It is important to stress that rectal absorption of tramadol showed low variability, which is in contrast to the rectal absorption of morphine, which can be poor and variable in children.16 17
The analgesic activity of O-desmethyltramadol (M1) is one to four times higher than the parent compound,9 which is reflected by the lower analgesic efficacy of tramadol in patients with low serum concentrations of M1.18 The mean maximum serum concentration of the active metabolite in our patients was 35 ng ml1, which is roughly similar to concentrations found in adults9 and older children.19 It is likely that in young children, O-desmethyltramadol also contributes to the analgesic effect of rectally administered tramadol. The contribution of the metabolite to the analgesic effect may be even larger, as the ratio of the AUC of the metabolite over the AUC of parent tramadol is twice as high in children (28%) as in adults (14%).12
Tramadol is metabolized primarily by cytochrome P4502D6 for which polymorphisms have been described. Variability in enzyme capacity may thus play a role in the disposition and effects of tramadol.18 20 Variability in the AUC in this pharmacokinetic study was limited, but it must be appreciated that only a limited population was studied. In future studies of rectal tramadol in children, this issue should be considered.
The population model was also used to simulate the pharmacokinetics of different doses of tramadol and to estimate the period during which analgesia was likely to be achieved. This simulation suggested that rectal administration of tramadol 1 mg kg1 is unlikely to result in therapeutic concentrations. This is corroborated by the observation that the analgesic effect of i.v. tramadol 2 mg kg1 is superior to a dose of 1 mg kg1 in children.4
The design of the study only allows the reporting of major adverse events. In two patients vomiting occurred. However, it is important to realise that vomiting is frequently reported after anaesthesia and that this study was not appropriate to assess the true incidence of vomiting.
In conclusion, the results of this study indicate that in small children, rectally administered tramadol is absorbed with low variability and shows limited variability in its clearance. Simulations suggest that at least tramadol 1.5 mg kg1 is necessary to reach therapeutic concentrations. Further prospective studies are warranted in which the efficacy of rectal tramadol vs standard treatment analgesic in children is investigated.
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
2 Barsoum MW. Comparison of the efficacy and tolerability of tramadol, pethidine and nalbufine in children with postoperative painan open randomized study. Clin Drug Invest 1995; 9: 18390[ISI]
3 Schäffer J, Piepenbrock S, Kretz FJ, Schonfeld C. Nalbuphine and tramadol for the control of postoperative pain in children. Anaesthesist 1986; 35: 40813[ISI][Medline]
4 Schäffer J, Hagemann S, Holzapfel S, Panning B, Piepenbrock S. Investigation of paediatric postoperative analgesia with tramadol. Fortschr Anästhesiol Notfall-Intensivmed 1989; 3: 425
5 Finkel JC, Rose JB, Schmitz ML, et al. An evaluation of the efficacy and tolerability of oral tramadol hydrochloride tablets for the treatment of postsurgical pain in children. Anesth Analg 2002; 94: 146973
6 Silvasti M, Tarkkila P, Tuominen M, Svartling N, Rosenberg PH. Efficacy and side effects of tramadol versus oxycodone for patient-controlled analgesia after maxillofacial surgery. Eur J Anaesthesiol 1999; 16: 8349[CrossRef][ISI][Medline]
7 Petrone D, Kamin M, Olson W. Slowing the titration rate of tramadol HCl reduces the incidence of discontinuation due to nausea and/or vomiting: a double-blind randomized trial. J Clin Pharm Ther 1999; 24: 11523[CrossRef][ISI][Medline]
8 Lintz W, Barth H, Osterloh G, Schmidt-Bothelt E. Pharmacokinetics of tramadol and bioavailability of enteral tramadol formulations. 3rd Communication: suppositories. Arzneimittelforschung 1998; 48: 88999[ISI][Medline]
9 Lehmann KA, Kratzenberg U, Schroeder-Bark B, Horrichs-Haermeyer G. Postoperative patient-controlled analgesia with tramadol: analgesic efficacy and minimum effective concentrations. Clin J Pain 1990; 6: 21220[ISI][Medline]
10 Nobilis M, Pastera J, Anzenbacher P, Svoboda D, Kopecky J, Perlik F. High-performance liquid chromatographic determination of tramadol in human plasma. J Chromatogr B Biomed Appl 1996; 681: 17783[CrossRef][Medline]
11 Paar WD, Frankus P, Dengler HJ. High-performance liquid chromatographic assay for the simultaneous determination of tramadol and its metabolites in microsomal fractions of human liver. J Chromatogr B Biomed Appl 1996; 686: 2217[CrossRef][Medline]
12 Lee CR, McTavish D, Sorkin EM. Tramadol. A preliminary review of its pharmacodynamic and pharmacokinetic properties, and therapeutic potential in acute and chronic pain states. Drugs 1993; 46: 31340[ISI][Medline]
13 Murry DJ, Crom WR, Reddick WE, Bhargava R, Evans WE. Liver volume as a determinant of drug clearance in children and adolescents. Drug Metab Dispos 1995; 23: 11106[Abstract]
14 Murthy BV, Pandya KS, Booker PD, Murray A, Lintz W, Terlinden R. Pharmacokinetics of tramadol in children after i.v. or caudal epidural administration. Br J Anaesth 2000; 84: 3469[Abstract]
15 Hunfeld JA, Passchier J. Pain and pain measurement in children. Ned Tijdschr Geneeskd 1995; 139: 12227[Medline]
16 Westerling D. Rectally administered morphine: plasma concentrations in children premedicated with morphine in hydrogel and in solution. Acta Anaesthesiol Scand 1985; 29: 6536[ISI][Medline]
17 Gourlay GK, Boas RA. Fatal outcome with use of rectal morphine for postoperative pain control in an infant. BMJ 1992; 304: 7667[ISI][Medline]
18 Poulsen L, Arendt-Nielsen L, Brosen K, Sindrup SH. The hypoalgesic effect of tramadol in relation to CYP2D6. Clin Pharmacol Ther 1996; 60: 63644[ISI][Medline]
19 Wilson JT, Jeroudi M, Kearns G, Liao S, Medve RA. Disposition of tramadol in children. 30th Eur Symp Clin Pharmacy 2001; Abstract 542
20 Stamer UM, Lehnen K, Hothker F, et al. Impact of CYP2D6 genotype on postoperative tramadol analgesia. Pain 2003; 105: 2318[CrossRef][ISI][Medline]