Tramadol is the most commonly used centrally acting analgesic for moderate and severe pain in Finland  and is widely used elsewhere in the world. Its main analgesic action is mediated by μ-opioid receptors . An active metabolite of tramadol, O-desmethyltramadol, is also a weak μ-opioid receptor agonist. Tramadol has another mechanism of action: promoting serotonin release and inhibiting norepinephrine reuptake in neural tissues, thereby enhancing inhibition of pain transmission in the spinal cord .
Interactions between tramadol and anaesthetic drugs have been suspected. Tramadol has been implicated in cases of awareness during enflurane anaesthesia . In patients with isoflurane-nitrous oxide-oxygen anaesthesia, activation in the EEG suggesting lightening level of hypnosis has been reported [5,6]. These features included increased frequencies (median power frequency, spectral edge) and decreased amplitudes. The Swedish Medical Products Agency has received two reports concerning a suspected pharmacodynamic interaction between tramadol and propofol. Two patients on regular tramadol medication needed excessive doses of propofol for induction of anaesthesia. A 31-yr-old female and a 45-yr-old male were both reported to have been only mildly sedated after 10 mg kg−1 of propofol. The doses were fivefold higher than that recommended for induction of anaesthesia in the summary of product characteristics (1.5-2 mg kg−1) and should have induced deep general anaesthesia in these patients .
Entropy monitoring is a novel method to measure the hypnotic component of anaesthesia. The Entropy Module (GE Healthcare Finland, Helsinki, Finland) collects an electrical biosignal from the forehead consisting of both EEG and electromyogram (EMG). The analysis and processing results in two numbers, state entropy (SE) and response entropy (RE) . Entropy has been evaluated for propofol anaesthesia [9,10].
The effect of regular tramadol medication on anaesthetic requirements has not been studied previously. The aim of this study was to investigate whether patients with regular tramadol medication need more propofol in the induction of general anaesthesia to reach a similar level of hypnosis than their age and gender matched controls. The primary end-points of our study were propofol doses needed to reach two clinical stages: loss of consciousness and appearance of burst suppression pattern in EEG.
A two-centre, prospective, open study was approved by the Ethics Committees of both participating hospitals, Kuopio and Helsinki University Hospitals. All studied patients gave their written informed consent. Forty-six ASA Physical status I-II patients, scheduled for elective orthopaedic surgery under general anaesthesia, were studied. Out of these, 23 had used tramadol regularly for a minimum of 2 weeks, and at least 100 mg daily during the last week before surgery. The patients scheduled for elective surgery under general anaesthesia were asked on a preoperative visit about their medication. Those patients who used tramadol regularly, and had no contraindications, were asked to participate in the study. The 23 controls did not use tramadol. Exclusion criteria were body mass index over 30 kg m−2, disease affecting the central nervous system, alcohol or drug abuse and use of any drug affecting the central nervous system or propofol or tramadol pharmacokinetics or pharmacodynamics.
EEG and entropy recordings
Entropy is measured from two different frequency bands, SE from 0.8-32 Hz representing mainly the EEG range and RE from 0.8-47 Hz, containing some frontal EMG activity as well. Entropy parameters for RE range from 0 (suppression of EEG) to 100 (awake) and for SE from 0 to 91. During light anaesthesia, a strong stimulus may increase frontal EMG activity causing RE to increase above SE (10).
The entropy module of the S/5™ Monitor (GE Healthcare Finland, Helsinki, Finland) was used to collect the data of SE and RE. A disposable Entropy™ Sensor composed of a self-adhesive flexible band incorporating three EEG electrodes was applied to the forehead and the right temple according to the manufacturer's recommendations. Before electrode application, the skin was wiped with an alcohol swab and allowed to dry. Electrode impedance was considered acceptable if below 7.5 kΩ. The EEG sampling rate was 400 Hz. RE, SE and time domain EEG were downloaded to a laptop computer using S5/Collect software (GE Healthcare Finland, Helsinki, Finland) for offline analysis.
No premedication was used, but patients on tramadol medication were given their regular tramadol dose according to their customary timetable. An intravenous (i.v.) infusion was started with Ringer's acetate solution. Non-invasive blood pressure, pulse oximetry and heart rate were continuously monitored with the S/5™ Monitor, as were inspired fraction and end-tidal concentrations of oxygen and carbon dioxide. Entropy recording was started before induction of anaesthesia.
