- © 2007 Canadian Medical Association or its licensors
One problem with research on opioids for noncancer pain, in addition to those outlined by Andrea Furlan and associates,1 is the cognitive effect of these drugs, which makes it difficult to compare them with either placebo or active placebo such as NSAIDs. Even a nominal dose of opioids will improve mood through a euphoric effect. Tolerance develops such that the patient experiences intermittent withdrawal symptoms and smaller euphoric effects with trough and peak drug levels.
Under such conditions, patients will generally endorse the benefit of opioids for chronic noncancer pain. Euphoria is associated with pain relief and a greater sense of well-being, and the dysphoria and pain of withdrawal are associated with worsening chronic noncancer pain.
Further, it is often difficult for patients to recall how bad their pain was before they started treatment with opioids, and people taking opioids for chronic noncancer pain are likely to report an improvement in their overall condition even if they are worse off with the peak and trough effect of the opioid regimen. Perhaps they think that going without opioids altogether would feel like a trough: a state of withdrawal.
How else could research be done in this area? Delivering opioids via pumps that provide continuous infusions or fentanyl patches might overcome some of these problems. The dose would have to be titrated gradually, and a suitable placebo would be necessary. In trials comparing opioids in pill form with active placebo, perhaps the active placebo should contain one ingredient that gives nociceptive relief (e.g., NSAIDs) and a second ingredient that produces a calming but euphoric effect (e.g., a benzodiazepine such as lorazepam).
In traditional trials comparing oral opioids with placebo or NSAIDs, objective and detailed assessments of level of functioning and overall quality of life are better primary outcomes than pain. Participants' functioning, quality of life and pain should also be measured at a single point in time rather than being compared with baseline values.
REFERENCE
- 1.↵