Barriers to the Use of Tricyclic Antidepressants in the Management of Pain

The evidence that tricyclic antidepressants (TCAs) are effective for the management of pain is incontrovertible. TCAs, and in particular amitriptyline, have more studies demonstrating effectiveness for a broader range of pain syndromes than any other treatment modality. The evidence is strong and consistent across different classes of pain. In a series of meta-analyses, we found that TCAs are effective in the treatment of back pain [1], idiopathic pain [2], fibromyalgia [3], functional gastrointestinal disorders [4], chronic fatigue [2], tinnitus [2], tension headaches [5] and, most recently, migraine headaches [6]. While TCAs have generally been reported to be effective in a broad array of pain syndromes, the evidence for effectiveness in neuropathic pain is particularly strong [7,8]. However, while effective, the use of TCAs for pain is not as common as it should be [9].

This is because there are a number of barriers to their use. TCAs have a number of side effects. In our recent analysis, we found that patients taking TCA, compared with placebo or selective serotonin reuptake inhibitors (SSRIs), were more likely to experience drowsiness and a dry mouth [6]. Other side effects commonly attributed to TCAs, such as dizziness, constipation and blurred vision, occurred no more commonly than either placebo or SSRIs. The side effects from TCAs tend to decrease over time and can be reduced if the medications are titrated. However, for many pain syndromes, alternative treatment modalities have a reputation for having fewer side effects. Consequently, many providers turn to other agents as their first-line therapy; for example, topiramate or neurotin for headaches. Interestingly, in our studies, we have consistently found no difference in the rate of patient withdrawal from treatment between TCAs and other modalities, including placebo or SSRIs, and some side effects attributed to TCAs were no more common than in placebo. It is likely that many of the reputed side effects from TCAs are overstated. Another barrier is that, while TCAs are effective, they are not a panacea. TCAs have consistently resulted in a twofold increase in the odds of experiencing at least 50% improvement. The amount of pain reduction has been demonstrated to be large (∼2 standard deviations). TCAs reduce pain, but generally do not provide complete pain relief. This is not satisfying to patients or providers who want a cure, not improvement. For TCAs to be effective, a reasonable dose over a long period of time is required. Patients often ‘fail‘ TCAs because of inadequate treatment dose and duration. In addition, the studies of pain relief with TCAs are relatively short in duration, usually only 4–6 weeks. This is unfortunate as most patients being considered for TCA treatment have chronic pain. It is unclear how well TCAs work over the long term because nearly all systematic reviews pooling data on TCA effectiveness have not examined the effects of TCA over time. Our review of TCA and migraine headaches found that longer duration of treatment (up to 26 weeks) was associated with greater improvement [6]. A further barrier is that there are very few head-to-head comparisons between TCA and other treatment modalities. For some pain syndromes, such as headache, it is clear that TCAs are more effective than SSRIs. For other pain syndromes, the relative efficacy of different options is unclear. In the absence of a direct comparison demonstrating the relative effectiveness of various treatment options for the various pain syndromes, clinicians struggle to know which drug is most likely to make their patients better. Furthermore, providers may misunderstand TCAs‘ mechanism of action, believing that the benefit from TCA may be its effects on depression rather than direct pain modulation. In the several reviews we have conducted, while TCA improved depressive symptoms, its effects on pain appear to be independent of its effect on depression. In addition, there are concerns about using TCA in patients with certain comorbidities, such as coronary artery disease, and a belief that drug–drug interaction with other medications is more significant with TCAs than other medication classes that might be used for pain syndromes. There is concern regarding the potential mortality associated with TCA overdose. As with all medical interventions, this highlights that appropriate patient selection is paramount. Finally, most TCAs are generic, and the emphasis in pharmaceutical advertising, either directly to patients or, in its various manifestations, to providers, is on profitable drugs.

The consequence is that TCA is probably underutilized for treatment of pain. This is problematic because clinicians tend to turn to treatment modalities with which they are familiar [6]. Since clinicians are increasingly using drugs other than TCAs, they have less experience with them, and unfamiliarity breeds distrust. One study reported that, while TCA use was low in primary care, it was more frequently used in specialty clinics focused on specific pain syndromes [9].

Therefore, when seeing a patient with a chronic, painful syndrome, what should clinicians do? There is no one-size-fits-all plan; pain management is individualized based on the specific pain syndrome and patient characteristics. For nearly any type of pain, clinicians have a variety of treatment options. If the decision is made to begin pharmacological treatment, an important first step is to assess what medications have been previously tried and if the previous trials were adequate in either duration or dose. It is important that TCAs remain one option in the armamentarium of potential treatments. For chronic, stable syndromes, an excellent option is an n-of-one trial [10]. In its simplest form, an n-of-one trial involves carefully measuring the pain syndrome at baseline then after a trial period of a medication. This measurement could be as simple as asking patients to rate their symptom on a seven-point scale. For example, if the symptom is muscle pain, one could ask: how much muscle pain have you experienced over the last 24 h (extreme pain, very painful, quite a bit of pain, moderate amount of pain, mild pain, a little bit of pain or no pain at all)? For most pain syndromes, there are standard, readily available instruments that could be used to assess pain. At its most rigorous, the provider and patient decides what treatment options they want to compare (TCA vs placebo, TCA vs other active treatment or TCA vs more than one treatment option). The medications can be crushed and put into matching capsules by nearly any pharmacy and patients can be randomized to treatment with both the provider and patient blinded. After a certain period of time, patients can then cross over to the other treatment (or, in the case of more than one treatment comparison, crossed over more than once). Baseline and follow-up measurements are collected on patient symptoms and the results are revealed when the study is complete and unblinding occurs. In one study of 57 n-of-one trials, 50 (88%) provided a definitive clinical answer [10]. While these trials sound difficult to accomplish, clinicians who have tried them have found them helpful and relatively easy to conduct. Patient satisfaction is also high with this approach; they feel intimately involved in trying to find the best treatment option for them. An additional reason to consider this approach is that studies have consistently shown a significant placebo effect in trials of treating pain syndromes [11,12].

Regardless of clinical or scientific reasoning, and despite real and perceived barriers, we owe it to our patients who are suffering from pain syndromes to consider TCAs among the list of potentially efficacious treatment modalities. The use of TCA should be considered for pain management, whether by an n-of-one trial or simply as a first-line alternative based on available evidence. The scientific evidence for the efficacy of TCAs in the control of pain is strong across different cultures and pain syndromes. Not including TCA as an option in our decision-making discussions with patients means that the choices we offer them to improve the quality of their lives is incomplete.

Disclosure

The opinions expressed in this editorial are those of the authors and should not be construed to reflect, in any way, those of the Zablocki VA Medical Center or the Department of Veteran Affairs.

Financial & competing interests disclosure

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.