Innovations

Psychother Psychosom 2006;75:139–153 DOI: 10.1159/000091771

Current Status of Augmentation and Combination Treatments for Major Depressive Disorder: A Literature Review and a Proposal for a Novel Approach to Improve Practice

a b Maurizio Fava A. John Rush

a Depression Clinical and Research Program, Massachusetts General Hospital, Harvard Medical School, b Boston, Mass. , and UT Southwestern Medical School, Dallas, Tex., USA

Key Words rates since the lack of response with antidepressant Augmentation therapy Combination therapy monotherapy may lead many depressed patients with Antidepressant-resistant depression Remission little or no benefi t to drop out of treatment, precluding the subsequent use of augmentation or combination strategies altogether. In addition, the emergence of cer- Abstract tain side-effects (e.g., agitation, insomnia) or the persis- Most patients with major depressive disorder (MDD) do tence of some initial baseline symptoms (e.g., anxiety, not reach symptom remission. These patients with re- insomnia) may lead to premature discontinuation from sidual symptoms have worse function and worse prog- monotherapy in the absence of concomitant use of aug- nosis than those who remit. Several augmentation and menting pharmacological options targeting these symp- combination treatments are used to either increase the toms. chances of achieving remission or to eliminate/minimize Copyright © 2006 S. Karger AG, Basel residual depressive symptoms. Evidence for these phar- macological approaches rests primarily on open, uncon- trolled studies, and there are clearly not enough con- Treatment-resistant depression typically refers to the trolled studies. Clinicians should carefully weigh these presence of an inadequate clinical response following ad- different treatment options to increase their patients’ equate antidepressant therapy among patients suffering chances of achieving and sustaining remission from de- from major depressive disorder (MDD). While the more pression. This paper will review the pertinent studies and traditional view of treatment resistance has focused on will propose a novel approach to improve practice in- nonresponse (i.e., patients with minimal or no improve- volving the use of augmentation or combination strate- ment), ‘inadequate response’ is now viewed as the lack of gies at the outset of initial treatment to primarily enhance symptom remission. Achieving remission is a signifi cant the chances of remission through synergy and/or a clinical challenge [1] , as a signifi cant proportion of pa- broader spectrum of action. This novel approach could tients with MDD do not achieve full remission despite potentially enhance retention and/or increase remission adequate treatment and (for some) signifi cant symptom

© 2006 S. Karger AG, Basel Maurizio Fava, MD Depression Clinical and Research Program, Massachusetts General Hospital Fax +41 61 306 12 34 Bulfi nch 351-55 Fruit Street E-Mail [email protected] Accessible online at: Boston, MA 02114 (USA) www.karger.com www.karger.com/pps Tel. +1 617 724 0838, Fax +1 617 726 7541, E-Mail [email protected] Downloaded by: Llumc.CP 11105 Loma Linda Univ. 151.112.124.130 - 3/13/2014 10:04:42 PM improvement [2]. In fact, pooled analyses of double-blind Psychoeducational materials and an emphasis on the im- antidepressant effi cacy trials in typically uncomplicated portance of communication and collaboration may help (i.e., minimal general medical and psychiatric comorbid- set the stage for meaningful dialogue. Psychoeducation ity), nonchronic depressions treated under research clin- may therefore increase the degree of collaboration be- ic conditions reveal that remission rates range between tween the patient and the treating clinician and may en- 30 and 45%, so that the majority of MDD patients are hance the acceptability of any subsequently proposed expected to fail to achieve remission with a single trial of treatment approach. monotherapy with antidepressants [3]. Patients who do (2) Enhancing treatment adherence. Adequate follow- not achieve remission, including those who respond (e.g., up with patients (offi ce visits or phone contacts) is known experience a 650% reduction of symptoms) but do not to lead to better adherence to treatment [9] . In certain remit, continue to be affected by psychological, behav- settings (e.g., primary care), improved care management ioral, and somatic residual symptoms [4], including de- can greatly improve treatment adherence [10]. Antide- pressed mood, reduced levels of interest, excessive guilt, pressants with relatively greater tolerability, fewer side- fatigue, sleep disturbances, appetite changes, and pain. effects, or that require only once-a-day dosing, also can Even among 108 patients who had reached ‘remission’ enhance adherence. It is important to discuss possible based on the convention of a 17-item Hamilton Rating side-effects that may occur during antidepressant treat- Scale for Depression (HAM-D) score !8, 25.9% had 1 ment and possible approaches to their management, residual symptom, and 56.5% had 2 or more symptoms should they occur, even before prescribing a treatment. [4]. Engaging the patient in the choice of the medication and Relapse and recurrence after successful treatment of in discussing the relative risk of potential side-effects is MDD is both common and debilitating [5]. Patients with thought to also enhance patient engagement in the treat- MDD who do not achieve complete symptom remission ment process [11]. are particularly vulnerable to relapse [6–8]. For example, (3) Ensuring adequacy of antidepressant dose. Al- in a 15-month study of long-term outcome of treatment though antidepressant medications are typically admin- of depression, Paykel et al. [6] followed 60 patients with istered at doses within recommended therapeutic ranges, major depression in remission. Relapses occurred within some patients may require doses well above the therapeu- the fi rst 10 months of follow-up in 76% (13/17) of patients tic range in order to remit. Monitoring antidepressant with residual symptoms but in only 25% (10/40) of pa- blood levels may be useful for patients who are not re- tients without residual symptoms. Taken together, these sponding and do not report side-effects, even with the data strongly support the need to improve the treatment newer antidepressants for which there is no clear blood of depression to achieve higher rates of remission and to level-response relationship. eliminate residual symptoms. (4) Ensuring adequacy of antidepressant treatment du- ration. Most patients require 6–12 weeks of treatment to achieve adequate response [12] . On the other hand, stud- Treatment Tactics to Increase Remission ies have shown that minimal improvement by week 4 or Rates with Antidepressant Monotherapy 5 leads to a very small chance of response [12, 13]. In fact, a study from our group has demonstrated that nonre- Clinicians can employ several treatment tactics to in- sponse as early as week 4 or 6 predicted poor outcome at crease the chances of achieving full remission with anti- week 8 [13] . These fi ndings suggest that, in general, clini- depressant monotherapy. These include (1) psycho-edu- cians must consider taking action, if at least minimal cation, (2) enhancing treatment adherence, (3) ensuring symptom improvement has not occurred by weeks 5 or 6 adequacy of antidepressant dose, (4) ensuring adequacy with an adequate dose. Improved longer-term outcomes of antidepressant treatment duration, (5) choosing anti- may be achieved with longer courses of treatment, per- depressant medications with relatively greater effi cacy in haps especially for the chronic or comorbidly ill pa- specifi c subtypes or populations, and (6) adding psycho- tients. therapy. (5) Choosing antidepressant treatments with relatively (1) Psychoeducation. It is considered quite helpful to greater effi cacy in specifi c subtypes or populations. Re- explain to patients that depression is a medical illness that mission rates tend to be comparable across the different is associated with changes in brain functioning and that antidepressant medications and classes. The chances of antidepressants are used to help improve brain function. remission, however, may be enhanced by choosing agents

