An Evaluation of the Efficacy and Tolerability of Oral Tramadol Hydrochloride Tablets for the Treatment of Postsurgical Pain in Children : Anesthesia & Analgesia

Secondary Logo

Journal Logo


An Evaluation of the Efficacy and Tolerability of Oral Tramadol Hydrochloride Tablets for the Treatment of Postsurgical Pain in Children

Finkel, Julia C. MD*,; Rose, John B. MD†,; Schmitz, Michael L. MD‡,; Birmingham, Patrick K. MD§,; Ulma, George A. MD∥,; Gunter, Joel B. MD¶,; Cnaan, Avital PhD#**,; Coté, Charles J. MD§,; Medve, Robert A. MD††,; Schreiner, Mark S. MD†**

Author Information
doi: 10.1213/00000539-200206000-00017
  • Free


Tramadol, a synthetic 4-phenyl-piperidine analog of codeine, is a centrally acting atypical opioid (1). Although tramadol’s mode of action is not completely understood, at least two complementary mechanisms contribute to its effect. Tramadol’s opioid activity results from low affinity binding of the parent compound to μ-opioid receptors and higher binding of the M1 (0-desmethylated) metabolite (2). Tramadol is also a weak inhibitor of norepinephrine and serotonin reuptake (3). The opioid and monoaminergic mechanisms of action are thought to extend the analgesic benefit to both opioid sensitive and insensitive pain. Clinical experience indicates that tramadol lacks many of the side effects typically associated with opioid agonists.

Tramadol’s apparently negligible effect on respiration, as demonstrated in studies in adults and children (4–8), suggests that tramadol may offer a distinct advantage over typical opioid analgesics for the relief of postoperative pain in children. The objective of this randomized, multicenter study was to evaluate the tolerability and efficacy of a single administration of one of two dose levels of oral tramadol tablets for the treatment of postsurgical pain in children and adolescents aged 7–16 yr. We hypothesized that the larger tramadol dose would have a more morphine-sparing effect and commensurate decrease in side effects, reflecting the decreased need for rescue morphine analgesia.


This double-blinded, randomized, multicenter study was conducted at six different institutions, and the IRB at each site approved the study. Parental consent and patient assent were obtained before trial enrollment. Patients were ASA physical status I or II, aged 7–16 yr old, weighing at least 20 kg, with moderate to moderately severe postoperative pain requiring patient-controlled analgesia (PCA) in the immediate postoperative period. Patients were eligible for inclusion onto the study when they were able to tolerate oral intake and when their pain was manageable with PCA bolus morphine without a basal infusion. Patients were excluded if they required the use of sedatives or diphenhydramine, received acetaminophen for fever in the 6 h before dosing, received an injectable analgesic or basal infusion of morphine within 2 h of dosing, or received medications known to inhibit CYP2D6 or CYP3A4 within 1 wk of dosing. Eligible patients were randomized to receive 1 or 2 mg/kg of tramadol as a single oral dose. The study used scored 50 mg tablets in combinations of half and whole tablets.

The Wong-Baker Faces Pain Rating Scale (9) was used to assess pain intensity. If adequate analgesia was not achieved with tramadol, patients received rescue medication with supplemental PCA morphine (0.015–0.025 mg/kg every 7–8 min; 4-h maximum dose of 0.4 mg/kg). If IV access was no longer available after study medication dosing, oral oxycodone tablets or a liquid equivalent (0.15 mg/kg [rounded to the closest 2.5 mg]) every 2 h was substituted as the rescue medication. Patients were encouraged, but not required, to wait 1 h after the tramadol administration before receiving supplemental analgesics.

Safety was assessed by measuring respiratory rate (RR) and Spo2 throughout the study. An Spo2 of ≤90% was considered clinically significant. A RR of ≤10 breaths/min (12–16 yr) or ≤12 breaths/min (7–11 yr) was defined as hypoventilation.

Patients remained in the study facility for the entire 8-h evaluation period after the administration of tramadol, whether or not they received supplemental analgesic medication. Pain assessment, RR, and Spo2 values were recorded at 0.5, 1, 2, 3, 4, 6, and 8 h after dosing or at the time of early termination. At the end of the 8-h evaluation period, or early termination, whichever occurred first, the patient, parent (if available), and the study observer made an overall assessment (selected from excellent, very good, good, fair, and poor) of the quality of pain relief achieved from tramadol.

At the conclusion of the study, patients could elect to continue to receive tramadol, approximately 1–2 mg/kg (maximum single dose 100 mg), every 4–6 h (maximum daily dose was the lesser of either 8 mg/kg or 400 mg) for an additional 24 h.

