Secondary Logo

Journal Logo

Technology, Computing, and Simulation: Research Report

The Efficacy and Safety of Oral Immediate-Release Oxymorphone for Postsurgical Pain

Gimbel, Joseph MD*; Ahdieh, Harry PhD

Author Information
doi: 10.1213/01.ANE.0000132548.91622.B3
  • Free

Most patients who undergo major orthopedic surgery experience moderate-to-severe pain and receive inadequate pain relief (1). Pain causes anxiety, sleeplessness, and other pathophysiologic responses that can adversely affect the patient and extend hospitalization (2,3). Effective postsurgical pain management is essential to the recovery and rehabilitation process. Oxymorphone hydrochloride has a long-standing and well established role in relieving postsurgical pain (4). A semisynthetic opioid agonist, oxymorphone has several times the analgesic potency of morphine (5). Oxymorphone binds to specific receptors in the supraspinal and spinal pain pathways and acts with significant specificity at the μ-receptor site (3,6,7).

In the United States, oxymorphone hydrochloride is currently formulated for parenteral or rectal administration. Oxymorphone immediate release (IR) is a new oral tablet formulation of oxymorphone hydrochloride. The pharmacokinetic characteristics of this formulation include a median time to maximum drug concentration (tmax) of 0.5 h, suggesting a rapid onset of analgesia (data on file, Endo Pharmaceuticals, Inc., 2003). However, this new formulation has yet to be tested for clinical efficacy in an acute pain setting.

The objectives of this study were to evaluate the analgesic efficacy and dose response of three doses of oxymorphone IR compared with placebo and to assess the safety of oxymorphone IR compared with oxy-codone IR and placebo in acute moderate to severe postsurgical pain.


This multicenter, double-blind, randomized, placebo-controlled, parallel-group, dose-ranging study consisted of two segments: an 8-h single-dose phase and a multiple-dose phase that extended the study to 48 h. The protocol and informed consent form were reviewed and approved by each site’s IRB. The study was conducted in accordance with the provisions of the Declaration of Helsinki, the US Food and Drug Administration, principles of good clinical practice, and International Conference on Harmonization guidelines. All patients provided written informed consent.

Men and nonpregnant, nonlactating women (aged 18–75 yr) receiving primary total hip or knee replacement surgery (including an osteotomy) and scoring I to III on the ASA physical status classification system were enrolled. Patients must have developed moderate pain intensity (≥45 mm on a 100-mm visual analog scale [VAS]) within 6 h of discontinuing patient-controlled analgesia or within 9 h of the last postsurgical dose of IM opioid.

Patients were excluded if they had a history of allergy to opioids; had any physical, medical, or psychological condition contraindicated for opioid use, including drug or alcohol dependence; or were undergoing medical procedures or treatments that might adversely affect the study or interpretation of the results. Use of nonsteroidal antiinflammatory drugs other than celecoxib or rofecoxib was prohibited within 48 h of surgery. Use of corticosteroids was not permitted for 7 days before surgery unless the drugs were taken through topical, inhaled, or intraarticular routes.

For 4 wk before the study, anticonvulsants were prohibited. During the study, medications with analgesic effects, cough syrups with opioids or dextromethorphan hydrobromide, and other investigational drugs were prohibited. Monoamine oxidase inhibitors were not permitted beginning 2 wk before surgery. Other antidepressants were permitted if treatment had been stable for 4 wk before the study. After surgery, patients received only parenteral opioid medications, except for acetaminophen and aspirin, which were permitted for fever and prevention of thrombosis, respectively. Tranquilizers, muscle relaxants, antihistamines, and antiemetics were prohibited beginning 4 h before the parenteral opioid medication was stopped.

Icing of the surgical incision was prohibited between discontinuation of opioid pain medication and 2 h after the first dose of study medication. Each patient had a 15-min washout period from physical therapy before each evaluation.

During the 8-h single-dose phase, patients were administered a single dose of oxymorphone IR 10, 20, or 30 mg; oxycodone IR 10 mg; or placebo. Pain was assessed by determining intensity on the categorical scale (0 [none] to 3 [severe]) and VAS (0 [none] to 100 [worst pain imaginable]); by using a pain relief scale (categorical from 0 [none] to 4 [complete]); and by determining whether pain was half gone. Assessments were performed just before the first dose; at 15, 30, and 45 min; and at 1, 1.5, 2, 3, 4, 5, 6, 7, and 8 h after the first dose or until remedication was required. Onset and duration of analgesia were determined by measuring the time to first perceptible pain relief and remedication, respectively. The single-dose phase ended when additional medication was requested or the 8-h assessments were completed, and patients made a global evaluation of the study medication.

