Katharine N. Gurba1 & Simon Haroutounian1 1 Washington University Pain Center and Department of Anesthesiology, Washington University School of Medicine, St. Louis, Missouri, USA Antidepressants are important analgesic adjuncts in the treatment of chronic non‐cancer pain. This chapter is based on systematic reviews of high‐quality randomized clinical trials of antidepressant agents in multiple chronic non‐cancer pain conditions [1–9]. Three classes of antidepressant agents have been widely tested for efficacy in chronic pain: tricyclic antidepressants (TCAs), serotonin‐norepinephrine reuptake inhibitors (SNRIs) and selective serotonin reuptake inhibitors (SSRIs). Currently, several TCAs and SNRIs are FDA‐approved for treatment of chronic pain. TCAs currently in common clinical use include amitriptyline, nortriptyline, imipramine and desipramine. Additionally, there are also closely‐related tetracyclic antidepressant medications, mirtazapine and maprotiline. The first TCA, imipramine, was approved for treatment of depression in 1959 [10]. Within two years, publications also reported efficacy in myalgia, facial pain and headache. TCAs are now considered first‐line treatment in several chronic pain conditions. TCAs act at multiple molecular targets; they inhibit presynaptic serotonin and norepinephrine reuptake transporters and block additional targets including voltage‐gated sodium channels [11], adrenergic α1 and α2 receptors, histamine H1 receptors and muscarinic receptors [10]. Each of these receptor targets has been associated with increased pain or neuronal hypersensitivity [12], and the non‐selective mechanisms of TCAs may contribute to analgesic efficacy in multiple conditions. However, due to this broad pharmacologic profile, TCAs are often associated with multiple side effects including xerostomia, urinary retention, orthostasis and sedation. Therefore, interest grew in developing antidepressants with fewer side effects. Emerging evidence on the monoamine hypothesis of depression in the 1960s suggested that low serotonin levels played a role in major depressive disorder [13,14]. Consequently, pharmaceutical companies focused efforts on developing inhibitors of the serotonin reuptake transporter, culminating in the approval of fluoxetine in 1987. Other SSRIs tested in chronic pain include sertraline, paroxetine, citalopram and escitalopram. Several small trials found that SSRIs were more effective than placebo in conditions including diabetic peripheral neuropathy, headache and somatoform disorder; however, a 2016 narrative review concluded that a meta‐analysis of SSRI chronic pain trials was impossible due to trial heterogeneity, quality and risk of bias [15]. Serotonin‐norepinephrine reuptake inhibitors, developed shortly after SSRIs, proved more efficacious in treatment of chronic pain. This pharmacologic class includes the medications duloxetine, venlafaxine, desvenlafaxine and milnacipran. Their most common side effects include nausea, constipation, fatigue and xerostomia. However, these effects often tend to abate with continuing therapy [16]. There is well‐established, complex interplay between pain and emotion. However, trials have clearly demonstrated that the analgesic effects of antidepressants are not solely due to their effects on mood [17]. Additionally, analgesic effects tend to occur more quickly and at lower doses than antidepressant effects [18]. One important analgesic mechanism seems to involve modulation of ascending pain signals by central monoaminergic pathways [12,19,20]. Descending noradrenergic pathways arise from the locus coeruleus and project to the spinal cord. Norepinephrine (NE) binds to multiple subtypes of adrenoreceptors, but inhibitory α2 receptors have been most commonly implicated for pain modulation. Of note, these receptors are also highly expressed in brain regions important for pain processing, including periaqueductal gray, rostral ventromedial medulla (RVM), thalamus, prefrontal cortex and amygdala. Central serotonergic pain modulation is rather complex. Descending projections from the RVM evoke spinal release of serotonin (5‐HT), which can inhibit or facilitate pain signaling depending on which 5‐HT receptor is activated. Preclinical research suggests that 5‐HT1 receptor subtypes, which are Gi/Go coupled G‐protein coupled receptors (GPCRs), may inhibit nociceptive pathways. Conversely, 5‐HT2 receptors (which are Gq‐coupled GPCRs) and 5‐HT3 receptors (which are cation channels) may facilitate nociceptive pathways. TCAs share a tricyclic chemical structure and can be divided to secondary amines and tertiary amines, depending the on the number of methyl (‐CH3) groups on the side chain nitrogen. Secondary amine TCAs (nortriptyline, desipramine) result from the demethylation of tertiary amines (amitriptyline and imipramine, respectively), and typically have higher affinity to norepinephrine transporter than to serotonin transporter. In tertiary amines, this ratio is reversed. In addition, secondary amines have lower affinity to muscarinic acetylcholine receptors and thus have a lower likelihood of anticholinergic side effects, especially at low doses. There are some differences among tertiary amine TCAs as well. The affinity of imipramine to block histaminergic H1 receptors, for example, is lower than of amitriptyline [21]; therefore, it might be less sedating. TCAs are metabolized by the hepatic cytochrome P450 (CYP450) system. Tertiary amine TCAs can be metabolized by multiple CYP450 systems (e.g., CYP1A2, CYP2C19 and CYP2D6). Desipramine and nortriptyline are metabolized almost exclusively by CYP2D6, and thus may be substantially affected by genetic polymorphisms of the CYP2D6 enzyme [21]. All TCAs have excellent oral bioavailability, and have long enough plasma elimination half‐lives to allow once‐daily administration. The analgesic mechanism of SNRIs is assumed to be similar to that of TCAs in terms of serotonin and norepinephrine reuptake inhibition, but