© 2024 MJH Life Sciences™ and Patient Care Online. All rights reserved.
Antidepressants, although guideline-recommended and widely prescribed for common pain conditions, may not be as effective as perceived. Author of a new meta-analysis explains.
Antidepressant medications as treatment options to help manage chronic musculoskeletal pain and neuropathic conditions such as sciatica may need considerable re-evaluation, according to research published this month in the British Medical Journal.
Recommended by international pain management guidelines and widely prescribed, agents from drug classes including tricyclic antidepressants and selective serotonin and norepinephrine reuptake inhibitors are often chosen to treat longstanding pain from osteoarthritis and back injury after nonpharmacologic therapy and non-prescription and/or prescription anti-inflammatories have failed to help. These agents have become even more popular as clinicians continue to avoid prescribing opioid analgesics.
But results of the comprehensive meta-analysis just published conclude that not only is evidence for the efficacy of these agents of low to poor quality, the drugs may have worrisome side effects that should preclude their use.
Meta-analysis lead author Giovanni Ferreira, PhD, a postdoctoral research fellow at the University of Sydney School of Public Health, spoke with Patient Care Online in more detail about his team's findings.
The following transcript has been lightly edited for clarity.
Patient Care Online: I'm Grace Halsey, Senior Editor of Patient Care Online, and our clinical focus this month for primary care is on pain and pain management. My guest today is Dr Giovanni Ferreira. He's a member of the faculty of Medicine and Health and a research fellow at the University of Sydney School of Public Health. Dr. Ferreira is the author of a recent meta-analysis published in the British Medical Journal in January on the safety and efficacy of antidepressants used for the treatment of back pain and osteoarthritis. Welcome.
Giovanni Ferreira, PhD: Grace, thanks for having me.
PCO: So why don't you tell us a bit first about your background and your research focus and what led you and your colleagues to look more closely at the efficacy of these drugs for chronic pain?
Ferreira: Yeah, so I'm a physiotherapist by background and I joined the Institute of Musculoskeletal Health in 2017. Went there to take a PhD. Our group is a multidisciplinary collaboration that involves clinicians and researchers from lots of different backgrounds, including rheumatology, orthopedic surgery, pharmacy, and physiotherapy. We also have a strong tradition in testing/examining the efficacy of some pharmacological treatments for back pain. So for example, our groups published a definitive trial of paracetamol for acute back pain in 2014 in The Lancet, and most recently, the PRECISE trial, a trial for pregabalin for sciatica published in the New England Journal of Medicine. So we've got a strong tradition in, a strong background in the area of testing the efficacy of medicines. So, we decided to do this review, because we realize that many depressants are commonly used in clinical practice to reduce pain in people with back pain and osteoarthritis, and, also several guidelines recommend their use. So we wanted to understand what was the evidence out there for making those recommendations. And we found out that all the systematic reviews that were out there, were either outdated or had some methodological problems that didn't allow a comprehensive assessment of the efficacy of those medicines.
PCO: Hmm. I found 2 sets of facts interesting in your introduction. One is that in the US, antidepressants are the fourth most prescribed medication for low back pain, with more than a quarter of Americans prescribed an antidepressant within 3 months of a first diagnosis of back pain. And also that they are now endorsed, by about, I think you said three-quarters/75% of clinical practice guidelines for low back pain and also by the American College of Physicians for osteoarthritis, and UK National Institute for Health and Care Excellence (NICE). So that is pretty widespread use and recommendation for agents that don't particularly appear to perform as well as they are purported to. So why don't you tell us a little bit specifically about the results in back pain and osteoarthritis and what you found?
Ferreira: So we found so we've grouped our findings into different classes of antidepressants. So when I say the efficacy of antidepressants, I'm talking about different classes of medications, right. So we found that SNRI antidepressants had a small effect, there was statistically significant, but probably not clinically important for people with back pain and osteoarthritis. There's a little caveat here. So we found that for back pain, the benefits are very unlikely to be clinically important. But we did find that we can't say the same for osteoarthritis. So a clinically important benefit cannot be excluded. And that's because the effects for osteoarthritis seem to be a little bit more important or bigger than for people with back pain. We've also studied the effects the efficacy of these medicines for people with sciatica, pain going down the leg, back pain going down the leg. And we found that for some of the comparisons, SNRI antidepressants and tricyclics they could provide they had a significant effect in relationship with placebo, but they also provided effects that were considered to be clinically important, but they wouldn’t be considered to be clinically important by most patients and that I think was was an interesting finding from our review. We've also tested we've also summarized the efficacy of different classes of antidepressants, and that includes SSRIs and other medicines and those particular classes we didn't find any effects for back pain, and they weren't tested for osteoarthritis and sciatica.
PCO: Is it your finding that the guideline recommendations for their use are based on a similar collection of the studies that you looked at as well?
Ferreira: I think it depends on the guideline, there's some guidelines, international guidelines, there are a little bit outdated. Even the NICE guidelines, the latest version of the NICE guidelines for back pain was published in 2016. But since then, we've seen a good amount of trials being published in in back pain and in people with sciatica as well. So I think our review will provide some update for guidelines to reassess the recommendations later on and consider whether or not to recommend these medicines for people with chronic back pain and osteoarthritis.
PCO: What were the primary shortcomings of the research studies?