In all patients, anaesthesia was induced with propofol (Propofol Lipuro, 10 mg mL−1; B. Braun Medical, Espoo, Finland) infused constantly at a rate of 1 mg kg−1 min−1 with an automated syringe driver pump. During the preanaesthetic preparations, the patients were instructed to hold a 100 mL plastic saline bottle in their hand, hanging downward, as long as possible. This instruction was repeated just as the propofol infusion was initiated. Dropping the bottle was defined as loss of consciousness. At that point, propofol dose, SE and RE values were recorded. Propofol infusion was continued until SE and RE were below 50. At that point, an 80 mA tetanic stimulation was applied for 30 s over the ulnar nerve at the wrist to estimate purposeful reaction to a constant nociceptive signal. If SE and RE remained below 50 after tetanic stimulation, laryngoscopy and tracheal intubation was performed. If SE or RE increased over 50 during tetanic stimulation or laryngoscopy, the procedure was ceased, and additional propofol bolus doses of 0.5 mg kg−1 were given until RE and SE remained below 50. Tracheal intubation was facilitated with rocuronium 0.5 mg kg−1. After intubation, propofol infusion was continued until a burst suppression pattern was detected in the EEG by the entropy algorithm and burst suppression rate% increased above zero on the monitor display. The total propofol dose from induction of anaesthesia to burst suppression was recorded. For the screening of possible awareness and explicit recall, the patients were interviewed in the recovery room.
An estimate of the sample size required was based on data from the study of Goyal and colleagues , which showed a mean propofol dose of 1.4 (SD 0.4) mg kg−1 was needed to achieve a bispectral index of 50. A 30% change in the propofol requirement was estimated to be clinically significant. With power analysis, it was calculated that 20 patients per group were needed to detect a 30% change in propofol dose requirement (two-sided) with 80% power and an α level of 0.05.
The statistical analyses were performed using SPSS version 11.5 (SPSS Inc., Chicago, USA). To compare the two groups, χ2 analysis was used for categorical variables, U-test for continuous data and Wilcoxon signed rank sum test for paired comparisons within the group. To see the effect of gender on the dose requirement for propofol, χ2-test was used. A P-value <0.05 was considered statistically significant. The data are expressed as median with range unless otherwise stated.
Forty-nine patients were asked and 47 agreed to participate in the study. One patient was not studied because of the busy surgical schedule on that day. The groups were similar in age, height, weight, ASA Physical status and gender. The patient characteristics are shown in Table 1. A protocol deviation was noted in the tramadol group where one patient had used tramadol only for 1 week (300 mg daily for 7 days).
The median (range) dose of propofol required to produce loss of consciousness was similar in the tramadol group and controls, 2.0 (1.0-5.5) mg kg−1 and 2.4 (0.9-8.3) mg kg−1 (P = 0.95), respectively. The mean dose of propofol in the tramadol group was 2.4 (SD 1.1) mg kg−1 and 2.4 (1.5) mg kg−1 in the controls. The amount of propofol required for a burst suppression pattern to appear in the EEG was 5.8 (3.9-12.7) mg kg−1 and 6.4 (2.9-15.1) mg kg−1 (P = 0.89) in the tramadol group and controls. In the tramadol group, 10 out of 23 and in the control group eight out of 23 needed a rescue dose of propofol (0.5 mg kg−1 (0.5-1.0 mg kg−1)) after the tetanic stimulus, because SE or RE increased above 50. RE-SE difference before applying the nociceptive tetanic stimulation was similar in the patients reacting (7 (1-12)) and in those not reacting to the nociceptive stimulus (5 (0-21)) (P = 0.84). There were no differences between genders in the amounts of propofol to reach the end-points.
The SE and RE values during induction of anaesthesia are shown in Table 2, and in Figures 1 and 2.
None of the patients had explicit recall of any events under anaesthesia. In three control patients, SE and RE increased over 80 after tracheal intubation. These patients received propofol 0.5 mg kg−1 as a rescue dose. Three patients in the tramadol group and one in the control group showed myoclonic activity during induction of anaesthesia. The raw EEG of these three patients showed mainly normal patterns consistent with anaesthesia, with a few solitary spikes. None of these four patients had any concomitant medication.
The present study indicates that regular use of tramadol did not influence the dose of propofol required for anaesthetic induction, i.e. to reach the predefined pharmacodynamic end-points, loss of consciousness and appearance of a burst suppression pattern in the EEG. The required doses did not differ between tramadol users and controls and they were in line with those of Mustola and colleagues . However, there was a substantial inter-individual variation in the amount of propofol needed to achieve the end-points in both groups. In controls, the difference between the lowest and the highest dose of propofol to achieve loss consciousness was 9.2-fold and to achieve a burst suppression pattern was 5.2-fold. This considerable variability is currently not sufficiently appreciated in clinical practice, and the individual response to propofol should be observed by means of EEG monitoring to avoid unintentional awareness during induction of anaesthesia and tracheal intubation.
