Tramadol allows reduction of naproxen dose among patients with naproxen-responsive osteoarthritis pain: a randomized, double-blind, placebo-controlled study

Arthritis Rheum. 1999 Jul;42(7):1370-7. doi: 10.1002/1529-0131(199907)42:7<1370::AID-ANR10>3.0.CO;2-T.


Objective: To demonstrate that in patients receiving naproxen for the pain of osteoarthritis (OA), the addition of tramadol will allow a reduction in the naproxen dosage without compromising pain relief.

Methods: This trial consisted of a 5-week open-label run-in and an 8-week double-blind phase. Patients with at least moderate pain (> or =40 mm on a 100-mm visual analog scale) of OA of the knee after a 1-week medication washout were treated with naproxen 500 mg/day for 1 week. Patients whose pain scores were reduced to <20 mm were discontinued. The remaining patients received naproxen 1,000 mg/day for 3 weeks. Tramadol 200 mg/day was added during the third week. Patients were then randomized in a double-blind manner to continue tramadol 200 mg/day or to begin placebo in addition to naproxen. Randomization was stratified based on response to naproxen 1,000 mg/day. During the double-blind phase, the naproxen dose was reduced by 250 mg every 2 weeks. The primary efficacy end point was the minimum effective naproxen dose (MEND). The MEND was defined as 250 mg above the naproxen daily dosage at which pain relief was no longer adequate. Patients discontinuing the double-blind phase of the study for reasons other than lack of efficacy were assigned a MEND equal to the last naproxen dose received. If the effect of treatment between the responder and nonresponder groups was statistically different, the difference in the MEND was assessed separately within the groups.

Results: Of 236 patients randomized (mean age 61 years; 147 females), 90 were stratified as naproxen responders and 146 as naproxen nonresponders. There was a significant difference (P = 0.040) in the treatment effect between the naproxen responders and nonresponders, thus demonstrating a difference in the way responders and nonresponders react to a decrease in naproxen dosage after the addition of tramadol. Among naproxen responders, the MEND was significantly lower in patients receiving tramadol (n = 36) than in patients receiving placebo (n = 54), 221 mg versus 407 mg, respectively (P = 0.021). For the naproxen nonresponders, the mean MEND was 419 mg in the tramadol group and 396 mg in the placebo group (P = 0.706).

Conclusion: In patients with painful OA of the knee responding to naproxen 1,000 mg/day, the addition of tramadol 200 mg/day allows a significant reduction in the dosage of naproxen without compromising pain relief.

Publication types

  • Clinical Trial
  • Randomized Controlled Trial
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Aged
  • Analgesics, Opioid / pharmacokinetics
  • Analgesics, Opioid / pharmacology*
  • Anti-Inflammatory Agents, Non-Steroidal / administration & dosage*
  • Anti-Inflammatory Agents, Non-Steroidal / pharmacokinetics
  • Dose-Response Relationship, Drug
  • Double-Blind Method
  • Female
  • Humans
  • Male
  • Middle Aged
  • Naproxen / administration & dosage*
  • Naproxen / pharmacokinetics
  • Osteoarthritis / drug therapy*
  • Osteoarthritis / physiopathology
  • Pain / drug therapy*
  • Placebos
  • Single-Blind Method
  • Therapeutic Equivalency
  • Tramadol / pharmacokinetics
  • Tramadol / pharmacology*


  • Analgesics, Opioid
  • Anti-Inflammatory Agents, Non-Steroidal
  • Placebos
  • Tramadol
  • Naproxen