140 Psychother Psychosom 2006;75:139–153 Fava /Rush

Downloaded by: Llumc.CP 11105 Loma Linda Univ. 151.112.124.130 - 3/13/2014 10:04:42 PM with relatively greater effi cacy in a specifi c depressive treatment, precluding the use of augmentation or combi- subtype. For example, dual-action antidepressants, act- nation altogether, and (4) the emergence of certain side- ing to inhibit the reuptake of both serotonin and norepi- effects (e.g., agitation, insomnia) or persistence of some nephrine, have been associated with higher remission initial baseline symptoms (e.g., anxiety, insomnia) may rates than certain single-action selective serotonin reup- lead to premature discontinuation from monotherapy take inhibitors (SSRIs) among patients with melancholic/ without pharmacological options to deal with these symp- endogenous depression [3, 14]. toms. In addition, such initial combinations may create (6) Adding psychotherapy. Recent evidence indicates a broader spectrum of action (i.e., a larger proportion of that for those who respond but do not remit to a medica- patients initially treated may at least respond if not re- tion, cognitive therapy not only removes the residual mit). Finally, the pharmacological synergism achieved symptoms but is associated with an improved prognosis with combinations may convert monotherapy responders [15]. Beginning with a combination of depression-target- into combination treatment remitters. In the following ed psychotherapy and medication substantially increases sections, we are going to review the studies concerning remission rates in clinically depressed patient [16]. the effi cacy of combination and augmentation strategies All these treatment tactics may increase the chance of in the treatment of MDD. All these strategies are off-label a depressed patient achieving full remission with antide- as no treatment has been approved for augmentation of pressant monotherapy, although it is not clear whether antidepressants in MDD. there is synergy across them or whether each tactic affects treatment outcome independently. However, even with enhanced care, remission rates in ‘real world’ patients not Augmentation Studies preselected to be treatment resistant, as in the case of the IMPACT study (25% at 1 year) [10] , remain rather mod- Over the past few decades, numerous compounds have est. For these reasons, clinicians often choose to augment been used to augment antidepressant medications. Most or combine antidepressant medications. of these studies are uncontrolled and open label, though some investigations are double blind and placebo-con- trolled. The latter studies allow us to draw relatively fi rm Augmenting or Combining Medications to conclusions about the effi cacy of some augmenting agents Achieve Remission (e.g., lithium and thyroid hormones). When the augmen- tation strategy is implemented after the patient with Augmentation strategies involve the use of medica- MDD has not achieved remission with antidepressant tions that are not standard antidepressant monotherapies monotherapy, improvement, if it is to occur, will follow in order to enhance the antidepressant effect. Combina- initiation of antidepressant augmentation within 3–4 tion strategies involve the use of two antidepressant med- weeks. Thus it is premature to decide about the effi cacy ications, typically of different classes. Augmentation and of augmentation within less than several weeks. Nearly combination treatments may be implemented either: all augmentation studies have focused on the short-term (a) at the onset of treatments to enhance the chances of outcomes (4–8 weeks). Little is known about the mini- achieving remission or (b) following the use of antidepres- mum duration of the augmentation trial for responders sant monotherapy initially, if it did not result in remis- to such a strategy. Augmentation strategies have also been sion. Most augmentation and combination studies have used to hasten the onset of antidepressant effect. A pleth- used the latter approach. Despite this fact and the fact ora of augmentation strategies has been reported [17] , but that current treatment guidelines recommend a single the most-studied strategies for depression are lithium, medication to initially treat MDD, several reasons sup- thyroid hormone, pindolol, and buspirone. port the use of augmented or combined treatment at treat- ment initiation to either enhance retention or to increase Lithium remission rates. (1) First, most patients with MDD do not Lithium augmentation, most popular in the 1980s, is remit with initial antidepressant monotherapy; (2) no ini- supported by controlled studies that have clearly shown tial monotherapy medication is robustly different from that the addition of a dose of 600 mg/day or more of lith- others in achieving remission [3]; (3) the lack of response ium, typically in divided doses, and with reasonably good with antidepressant monotherapy may lead many de- blood levels, leads to robust increases in response chanc- pressed patients with little or no benefi t to drop out of es in patients not responding to tricyclic antidepres-

Augmentation and Combination Psychother Psychosom 2006;75:139–153 141 Treatments for MDD Downloaded by: Llumc.CP 11105 Loma Linda Univ. 151.112.124.130 - 3/13/2014 10:04:42 PM sants (TCAs), monoamine oxidase inhibitors (MAOIs), studies of thyroid hormone augmentation with SSRIs or SSRIs. Twelve double-blind controlled trials of lithium have been carried out this far [36–38] . Common side-ef- augmentation in depression have reported response rates fects with thyroid hormone augmentation include palpi- that averaged 52% in a total of 255 lithium-augmented tations, sweating, nervousness, and tremor [39] . Six dou- patients [18–32] . A meta-analysis of placebo-controlled ble-blind, placebo-controlled studies assessed the effi cacy trials by Bauer et al. [33] revealed a response rate in the of concomitant administration of T3 and TCAs in accel- lithium-augmented group of 45 vs. only 18% in the pla- erating clinical response among patients with nonrefrac- cebo group (p ! 0.001). This meta-analysis, however, did tory depression. Five of the six studies found T3 to be not include the largest placebo-controlled study of lithi- signifi cantly more effective than placebo in accelerating um augmentation so far. In this study among 142 MDD clinical response, with the pooled, weighted effect size patients who had not responded to clomipramine mono- index being 0.58, and the average effect being highly sig- therapy [32], study exit remission and response rates were nifi cant [40] . Despite apparent acceleration of response not statistically different between lithium and placebo with T3, there are no published placebo-controlled stud- augmentation. Lithium augmentation was not particu- ies assessing the effi cacy of thyroid hormone augmenta- larly effective when added to SSRIs or serotonin norepi- tion in nonresistant depressed patients to enhance the nephrine reuptake inhibitors (SNRIs) [18, 19, 26, 32]. chances of achieving remission as a fi rst-step treatment. Similarly, the effi cacy of lithium augmentation in the only two double-blind studies in partial responders to SSRIs Buspirone in the literature is rather modest, with an overall response Buspirone is typically a well-tolerated anti-anxiety rate of 27% (7/26) [18, 19]. Therefore, the role of lithium drug, with serotonin (5-HT 1A ) receptor partial agonist augmentation among MDD patients treated with the properties. Open studies using 5–15 mg b.i.d. of buspi- newer antidepressants remains to be established. Further- rone have shown signifi cant improvement in refractory more, despite the relative robust early fi ndings concern- patients [24, 41–43]. The fi rst placebo-controlled study ing lithium augmentation of TCAs in the literature, the in resistant depression compared buspirone and placebo three most recent double-blind studies of lithium aug- augmentation among 117 patients who had not respond- mentation have all failed [18, 31, 32] , suggesting that per- ed to a minimum of 4 weeks of treatment with either par- haps lithium’s therapeutic success in the 1980s might oxetine or citalopram (mean treatment duration prior to have been partly due to the fact that bipolar spectrum augmentations: 5 months) and found no statistically sig- patients may have been enrolled in some of the early stud- nifi cant difference in response rates between these two ies of unipolar depression. The use of lithium is further treatments (51% buspirone versus 47% placebo) [44]. limited by high rates of side-effects (e.g., sedation and However, the extremely long duration of SSRI treatment weight gain), its potential toxicity (e.g., renal and thy- prior to randomization raises the possibility that many of roid), frequent blood monitoring requirements, and sig- the randomized patients may have been relapsers and not nifi cant toxicity or death in overdose. Perhaps for these nonresponders to SSRIs. A more recent double-blind reasons, there are no placebo-controlled studies to assess study showed that, among 102 depressed outpatients who the effi cacy of lithium augmentation in nonresistant de- had failed to respond to a minimum of 6 weeks of treat- pressed patients to enhance the chances of achieving re- ment with either fl uoxetine or citalopram, in the SSRI- mission (i.e., at the start of initial treatment). resistant patients with severe depression buspirone (20– 60 mg/day) augmentation was more effective than pla- Thyroid Hormone cebo and very well tolerated, even though the difference In treatment-resistant depression studies, L-triiodo- between buspirone and placebo was not signifi cant in the thyronine (T3) has been shown to be superior in effi cacy overall sample [45]. A randomized, open-label study in to thyroxine (T4) [34]. Among four randomized double- 120 nonresistant depressed patients showed compara- blind studies of T3 augmentation of antidepressants, ble response rates among patients treated with fl uox- pooled effects were not signifi cant (relative response, etine alone versus fl uoxetine plus buspirone [46] . How- 1.53; 95% CI, 0.70–3.35; p = 0.29), but one study with ever, there are no placebo-controlled studies in the negative results accounted for most of this effect [35]. literature assessing the effi cacy of buspirone augmenta- Since all published controlled studies involved T3 aug- tion in nonresistant depressed patients to enhance the mentation of TCAs, the effi cacy of T3 augmentation in chances of achieving remission. This is somewhat sur- SSRI-resistant patients is unknown. Only uncontrolled prising, as buspirone has shown effi cacy in anxiety dis-