A limitation of our study is that we did not measure the tramadol plasma concentrations. Thus, we are unable to evaluate possible correlation between propofol requirement and tramadol concentration in blood.
The propofol dose required to induce anaesthesia is dependent on both recognized and unknown patient factors [13,14]. In our study the inter-individual variation was larger than usually reported. A possible reason for this may be the heterogeneity of our material, concerning both age (range 21-84 yr) and weight (range 45-112 kg), although excessively obese patients were excluded. Our study was neither designed nor powered to detect age-dependent dose requirements of propofol which are known to exist , and these may have affected the results. However, there was no correlation between propofol requirement and age or weight of the patients.
One of our end-points was the burst suppression pattern in EEG recorded and detected by the entropy monitor. Elderly people may sometimes have very low-amplitude EEG during anaesthesia. Whether such low-amplitude EEG could become erroneously interpreted as burst suppression by the entropy algorithm has not been studied. Neither did we study, how accurate the entropy module was in detecting true suppression periods and what was the delay.
Previous clinical studies have raised a question about the effects of tramadol on EEG activity. Dose-dependent activation of the EEG after a single i.v. dose of tramadol has been reported. Features shown in time domain EEG related with tramadol are typical to light level of hypnosis, such as increased frequencies and decreased amplitudes. However, these changes were small and supposed to be clinically insignificant .
As in our study on long-term tramadol use, there was no evidence of increased risk of awareness during general anaesthesia in patients who were given tramadol acutely [16,17]. In these studies, tramadol was administered during sevoflurane- or propofol-remifentanil anaesthesia and no changes were seen in bispectral index.
Surgical stress significantly modifies the depth of anaesthesia. In the present study, the target level of hypnosis was set to SE and RE equal to 50, which is considered adequate for surgical anaesthesia. During the study period, nociceptive stimuli were limited to tetanic stimulation and tracheal intubation, which may have caused lower stress level than skin incision. It has been suggested that EEG-activating effect of tramadol depends on the concentration of the volatile anaesthetics . The fact that we did not find any significant EEG changes associated with regular tramadol use may be partly due to the study protocol resulting in rapid progression to a relatively deep level of propofol-induced hypnosis.
Adequate analgesia reduces the need for hypnotic medication during anaesthesia. In experimental studies, the analgesic effect of tramadol has been shown to reduce the minimum alveolar concentration of isoflurane in rats, and this reduction is inhibited by the opioid antagonist, naloxone . In clinical practice, many more interactions may exist, for example, the analgesic efficacy of tramadol is reduced by the concurrent use of ondansetron, a 5-HT antagonist , and enhanced by i.v. magnesium sulphate . Tramadol has also been shown to have potential for important drug interactions with monoamine oxidase inhibitors, selective serotonin reuptake inhibitors [21,22] and tricyclic antidepressants . These interactions may result in serotonin syndrome. Manifestations may include confusion, psychosis, agitation, tachycardia, severe shivering, myoclonus, hyperreflexia and fever. Concurrent use of other central nervous system depressants may fortify the effects of tramadol.
Both tramadol and propofol may induce convulsive movements [22,24-26]. Four of our patients (three in the tramadol group and one in the control group) developed myoclonic activity in all limbs during propofol induction. During these movements, there were mainly normal anaesthesia patterns, with only some occasional single spikes as a slightly abnormal feature in the raw EEG recordings. Tramadol use may also provoke true convulsions in patients with a lowered convulsion threshold. Predisposing factors are concomitant use of other drugs, which lower the convulsion threshold like selective serotonin reuptake inhibitors, tricyclic antidepressants and antipsychotics [21-23]. None of the four patients with myoclonic activity in the present study had any concomitant medications.
We used entropy monitoring to quantify similar EEG effects for all patients. Three patients in the control group responded to tracheal intubation by an increase in SE and RE exceeding 80 after laryngoscopy and endotracheal intubation, despite the fact that an 80 mA tetanic stimulation for 30 s prior to tracheal intubation had not increased entropy indices or provoked any reactions. None of these patients reported memories of the incident. However, we interviewed the patients only once after the anaesthesia to detect possible awareness. It has been shown that three successive interviews may provide more accurate information concerning the incidence of awareness . Examples of unintentional awareness during propofol anaesthesia with paralysis have been reported [28,29].
In conclusion, regular tramadol use seems not to affect the propofol requirement for induction of anaesthesia. However, a very large inter-individual variation was found in the amount of propofol needed to achieve loss of consciousness and a burst suppression pattern in EEG. Substantially greater amounts of propofol than the recommended dose is needed to achieve unconsciousness in many patients.
