Antidepressants: The spectrum beyond depression
Referring to certain groups of drugs as antidepressants does them a great disservice; their potential uses range far beyond mood disorders.
A molecule is a molecule is a molecule—until it becomes identified with a purpose. Consider, for example, (-)-trans-4R-(4’-fluorophenyl)-3S-[(3’,4’-methylenedioxyphenoxy) methyl] piperidine. You probably know this molecule as paroxetine—an antidepressant, of course, but it is more than that. If you examine paroxetine’s FDA-approved indications, it also has anti-panic, anti-social anxiety, anti-obsessive-compulsive disorder, anti-posttraumatic stress disorder, and anti-premenstrual dysphoric disorder effects.
“Antidepressants” have achieved fame as antidepressants; one could say these molecules’ search for meaning has been fulfilled. Yet even within psychiatry, their many other uses (Table) can create semantic misunderstandings. Beyond psychiatry, consider the nondepressed patient with neurocardiogenic syncope who wonders why he’s being treated with an antidepressant.
Rather than calling antidepressants “panaceas,” the better choice is to educate patients about the drugs’ wide spectrum of activity. Let’s look broadly across the so-called antidepressants and examine their varied uses in psychiatry and other medical specialties.
FDA-approved psychiatric indications for serotonin uptake inhibitors*
SSRIs: selective serotonin reuptake inhibitors; SNRIs: serotonin-norepinephrine reuptake inhibitors; MDD: major depressive disorder; PD: panic disorder; SAD: social anxiety disorder; PTSD: posttraumatic stress disorder; GAD: generalized anxiety disorder; OCD: obsessive-compulsive disorder; PMDD: premenstrual dysphoric disorder; BUL: bulimia
* The absence of an X does not necessarily imply that a drug is ineffective for a given indication but, more likely, that definitive studies are lacking.
Peripheral neuropathy. The only antidepressant with an FDA-approved pain indication is duloxetine, a serotonin-norepinephrine reuptake inhibitor (SNRI). Its approval for diabetic peripheral neuropathic pain (DPNP) was based on two 12-week, randomized, double-blind, placebo-controlled studies using fixed doses of 60 mg once or twice daily.1,2 Another SNRI—venlafaxine XR, 150 to 225 mg/d, but not 75 mg/d—also was found to be more effective than placebo for this indication in a 6-week, double-blind study (Box 1).3
Using antidepressants to treat pain syndromes is neither new nor restricted to SNRIs, however. In combined double-blind, cross-over studies of patients with DPNP, Max et al4 found:
- moderate or greater pain relief in 74% and 61% of subjects, respectively, from the tricyclics amitriptyline, mean 105 mg/d, and desipramine, mean 111 mg/d—with pain reduced by equal amounts in depressed and nondepressed patients
- no statistically significant difference in pain relief between the selective serotonin reuptake inhibitor (SSRI) fluoxetine, 40 mg/d, and placebo.
Sindrup et al5 concluded in a 2005 review that antidepressants relieve DPNP according to this hierarchy:
- tricyclics: 1 in every 2 to 3 patients
- SNRIs: 1 in every 4 to 5 patients
- SSRIs: 1 in every 7 patients.
Bupropion SR—a norepinephrine dopamine reuptake inhibitor—also may be more effective than placebo in relieving neuropathic pain, as shown in a small (N=41) 6-week, double-blind study.6
Chronic headache. A meta-analysis7 of randomized, placebo-controlled studies found antidepressants more effective than placebo for chronic migraine and tension headache prophylaxis. Although a subgroup meta-analysis found similar effects for tricyclics and SSRIs, the authors characterized the tricyclics’ results as well established and the SSRIs’ as “less certain.”
The results of this meta-analysis might not accurately reflect bona fide antidepressants, however. Some of the 38 studies (25 of migraine, 12 of tension headache, 1 of both) included treatment with serotonin antagonists—most commonly pizotifen, which is not available in the United States and does not appear to be an antidepressant.
Back pain. Patients with chronic low back pain (average 10 years) seem to benefit from antidepressants, according to a meta-analysis of 9 randomized, controlled trials by Salerno et al.8 The effect on pain in the total 504 patients was “small but significant,” and improvement in function was “small but nonsignificant.” Individual sample sizes also were small, however, and only 2 studies excluded depressed patients.
Fibromyalgia, with chronic generalized musculoskeletal pain and tenderness, has been a focus of antidepressant drug therapy. Goldenberg et al9 concluded from an ambitious literature review (505 articles) that evidence of efficacy was strong for amitripty-line and modest for SSRIs and SNRIs.
On the other hand, Littlejohn and Guymer10 concluded in a clinical review that trials of SSRIs “have been somewhat disappointing,” that tricyclics are, at best, “only moderately effective,” and that more balanced dual uptake inhibitors such as duloxetine and the investigational agent milnacipran “are showing more promise.” Two placebo-controlled studies by Arnold et al11,12 of women with fibromyalgia showed benefit from duloxetine that appeared independent from its effect on depression and anxiety.
Overall, antidepressants are generally understood to have analgesic effects in the absence of depression. Benefits for patients with pain syndromes are well established for tricyclics (especially amitriptyline) and recently with SNRIs, whereas SSRIs are less effective.
Dual-action antidepressants ease neuropathic pain
Serotonin and norepinephrine are involved in pain modulation via descending inhibitory pathways in the brain and spinal cord. Serotonin-norepinephrine reuptake inhibitors (SNRIs) have been shown to reduce the severity of diabetic peripheral neuropathic pain (DPNP) in randomized controlled trials.
Duloxetine. In 2 double-blind studies,1,2 nondepressed patients with DPNP received duloxetine, 60 mg once daily; duloxetine, 60 mg bid; or placebo for 12 weeks. They rated the severity of neuropathic pain every 24 hours on an 11-point Likert scale, and weekly mean scores were the primary outcome measure. Average pain scores improved more in both duloxetine groups vs placebo. Duloxetine treatment did not interfere with diabetic control, and both dosages were well tolerated.
The FDA approved an added indication for duloxetine in the management of DPNP.
Venlafaxine. In a double-blind study,3 244 adult outpatients with moderately severe DPNP received venlafaxine ER, 75 or 150 to 225 mg/d, or placebo for 6 weeks. Daily scores on the Visual Analog Pain Intensity (VAS-PI) and Pain Relief (VAS-PR) scales were primary efficacy measures.
Patients receiving the higher venlafaxine dosage—but not 75 mg/d—showed statistically significant less-intensive pain vs placebo. VAS-PI scores were 27% lower than at enrollment with placebo, 32% lower with venlafaxine, 75 mg/d, and 50% lower with venlafaxine, 150 to 225 mg/d (P<0.001 vs placebo). VAS-PR scores also were significantly greater with venlafaxine, 150 to 225 mg, compared with placebo (P<0.001).
Nausea and somnolence were the most common side effects; clinically important ECG changes occurred in 7 patients treated with venlafaxine, 150 to 225 mg/d.
Bupropion SR is FDA-approved to aid smoking cessation, and this effect is independent of the drug’s antidepressant activity. Bupropion may act as a nicotine receptor antagonist as well as a norepinephrine dopamine reuptake inhibitor.
Other antidepressants have been studied for smoking cessation, with nortriptyline showing benefit in 2 large placebo-controlled trials. Studies with doxepin, fluoxetine, and moclobemide found little or no benefit for this indication.
Angina. Monoamine oxidase inhibitor (MAOI) antidepressants were used to treat angina pectoris in the late 1950s and early 1960s. This practice stopped after evidence showed that whereas angina pain may have improved with MAOIs, stress-induced ischemia on ECG did not.
Antiarrhythmia. Tricyclics had a brief fling in cardiovascular therapeutics when their quinidine-like class I antiarrhythmic activity was recognized. Imipramine was one of several drugs included in the Cardiac Arrhythmia Pilot Study in the 1980s that involved 502 postmyocardial infarction patients with ventricular arrhythmias. Imipramine was the least effective of the 4 drugs studied and the least well tolerated.13
Syncope. Neurocardiogenic (vasovagal) syncope—a common cause of fainting— can be treated when necessary with a
variety of medications. Options include the vasopressor midodrine, fludrocortisone, beta blockers, and SSRIs— none
FDA-approved for this indication. Paroxetine, 20 mg/d, was considerably more effective than placebo in preventing
recurrent syncope in 68 patients who had been unresponsive
to or intolerant of traditional medications. During a mean 25 months of treatment, 82% of patients remained syncope-free on paroxetine vs 47% on placebo.14
Nondepressed patients with IBS feel better when taking SSRIs
Selective serotonin reuptake inhibitors (SSRIs) often are used to treat irritable bowel syndrome (IBS), though evidence of their effectiveness is scarce. SSRIs can improve IBS patients’ quality of life, but effects on abdominal pain and bloating are less clear.