Ferreira: Yeah, so there were quite a few limitations in the in, in our review that word. Because of the included studies, the majority of studies were at high risk of bias for several reasons. One of them is industry funding. So lots, lots of trials, a good proportion of them, I can't remember off the top of my head add on industry funding. And there's plenty of research out there showing that whenever pharmaceutical companies sponsoring conduct trials, there is a tendency for these trials to show effects that are a little bit overestimated in relation to trials conducted by independent researchers. And this was a particular problem in osteoarthritis as well. So for osteoarthritis, we've included eight trials, and six of those trials were industry sponsored, they were also the biggest trials with the most patients enrolled. So and it's no surprise that I think that we found an effects that sick that is statistically significant, but also borderline clinically important for people. So I guess, one of the biggest recommendations that we made in this review is that we need trials, particularly in osteoarthritis that are conducted independent from pharmaceutical companies to actually ascertain whether the effect that we saw in our review is, will be reproducible and independently conducted trials.
PCO: There was an editorial along with your study by Dr. Martin Underwood. And he said that you and your colleagues used the standard method for threshold for the smallest worthwhile difference between groups to evaluate efficacy, so a 10 point difference on a scale of zero to 100. And he says that it's important because “…modest overall benefit at a group level could still mean that some treated individuals gain a worthwhile benefit.” So essentially, I'm thinking don't, you know, discard the drug class because the big picture says that it's not highly effective; it probably still work will work in some of your patients.
Ferreira: Yeah, I think I think overall professors, Professor Underwood's interpretation of our findings is correct. Like, like you mentioned, and like you mentioned our results, they represent the average effect across the population. So our, but I think we need to be clear about what classes of antidepressants we're talking about, for what conditions. So for example, like I've mentioned before, we showed that SNRI antidepressants, they were the only ones that provided a statistically significant benefit for back pain. But that benefit was, like I said, it was small and below what we defined as a clinically important threshold for most patients. For this particular antidepressant class, I think Professor Underwood's quote is indeed correct. But I would like to point out that there were other classes of antidepressants that did not show any benefit in relation to placebo, particularly for back pain, and that includes SSRIs for example, fluoxetine or paroxetine. And I think in this situation, the correct interpretation would be a more cautious one that we do not have the evidence to recommend they're used for people with back pain at this point time. So we need to be clear think about which antidepressants we were talking about.
PCO: And you did look at the adverse event profiles of most of them as well. And what are some of the more dangerous downsides of prescribing these when there's little evidence for their efficacy?
Ferreira: Yeah, so this is this is a very good question. We had evidence that we classified as being of low to very low quality for safety. And that is because the majority of trials they weren't big enough to detect to assess safety in the correct way. So like we're seeing now with the COVID trials that are enrolling dozens of 1000s of people and to detect only a few side effects. So you need big trials to be able to, to say, with more certainty whether the drug is, is safe, it causes adverse events or serious adverse events. So, that point that being made, the majority of trials didn't have enough power, didn't have enough people to detect, for example, serious adverse events that have been associated with the use of some antidepressants for other conditions. We did find, we also found that the reporting of these adverse events was rather poor across the studies. But some that I can mention here was, I think nausea was the most mentioned adverse event related to the use of SNRIs, for example, But we don't have, I think we have to rely on the evidence that we have already for the use of this medicines for other conditions like depression to kind of be aware of what the patient could expect. But there's also an issue with dosing and dosage; these people are usually for pain given antidepressants at a lower dose. So we might not expect as many side effects or side effects that would be that important in this population, given the small the smaller dose that they're receiving. So there was a, there was a tricky thing for us to summarize, mostly because the studies had some problems with how they reported their safety outcomes.
PCO: Where does this take us in terms of people with really difficult, chronic conditions, who have probably tried everything in the book, including non-pharmacologic interventions? Where are we in the in the research?
Ferreira: I think this is, as we discussed above, some antidepressants, they do provide small benefits to people with chronic pain. So, some patients, I think it's perfectly fine to have a discussion with your doctor, because some people will be willing to accept taking these drugs that have a range of adverse events to obtain some pain relief. And I think it's our data eventually, which showed that some people might experience some benefit with these medications, like Professor Underwood has pointed out in his editorial.
But I guess there's also a different side to this story, which is, we need to understand, we need to recognize that pain management that deals with chronic, severe, intractable pain hasn’t…is very limited, the kind of knowledge that we have so far. And we think we need to recognize that. And I think that will, that that's the sort of mindset that will move us forward into thinking about, you know, developing and discovering new new treatments for people who suffer with these conditions down the track.
PCO: So would your guidance to a primary care physician, who in many cases is the first line treating these patients to, you know, do your homework, do some research, find out what you're comfortable with, and then have that conversation with your patient?
Ferreira: I think this is a very important part of, of back pain and osteoarthritis care—of presenting patients with accurate information about the sorts of benefits that that people can expect from these medications, but also highlighting what the possible adverse events are these drugs in particular. And, we've got some evidence that doctors and patients tend to overestimate the effects of interventions, and underestimate the role of cons, of the adverse events. So I think having a balanced discussion about the benefits and the harms of treatments in general, including antidepressants is a very good starting point for people to be able to make an informed decision about their own health.
PCO: Amen. And what's next for you? What's next on the research front?
Ferreira: Yeah, so I'm still interested in exploring the role of the president a little bit more for people with with back pain in particular, which is my main area of research. I think, for me, a very interesting finding from this review was that we did find some clinically important benefits for people with sciatica with some antidepressant classes—SNRIs and classic tricyclics in particular, but because the trials were very small, and there was lots of uncertainty in the effects in the estimates of effect, we could not make a strong recommendation about their use in clinical practice. So I think there's definitely room for a well designed trial that's independent from pharmaceutical companies to actually test whether those drugs can be efficacious effective in in people with sciatica. And I think this would be an important contribution to research because sciatica is a particularly difficult condition to manage in primary care. People don't respond well to lots of medications and nonpharmacological treatments are also quite limited. So I think people with sciatica need to we need to find different solutions and effective solutions for this t this population.
PCO: Especially since, I'd say, you know, 475% of us are sitting a lot more than we ever used to. That's a very timely topic of research.
Related Content: