Add bupropion (Wellbutrin). It’s easy, generally well-tolerated, and it works…or so we hope. One open trial found that 15 out of 28 patients (54%) responded to bupropion augmentation of SSRIs (DeBattista C, J Clin Psychopharm 2003 Feb;23(1):27-30.) In the Star-D trial, adding bupropion SR (mean dose, 267.5 mg/day) to Celexa 40 mg/day resulted in a 30% remission rate, no better than adding BuSpar (mean, 40.9 mg/day). There was no placebo comparison, and no double-blinding, so the results are not a sure bet, particularly since BuSpar has been shown to be ineffective as an augmenter.

Add lithium. Let’s face it, few of us ever use this strategy, even though we are constantly berated by academia for  not doing so. The latest scolding came in the form of a meta-analysis published in the Journal of Clinical Psychiatry (Crossley NA, et al., 2007;68(Jun):935-940). Researchers examined 10 high-quality, placebo-controlled trials and concluded that “there is firm evidence for lithium as an effective augmentation strategy.” Nonetheless, most psychiatrists and their patients are leery of lithium’s side effects and the need for monitoring. In addition, all those lithium-boosting studies are old (none more recent than 1996) and they mainly examined lithium/tricyclic combinations. The most recent double-blind study showed no efficacy of lithium added to nortriptyline (Nierenberg AA et al., J Clin Psychopharmacol 2003;23:92-95), and the Star-D study made Li look like an underwhelming choice. In Step 3 of that trial, patients who were randomized to lithium augmentation of citalopram (mean dose of lithium, 860 mg/day, mean serum level 0.6 meq/liter, mean duration 9.6 weeks) achieved only a 16% remission rate (Nierenberg AA et al., Am J Psychiatry 2006;163:1519-1530.)

Add thyroid. Like lithium, most of the placebo-controlled studies for T3 (triodothyronine) augmentation were done with tricyclics (see a brief but excellent literature review in Iosifescu DV et al., J Clin Psychiatry 2005;66:1038-1042). Three of the five tricyclic studies endorsed T3’s effectiveness; the other two did not. There have been two large controlled studies of T3 augmentation of SSRIs, both with non-treatment resistant depressed outpatients. The first showed no benefit of adding T3 25 or 50 µg/day to paroxetine 30 mg/day (Appelhof, BC et. al., J Clin Endocrinol Metab 2004;89: 6271-6276). Patients assigned to T3 50 µg/day were more likely to have side effects of palpitations, sweating, and nervousness. On the other hand, a recent study conducted in Israel found T3 (mean, 35 µg/day) plus sertraline (mean 88 mg/day) led to significantly better response and remission rates than sertraline plus placebo (Cooper-Kazaz R et al., Arch Gen Psych 2007; 64(6):679-688). Meanwhile, Star-D showed a 25% remission rate, not statistically higher than lithium’s 16% rate, but at least T3 was better tolerated (average dose was 45 microgms/day) (Nierenberg AA et al., Am J Psychiatry 2006;163:1519-1530.) Confused? You're not alone. The evidence is mixed, but it is a reasonable option to keep in the back of your mind.

Add mirtazapine (Remeron). We don’t normally think much about poor neglected mirtazapine. It’s generic, so no one is marketing it, and it can cause both weight gain and sedation. Nonetheless, using mirtazapine as an augmentation strategy has recently achieved a bit of buzz, and has been called “California Rocket Fuel” by Stephen Stahl (see page 290 of his Essential Psychopharmacology, 2nd Edition). The rationale is that mirtazapine increases both serotonin and norepinephrine via a different mechanism than SSRIs/SNRIs, turbo-charging their effects. This is a European favorite. For example, after an SSRI failure, Spanish psychiatrists favorite next move is to add mirtazapine (J. de la Gándara et al., Acta Psychiatrica Scandinavica 2005;112 (s428), 32-35.) An open label study in Ireland of 32 patients with refractory depression reported a 44% response rate at four weeks (Hannan N et al., J Psychopharm 2007;21(2):161-164). Meanwhile, back in the states, the STAR-D study found that this combination outperformed the MAOI Parnate (tranylcypromine), though the difference was not statistically significant (McGrath PJ et al., Am J Psychiatry 2006; 163(9): 1531-1541).

Add an atypical. We covered some recent data on atypicals as antidepressant augmentation in the March 2007 issue of TCPR (“Do Second-Generation Antipsychotics Treat Depression? An Update”). The bottom line is that there is positive evidence (of varying quality) for all the atypicals. The “news” is that the FDA has fast-tracked Bristol-Myers Squibb/Otsuka’s application to approve aripiprazole (Abilify) for adjunctive use in treatment resistant depression. They apparently have some placebo controlled data, although the numbers presented at the latest APA meeting were not very impressive (see coverage on the web at http://www.medicalnewstoday.com/articles/71865.php).

Add a stimulant or modafinil. In one study of the elderly, 6 of 11 patients given a combination of methylphenidate (mean 12.2 mg/day) and citalopram (27.5 mg/day) had an accelerated antidepressant response (response within 2 weeks) (Lavretsky H, et al., J Clin Psychiatry 2003;64(Suppl 12):1410-1414). If you are particularly daring, you’ll add a stimulant to an MAOI – that’s right, an MAOI. The combination is used by experienced psychopharmacologists, and in one case series of 32 patients, it was well tolerated, with only one case of elevated blood pressure (see the excellent review by Feinberg SS, J Clin Psychiatry 2004;65:1520-1524). If you use it, start low, with either methylphenidate 5 mg or d-amphetamine 2.5 mg and work your way up gradually. In two trials, adding modafinil (Provigil) to an antidepressant did a good job of waking people up but did not help with the core symptoms of depression. However, one recent study found it effective as an augmentation agent in bipolar depression (see this month’s update section for details).