RESPONSE LETTER TO NACHAMIE ET AL.

To the Editor: We are grateful to Drs. Nachamie, McNicoll, and Dosa for their comments about our subset analysis of elderly patients with osteoarthritis flare pain who participated in a placebo-controlled study of tramadol/acetaminophen. As they noted, clinical treatment guidelines recommend acetaminophen as first-line treatment for osteoarthritis,1,2 but these guidelines go on to recommend nonsteroidal antiinflammatory drugs (NSAIDs) and then tramadol alone or tramadol/acetaminophen when patients experience inadequate pain relief. Thus, we disagree that enrolling patients who had inadequate pain relief with a nonspecific or cyclooxygenase-2–specific NSAID was a shortcoming of our study. It is possible that a multiarm study comparing tramadol/acetaminophen with tramadol or acetaminophen alone would have provided additional information as they suggested, but several placebo-controlled trials established the superior efficacy of tramadol/acetaminophen to either drug alone in the management of acute pain,3–5 and tramadol/acetaminophen was tolerated better than tramadol alone in clinically recommended doses.5 We agree with their statement that the study cannot be generalized to cognitively impaired, frail, and medically complex patients. Of the 326 patients who were screened, only 18 failed the inclusion/exclusion criteria. Although the reasons for screen failure were captured, the ages of the patients were not, so we cannot determine how many of these patients were elderly. In addition, investigators did not report how many elderly patients were not screened because they were clearly too frail to participate. Nonetheless, we do not believe that our study design was inconsistent with other placebo-controlled trials, as suggested by Nachamie et al.'s final statement, because medically complex patients are usually considered inappropriate for initial studies of new treatments. As we noted in the original paper, the low incidences of constipation and somnolence and the lack of gastrointestinal toxicity in the study, combined with the low potential for dependence or abuse with tramadol,6 make tramadol/acetaminophen an attractive alternative to traditional opioid analgesics for elderly patients. Tramadol/acetaminophen might be particularly attractive for frail and medically complex patients who are highly susceptible to adverse events, but additional studies are required to confirm this. Despite the concerns raised by Nachamie et al., we feel that our study answered the question of whether tramadol/acetaminophen is an appropriate treatment option in elderly patients with osteoarthritis flare-up pain. We join them in urging researchers to continue evaluating tramadol/acetaminophen and other effective treatments for the wide variety of elderly patients who deserve improved control of osteoarthritis pain.