Antidepressants are of little help and may possibly be harmful in the treatment of musculoskeletal pain, new research suggests.
Results of a systematic review and meta-analysis show that serotonin and norepinephrine reuptake inhibitors (SNRIs) or tricyclic antidepressants (TCAs) had little or no effect on back pain, sciatica, or osteoarthritis (OA). In addition, investigators found that SNRIs were associated with a significantly increased risk for adverse events (AEs).
“We have shown that the benefit of antidepressants for pain appears to be very limited, and even for those who have not responded to other analgesic medicines, antidepressants may not be the answer,” study investigator Giovanni E. Ferreira, PhD, postdoctoral research fellow, Institute for Musculoskeletal Health, University of Sydney, Sydney, Australia, told Medscape Medical News.
“There has to be a clear understanding about the limited role of medicines for treating pain in general, and perhaps we should start looking more into nonpharmacological options that have fewer side effects and may have good outcomes,” said Ferreira.
The study was published online January 20 in The BMJ.
Back pain and OA are leading causes of disability. It has been estimated that back pain affects about 7.3% of the global population and that neck pain affects about 5.0%. OA-related hip and knee symptoms affect 12% of the population.
Antidepressants are frequently prescribed for pain, particularly chronic pain. The authors note that in the United States, these drugs are the fourth most prescribed medications for low back pain.
About 75% of clinical practice guidelines endorse antidepressants for low back pain, as do two recently published OA guidelines, the authors note.
The American College of Physicians recommends the SNRI duloxetine for low back pain, and the Osteoarthritis Research Society International and the American College of Rheumatology recommend duloxetine for pain management.
To determine the efficacy of these agents for pain, the investigators conducted a comprehensive literature search for randomized controlled trials that compared any antidepressant with placebo for patients with neck or low back pain with or without radicular symptoms, as well as for patients with hip or knee OA, or both.
The study’s primary outcomes were pain intensity and disability. The secondary outcome was adverse events.
The analysis included 33 trials with 5318 participants. Most trials (84.9%) had a parallel-group design. The others had a crossover design with washout periods ranging from 1 to 2 weeks. All but one trial reported data from participants with chronic pain.
Fourteen trials were sponsored by industry; the sponsorship source was unclear in five trials. The median duration of the drug regimen was 8 weeks.
Not Clinically Relevant
The researchers evaluated six antidepressant drug classes: SNRIs, TCAs, selective serotonin reuptake inhibitors (SNRIs), norephinephrine dopamine reuptake inhibitors, and serotonin antagonist reuptake inhibitors.
Researchers classified the certainty of evidence as high, medium, low, or very low. Classifications were based on the degree of confidence that the true effect was close to that of the estimated effect.
The threshold for clinical efficacy was a difference between groups of at least 10 points on a 0–100 scale for pain or disability, a threshold that is commonly used in studies of chronic pain.
Nineteen trials examined the efficacy of antidepressants for back pain. All studies that examined the effects of SNRIs were sponsored by industry and involved duloxetine.
Evidence of moderate certainty showed that SNRIs reduced pain at 2 weeks or less (mean difference, -3.67; 95% CI, -5.91 to -1.42; three trials with 1068 participants) and 3–13 weeks (mean difference, -5.30; 95% CI, -7.31 to -3.30; four trials with 1415 participants). SNRIs also reduced disability at 3–13 weeks (mean difference, -3.55; 95% CI, -5.22 to -1.88; four trials with 1423 participants).
However, the effect was below the review’s predetermined threshold of clinical importance.
“In terms of back pain, we found moderate-quality evidence that SNRIs work just a little, but that effect is very small, too small to be considered clinically important,” said Ferreira.
The researchers found that the other classes of antidepressants had no effect, although Ferreira noted that only a few studies examined the efficacy of these agents for back pain.
“High Risk of Bias”
Six trials reported data for sciatica. All of the studies were “very small and had very important methodological limitations,” including “high risk of bias,” said Ferreira. For these reasons, the researchers determined that the level of certainty of the evidence for the sciatica comparisons was low to very low.
This evidence showed that SNRIs reduced pain at 2 weeks or less (mean difference, -18.60; 95% CI, -31.87 to -5.33) but not at 3–13 weeks (mean difference, -17.50; 95% CI, -42.90 to 7.89).
Evidence with low to very low certainty showed that TCAs did not reduce pain at 2 weeks or less but did at 3–13 weeks and 3-12 months. TCAs did not reduce disability at 2 weeks or less and at 3–13 weeks but did at 3–12 months.
“There might be a clinically important effect for some antidepressants in sciatica, particularly the SNRIs and tricyclic antidepressants, but we can’t make a strong recommendation right now because of the uncertainty of the evidence,” said Ferreira. He noted that this could change with well-conducted trials.
