• Posted February 2, 2023

Doctors Often Prescribe Antidepressants for Pain, But Do They Really Work?

Antidepressants are often prescribed to people suffering from chronic pain, but a new evidence review argues that the science behind these prescriptions is shaky at best.

These drugs helped people in chronic pain in only a quarter of potential uses tested, and even then the effect ranged from low to moderate, according to a combined analysis of 26 prior reviews.

"We found that, for most pain conditions and types of antidepressants, the evidence of their effectiveness was either inconclusive or they were ineffective," said lead researcher Giovanni Ferreira, a research fellow at the University of Sydney Institute for Musculoskeletal Health in Australia.

In particular, the review found scant evidence supporting the use of tricyclic antidepressants, he said.

Previous studies have found that as many as 3 out of 4 antidepressant prescriptions for pain involved a tricyclic antidepressant, researchers said in background notes.

"Tricyclic antidepressants such as amitriptyline are by far the most commonly used antidepressant for the treatment of pain," Ferreira said. "But to our surprise most evidence for tricyclic antidepressants showed that the effectiveness of these antidepressants is inconclusive. We think this is a concerning finding."

The U.S. opioid epidemic has led doctors to look to non-opioid drugs as a means of pain relief.

For example, a 2021 guideline for chronic pain management from the U.K's National Institute for Health and Care Excellence (NICE) explicitly recommends against using pain medicines, but carves out an exception for antidepressants, researchers noted.

For this new report, Ferreira's team analyzed a series of reviews that distill evidence from 156 clinical trials involving more than 25,000 people.

The researchers used the combined evidence to judge the effectiveness of eight antidepressant classes on 22 separate pain conditions, totaling 42 distinct comparisons in all.

"This review, for the first time, brings together all the existing evidence about the effectiveness of antidepressants to treat chronic pain in one comprehensive document,"Ferreira said.

In all, 11 of the 42 comparisons (26%) yielded evidence supporting the effectiveness of an antidepressant to treat a type of chronic pain.

"For the other 31 [74%] comparisons, antidepressants were either inefficacious or evidence on their efficacy was inconclusive,"the researchers wrote.

They listed nine specific pain conditions for which one antidepressant or another appeared effective. These were back pain, postoperative pain, fibromyalgia, nerve pain, pain related to breast cancer treatment, pain exacerbated by depression, knee arthritis, irritable bowel syndrome, and headache caused by chronic tension.

"For most of these conditions, SNRI [serotonin-norepinephrine reuptake inhibitor] antidepressants were effective," Ferreira said.

Moderate certainty evidence suggested that SNRIs were particularly helpful in managing back pain, postoperative pain, fibromyalgia and neuropathic pain, results showed.

On the other hand, researchers were taken aback by the limited effectiveness of tricyclic antidepressants, despite their widespread use.

"There were only three conditions for which they were effective: neuropathic pain [nerve pain], irritable bowel syndrome and tension-type headache," Ferreira said. "However, for those three conditions the certainty of evidence was low."

People should carefully weigh with their doctor whether an antidepressant could help their pain, he concluded.

"We hope that the findings from this review will support both clinicians and patients to weigh up the benefits and harms of antidepressants for various pain conditions so that they can make informed decisions about whether and when to use them," Ferreira said.

There are many different means by which an antidepressant could work to help someone in chronic pain, said Dr. Glenn Treisman, a professor of psychiatry and behavioral sciences at Johns Hopkins University School of Medicine in Baltimore.

Some might directly affect the pathways that report pain to the brain, while others might reduce a person's perception of their pain or ease an emotional condition like anxiety or depression that makes someone more sensitive to pain, said Treisman, who wasn't part of the study.

Both Treisman and Ferreira doubt that a placebo effect could explain the results found in these clinical trials.

"It is not well understood why some trials show that certain types of antidepressants are effective for one condition but not for other, but this is unlikely to be due to placebo effects," Ferreira said. "And contrary to population opinions, placebo effects in pain research are not actually very powerful."

For his part, Treisman said it's difficult to draw broad conclusions from an analysis that combines data involving many different pain conditions and many different antidepressants.

"When you do one of these studies, what you want is a homogeneous group of patients who have the same thing. And that's very difficult to find in chronic pain," Treisman said.

Everybody's quite different, he noted.

"Their personalities are different. They have different pain threshold set points. They have different abilities for coping, different ways of distracting themselves from pain," Treisman said. "Then when you try to do the experiment, it's very difficult to get the experiment to show what you want it to show."

He noted, for example, that some people are what is known as symptom amplifiers -- that is, they experience symptoms more intensely. Others are symptom minimizers. "Even when they're having a fair amount of pain, they say 'I'm OK,'" Treisman said. "And when you go to do studies on people, if you don't match them, it's very hard to get the studies to come out right."

Studies that show no effect aren't necessarily ruling out the potential of antidepressants to help pain, he noted.

"The studies that are negative aren't telling you that nothing's happening," Treisman said. "What they're telling you is that they weren't able in that study to show that anything happened. And the reason they were unable to show it often is because those studies are really hard to do well. A failed experiment doesn't prove the negative."

The research review was published Feb. 1 in the BMJ.

More information

Harvard Medical School has more on antidepressants for pain relief.

SOURCES: Giovanni Ferreira, PhD, research fellow, Institute for Musculoskeletal Health, University of Sydney, Australia; Glenn Treisman, MD, PhD, professor, psychiatry and behavioral sciences, Johns Hopkins University School of Medicine, Baltimore; BMJ, Feb. 1, 2023