both venlafaxine and duloxetine have very low affinity to cholinergic, adrenergic and histaminergic receptors. In addition, venlafaxine is more selective to the serotonin transporter in lower doses. Common side effects of TCAs include drowsiness, dry mouth, urinary retention, constipation, weight gain and orthostatic hypotension [9]. Other effects can include increased intraocular pressure, increased risks of falls in elderly, palpitations, QT prolongation and arrhythmias at high doses. TCAs should not be combined with other QT‐prolonging drugs. They should be used with caution when combined with other serotonergic agents to minimize the risk of serotonin syndrome [22] and in epileptic patients as TCAs can reduce the threshold for seizures [23]. The most common side effects of SNRIs include nausea, drowsiness and dizziness. Gastrointestinal side effects such as constipation, diarrhea and dry mouth are often reported9. SNRIs, similar to SSRIs, may affect the effects of serotonin on platelets, and increase the risk of bleeding, particularly in the GI tract in patients on chronic anticoagulant or antiplatelet therapy24. Venlafaxine may also cause sweating, weight loss, and headaches. Hypertension has been reported in 3–13% of subjects, especially with high doses of 375 mg/day [25]. Neuropathic pain is defined as pain caused by a lesion or disease of the somatosensory nervous system. Most antidepressant pain research has focused on neuropathic pain, and both TCAs and SNRIs are considered first‐line agents in multiple neuropathic pain conditions. Current IASP Neuropathic Pain Special interest group (NeuPSIG) guidelines regarding pharmacologic treatment of neuropathic pain were updated based on a systematic review and meta‐analysis of randomized controlled trials published (or unpublished but with results available) between 1966 and 2014 [1]. Subjects’ diagnoses included post‐herpetic neuralgia, diabetic and non‐diabetic painful polyneuropathy (DPPN, PPN), post‐amputation pain, post‐traumatic or post‐surgical neuropathic pain (including plexus avulsion and complex regional pain syndrome type II), central post‐stroke pain (CPSP), spinal cord injury (SCI) pain and multiple sclerosis (MS)‐associated pain. Studies included in the meta‐analysis were randomized, double‐blind, placebo‐ or first‐line drug‐controlled trials with a parallel group or crossover design, at least ten patients per group and a primary outcome measure of neuropathic pain intensity. For each antidepressant class, the authors calculated the number needed to treat (NNT) for 50% pain intensity reduction (or 30% pain reduction, or at least moderate pain relief) and number needed to harm (NNH) for one patient to drop out of the study due to adverse effects. Fifteen studies of tricyclic antidepressants were included in the meta‐analysis: nine with amitriptyline (75–150 mg), two with desipramine alone (25–250 mg), one with nortriptyline alone (100 mg), one with nortriptyline and desipramine (160 mg), one with maprotiline (75 mg) and one with imipramine (150 mg). The majority of studies focused on painful polyneuropathy (seven studies) or post‐herpetic neuralgia (three studies). One study focused on each of: central post‐stroke pain, spinal cord injury, peripheral nerve injury, painful radiculopathy and multiple sclerosis. All studies demonstrated analgesic effects except two: amitriptyline 100 mg was ineffective for HIV‐associated neuropathy and nortriptyline 100 mg was ineffective for painful radiculopathy. Overall, these studies provided moderate‐quality evidence for efficacy of tricyclic antidepressants in most studied neuropathic pain conditions, with a combined NNT of 3.6 and NNH of 13.4. There was no apparent dose‐response effect. Ultimately, TCAs were strongly recommended as first‐line treatment. Ten studies of SNRIs were included in this meta‐analysis: seven with duloxetine (40–120 mg), two with venlafaxine (150–225 mg) and one with desvenlafaxine (50–400 mg). Again, most studies focused on subjects with DPPN, while one focused on mixed PPN and one on multiple sclerosis. Duloxetine and venlafaxine were efficacious in all included studies, while desvenlafaxine results were unavailable. Of note, two studies included in the systematic review but excluded from meta‐analysis showed no benefit of venlafaxine over placebo in peripheral nerve injury or mixed neuropathic pain conditions. Overall, the studies included in this meta‐analysis provided high‐quality evidence for SNRI efficacy in neuropathic pain, with a combined NNT of 6.4 and NNH of 11.8. Therefore, SNRIs were also recommended as first‐line agents in neuropathic pain. Several placebo‐controlled randomized controlled trials (RCTs) of SNRIs in neuropathic pain have been conducted since 2014. Both duloxetine and (to a lesser extent) venlafaxine reduced pain associated with chemotherapy‐induced peripheral neuropathy [26]. Duloxetine was effective in patients with chronic low back pain and concomitant radicular pain [27]. Two other studies found that duloxetine improved DPPN [28,29], and a systematic review concluded that 84% of reviewed studies supported use of duloxetine as a first‐line agent [16]. With regard to SSRIs, only two studies qualified for inclusion in the systematic review. Fluoxetine (40 mg) was ineffective in treatment of diabetic painful peripheral neuropathy, whereas escitalopram (20 mg) was effective in treatment of mixed painful peripheral neuropathy, with a NNT of 5.1. Fibromyalgia is characterized by widespread, persistent musculoskeletal pain of unclear etiology, often associated with other nonspecific symptoms such as fatigue and difficulty with sleep and cognition [30,31
Chapter 16
Antidepressant analgesics in the management of chronic pain
Introduction
Pharmacological mechanisms
Safety profile
Tricyclic antidepressants
Serotonin‐norepinephrine reuptake inhibitors
Chronic pain conditions
Neuropathic pain
Fibromyalgia
Stay updated, free articles. Join our Telegram channel