Antidepressants for Fibromyalgia

Source: Arnold LM, et al. Psychosomatics 2000;41(2):104-113.

Although antidepressants have been used as treatment for fibromyalgia, there is no consensus in the literature as to the efficacy of various agents. The current report is a meta-analysis that examines available clinical trials of antidepressants and cyclobenzaprine for the treatment of fibromyalgia. The idea of antidepressants in the treatment of fibromyalgia was initially based, in part, on the finding of abnormal sleep architecture, specifically insertion of alpha waves into deep sleep, which may be due to a central serotonergic mechanism. Arnold and colleagues reviewed randomized, controlled trials of antidepressants for treatment of fibromyalgia.

The methodology, results, and potential predictors of response of suitable controlled trials were studied by meta-analysis. Of 21 controlled trials, 16 involving tricyclic agents were identified, and nine were suitable for meta-analysis. Suitability for inclusion in meta-analysis was based on the presence of sufficient statistical data for effect size computations (i.e., means and standard deviations for continuous outcomes, proportions for binary outcomes). Effect sizes were calculated for measurements of physician and patient overall assessment, pain, stiffness, tenderness, fatigue, and sleep quality.

The antidepressants that were included in the analysis were mostly tertiary amine tricyclics, including amitriptyline and dothiepin. Cyclobenzaprine (Flexeril) was also included in the analysis due to its similarity to the tricyclic antidepressants and its reported effects on serotonin and norepinephrine. Compared with placebo, tricyclic agents were associated with effect sizes that were substantially larger than zero for all measurements. The mean effect size for all studies was approximately normally distributed, with a mean treatment effect of 0.44 of a standard deviation. Based on universally accepted principles of inferential statistics, an effect of this magnitude would be classified as "medium" and would be recognized clinically.

Arnold et al reported that the largest improvements were associated with measures of sleep quality; the most modest improvement was found in measures of stiffness and tenderness. Amid these extremes, global measures, as well as measures of pain and fatigue, showed moderate improvement.

In the review of reports of antidepressant treatments that were not suitable for meta-analysis, Arnold et al found results that were fairly consistent with the findings in the meta-analysis. The selective serotonin reuptake inhibitors (SSRIs) citalopram and fluoxetine were studied. These were associated with high dropout rates and did not demonstrate efficacy. Additionally, past history of depressive episodes were found to be possible predictors of antidepressant treatment response.

Comment by Michael F. Barber, PharmD

Fibromyalgia is a debilitating disease with many symptoms that may either overlap or exhibit similarities with depressive symptoms. This meta-analysis is particularly useful because it helps sort out the actual effects of treating fibromyalgia with antidepressants. Clearly, if the symptoms of fibromyalgia approached complete resolution after 4-6 weeks of treatment, one might question whether the patient is actually being treated for depression. However, the results from the meta-analysis showed only medium treatment effects, most prominently on sleep disturbance. This suggests that there are target symptoms of fibromyalgia that may respond somewhat to antidepressant therapy, even in the absence of the commonly reported symptoms of depression (depressed mood, reduced energy, impaired concentration, helplessness, hopelessness, or suicidality).

Since the studies that showed the most improvement involved tertiary amine tricyclics (e.g., amitriptyline), combined with the lack of efficacy of SSRIs, amitriptyline should be considered as one of the preferred agents for this purpose. However, studies involving secondary amines such as desipramine or nortriptyline have not been conducted and cannot be ruled out in terms of efficacy. Since they are relatively better tolerated than the tertiary amines, clinicians may opt to use these agents as well. Of course, these agents should typically be avoided in patients with active suicidal thoughts. Further, a great deal of caution should be used in patients with a history of cardiovascular disease due to the propensity of tricyclics to induce arrhythmias. In such cases, it may be advisable to use cyclobenzaprine.

Dr. Barber is Assistant Professor of Clinical Sciences and Administration, University of Houston College of Pharmacy, Houston, Texas.