Patients who completed the single-dose phase entered the multiple-dose phase, and those who previously received placebo were rerandomized to active treatment. During the multiple-dose phase, all patients received study medication every 4 to 6 h for the remainder of the 48 h.

Efficacy end-points included total pain relief during the 8-h single-dose interval (TOTPAR0–8), defined as the area under the current pain relief scores. Additional efficacy variables for the single-dose phase included TOTPAR0–4 and TOTPAR0–6 and the sum of pain intensity differences from baseline to 4, 6, and 8 h (SPID0–4,0–6,0–8). These were assessed by using both categorical and VAS scales. In addition, the sum of combined pain relief and pain intensity difference during the intervals from 0 to 4 h, 0 to 6 h, and 0 to 8 h (SPRID0–4, 0–6, 0–8) was assessed categorically. The time to meaningful pain relief and remedication and the proportion of patients experiencing 50% pain relief were evaluated.

During the multiple-dose phase, efficacy variables included patients’ recall at bedtime and in the morning of their worst pain intensity during that day and the previous night, respectively. Global evaluations included patient and physician satisfaction with pain medication.

An analysis of covariance model was used to evaluate TOTPAR and SPID. All statistical tests were two sided, and significance was denoted by P ≤ 0.05. Patient and physician global evaluations of study medication were assessed with a ranked sum test stratified by center. The efficacy-evaluable population included patients randomized to treatment who received one dose of study medication and completed the first hour of primary efficacy assessments without remedication, vomiting, or protocol violations. The last-observation-carried-forward method was used for early withdrawals.

Safety was assessed by physical examinations, vital signs, and clinical laboratory data. Adverse events were coded by standard methods (8) and were recorded by severity, relationship to study drug, and outcome.


The safety analysis included 300 patients: 42 (14%) did not complete the first efficacy evaluation without remedication, vomiting, or protocol violation; 258 patients (86%) were evaluated for efficacy; and 165 patients (55%) completed the single-dose phase of the study. Withdrawals for lack of efficacy totaled 27% among patients receiving oxymorphone IR (value is the mean across all 3 doses, with a range of 15.3% to 33.8%), compared with 41.7% who received oxy- codone 10 mg and 47.4% who received placebo. Adverse experiences accounted for 3.4%, 8.5%, and 12.3% of withdrawals among patients treated with oxymorphone 10, 20, and 30 mg, respectively, compared with 0% of those who received oxycodone 10 mg and 3.5% of those who received placebo. Other reasons for withdrawal from the study included withdrawal of consent and protocol violation.

The multiple-dose phase included 164 patients; 118 (72%) completed this phase. The most common reasons for withdrawal were typical opioid-related adverse events (n = 19; 12% and 10% of patients treated with oxymorphone IR and oxycodone IR, respectively) and lack of efficacy (n = 17; 11.2% and 8% of patients treated with oxymorphone IR and oxycodone IR, respectively). In this study phase, 161 patients (98%) were evaluable for efficacy.

More patients were women (mean, 52%–66%) and white (mean, 84%–88%). The mean age ranged from 62 to 67 yr. Most patients were experiencing moderate pain at baseline (mean, 64%–82%) (Table 1).

Table 1
Table 1:
Demographics of Patients

Mean TOTPAR0–4, 0–6, and 0–8 scores for all doses of oxymorphone IR were statistically superior to placebo (Table 2). Oxymorphone showed a statistically significant dose-response relationship in a regression model (TOTPAR0–8) by using the arithmetic dose as the regressor (slope estimate, 0.184; P < 0.001; 95% confidence interval, 0.089–0.279) and reached an analgesic plateau at the 20-mg dose (Fig. 1).

Table 2
Table 2:
Summary of Total Pain Relief Scores for Efficacy-Evaluable Patients
Figure 1.
Figure 1.:
Dose response for TOTPAR0–8 (total pain relief from 0 to 8 h) for efficacy-evaluable patients. Slope estimate = 0.184; P < 0.001; 95% confidence interval (0.089, 0.279).

Oxymorphone IR at 10, 20, and 30 mg was statistically superior to placebo for SPID0–4, 0–6, and 0–8, as well as SPRID0–4, 0–6, and 0–8 (Table 2). All 3 doses of oxymorphone IR were also significantly better than placebo for the time-specific end points, pain intensity difference, and pain relief and pain intensity difference at most time points beginning at 45 min (Figs. 2 and 3). Although oxycodone IR was generally numerically superior to placebo, the differences were not significant for any efficacy measures (Table 2, Figs. 2 and 3).

Figure 2.
Figure 2.:
Summary of pain intensity differences (PID; categorical) for efficacy-evaluable patients. *P < 0.05 versus placebo. CAT = categorical; IR = immediate release.
Figure 3.
Figure 3.:
Summary of pain relief and pain intensity differences (PID; categorical) for efficacy-evaluable patients. *P < 0.05 versus placebo. CAT = categorical; IR = immediate release.