As for OA, eight trials evaluated the efficacy of antidepressants, all in participants with knee OA. Every trial examined SNRIs. Of the eight studies, six were sponsored by industry.
Evidence of moderate certainty showed that SNRIs reduced pain at 2 weeks or less (mean difference, -4.66; 95% CI, -6.28 to -3.04; four trials with 1328 participants). Evidence of low certainty showed that SNRIs reduce pain at 3–13 weeks (mean difference, -9.72; 95% CI, -12.75 to -6.69; eight trials with 1941 participants).
Low-certainty evidence showed that SNRIs reduce OA disability at 2 weeks or less and at 3–13 weeks of follow-up.
Here again, the effect of SNRIs was small and was below the review’s predetermined threshold of clinical importance. The lower limit of the confidence interval did contain clinically important effects for pain but not for disability.
This means that a clinically important benefit of an SNRI for patients with knee OA can’t be excluded. “The effects for osteoarthritis were a little bit better than we found for back pain,” said Ferreira.
Overestimation of Benefit?
Ferreira emphasized that most of the trials were sponsored by industry. “We have plenty of evidence showing that industry-sponsored trials tend to overestimate effects of treatment,” he said.
Low-certainty evidence showed that SNRIs significantly increased the risk for any AE, but not for serious ones. In many trials, the most common AE was nausea.
Other antidepressants did not seem to increase the risk for AEs. That could be because there were too few trials and because the ones that were included “were not powered enough to detect any potential harms with those medicines,” said Ferreira.
Participants who had been diagnosed with depression did not benefit more from antidepressants for pain than did those without depression. Ferreira noted that most study participants did not have clinical depression.
“Only a handful of studies in our review actually included patients with depression, so we don’t have enough evidence to make a firm recommendation that if you have both depression and pain, taking an antidepressant will reduce the pain,” he said.
Ferreira noted that none of the studies in the review evaluated outcomes beyond 12 months and that most trials only assessed participants for up to 3 months.
“That’s a big problem, because antidepressants are commonly used as a long-term medication,” he said. “We don’t know what happens to these people in the long run in terms of clinical effects but also in terms of safety,” he said.
“Our findings show there’s little justification for recommending antidepressants for people with back pain at the moment, and for sciatica and osteoarthritis, while these medications may be effective, guidelines should emphasize that the current evidence is uncertain,” Ferreira added.
For patients who have pain, clinicians should provide “both sides of the story” when it comes to antidepressants to enable decisions as to whether these medications “might work” for them, said Ferreira.
“This includes the likely small effects for some classes of antidepressants, the absence of effects for some other classes, plus the possibility patients might experience some adverse events,” he said.
Clinicians should also inform patients about effective nonpharmacologic approaches for treating pain, including weight control, avoiding smoking, and regular physical activity, said Ferreira.
“There’s quite a lot of evidence showing the benefits of exercise, particularly for chronic back pain and chronic osteoarthritis, and for sciatica, but to lesser degree.”
In an accompanying editorial, Prof Martin Underwood and a colleague at the Warwick Clinical Trials Unit, University of Warwick, Coventry, United Kingdom, describe the review as “well conducted” and “timely.”
Making sense of “inconsistent” guidance on the use of antidepressants for pain as well as the “various sources of evidence” is “challenging,” they write.
“For example, draft NICE [National Institute for Health and Care Excellence] guidance is to consider antidepressants for chronic pain but not for chronic sciatica,” they add.
The editorialists note that a “robust overview” is needed “to clarify guidance and to inform a consistent approach to use of antidepressants for people with painful disorders.”
Because many people with chronic pain also have symptoms of depression, such an overview “should consider the potential for reducing depressive symptoms,” they write.
They also point out modest overall benefit at group level could still mean that some individuals derive benefit.
“Despite the reported small effects at group level, some individuals with back pain or osteoarthritis may gain a personal benefit from serotonin-noradrenaline reuptake inhibitors,” they add.
Overall, though, the editorialists note that drug treatments are largely ineffective for back pain and OA and can cause serious harm.
“We need to work harder to help people with these disorders to live better with their pain without recourse to the prescription pad,” they write.
The study did not receive funding from the public, commercial, or not-for-profit sectors. Ferreira has disclosed no relevant financial relationships. Underwood is a director and shareholder of Clinvivo, which provides electronic data collection for health services research. He is part of an academic partnership with Serco, funded by the European Social Fund, related to return-to-work initiatives. He is a coinvestigator on three studies funded by the National Institute for Health Research and that were additionally supported from Stryker outside the scope of the editorial.