Am Fam Physician. 2005;72(10):1964-1976
Tramadol for Acute Pain: A Review of the Evidence
TO THE EDITOR: Kudos to Dr. Sachs for her attempt to develop an evidence-based approach to the use of oral analgesics for acute pain.1 However, her recommendation against the use of tramadol (Ultram) seems to be the result of selective reporting of the research literature, based on a randomized trial2 com-paring tramadol with hydrocodone (Vicodin) and a randomized trial3 comparing tramadol with placebo or acetaminophen and codeine. Dr. Sachs does not cite a double-blind ran-domized trial4 of 200 adults that showed that tramadol (75 mg) plus acetaminophen was as effective as hydrocodone (10 mg) plus acetaminophen in relieving pain caused by extraction of impacted molars, and was less likely to cause adverse drug events. She also did not cite a meta-analysis5 of individual patient data that showed that, compared with placebo, tramadol (75 mg) plus acetamino-phen had a number needed to treat (NNT) of 2.6 to reduce pain by 50 percent, which was comparable to the NNT of 2.4 for 5 mg of oxycodone (Oxycontin) plus acetaminophen in a separate meta-analysis.6 However, the meta-analysis5 of tramadol should be viewed cautiously because it is based on unpublished pharmaceutical company data.
Using a MEDLINE search strategy (tramadol and postoperative pain/drug therapy) limited to randomized controlled trials, I identified 104 separate trials comparing tramadol with a variety of other medications for the treatment or prevention of postoperative pain. The evidence from those trials is mixed, with some trials showing substantial benefit and others showing little benefit for tramadol.
I believe the evidence on tramadol for acute pain is mixed. It would be useful to have a trial comparing multiple doses of tramadol plus acetaminophen with multiple doses of hydrocodone plus acetaminophen. Until that time, it is prudent to keep tramadol plus acetaminophen as a therapeutic option for treating patients with acute pain.
It appears to have at least moderate painrelieving effects and may have fewer side effects than narcotic analgesics.
IN REPLY: I thank Dr. Sonis for the opportunity to further discuss important caveats that govern the interpretation of studies about analgesic efficacy. As Dr. Sonis states, our review1 cited only a sample of the literature pertaining to tramadol (Ultram). This series in American Family Physician aims to provide “succinct, evidence-based, authoritative clinical reviews”2 rather than an exhaustive review of every related paper that is published. In addition to analgesic efficacy, our literature selection pays careful attention to appropriateness of comparison medication, adverse effects, and research sponsor motivation. Based on these guidelines, I believe that quality studies with appropriate comparison groups do not support the use of tramadol in the initial treatment of acute pain.
Dr. Sonis cites two specific studies3,4 supporting the efficacy of a combination of tramadol and acetaminophen (Ultracet). Rather than equality, the first study3 actually showed greater efficacy in time to pain relief and time to remedication for a combination of hydrocodone and acetaminophen when compared with the combination of tramadol and acetaminophen. Additionally, two of the four study authors3 were employed by the makers of Ultracet.
The second cited article4 is a Cochrane review of the combination of tramadol and acetaminophen versus acetaminophen alone or ibuprofen.4 This review reports efficacy of the combination in some studies but not in others. With the addition of acetaminophen, tramadol demonstrated superior analgesia to acetaminophen alone, but in no study was it superior to 400 mg of ibuprofen, and in several studies it was inferior. Furthermore, in the five studies that used a dose of 975 mg of acetaminophen as a comparison, the investigators failed to demonstrate superiority of the tramadol combination over acetaminophen alone.
The Cochrane review also reported significantly more side effects with tramadol than with the recommended first-line analgesics. For single-dose oral tramadol (75 mg) plus acetaminophen (650 mg), the number needed to treat to harm for a patient to report any adverse effect was 5.4 (4.0 to 8.2). Neither acetaminophen (650 mg) nor ibuprofen (400 mg) demonstrated any increase in adverse effects compared with placebo (relative risk: 0.9 [0.7 to 1.3] and 0.7 [0.5 to 1.01], respectively). For dental pain patients, this review4 reported vomiting in one out of every six patients receiving tramadol.
In summary, the literature fails to demonstrate increased efficacy of the combination of tramadol and acetaminophen over other recommended first-line analgesics (i.e., ibuprofen [400 mg] or acetaminophen [1,000 mg]) but does demonstrate a striking increase in adverse events.