142 Psychother Psychosom 2006;75:139–153 Fava /Rush

Downloaded by: Llumc.CP 11105 Loma Linda Univ. 151.112.124.130 - 3/13/2014 10:04:42 PM orders such as generalized anxiety [47] , and anxious Psychostimulants depression accounts for approximately half of depressed Psychostimulants also have signifi cant effects on do- outpatients [48] . pamine neurotransmission, and they have been used to augment TCAs, MAOIs, SSRIs, and SNRIs [71–77] . Cli- Pindolol nicians typically use methylphenidate (20–80 mg/day) or Pindolol augmentation, while rarely used in clinical dextroamphetamine (10–40 mg/day) in divided doses. A practice in the US, is relatively more popular in Europe small, uncontrolled study (n = 11) suggested the useful- and Canada in the management of resistant depression. ness of methylphenidate in accelerating antidepressant Pindolol is a -blocker and a serotonin 5-HT 1A antago- response [78] . The main issues concerning the use of psy- nist. A dose of 2.5 mg t.i.d. has been used in most depres- chostimulant augmentation are the potential for abuse in sion studies, yet PET imaging studies suggest that this some patients with history of substance abuse, the possible dose is likely suboptimal for an adequate occupancy of emergence of anxiety and irritability, and their relatively 5-HT 1A receptor [49] . This agent has generated a lot of short half-life [17]. There are no controlled data on the interest because it has been shown to accelerate antide- effectiveness of these augmenting agents in treatment- pressant response when combined with SSRIs in most resistant MDD or in nonresistant MDD to enhance the [50–55], but not all [56] studies. In some studies, higher chances of achieving remission at the onset of treatment. response rates were reported at endpoint with pindolol augmentation than with placebo augmentation [50, 52, Modafi nil 55, 57, 58] . However, a study by Moreno et al. [59] found Modafi nil, a novel psychostimulant, has pharmacolog- no response among 10 treatment-resistant depressed pa- ical actions that are distinct from those of amphetamines. tients, and two separate studies [60, 61] showed no dif- In a small (n = 7) retrospective case series, Menza et al. ference from placebo in augmenting antidepressants in [79] fi rst reported the usefulness of modafi nil (in doses up resistant depressed populations, despite good tolerability to 200 mg/day) as an adjunct to antidepressants in re- [61]. On the other hand, a study in 31 hospitalized de- sistant depression. More recently [80] , 8 of 14 patients pressed patients showed a 60% response rate in patients not responsive to SSRIs or venlafaxine rated themselves treated with fl uoxetine plus pindolol compared with a 9% as much improved following augmentation with up to response rate in patients treated with fl uoxetine alone 400 mg/day of modafi nil. A preliminary double-blind, [58]. placebo-controlled, 6-week study found that modafi nil rapidly improved fatigue and daytime wakefulness, with Dopaminergic Drugs signifi cantly greater mean improvements from baseline Given the putative role of dopaminergic neurotrans- than placebo in fatigue (Fatigue Severity Scale) scores at mission in MDD [62], augmentation with dopaminergic week 2 (p ! 0.05) and sleepiness (Epworth Sleepiness drugs is a potentially useful strategy. In an open trial (n = Scale) scores at week 1 (p ! 0.01), but no signifi cant dif- 20), Bouckoms and Mangini [63] used, with some suc- ferences between modafi nil and placebo at endpoint [81] . cess, the antiparkinsonian drug pergolide (0.25–2 mg/ A subsequent, placebo-controlled multicenter study eval- day) in resistant unipolar and bipolar depression. Simi- uated the effi cacy of modafi nil augmentation in MDD larly, there are uncontrolled, small studies of antidepres- patients with fatigue and excessive sleepiness despite sant augmentation with the dopaminergic drugs amanta- SSRI monotherapy. Of the 311 patients who received at dine (100–200 mg b.i.d.) [64, 65], pramipexole (0.125– least 1 dose of study drug, modafi nil signifi cantly im- 0.50 mg t.i.d.) [66–68] , and ropinirole (0.5–1.5 mg t.i.d.) proved patients’ overall clinical condition compared with [69]. The dopamine D 2 and D 3 receptor agonists prami- placebo (based on CGI-I scores, p = 0.02), with trends pexole and ropinirole are typically much better tolerated toward greater reductions in sleepiness and depression than the older dopaminergic drugs and are associated pri- (HAM-D and MADRS) severity scores versus placebo. marily with somnolence and very mild nausea [69, 70]. Modafi nil also signifi cantly reduced Brief Fatigue In- Without controlled studies, the effectiveness of antide- ventory scores for worst fatigue at exit (p ! 0.05 versus pressant augmentation with dopaminergic agents re- placebo). Only nausea and jitteriness were signifi cantly mains to be established for treatment resistant depression more common with modafi nil than placebo [82] . A small, for residual symptoms of depression, and as a treatment uncontrolled study of modafi nil has also suggested its use- for nonresistant depressed patients to enhance remission fulness in accelerating response and enhancing the chanc- rates. es of achieving remission [83] . The relatively greater user

Augmentation and Combination Psychother Psychosom 2006;75:139–153 143 Treatments for MDD Downloaded by: Llumc.CP 11105 Loma Linda Univ. 151.112.124.130 - 3/13/2014 10:04:42 PM friendliness of modafi nil (compared to psychostimulants) Opiates has made it a relatively popular augmentor of antidepres- Minimal evidence (mostly case reports and case series) sants for MDD residual symptoms, especially fatigue, suggests a role for antidepressant augmentation with opi- sleepiness, and lethargy [80, 82]. ates, such as oxycodone, oxymorphone [98], and bu- prenorphine [99]. The lack of defi nitive evidence and the Atypical Antipsychotics potential risk of abuse do not recommend the use of these Risperidone [84], olanzapine [85] , quetiapine [86] , ar- agents presently. ipiprazole [87, 88] and ziprasidone [89, 90] have all shown good responses in small, uncontrolled trials with SSRI Estrogen nonresponders, with the exception of the Shelton et al. Estrogen exerts profound effects on behavior by inter- study [85]. The typical doses in augmentation of antide- acting with neuronal estrogen receptors [100] . Mostly an- pressants are 0.5–2 mg/day for risperidone, 5–20 mg/day ecdotal evidence suggests the effi cacy of estrogen augmen- for olanzapine, 50–300 mg/day for quetiapine, 10–30 mg/ tation of antidepressants in resistant depression among day for aripiprazole, and 40–160 mg/day for ziprasidone. postmenopausal women. Two early studies [101, 102] A large study (n = 386) of citalopram monotherapy non- failed to fi nd a benefi t of estrogen augmentation of TCAs. responders showed a 63% remission rate after 4–6 weeks Four nonrandomized studies of hormone replacement open phase of risperidone augmentation (0.5–2 mg/day). therapy (HRT) to treat SSRI nonresponse [103–106] sug- During the double-blind discontinuation phase (n = 241), gested that estrogen may augment the antidepressant ef- however, 53% of patients randomized to remain on ris- fect of SSRIs. In addition, as pointed out by Stahl [107] , peridone relapsed versus 55% of those switched to pla- there are no guidelines on how to optimize antidepressant cebo (ns) [91]. A meta-analysis of two large controlled administration with estrogen, especially in women insuf- studies found olanzapine augmentation of fl uoxetine to fi ciently responsive to antidepressants. be effective over 8 weeks [92] . The apparently rapid onset of olanzapine augmentation [85] has made it relatively Dehydroepiandrosterone and Testosterone popular among clinicians for treatment-resistant depres- Dehydroepiandrosterone (DHEA), a major circulating sion, although very little is known about its effi cacy in adrenal androgen in humans, plays an unclear physiolog- managing residual symptoms of MDD or in its use for ic role. In addition to serving as a precursor to testoster- nonresistant depressed patients to enhance the chances one and estrogen, DHEA and its sulfated metabolite, of achieving remission. The potential risk for treatment- DHEA-S, most likely play important biological roles and emergent weight gain, sedation, extrapyramidal symp- have been hypothesized to be involved in regulating toms, metabolic disturbances (e.g., diabetes, and hyper- mood and sense of well-being [108] . A very small (n = 22), lipidemia), and hyperprolactinemia [93–95], although preliminary, double-blind study suggested the utility of variable among atypical antipsychotic drugs, has some- up to 90 mg/day as an adjunct to antidepressants in re- what limited the use of these drugs for nonpsychotic de- sistant depression [108]. Further studies are clearly neces- pression. Whether these agents are needed in the longer sary, given the small number of patients studied. Simi- term is unknown. The above-mentioned risperidone aug- larly, in an 8-week randomized, placebo-controlled trial mentation results suggest they may not be effective in the of testosterone transdermal gel among 23 men, aged 30– long term. 65 years, with resistant depression and low or borderline testosterone levels, testosterone was signifi cantly better Inositol than placebo in treating depressive symptoms [109]. A Despite initial anecdotal positive reports of inositol subsequent small study [110] did not show signifi cant dif- (up to 12 g/day) to augment antidepressants, a con- ferences between testosterone and placebo in the augmen- trolled, double-blind augmentation trial did not sup- tation of antidepressants among older hypogonadal men. port its use in SSRI nonresponders [96] . Another study However, there are no published studies that have exam- assessed the ability of inositol augmentation in enhanc- ined specifi cally the role of DHEA or testosterone aug- ing or speeding up response to SSRIs [97] and found no mentation in the management of residual MDD symp- difference in outcome between patients treated with toms or in nonresistant depressed patients to enhance the SSRIs and placebo versus those treated with SSRIs and chances of achieving remission with initial treatment. inositol. The need to monitor the neuroendocrine effects of these treatments limits their use in depression.

