An SSRI antidepressant is the initial choice for standard treatment of major depression, and often another SSRI antidepressant is the second choice if the first is ineffective. The overwhelming majority of antidepressant prescriptions in the United States are for SSRIs.1 They are considered safer and better tolerated than the antidepressant medications that preceded them, including monoamine oxidase inhibitors (MAOIs) and tricyclic antidepressants (TCAs).2-4
Yet, a pivotal recent study that measured the rate and degree of clinical response to SSRI therapy in 2876 typical outpatients in conventional "real-world" treatment settings, found a 47% response rate (defined as 50% or more improvement), and a 28% remission rate (conventionally defined as a 17-item Hamilton Depression Rating Scale Score of 7 or less).5 Although remission rate was not commonly reported in early antidepressant trials and so direct comparisons cannot be made with the early literature, this is a markedly lower rate of response than had been consistently observed with TCAs6 and in a subgroup of outpatients with MAOIs.7,8
Various explanations have been offered for the decreasing response rates to antidepressants reported in clinical trials over time. These include changes in placebo response, study settings (inpatient vs outpatient), and populations studied (more treatment-refractory persons presenting for studies).9 Whatever the role of the comparative efficacy of older versus newer antidepressants may be, a review of the recent literature clearly indicates that a large percentage of patients will not respond, have only partial response, or lose an initial response to the newer antidepressants.10 Thus, it is important to be familiar with, and comfortable using, all classes of antidepressant medications.
In lecturing to medical students, residents, and psychiatrists during the past several years, we have encountered widespread hesitancy in the use of MAOIs and even TCAs, mainly because of concerns about their safety but also because of doubts about their effectiveness compared with newer alternatives. Thus, it is timely to review the literature on the efficacy and safety of TCAs and MAOIs, with a view to maintaining an appropriate place for these 2 drug classes in the pharmacotherapy of depression.
Tricyclic antidepressants
The prototype TCA is imipramine(Drug information on imipramine), the antidepressant activity stumbled on in the late 1950s in the search for new phenothiazine antipsychotics, which resemble imipramine on the molecular level.11 The second TCA to be found effective was amitriptyline(Drug information on amitriptyline). Subsequently, the active metabolites of each of these, desipramine and nortriptyline(Drug information on nortriptyline), were marketed and found to be somewhat better tolerated than their parent compounds.11 Several other TCAs were also introduced, including clomip- ramine; doxepin; trimipramine; protriptyline; and the closely related tetracyclic compounds, maprotiline(Drug information on maprotiline) and amoxapine(Drug information on amoxapine).11
TCAs were the precursors of the current SSRIs and the serotonin-norepinephrine reuptake inhibitors (SNRIs). They share the basic mechanism of action of inhibition of reuptake of monoamine neurotransmitters from the synaptic cleft. They vary across a spectrum of relative affinity for serotonin and norepinephrine(Drug information on norepinephrine) reuptake sites (Table 1), with clomipramine(Drug information on clomipramine) showing the most selective affinity for the serotonin reuptake site, desipramine the most potency at the norepinephrine reuptake site, and maprotiline the most selectivity for the norepinephrine reuptake site.4,12-14 It is notable that the major metabolite of clomipramine, desmethylclomipramine, is also an SNRI, giving clomipramine a combined serotonergic-noradrenergic effect clinically.4
Apart from serotonin and norepinephrine reuptake blockade, TCAs share varying degrees of direct receptor antagonist activity (Table 2), which is considered responsible for their side effects, although it has been speculated that this activity may contribute to their therapeutic effects as well. In addition to blocking adrenergic, histaminergic, and cholinergic receptors, these drugs all show quinidine(Drug information on quinidine)-like effects on cardiac conduction, which contribute to their potentially lethal toxicity when taken in overdose.
TCA efficacy
Placebo-controlled studies have consistently found TCAs to be highly effective in treating major depression. In a 1969 review of all studies published in English with sufficient data to be analyzed, the prototype TCA, imipramine, was shown to be superior to placebo by a large margin.6 In patients treated with imipramine, the percentage of those who were moderately improved or better was 70% versus 39% for placebo.6
Early clinical trials characterized TCAs as most effective in endogenous depression, a diagnosis similar to the current DSM-IV subtype of depression with melancholic features. This form of depression manifests as a marked loss of pleasure or interest with insomnia, weight loss, and guilt. These studies compared endogenous depression with reactive and neurotic depression subtypes.15-17 Neither subtype has current-day equivalents, but they can perhaps best be considered forms of nonmelancholic depression.A consistent observation has been that severe depression showed the most reliable response to tricyclics.18
Since the introduction of the SSRIs, it has been repeatedly observed that severely depressed inpatients, usually those with melancholia, show a more robust response to TCAs than to SSRIs. The first studies to demonstrate this were by the Danish University Antidepressant Group, which published 2 widely cited studies showing the superiority of the TCA clomipramine to 2 SSRIs, citalopram(Drug information on citalopram) and paroxetine(Drug information on paroxetine), in hospitalized patients who had depression.19,20 Subsequent studies replicated the finding that TCAs are superior to SSRIs in specific populations.21-23
Akhondzadeh and colleagues21 reported that nortriptyline was more effective than fluoxetine(Drug information on fluoxetine) for the treatment of moderate to severe depression. Roose and colleagues22 demonstrated the superiority of nortriptyline over fluoxetine in elderly hospitalized patients with melancholic depression with comorbid heart disease. Joyce and colleagues23 found melancholia more responsive to nortriptyline than fluoxetine treatment in older men, but the opposite was true in younger women. Increased benefits have been demonstrated by Robinson and colleagues24 with nortriptyline compared with fluoxetine in patients with poststroke depression. Interestingly, in a study designed to rate patients' perceived benefit from treatment, patients reported older medications and electroconvulsive therapy (ECT) to be more helpful than the newer antidepressants.25 The authors also reported that ECT and older antidepressant treatments were more strongly correlated with acute recovery.25
Meta-analyses of the efficacy of TCAs compared with other antidepressants have tended to find TCAs somewhat more effective than SSRIs.3,26 However, because of heterogeneity of study designs and outcome measures, it is difficult to draw firm conclusions based on the meta-analyses currently available.9
One study looked at the effect of a placebo-controlled crossover to the alternative treatment between an SSRI and a TCA in cases of clinical nonresponse to an initial 6-week trial of paroxetine versus imipramine.27 It was found that 73% of paroxetine nonresponders recovered while taking imipramine, while only 50% of imipramine nonresponders recovered when treated with paroxetine.27 Interestingly, in the 1-year follow-up maintenance study, including initial responders and crossover responders, 26% of those who had recovered with paroxetine relapsed, while no patients who had recovered with imipramine relapsed.28
A much larger multicenter crossover study of treatment-resistant chronic depression found that imipramine was slightly less effective than sertraline as a second treatment (55% vs 63%), although completer analysis did not demonstrate a significant difference.29 In the STAR*D study, after failing to achieve remission from 2 previous courses of treatment for the index episode of illness, the first being 14 weeks of treatment with 40 mg of citalopram daily, 20% of patients were able to achieve remission by taking nortriptyline while only 12% achieved remission when treated with mirtazapine(Drug information on mirtazapine).30
Finally, there is a small database of literature that examined the risks and benefits of combining SSRIs with TCAs, which we have reviewed elsewhere.31 Although there are metabolic interactions to be careful of, and the published numbers are still quite small, when practiced with care, the combination has been demonstrated to be highly effective.31