Efficacy measurements included use of rescue analgesics (morphine or morphine equivalents), pain intensity assessed by the Faces scale, and overall assessment. The amount of morphine or the morphine equivalent dose of oxycodone used was recorded hourly. Both the hourly dose and the cumulative dose were compared between the two treatment groups. The amount of morphine or the morphine equivalent dose of oxycodone used and the amount of morphine or morphine equivalents used adjusted for body weight were summarized hourly using selected percentiles. In addition, treatment comparison using the Kruskal-Wallis test was performed hourly for these two variables. Summary statistics such as mean and sd were also calculated. The 95% confidence intervals (CIs) for each of the hourly doses were calculated.

An increase from baseline of more than one point on the pain scale was considered a failure for the purpose of efficacy evaluations. A comparison between the rates of failure between the two treatment groups was conducted using Fisher’s exact test. In addition, the frequency of each score at each time point, and the differences from baseline, were summarized at each time point. The intermittent missing Faces scores were filled in with linear interpolation using the nonmissing scores immediately preceding and immediately after the missing observation. For time points after the last observed Faces score, the poorest of the last score and the baseline score were used to fill in the missing observation. The scores at the 8-h time point were compared using an exact Kruskal-Wallis test for the values and changes from baseline. Analyses were performed separately for the two age groups (7–11 and 12–16 yr).

The sample size was based on the width of the 95% CIs for anticipated adverse events, which were anticipated to occur at a small rate. Because the largest observed event rate in the adult studies of multiple doses was 26%, a rate of 30% was chosen for the purpose of sample size calculations. The width of a CI around a rate of an adverse event within a treatment group would then be no more than 30% with 40 children within the treatment group. The width of the CI for the entire study sample (n = 80) is no more than 20%.


Eighty-one patients were enrolled. There were no significant differences in demographic variables between the two groups (Table 1). The patients randomized to 1 mg/kg received from 0.41 mg/kg to 1.43 mg/kg with a mean dose of 0.89 mg/kg. Patients randomized to the 2-mg/kg group received from 0.87 mg/kg to 2.5 mg/kg with a mean dose of 1.78 mg/kg.

Table 1:
Patient Demographics


Table 2:
Adverse Events

Eighty patients were included in the analysis of clinical efficacy. For every hour after the tramadol administration, more supplemental analgesics were used by patients in the 1-mg/kg group than in the 2-mg/kg group. In addition, at each hour, at least 50% of the 2-mg/kg group did not require supplemental analgesics. The hourly P values followed a diminishing trend over time, reaching statistical significance at Hour 4 (P = 0.041). At the end of the 8-h study period, a total of 0.16 ± 0.12 mg/kg of supplemental morphine or the morphine equivalent was used in the 1-mg/kg group and a total of 0.09 ± 0.09 mg/kg was used in the 2-mg/kg group (P = 0.006) (Fig. 1). Ten (12.5%) patients (2 in the 1-mg/kg group and 8 in the 2-mg/kg group) did not require any supplemental analgesia during the 8-h study period (P = 0.088, two-sided Fisher’s exact test).

Figure 1:
Cumulative patient-controlled analgesia morphine/morphine equivalents use at each time point, adjusted for body weight (mg/kg): 50th percentile. *P = 0.006; ✦P = 0.008; αP = 0.041; +P = 0.038; ▿P = 0.041; Kruskal-Wallis test for comparing the two treatments at each hour.

The median time to the first supplemental analgesic requirement was 1 h and 10 min in the 1-mg/kg group and 1 h and 43 min in the 2-mg/kg group (P = 0.0865, log-rank test). The rate of a treatment failure was comparable between the two treatment groups at all time points. At one or more time points during the 8-h study period, 12 (31%) patients in the 1-mg/kg group and 10 (24%)in the 2-mg/kg group were defined as treatment failures.

All investigators, and at least 80% of parents and patients, provided an overall assessment of the treatment as being good/excellent. A large percentage of all three assessors rated the 2-mg/kg dose as very good or excellent (investigators, 61% very good/excellent at 2 mg/kg versus 46% at 1 mg/kg; parent, 59% versus 36%; patient, 49% versus 39%). Using the worst overall assessment for each patient, there was no overall significant difference between the two groups. Only five (13%) of patient/parent couples rated tramadol 1 mg/kg as poor and six (15%) of patient/parent couples rated tramadol 2 mg/kg as poor. A comparison between the values show a significant difference between the treatment groups in the parents’ assessments (P = 0.047, Kruskal-Wallis test) and no significant differences in the investigators’ or patients’ assessments. There were no apparent differences between the three groups that were rating.

Of the 81 patients assessable for tolerability, 80 completed the 8-h evaluation period, and one was withdrawn from the study because of the patient’s desire to receive a different analgesic. The incidence of adverse events was small, with vomiting (10%), nausea (9%), pruritus (7%), fever (7%), and rash (4%), as expected, constituting the most frequently reported. For adverse events in which CIs were calculated for both treatment groups the CIs overlapped, indicating similar frequency in both the 1-mg/kg and 2-mg/kg treatment groups. There were no clinically important mean changes in RR or Spo2 measurements between the treatment groups.