144 Psychother Psychosom 2006;75:139–153 Fava /Rush

Downloaded by: Llumc.CP 11105 Loma Linda Univ. 151.112.124.130 - 3/13/2014 10:04:42 PM Folate and S-Adenosyl-Methionine pentin; blood dyscrasias with carbamazepine) [126, 127] Folate, in particular its active form methyltetrahydro- further limits the use of these drugs in treatment-resistant folate, and S-adenosyl-methionine (SAMe) are com- depression. pounds closely involved in the one carbon cycle and in methylation processes in the brain. These compounds Benzodiazepines/Hypnotics have been studied extensively in depression. Literature Benzodiazepine and sedative hypnotic augmentation suggests that they have antidepressant properties [111, in depression has been used primarily to reduce attrition 112]. An open trial of methylfolate (up to 30 mg/day) in during antidepressant treatment (partly by managing SSRI-refractory patients suggested its usefulness as an ad- side-effects such as insomnia and agitation) or to enhance junct [113] . A large randomized study of 127 patients who antidepressant response. Lormetazepam has been shown received either 500 g/day of folic acid or placebo (in ad- to be more effective than placebo in enhancing the anti- dition to 20 mg/day of fl uoxetine) revealed a signifi cantly depressant effect of maprotiline or nortriptyline [128] . greater improvement in the fl uoxetine plus folic acid Similarly, clonazepam augmentation of fl uoxetine was as- group, primarily among women [114]. A recent open sociated with nonsignifi cantly higher remission rates than study with SAMe [115] has also shown the usefulness of placebo augmentation in an 8-week acute trial with 80 this augmenting agent in SSRI nonresponders. A placebo- depressed patients [129] . A 4-week study on 190 patients controlled study of SAMe augmentation of imipramine with a history of depression and of effective and stable had previously shown an acceleration of the response treatment with SSRIs showed that augmentation with the [116]. The availability of SAMe over the counter and the nonbenzodiazepine hypnotic zolpidem was signifi cantly favorable side-effect profi le (most-common side-effects more effective than placebo in treating residual symp- are gastrointestinal symptoms and headaches) [115] have toms of insomnia [130]. A recent, large, double-blind made this agent a relatively popular augmentation agent study has shown signifi cantly higher remission rates at 8 among MDD patients with residual symptoms. However, weeks among patients treated with fl uoxetine plus the there are no controlled studies of SAMe or methyltetra- non-benzodiazepine hypnotic eszopiclone than for pa- hydrofolate augmentation in the management of residual tients treated with fl uoxetine plus placebo (42 versus MDD symptoms or in nonresistant depressed patients to 32.8% remission rates, respectively) [131] . Further stud- enhance the chances of achieving remission. ies are needed to evaluate the usefulness of hypnotic and benzodiazepine augmentation to enhance the chances of Anticonvulsants achieving and/or sustaining remission among nonresis- There are uncontrolled and anecdotal reports of the tant depressed patients, particularly given the fact that usefulness of anticonvulsant augmentation of antidepres- these drugs tend to be very well tolerated and to be wide- sants in major depression, with drugs such as gabapentin ly used, despite concerns about possible benzodiazepine [117], topiramate [118], carbamazepine [119–121], and dependence and abuse [132–134] . valproic acid [122]. In a study of 59 treatment-resistant depressed patients randomly assigned to the addition of either lithium or carbamazepine to ongoing antidepres- Combination Studies sant treatment, the therapeutic effi cacy of both strategies, assessed after 28 days, was not signifi cantly different While double-blind controlled studies of augmenta- [123]. Two small, placebo-controlled trials of lamotrigine tion strategies are scarce, there are even fewer double- vs. placebo as an adjunctive treatment to paroxetine [124] blind studies of combination strategies, refl ecting the or fl uoxetine [125] in treatment-resistant depression were need for further studies in this area. Since improvement suggestive of effi cacy, but neither study was defi nitive following the initiation of antidepressant combinations because of the inadequate sample size. Therefore, there tends to occur within 4–6 weeks, it is premature to decide are no adequately powered, prospective controlled stud- on the effi cacy of a combination before that time. Almost ies of anticonvulsant augmentation in the management all effi cacy studies of combination treatments have fo- of nonremission of MDD, nor as initial treatment with cused on short-term outcomes with patients who had ini- nonresistant depressed patients. In addition, the risk of tially not remitted and often not responded to an initial signifi cant side-effects (e.g., hepatotoxicity with valproic antidepressant monotherapy. Most combination strate- acid; hypersensitivity reactions with lamotrigine; seda- gies employ full doses of both antidepressant agents. Very tion and cognitive side-effects with topiramate and gaba- little is known about the minimum duration of the com-