Paroxetine. In a randomized, double-blind trial,16 gastroenterologists tested a highfiber diet plus paroxetine in nondepressed patients with IBS. Ninety-eight patients ages 18 to 65 who experienced IBS symptoms on low- or average-fiber diets were first put on high-fiber diets and assessed for well-being and abdominal pain and bloating. Of these, 81 symptomatic patients continued highfiber diets with added paroxetine, 10 to 40 mg/d (n=38) or placebo (n=43).
With paroxetine, patients’ overall well-being improved more than with placebo, but abdominal pain and bloating and social functioning did not.
Fluoxetine. In a double-blind, randomized trial,17 44 patients with pain and constipation-predominant IBS received fluoxetine, 20 mg/d, or placebo for 12 weeks. These patients met Rome II criteria for IBS—abdominal discomfort/pain for ≥12 weeks in past year that met 2 of 3 criteria:
- relieved by defecation
- onset associated with change in stool frequency
- onset associated with change in stool appearance.
Patients receiving fluoxetine had less abdominal discomfort, less bloating, more frequent bowel movements, and decreased consistency of stool vs placebo 4 weeks after treatment stopped. Mean number of symptoms per patient decreased from 4.6 to 0.7 in the fluoxetine group vs 4.5 to 2.9 in controls (P<0.001).
Citalopram. IBS symptom severity was the primary outcome in a crossover trial comparing citalopram (20 mg for 3 weeks and 40 mg for 3 weeks) with placebo in 23 nondepressed patients.18 Abdominal pain and bloating, impact of symptoms on daily life, and overall well-being improved significantly more with citalopram than with placebo after 3 and 6 weeks.
Symptom improvements were not related to changes in depression, anxiety, or colonic sensorimotor function.
Peptic ulcer disease was shown in the 1980s to respond to tricyclic antidepressants. At the time, both anticholinergic and antihistaminic effects were thought to be responsible, but the later observation that trimipramine inhibited Campylobacter pylori in vitro suggested an additional explanation. Today, tricyclics are only of historic interest as treatments for peptic ulcer.
Irritable bowel syndrome (IBS) patients have responded favorably to antidepressants, although it is often difficult to know if the benefit is independent of improved coexisting anxiety or depression. A meta-analysis of 12 randomized, placebo-controlled trials—mostly with tricyclics—found an odds ratio for improvement of 4.2 and a number needed to treat of 3.2.15
More recently, a few placebo-controlled studies have shown SSRIs to be beneficial for IBS,16-18 although not all symptoms improved and some IBS subtypes might be more responsive than others (Box 2). In an editorial, Talley19 concluded that antidepressant therapy of IBS was “at best only a ‘band-aid’ approach to management.”
Nocturnal enuresis. In the 1960s, imipramine was shown—in some but not all placebo-controlled studies—to be beneficial for nocturnal enuresis in children and adults. Although imipramine is not FDA-approved for this indication, it is thought to work by relaxing bladder muscle and contracting bladder neck smooth muscle. Imipramine appears to have a vasopressin-independent antidiuretic effect in enuretic patients with nocturnal polyuria.
Stress urinary incontinence. Placebo-controlled studies have shown duloxetine to be an effective treatment for stress urinary incontinence in women. A Cochrane Database Review of 9 randomized studies in adults (N=3,327) concluded that duloxetine significantly improved patients’ quality of life, although how long the benefits would last was unclear.20
Duloxetine is thought to improve stress urinary incontinence by increasing urethral sphincter tone and the force of sphincter contraction. This indication is not FDA- approved for duloxetine but is approved in the European Union.