2. Raffa RB, Friderichs E, Reimann W et al.
Opioid and nonopioid components independently contribute to the mechanism of action of tramadol
, an ‘atypical’ opioid analgesic. J Pharmacol Exp Ther
3. Grond S, Sablotzki A. Clinical pharmacology of tramadol
. Clin Pharmacokinet
4. Lehmann KA, Horrichs G, Hoeckle W. The significance of tramadol
as an intraoperative analgesic. A randomized double-blind study in comparison with placebo. Anaesthesist
5. Coetzee JF, Maritz JS, du Toit JC. Effect of tramadol
on depth of anaesthesia. Br J Anaesth
6. Vaughan DJ, Shinner G, Thornton C, Brunner MD. Effect of tramadol
on electroencephalographic and auditory-evoked response variables during light anaesthesia. Br J Anaesth
7. Hallberg E, Wadalius M. Minskad effect av propofol
vid kombination med tramadol
. Information från Läkemedelsverket
8. Viertiö-Oja H, Maja V, Särkelä M et al.
Description of the entropy
algorithm as applied in the Datex-Ohmeda S/5 Entropy
Module. Acta Anaesthesiol Scand
9. Vakkuri A, Yli-Hankala A, Talja P et al.
Time-frequency balanced spectral entropy
as a measure of anesthetic drug effect in central nervous system during sevoflurane, propofol
, and thiopental anesthesia. Acta Anaesthesiol Scand
10. Vanluchene AL, Vereecke H, Thas O et al.
as an electroencephalographic measure of anesthetic drug effect: a comparison with bispectral index and processed midlatency auditory evoked response. Anesthesiology
11. Goyal P, Puri GD, Pandey CK, Srivastva S. Evaluation of induction doses of propofol
: comparison between endstage renal disease and normal renal function patients. Anaesth Intensive Care
12. Mustola ST, Baer GA, Neuvonen PJ, Toivonen KJ. Requirements of propofol
at different end-points without adjuvant and during two different steady infusions of remifentanil. Acta Anaesthesiol Scand
13. Smith I, White PF, Nathanson M, Gouldson R. Propofol
. An update on its clinical use. Anesthesiology
14. McCleane GJ, Fogarty DF, Watters CH. Factors that influence the induction dose of propofol
15. Schnider TW, Minto CF, Shafer SL et al.
The influence of age on propofol
16. Fodale V, Tescione M, Roscitano C et al.
Effect of tramadol
on bispectral index during intravenous anaesthesia with propofol
and remifentanil. Anaesth Intensive Care
17. Fodale V, Pratico C, Tescione M et al. Tramadol
does not modify the bispectral index during anaesthesia with sevoflurane and remifentanil. Br J Anaesth
18. de Wolff MH, Leather HA, Wouters PF. Effects of tramadol
on minimum alveolar concentration (MAC) of isoflurane in rats. Br J Anaesth
19. De Witte JL, Schoenmaekers B, Sessler DI, Deloof T. The analgesic efficacy of tramadol
is impaired by concurrent administration of ondansetron. Anesth Analg
20. Tauzin-Fin P, Sesay M, Delort-Laval S, Krol-Houdek MC, Maurette P. Intravenous magnesium sulphate decreases postoperative tramadol
requirement after radical prostatectomy. Eur J Anaesthesiol
21. Gnanadesigan N, Espinoza RT, Smith R, Israel M, Reuben DB. Interaction of serotonergic antidepressants and opioid analgesics: is serotonin syndrome going undetected? J Am Med Dir Assoc
22. Mahlberg R, Kunz D, Sasse J, Kirchheiner J. Serotonin syndrome with tramadol
and citalopram. Am J Psychiatry
23. Gardner JS, Blough D, Drinkard CR et al. Tramadol
and seizures: a surveillance study in a managed care population. Pharmacotherapy
24. Islander G, Vinge E. Severe neuroexcitatory symptoms after anaesthesia - with focus on propofol
anaesthesia. Acta Anaesthesiol Scand
25. Walder B, Tramer MR, Seeck M. Seizure-like phenomena and propofol
: a systematic review. Neurology
26. DeFriez CB, Wong HC. Seizures and opisthotonos after propofol
anesthesia. Anesth Analg
27. Sandin RH, Enlund G, Samuelsson P, Lennmarken C. Awareness during anaesthesia: a prospective case study. Lancet
28. Ruppen W, Ruesch S, Shupfer G, Christen P. Propofol
and questionable lack of sedation. Acta Anaesthesiol Scand
29. Nordstrom O, Engstrom AM, Persson S, Sandin R. Incidence of awareness in total i.v. anaesthesia based on propofol
, alfentanil and neuromuscular blockade. Acta Anaesthesiol Scand