Augmentation and Combination Psychother Psychosom 2006;75:139–153 145 Treatments for MDD Downloaded by: Llumc.CP 11105 Loma Linda Univ. 151.112.124.130 - 3/13/2014 10:04:42 PM bination trial for those who respond or remit. Common blocks the serotonergic 5-HT 2 and 5-HT 3 receptors. The practice entails maintaining the combination for 6–9 combination of mianserin with tricyclic antidepressants months following remission, followed by an attempt to at initiation of treatment has been shown to augment gradually discontinue one of the two antidepressants. therapeutic effi cacy in 40 depressed patients, with 77% Combination strategies for resistant depression entail the of the imipramine-plus-mianserin group being consid- use of two different antidepressants. Thus one cannot de- ered responders compared with 27% of the imipramine- termine whether the effi cacy achieved is due to the com- only group [144]. Another double-blind study (n = 31) has bination (perhaps due to the synergy) or to the simple shown higher response rates among patients treated with exposure to another antidepressant. Combination strate- fl uoxetine plus mianserin than among those treated with gies have also been used to speed up the onset of the an- fl uoxetine alone at initiation of treatment [58] . Mianse- tidepressant effect. rin augmentation has been reported anecdotally to be ef- fective in depressed patients unresponsive to TCA alone Bupropion plus SSRIs [145], and fl uoxetine nonresponders achieved greater ef- Bupropion in sustained release formulations (SR or fi cacy when mianserin was added to fl uoxetine than when XL) (100–300 mg once a day or 150 mg b.i.d.) com- placebo was added [146]. A more recent study found add- bined with SSRIs was the most popular combination ing mianserin in sertraline nonresponders had no advan- strategy chosen by 400 psychiatrists surveyed by Fred- tage over adding placebo [147], but the initial trial with man et al. [135]. Even so, the evidence for this combi- sertraline monotherapy was too brief and a dose increase nation is primarily based on anecdotal reports, case se- of sertraline was carried out 2 weeks prior to randomiza- ries, or small open trials that also suggest that this strat- tion, thereby confounding the results. All these studies egy is relatively well tolerated [136–139]. In a small, suggest the potential usefulness of combining mirtazap- nonrandomized, open-label trial, the combination of ine or mianserin with SSRIs when remission is not bupropion-SR and citalopram was more effective than achieved with antidepressant monotherapy or when re- switching to other medication in patients who had not sidual symptoms persist (especially insomnia and weight responded to either one of these two medications [140] . loss) or even to enhance therapeutic effi cacy as an initial Future studies need to systematically assess the effec- treatment. tiveness of this combination in MDD patients with re- sidual symptoms or the enhancement of initial thera- TCAs plus SSRIs peutic effi cacy. An early study using a historical control [148] sug- gested that a combination of TCA plus SSRI may produce Mirtazapine or Mianserin plus SSRIs a more rapid onset of action. A more recent, prospective Mirtazapine, a dual-action antidepressant, increases randomized trial [149] found remission rates were sig- both serotonergic and noradrenergic activity by blocking nifi cantly higher with desipramine plus fl uoxetine than the 2-adrenergic auto- and hetero-receptors and blocks with either drug alone. The results are consistent with the serotonergic 5-HT 2 and 5-HT 3 receptors. Mirtazapine uncontrolled fi ndings that desipramine and other TCAs (15–30 mg q.h.s.) combined with SSRIs has been report- are effective in combination with SSRIs in small cohorts ed to be helpful in an open study of nonresponders to of resistant patients [150, 151]. The effi cacy of TCAs SSRIs [141] , and to be more effective than placebo plus combined with SSRIs has been put into question by two SSRIs in a subsequent double-blind study among 20 studies that found that adding low-dose desipramine to SSRI-resistant depressed patients, although sedation and fl uoxetine was less effective than increasing the dose of weight gain were signifi cant issues among mirtazapine- fl uoxetine in patients unresponsive to 8 weeks of treat- treated patients [142]. A study in nonresistant depressed ment with fl uoxetine 20 mg/day [18, 19] . Since TCAs are patients [143] found a signifi cantly higher response rate substrates of the cytochrome P450 2D6 isoenzyme, TCA to the combination of paroxetine and mirtazapine than blood levels may rise when coadministered with SSRIs monotherapy with either drug alone, and a 64% response that inhibit this pathway, with the potential for cardiac rate after the switch to combination therapy for patients toxicity, anticholinergic side-effects, and orthostatic hy- not responding to monotherapy. potension. Low doses of TCAs (25–75 mg/day) are there- Mianserin hydrochloride, a tetracyclic antidepressant, fore typically used, and monitoring of TCA blood levels also enhances noradrenergic and serotonergic transmis- is necessary. sion by presynaptic 2 -adrenoreceptor antagonism and