The median time to meaningful pain relief was statistically significantly shorter in all of the oxymorphone IR groups (1 h) than in the placebo group (1.5 h; P < 0.05). Fifty percent pain relief was achieved by 90.2% of patients in the oxymorphone IR 20 mg group (P < 0.001), 82.4% of patients in the oxymorphone IR 10 mg group (P = 0.022), 77.2% in the oxymorphone IR 30 mg group (not significant), and 69.1% in the oxycodone IR 10 mg group (not significant), compared with 59.1% in the placebo group. Approximately 40% to 50% of patients in the oxymorphone IR groups, 27% in the oxycodone IR group, and 21% in the placebo group rated their medication “very good” or “excellent.”

During the multiple dose phase, the worst pain recalled from the previous day or night ranged from 2.0 to 2.3 among the active treatment groups on Day 1, with improvements on Day 2 (range, 1.4–1.7) and Day 3 (range, 1.2–1.4) (Table 3).

Table 3
Table 3:
Worst Pain Score in the Multiple-Dose Phase

Both patient and physician global evaluations indicated that approximately 80% to 85% of patients and physicians ranked all doses of oxymorphone IR and oxycodone IR 10 mg as “excellent,” “very good,” or “good.”

When the Day 1 to Day 3 dose intervals were averaged for each patient, the longest median dose interval (9 h 39 min) was observed in the oxymorphone IR–30-mg group. Median dose intervals ranged from 7 h to 7 h 44 min in the three other active-treatment groups.

The most frequently occurring adverse events in the oxymorphone IR groups were mild-to-moderate opioid side effects (i.e., nausea, vomiting, somnolence, and pruritus). During the single-dose phase, the incidence of adverse events was more frequent among the oxymorphone IR groups than in the oxycodone IR 10 mg group (39%–50% versus 27%). In contrast, the incidence was somewhat more frequent in the oxy-codone IR 10 mg group (82%) during the multiple-dose phase compared with the oxymorphone IR groups (61%–71%).

Serious adverse events occurred in 2 patients during the single-dose phase and in 13 patients during the multiple-dose phase. Serious adverse events occurring in five of these patients were considered possibly related (postoperative ileus, hypotension, increased sweating, respiratory distress and related symptoms, depressed consciousness, and somnolence) or probably related (coma) to oxymorphone IR 20 or 30 mg. There were no clinically meaningful changes in any of the laboratory results, vital signs, or physical examinations.


This study establishes the analgesic efficacy and safety of single and multiple doses of oxymorphone IR. All three doses (10, 20, and 30 mg) were significantly superior to placebo as assessed by TOTPAR0–8 and most single-dose efficacy end-points. An analgesic dose-response relationship was observed for the three oxymorphone IR doses, with an analgesic plateau at the 20-mg dose, underscoring the assay sensitivity of the analgesic model. In some instances, the oxymorphone IR 20-mg dose appeared numerically superior to the 30-mg dose. However, the differences do not appear to be clinically significant; pairwise statistical comparisons failed to reveal significant differences between the two dose levels.

During the multiple-dose phase of the study, patient pain scores improved on Days 2 and 3, indicating that the significant pain relief obtained with oxymorphone IR during the single-dose phase was maintained with multiple doses over consecutive days.

The median time to meaningful pain relief for patients taking oxymorphone IR was one hour. The rapid and dose-proportional analgesia experienced by patients in this study parallels prior pharmacokinetic analyses showing that the mean plasma concentration of oxymorphone increases proportionally as the dose of oxymorphone IR is increased from 5 to 20 mg, with a tmax of 0.5 hours across all doses (Fig. 4) (data on file, Endo Pharmaceuticals, Inc., 2003). This value compares favorably with IR tablets of morphine and oxycodone, for which tmax has been reported to be 1.2 and 1.5 hours, respectively (10,11).

Figure 4.
Figure 4.:
Mean plasma concentrations of oxymorphone after single-dose administration.