The primary efficacy end point for tramadol in this study was the difference in the use of rescue medication between the two dosage groups. The study design gave patients in all groups the ability to self-titrate rescue analgesics to an acceptable pain level. Similar pain ratings on the Faces scale in this context confirmed that both groups were made equianalgesic. For this reason, the pain scores did not track the treatment groups, and rescue opioid use proved to be the more valuable measure. A clear demonstration of a dose-ranging effect was shown with the 2-mg/kg group, which required an average of 42% less morphine or oral oxycodone equivalent when adjusted for weight than the 1-mg/kg group to maintain the same pain rating score.

Tramadol has proven to be an effective analgesic for the treatment of postoperative pain for pediatric patients in a number of clinical trials. However, direct comparisons between trials are difficult because of differences in the routes of the administration (IV, IM, per os (PO), and caudal routes), painful conditions, and ages of patients (8,10–15). Despite this, the findings of this study do parallel those found in other single-dose studies in children with tramadol 2 mg/kg demonstrating increased analgesic efficacy and longer duration of action than 1 mg/kg over a 6-h observation period (8,11).

Tramadol was well tolerated in this single-dose study when administered as a 1-mg/kg or 2-mg/kg dose. Because tramadol was combined with rescue morphine, the reported side effects were those of the combination of tramadol and morphine. The side effect profile, whereas not significantly different between the two groups, did show an increased incidence of side effects at a dose of approximately 1 mg/kg. A potential criticism of the study design is that a placebo control group was not included. It was felt that there was sufficient evidence in the adult population that indicated tramadol’s efficacy that this study would simply examine the dose-ranging effects of the drug. However, a placebo arm would have better established the morphine equivalence of tramadol in children and may have better delineated which adverse effects were most appropriately attributed to rescue opioid versus tramadol.

The larger dose of tramadol, associated with less use of rescue opioids, did not result in an increased incidence of side effects. This is contrary to what has been found in the adult population. In a meta-analysis of 3453 postoperative and dental patients examining single-dose oral tramadol versus placebo, codeine, and combination analgesics, a dose response for analgesia, as well as incidence of side effects (tramadol 50 mg versus 100 mg PO), was demonstrated (16). A summary of adverse events reported in largely adult perioperative Phase 2 and Phase 4 trials demonstrated an incidence of nausea and vomiting of 16.2% and 6.2%, respectively, with tramadol administered IV and 17.8% and 7% when administered IM. In postmarketing surveillance studies, the incidence of nausea and vomiting was 4.2% and 0.5%, respectively (17). The differences between the clinical trials and the postmarketing surveys were thought to reflect the different populations examined, with the clinical trials including hospitalized patients and the postmarketing surveys including outpatients (17).

The most significant adverse events reported during this study and other pediatric trials of postoperative analgesics were nausea and vomiting. The incidence and severity of nausea and vomiting may be mitigated by slow intraoperative injection (where the parenteral formulation is available) or a prophylactic administration of an antiemetic such as metoclopramide or a phenothiazine (3,8,18,19). In a previously reported trial involving children ages 2–10 years who received tramadol 1 or 2 mg/kg IV, nausea and vomiting did not occur (8). The authors attributed this finding to slow IV injection, but it was more likely because of premedication with the phenothiazine trimeprazine. Ondansetron did not reduce the nausea associated with tramadol (20) and, in fact, reduced the analgesic efficacy via its antagonism of serotonin receptor (21). Concomitant use of ondansetron or other serotonin receptor antagonists is therefore contraindicated when choosing tramadol as an analgesic.

Tramadol has compared favorably with other opioids in children. In a study involving children ages 2–10 yr, tramadol 1–2 mg/kg IV caused significantly less respiratory depression than meperidine 1 mg/kg. The mean decrease in RR for patients receiving tramadol 1 mg/kg, tramadol 2 mg/kg, meperidine 1 mg/kg, and placebo was 7.3, 11.4, 31.4, and 1.7 breaths/min, respectively (P < 0.001 for meperidine versus all treatment groups) (8).

The tolerability and efficacy profile of tramadol has also been compared with nonsteroidal antiinflammatory drugs in adults and children. Tramadol has demonstrated better (22) or similar (23) analgesia when compared with dipyrone, ketorolac, naproxen sodium, and clonixin in adult populations. The incidence of nausea and vomiting was similar among the groups (23). Oral tramadol has been compared with oral diclofenac in children 11 years of age and older for posttonsillectomy analgesia and was found to have the same analgesic efficacy (24).