146 Psychother Psychosom 2006;75:139–153 Fava /Rush

Downloaded by: Llumc.CP 11105 Loma Linda Univ. 151.112.124.130 - 3/13/2014 10:04:42 PM Reboxetine/Atomoxetine plus SSRIs mentation seems to accelerate response to SSRIs and per- Reboxetine and atomoxetine are norepinephrine reup- haps enhance their effi cacy, but its usefulness in SSRI take inhibitors (NRIs). Reboxetine is available in Europe nonresponders has been questioned. Buspirone augmen- for the treatment of depression; atomoxetine is marketed tation in SSRI nonresponders certainly merits further in the US for the treatment of attention-defi cit disorder. study, but evidence for enhancing remission rates with Three open trials of reboxetine, using doses up to 8 mg/ SSRI nonresponders is lacking. Benzodiazepine and non- day, have suggested the usefulness of combining this benzodiazepine hypnotics have clear value in treating re- agent with SSRIs in resistant depression [152–154] . Giv- sidual sleep disturbances in antidepressant-treated pa- en the fact that reboxetine is not available in the US, a tients. Placebo-controlled, 6- to 8-week trials indicate en- number of clinicians have been using atomoxetine in hanced remission rates with the benzodiazepine combination with SSRIs. An open trial suggests its effi - lormetazepam and the nonbenzodiazepine hypnotic cacy in antidepressant nonresponders [155] . Future stud- eszopiclone compared to placebo. Whether differential ies are needed to evaluate this off-label use of atomox- remission rates would be present after 12–14 weeks is etine. unknown, however. Dopaminergic drugs and psycho- stimulants have shown to help antidepressant nonre- Nefazodone/Trazodone plus SSRIs sponders in uncontrolled studies in unipolar depression, Only anecdotal reports or case series suggest effi cacy but their ability to enhance remission at initiation of for combining SSRIs with nefazodone [156] or trazodone treatment has not been tested yet in a controlled fashion. [157], which are antidepressants with signifi cant sero- Controlled studies have shown modafi nil augmentation tonin 5-HT 2 receptor antagonism. One controlled study to be helpful in treating residual symptoms in SSRI par- of nonresistant depressed patients has shown higher re- tial responders with excessive sleepiness and fatigue. sponse rates among patients treated with trazodone plus Modafi nil augmentation to enhance remission rates clear- fl uoxetine than among those receiving trazodone plus pla- ly merits further investigation, given a preliminary, posi- cebo [158] . Furthermore, a study on patients with resid- tive open study. The one-carbon cycle compounds folate, ual symptoms of insomnia taking fl uoxetine or bupropion methylfolate, and, in particular, SAMe are promising has shown greater effi cacy of trazodone than placebo in augmentation strategies to enhance remission rates and treating residual sleep disturbances [159]. Nefazodone help antidepressant nonresponders, but additional ade- augmentation of SSRIs has been associated with sero- quately powered studies are needed. Atypical antipsy- tonin syndrome [160] and nefazodone treatment with chotic and anticonvulsant augmentations, while promis- rare fatal cases of hepatotoxicity [161]. Trazodone aug- ing, clearly need adequately powered studies to assess mentation may be limited by the risk of sedation and or- their acute and longer-term effi cacy and safety in resistant thostatic hypotension [162]. depressed populations. Finally, there is no clear evidence yet favoring the adjunctive use of estrogen, opiates, and inositol. Summary of the Literature Review As for combination strategies, bupropion augmenta- tion, while widely used in practice, is supported by only Augmentation and combination strategies have been uncontrolled studies. Systematic, controlled prospective primarily used to manage treatment-resistant depression evaluations of its ability to enhance remission rates and and/or residual symptoms in MDD patients, and, in some to help antidepressant nonresponders are needed. Mir- cases, to enhance therapeutic effi cacy at the initiation of tazapine or mianserin augmentation of SSRIs is support- treatment in nonresistant patients. Most of these studies ed by controlled studies suggesting that they enhance re- are uncontrolled and open label. There is a clear shortage mission rates and help antidepressant nonresponders. of double-blind placebo-controlled studies. Further studies are needed for these two agents. TCA and Among the augmentation strategies, lithium and thy- NRI augmentations of SSRIs have modest data that in- roid hormone have shown robust effects with TCA non- dicate they may both enhance remission rates and help responders and, in the case of thyroid hormone augmen- antidepressant nonresponders. Finally, the evidence in tation, acceleration of response with TCAs. However, the favor of the use of nefazodone and trazodone augmenta- usefulness of these strategies in patients with inadequate tion is very limited. responses to the newer antidepressants and in enhancing Clinicians must choose among the different combina- remission rates remains to be established. Pindolol aug- tion or augmentation treatment options to increase the

Augmentation and Combination Psychother Psychosom 2006;75:139–153 147 Treatments for MDD Downloaded by: Llumc.CP 11105 Loma Linda Univ. 151.112.124.130 - 3/13/2014 10:04:42 PM chances of achieving sustained remission from depres- from monotherapy in the absence of concomitant aug- sion and resolution of residual symptoms. The ongoing mentation of pharmacological options targeting these NIMH-sponsored Sequenced Treatment Alternatives to symptoms. Finally, since crossover studies have clearly Relieve Depression (STAR*D) study [163, 164] is com- shown that the spectrum of effi cacy of antidepressant paring augmentation and combination treatment options treatments does not typically fully overlap and that de- for patients not benefi ting adequately from an initial pressed patients who do not respond to an antidepressant treatment with an SSRI. The results of STAR*D will may respond to the switch to another [165, 166], aug- hopefully inform and guide future treatment options for menting and combining antidepressants at treatment ini- psychiatrists and primary care physicians alike in the tiation may lead to synergy and to a broader spectrum of medication management of depressed patients who do action (i.e., a larger proportion of patients initially treat- not achieve remission with an initial adequately deliv- ed may at least respond if not remit). This novel approach ered antidepressant monotherapy. STAR*D does not, is clearly different from the sequential approach, which however, evaluate the use of augmentation or combina- has been recently outlined [167]. In the sequential ap- tion strategies at the outset of initial treatment to enhance proach, monotherapy is carried out in two different phas- the chances of remission. es of treatment (acute and residual) and the addition of a second pharmacological treatment substitutes the fi rst one, with the aims of improving the level of remission A Proposal for a Novel Approach to Improve and preventing relapse. With this novel approach, clini- Practice cians would not wait to combine drugs until one of the two alone is proven ineffective or in a second phase of Given the relatively low remission rates with antide- therapy, but instead they would use augmented or com- pressant monotherapy in practice [10], the use of aug- bined antidepressant treatments right from the start of mentation or combination strategies at the initiation of treatment. Moncrieff and Cohen [168] have recently sug- treatment to enhance the chances of remission seems gested that we need to rethink our models of antidepres- quite reasonable, at least from a theoretical standpoint. sant drug action. We also need to rethink whether mono- Although in practice clinicians typically augment and therapy with antidepressants is still a justifi ed fi rst-line combine antidepressants only after there is incomplete treatment for MDD or whether a more aggressive treat- response to monotherapy, there is enough preliminary ment from the outset is justifi ed. Given the strong ratio- evidence from controlled studies that combination [58, nale and the preliminary support from the literature, a 144, 150] and augmentation [50, 52, 55, 57, 58, 128, 131] systematic evaluation of this novel approach is clearly strategies used at the beginning of antidepressant therapy called for. may indeed increase the rates of remission compared to antidepressant monotherapy. Therefore, despite the fact that current treatment guidelines recommend a single Acknowledgment medication to initially treat MDD, the use of augmented or combined antidepressant treatment at treatment ini- This work has been funded in whole or in part with Federal funds from the National Institute of Mental Health, National In- tiation could potentially improve practice and treatment stitutes of Health, under Contract N01MH90003 (STAR*D) and outcome, since most patients with MDD do not remit GMO N01MH90003 (NIMH Depression Trials Network). The with initial antidepressant monotherapy and no initial content of this publication does not necessarily refl ect the views or monotherapy medication is robustly different from oth- policies of the Department of Health and Human Services, nor does ers in achieving remission [3]. In fact, this novel approach mention of trade names, commercial products, or organizations imply endorsement by the US Government. could potentially enhance retention and/or increase re- mission rates, since the lack of response with antidepres- sant monotherapy may lead many depressed patients Disclosures with little or no benefi t to drop out of treatment, thus precluding the subsequent use of augmentation or com- Dr. Fava has received research support from Abbott Laborato- bination strategies altogether. In addition, the emergence ries, Lichtwer Pharma GmbH, Lorex Pharmaceuticals. He has re- ceived speaking/consulting honoraria from Bayer AG, BioVail, of certain side-effects (e.g., agitation, insomnia) or the Brain Cells, Compellis, Cypress Pharmaceuticals, Dov Pharmaceu- persistence of some initial baseline symptoms (e.g., anxi- ticals, Grunenthal GmbH, Janssen Pharmaceutica, Knoll Phar- ety, insomnia) may lead to premature discontinuation maceutical Company, Lundbeck, Sepracor, and Somerset Pharma-