As expected, opioid-related adverse events were similar for oxymorphone IR and oxycodone IR during the multiple-dose phase of the study. However, there were fewer adverse events with oxycodone IR 10 mg during the single-dose phase; this parallels the lack of single-dose efficacy of oxycodone IR 10 mg compared with placebo. Because the study was not designed to compare oxymorphone IR and oxycodone IR for efficacy, these results must be interpreted carefully. There may be significant differences in the analgesic potencies of oxymorphone IR and oxycodone IR, as observed in longer-acting formulations, in which oxycodone/oxymorphone equianalgesic dose ratios have been reported to be 2:1 (11). Therefore, it is possible that the more frequent adverse events seen with oxymorphone in the single-dose phase may be secondary to its greater relative potency versus the oxycodone dose used for safety comparisons in this study. Testing this hypothesis will require a study to directly assess the analgesic potency of oxymorphone IR relative to oxycodone IR, as well as to other IR oral opioids. In the case of morphine, a greater relative potency of oxymorphone IR would not be surprising because oxymorphone has greater lipid solubility (permitting greater penetration into the central nervous system), exhibits increased selectivity and affinity for μ-opioid receptors (3), and is several times more potent than morphine when administered parenterally (5). It must be emphasized, however, that appropriately designed studies are required to establish the relative potencies of the IR formulations.

In this study, medications that could confound analgesia were generally excluded, and aspirin and acetaminophen were permitted only for fever or thrombosis prevention. In the clinical setting, many patients may be receiving multiple analgesic and/or nonanalgesic medications. Under these conditions, a single analgesic may be preferable to combination drugs because of the decreased potential for metabolic drug interactions. In this respect, oxymorphone IR may offer some clinical advantages because oxymorphone is not significantly metabolized by the cytochrome P450 enzymes that are involved in many drug-drug interactions (data on file, Endo Pharmaceuticals, Inc., 2003).

The treatment of acute postsurgical pain is essential not only for the patient’s comfort but also for the contribution it makes to the patient’s recovery. Failure to manage pain can cause anxiety, loss of sleep (1), extended time in intensive care, or prolonged hospitalization, as well as a number of adverse outcomes, including thromboembolic and pulmonary complications (2). When improperly treated, acute pain can translate into chronic pain and can persist for years or decades (12–14). Use of effective treatment to reduce acute pain provides immediate relief and diminishes the likelihood of long-term chronic pain. This study demonstrates that oxymorphone IR 10, 20, or 30 mg provides effective, dose-related relief of moderate-to-severe acute pain that can be maintained over consecutive days with multiple dosing.


1. Sinatra RS, Torres J, Bustos AM. Pain management after major orthopaedic surgery: current strategies and new concepts. J Am Acad Orthop Surg 2002;10:117–29.
2. American Society of Anesthesiologists. A report by the American Society of Anesthesiologists Task Force on Pain Management, Acute Pain Section. Anesthesiology 1995;82:1071–81.
3. Sinatra RS, Hyde NH, Harrison DM. Oxymorphone revisited. Semin Anesth 1988;7:209–15.
4. Beaver WT, Feise GA. A comparison of the analgesic effect of oxymorphone by rectal suppository and intramuscular injection in patients with postoperative pain. J Clin Pharmacol 1977;17:276–91.
5. Eddy NB, Lee LEJ. The analgesic equivalence to morphine and relative side action liability of oxymorphone (4-hydroxydihydromorphinone). J Pharmacol Exp Ther 1959;125:116–21.
6. Metzger TG, Paterlini MG, Ferguson DM, Portoghese PS. Investigation of the selectivity of oxymorphone- and naltrexone-derived ligands via site-directed mutagenesis of opioid receptors: exploring the “address” recognition locus. J Med Chem 2001;44:857–62.
7. Mather LE. Pharmacokinetics and patient-controlled analgesia. Acta Anaesthesiol Belg 1992;43:5–20.
8. International Conference on Harmonization. Medical Dictionary for Regulatory Activities (MedDRA). Reston, VA: Northrup Grumman, 2003.
9. Thirlwell MP, Sloan PA, Maroun JA, et al. Pharmacokinetics and clinical efficacy of oral morphine solution and controlled-release morphine tablets in cancer patients. Cancer 1989;63:2275–83.
10. Leow KP, Smith MT, Williams B, Cramond T. Single-dose and steady-state pharmacokinetics and pharmacodynamics of oxycodone in patients with cancer. Clin Pharmacol Ther 1992;52:487–95.
11. Gabrail NY, Dvergsten C, Ahdieh H. Establishing the dosage equivalency of oxymorphone extended release and oxycodone controlled release in patients with cancer pain: a randomized controlled study. Curr Med Res Opin 2004;20:911–8.
12. Wilkins KL, McGrath PJ, Finley GA, Katz J. Phantom limb sensations and phantom limb pain in child and adolescent amputees. Pain 1998;78:7–12.
13. Arnstein PM. The neuroplastic phenomenon: a physiologic link between chronic pain and learning. J Neurosci Nurs 1997;29:179–86.
14. Coderre TJ, Katz J, Vaccarino AL, Melzack R. Contribution of central neuroplasticity to pathological pain: review of clinical and experimental evidence. Pain 1993;52:259–85.
© 2004 International Anesthesia Research Society