Antidepressants for Chronic Pain: Do They Work?

By Michael Rubin, MD

Professor of Clinical Neurology, Weill Cornell Medical College

SYNOPSIS: Antidepressant medications have been widely used for treating a variety of chronic pain disorders but strong evidence to support their efficacy is lacking. Some patients may respond, but available data do not help us to determine which agents may be helpful in a specific type of chronic pain condition.

SOURCE: Ferreira GE, Abdel-Shaheed C, Underwood M, et al. Efficacy, safety, and tolerability of antidepressants for pain in adults: Overview of systematic reviews. BMJ 2023;380:e072415.

Affecting 20% of the global population, chronic pain results from physiologic, psychologic, and social factors and often requires a multifactorial approach to management, with many patients requiring pharmacologic therapy. The optimal choice of medication for pain is determined by the type of chronic pain syndrome. For neuropathic pain, which affects 7% to 10% of the population, treatment is initiated with either antidepressants, including tricyclic antidepressants (TCA) or serotonin-norepinephrine reuptake inhibitors (SNRI), or antiepileptic medications. Are antidepressants efficacious and, if so, which ones, and for which disorders?

Following guidelines from the “preferred reporting items for overviews of reviews” (PRIOR) statement, PubMed, Embase, PsycINFO, and the Cochrane Central Register of Controlled Trials from inception to June 20, 2022, were searched for systematic reviews published in peer-reviewed journals that investigated the efficacy of any antidepressant drug, compared with placebo, used for any pain condition in adults. No restriction was placed on antidepressant class, dose, or regimen, nor on language of the review, but reviews that included children or adolescents or omitted pain or safety outcomes, were excluded.

Pain, as measured by any instrument and reported as between group differences, along with corresponding 95% confidence intervals, was the primary outcome measure. When multiple pain outcomes were available, the primary outcome selected by the review was chosen, and when multiple time points existed for the same pain outcome within a review, data were extracted from the time point closest to the end of treatment in > 50% of trials.

Safety outcomes, considered secondary outcomes, were defined as beneficial or harmful when risk ratio estimates included 0.75 or 1.25 values, respectively. Outcomes data, reported separately by pain condition and antidepressant class, was classified as efficacious when the difference between intervention and placebo groups was statistically significantly in favor of the antidepressant, not efficacious when the difference was not statistically significantly in favor of antidepressants, or inconclusive when the certainty of evidence was very low or the comparison had only one small trial (defined arbitrarily as having a sample size less than 100 per arm), or both.

Among 1,732 reviews identified, 948 were excluded as duplicates, leaving 784. Of those 784, 26 satisfied inclusion criteria, encompassing 156 unique trials with more than 25,000 patients covering 22 pain conditions representing 42 distinct antidepressant vs. placebo comparisons. Although none provided high-certainty evidence for any pain condition for any antidepressant, nine reviews provided evidence that some antidepressants were efficacious, compared with placebo for nine conditions in 11 distinct comparisons.

Moderate-certainty evidence suggested that SNRIs, mostly duloxetine at 60 mg to 120 mg, were efficacious for chronic back pain, postoperative pain (most trials were in orthopedic surgery), fibromyalgia, and neuropathic pain, but only low-certainty evidence supported their efficacy for aromatase inhibitor-induced pain in breast cancer, depression and comorbid chronic pain, and knee osteoarthritis. Only low-certainty evidence similarly supported selective serotonin reuptake inhibitors (SSRIs), mostly paroxetine at a median dose of 20 mg, for depression and comorbid chronic pain, and TCAs for irritable bowel syndrome, neuropathic pain, and chronic tension-type headache. When prescribing antidepressants for pain, a more nuanced approach seems warranted.

COMMENTARY

Less than 50% of patients with neuropathic pain respond to current therapeutic regimens, with multiple factors likely contributing, including the lack of specificity of current treatment, a lack of adequate experimental models, and insufficient consideration of the heterogeneity of neuropathic pain, with different symptoms possibly reflecting diverse pathophysiologic mechanisms. Personalization of neuropathic pain management would be ideal, with objective, measurable biomarkers, able to predict treatment response, offering the ultimate mode to match the patient with the medication.

Sadly, this quest has remained fruitless, despite efforts aimed at identifying genetic, genomic, functional (both electrophysiologic and neuroimaging), structural, or clinical (sensory phenotyping or patient-reported) markers. Powerful new techniques that map the brain “connectome” in patients with chronic pain may yet yield the answer.¹

REFERENCE

1. Bouhassira D, Attal N. Personalized treatment of neuropathic pain: Where are we now? Eur J Pain 2023; Apr 28. doi: 10.1002/ejp.2120. [Online ahead of print].