In summary, tramadol 1 or 2 mg/kg PO as a single dose demonstrated a dose-ranging effect when each group was permitted unrestricted access to rescue analgesia in this population of postoperative children 7–16 years of age. The adverse events profile demonstrated no significant difference between the two groups and generally resembled the incidence found in both younger patients (receiving other formulations and means of administration) and in adults (25). Tramadol seems to be both effective and well tolerated in this pediatric population (25).


1. Dayer P, Desmeules J, Collart L. Pharmacology of tramadol. Drugs 1997; 53: 18–24.
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 1992; 260: 275–85.
3. Bamigbade TA, Langford RM. The clinical use of tramadol hydrochloride. Pain Reviews 1998; 5: 155–82.
4. Houmes RM, Voets MA, Verkaaik A, et al. Efficacy and safety of tramadol versus morphine for moderate and severe postoperative pain with special regard to respiratory depression. Anesth Analg 1992; 74: 510–4.
5. Vickers MD, O’Flaherty D, Szekely SM, et al. Tramadol: pain relief by an opioid without depression of respiration. Anaesthesia 1992; 47: 291–6.
6. Tarkkila P, Tuominen M, Lindgren L. Comparison of respiratory effects of tramadol and oxycodone. J Clin Anesth 1997; 19: 582–5.
7. Tarkkila P, Tuominen M, Lindgren L. Comparison of respiratory effects of tramadol and pethidine. Eur J Anaesthesiol 1998; 15: 64–8.
8. Bosenberg A, Ratcliffe S. The respiratory effects of tramadol in children under halothane anaesthesia. Anaesthesia 1998; 53: 960–4.
9. Whaley L, Wong D. Nursing care of infants and children. 4th ed. St Louis: Mosby-Year Book, 1991: 1148.
10. Schaffer J, Piepenbrock S, Kretz FJ, et al. Nalbuphine and tramadol for control of postoperative pain in children. Anaesthesist 1986; 35: 408–13.
11. Schaffer J, Hagermann H, Holzapfel S, et al. Investigation of paediatric postoperative analgesia with tramadol. Fortschr Anasthesiol 1989; 3: 42–5.
12. Barsoum MW. Comparison of the efficacy and tolerability of tramadol, pethidine and nalbuphine in children with postoperative pain: an open randomized study. Clin Drug Invest 1995; 9: 183–90.
13. Roelofse JA, Payne KA. Oral tramadol: analgesic efficacy in children following multiple dental extractions. Eur J Anaesthesiol 1999; 16: 441–7.
14. Prosser DP, Davis A, Booker PD, Murray A. Caudal tramadol for postoperative analgesia in paediatric hypospadias surgery. Br J Anaesth 1997; 9: 293–6.
15. Batra YK, Prasad MK, Arya VK, et al. Comparison of caudal tramadol vs. bupivacaine for post-operative analgesia in children undergoing hypospadias surgery. Int J Clin Pharmacol Ther 1999; 37: 238–42.
16. Moore RA, McQuay HJ. Single-subject data meta-analysis of 3435 postoperative subjects: oral tramadol versus placebo, codeine, and combination analgesics. Pain 1997; 69: 287–94.
17. Cossmann M, Kohnen C, Langford R, et al. Tolerance and safety of tramadol: results of international studies and drug control data [in French]. Drugs 1997; 53: S50–62.
18. Lehmann KA. Tramadol in acute pain. Drugs 1997; 53: 25–33.
19. Vickers MD. The efficacy of tramadol in the treatment of post-operative pain. Rev Contemp Pharmacother 1995; 6: 499–506.
20. Broome IJ, Robb HM, Raj N, et al. The use of tramadol following day-case surgery. Anaesthesia 1999; 54: 289–92.
21. DeWitte JL, Shoenmaekers B, Sessler DI, Deloof T. The analgesic efficacy of tramadol is impaired by concurrent administration of ondansetron. Anesth Analg 2001; 92: 1319–21.
22. Putland AJ, McCluskey A. The analgesic efficacy of tramadol versus ketorolac in day-case laparoscopic sterilization. Anaesthesia 1999; 54: 382–5.
23. Rodriguez MJ, De La Torre MR, Perez-Iraola P, et al. Comparative study of tramadol versus NSAIDs as intravenous continuous infusion for managing postoperative pain. Curr Ther Res 1993; 54: 375–83.
24. Courtney MJ, Cabraal D. Tramadol vs. diclofenac for posttonsillectomy analgesia. Arch Otolaryngol Head Neck Surg 2001; 127: 385–8.
25. Wilder-Smith CH, Hill L, Wilkins J, et al. Effects of morphine and tramadol on somatic and visceral sensory function and gastrointestinal motility after abdominal surgery. Anesthesiology 1999; 91: 639–47.
© 2002 International Anesthesia Research Society