Antidepressants as Analgesics: Which Ones Are Best?

Antidepressants are often referred to as “adjuvant analgesics.” Although the name suggests that these agents don’t provide direct pain relief in the same way as opioids or NSAIDs, it is well established that antidepressants provide excellent analgesia for many pain conditions. In fact, they are considered first-line drugs for several disorders-including fibromyalgia; neuropathic pain conditions, such as diabetic neuropathic pain, and postherpetic neuralgia; and prophylactic treatment of migraine  and tension-type headaches.

Multiple studies have also demonstrated that drugs that increase the bioavailability of both serotonin and norepinephrine-the SNRIs-provide much more analgesia than do drugs that affect only one neurotransmitter (ie, the SSRIs).

Tricyclic antidepressants (TCAs) are the oldest of the SNRIs; their analgesic properties were proved over 30 years ago. More recently introduced SNRIs include duloxetine (Cymbalta), venlafaxine (Effexor), desvenlafaxine (Pristiq), and milnacipran (Savella). Duloxetine and milnacipran are the only SNRIs that are FDA-approved as analgesics; however, the analgesic properties of venlafaxine are also supported by multiple studies. There is little information on the potential analgesic properties of desvenlafaxine. Duloxetine is currently the only drug approved by the FDA for the treatment of depression and pain. Milnacipran is not approved as an antidepressant in the US, but it is approved for this indication in a number of European countries.
 
Which SNRIs are the best pain relievers?
Most of the published guidelines on neuropathic pain still recommend the TCAs as first-line drugs. Some have also more recently elevated the newer SNRIs to this same level.^1-3  

A recently published review of randomized control trials of venlafaxine, duloxetine, and milnacipran versus the TCAs provides some guidance in this area. Watson and colleagues⁴ found multiple RCTs that established the analgesic effects of venlafaxine and duloxetine in patients with diabetic neuropathic pain, postherpetic neuralgia, and fibromyalgia, and for prophylaxis of migraine and tension-type headache. Duloxetine was found to be beneficial for osteoarthritic pain and low back pain. The studies on milnacipran are essentially limited to those focusing on fibromyalgia.

Unfortunately, there are few studies that compare the newer SNRIs with one another or with the TCAs. Based on those that are available, Watson et al⁴ conclude that TCAs are at least as analgesic as venlafaxine, duloxetine, or milnacipran, and that overall TCAs may actually be more efficacious than the newer drugs. 

The authors note, however, that the potential for more analgesic effect with the TCAs needs to be weighed against their adverse-effect profile. Because of their anticholinergic effects, TCAs are commonly associated with dry mouth, constipation, and difficulty in urinating and, especially in geriatric patients, they may be associated with problems with cognition, and their antihistaminic properties can cause sedation. Nortriptyline and desipramine have less anticholinergic and antihistaminic effects than the more widely used amitriptyline. TCAs are also associated with the risk of impaired cardiac conduction.

Although all of these adverse effects can be problematic in persons of any age, they limit use of TCAs in the geriatric population-a group especially at risk for disorders such as diabetic neuropathic pain and postherpetic neuralgia. The newer SNRIs do not have these properties and are usually well tolerated, even by older patients.

Watson and colleagues do not discuss the issue of cost. All of the TCAs are available in generic forms, as is venlafaxine. Duloxetine and milnacipran are still only available in the US in brand-name formulations, which makes them more expensive. 

Conclusions
The review by Watson et al generally confirms an approach I recommend when the decision is made to try an antidepressant for pain. If a patient can tolerate a TCA and if there are no contraindications, I would still consider these first-line drugs. In fact, one of their adverse-effects-sedation-may actually be beneficial in that many patients with pain have difficulty in sleeping. TCA therapy may reduce the need for an additional sleep medication.

For patients who cannot take a TCA, consider using venlafaxine before proceeding to duloxetine. Some physicians may prefer the latter because the dosing requires less titration. At lower doses, venlafaxine acts primarily as an SSRI: to obtain a noradrenergic effect, the dosage must be titrated up. 

I have used milnacipran much less frequently than these other drugs and have generally limited it to patients with fibromyalgia who haven’t responded to one of the others.

I certainly agree with the authors of the review that our confidence in these recommendations is still somewhat limited because of the dearth of studies that compare these drugs with each other and also with other analgesics-especially those used for neuropathic pain, including the anticonvulsants and opioids. 

One final comment: TCAs, venlafaxine, and duloxetine not only provide analgesia but also are excellent antidepressants. If patients have comorbid pain and depression, there is no reason why one of these drugs should not be used to address both problems.
 
References:
1. Dworkin RH, O’Connor AB, Audette J, et al. Recommendations for the pharmacological management of neuropathic pain: an overview and literature update. Mayo Clin Proc. 2010;85:S3-S14.
2. Moulin DE, Clark AJ, Gilron J, et al. Pharmacological management of chronic neuropathic pain: consensus statement and guidelines from the Canadian Pain Society. Pain Res Manage. 2007;12:13-21. 
3. Attal N, Cruccu G, Baron R, et al. European Federation of Neurological Societies (EFNS) guidelines on the pharmacological treatment of neuropathic pain: 2010 revision. Eur J Neurol. 2010;17:113-123.
4. Watson CPN, Gilron I, Sawynok J, et al. Nontricyclic antidepressant analgesics and pain: are serotonin norepinephrine reuptake inhibitors (SNRIs) any better? Pain. 2011;152:2206-2210.