148 Psychother Psychosom 2006;75:139–153 Fava /Rush

Downloaded by: Llumc.CP 11105 Loma Linda Univ. 151.112.124.130 - 3/13/2014 10:04:42 PM ceuticals. In addition, Dr. Fava has received both research support Dr. Rush has provided scientifi c consultation to or served on and honoraria from Aspect Medical Systems, AstraZeneca, Bristol- Advisory Boards for Advanced Neuronetic Systems Inc., Bristol- Myers Squibb Company, Cephalon, Eli Lilly & Company, Forest Myers Squibb Company, Cyberonics Inc., Eli Lilly & Company, Pharmaceuticals Inc., GlaxoSmithKline, J & J Pharmaceuticals, Forest Pharmaceuticals Inc., GlaxoSmithKline, Merck Co. Inc., Novartis, Organon Inc., Pharmavite, Pfi zer Inc, Roche, Sanofi / Neuronetics, Organon Inc., and Wyeth-Ayerst Laboratories. He has Synthelabo, Solvay Pharmaceuticals, Inc., and Wyeth-Ayerst Lab- received speaker honoraria from Cyberonics Inc., Forest Pharma- oratories. ceuticals Inc., GlaxoSmithKline, and Merck Co. Inc.

References

1 Fava M: Diagnosis and defi nition of treat- 11 Crismon ML, Trivedi M, Pigott TA, Rush AJ, 20 Baumann P, Nil R, Souche A, Montaldi S, Baet- ment-resistant depression. Biol Psychiatry Hirschfel RM, Kahn DA, DeBattista C, Nelson tig D, Lambert S, Uehlinger C, Kasas A, Amey

2003; 53: 649–659. JC, Nierenberg AA, Sackeim HA, Thase ME: M, Jonzier-Perey M: A double-blind, placebo- 2 Fava M, Davidson KG: Defi nition and epide- The Texas medication algorithm project: re- controlled study of citalopram with and without miology of treatment-resistant depression. port of the Texas consensus conference panel lithium in the treatment of therapy-resistant de-

Psychiatr Clin North Am 1996; 19: 179–200. on medication treatment of major depressive pressive patients: a clinical, pharmacokinetic,

3 Thase ME, Entsuah AR, Rudolph RL: Remis- disorder. J Clin Psychiatry 1999; 60: 142–156. and pharmacogenetic investigation. J Clin Psy-

sion rates during treatment with venlafaxine or 12 Quitkin FM, McGrath PJ, Stewart JW, Oce- chopharmacol 1996; 16: 307–314. selective serotonin reuptake inhibitors. Br J pek-Welikson K, Taylor BP, Nunes E, Deliyan- 21 Cournoyer G, De Montigny C, Oulette J, et al:

Psychiatry 2001; 178: 234–241. nides D, Agosti V, Donovan SJ, Petkova E, Lithium addition in tricyclic-resistant unipolar 4 Nierenberg AA, Keefe BR, Leslie VC, Alpert Klein DF: Chronological milestones to guide depression: a placebo-controlled study. Pre- JE, Pava JA, Worthington JJ 3rd, Rosenbaum drug change. When should clinicians switch sented at the XIVth Congress of the Collegium

JF, Fava M: Residual symptoms in depressed antidepressants? Arch Gen Psychiatry 1996; Internationale Neuropsychopharmacologicum,

patients who respond acutely to fl uoxetine. J 53: 785–792. Florence, Italy, June 19–23, 1984.

Clin Psychiatry 1999; 60: 221–225. 13 Nierenberg AA, McLean NE, Alpert JE, 22 de Montigny C, Cournoyer G, Morissette R, 5 Fava M, Kaji J: Continuation and mainte- Worthington JJ, Rosenbaum JF, Fava M: Ear- Langlois R, Caille G: Lithium carbonate addi- nance treatments of major depressive disorder. ly nonresponse to fl uoxetine as a predictor of tion in tricyclic antidepressant-resistant unipo-

Psychiatr Ann 1994; 24: 281–290. poor 8-week outcome. Am J Psychiatry 1995; lar depression. Correlations with the neurobio-

6 Paykel ES, Ramana R, Cooper Z, Hayhurst H, 152: 1500–1503. logic actions of tricyclic antidepressant drugs Kerr J, Barocka A: Residual symptoms after 14 Anonymous: Paroxetine: a selective serotonin and lithium ion on the serotonin system. Arch

partial remission: an important outcome in de- reuptake inhibitor showing better tolerance, Gen Psychiatry 1983; 40: 1327–1334.

pression. Psychol Med 1995; 25: 1171–1180. but weaker antidepressant effect than clomip- 23 Heninger GR, Charney DS, Sternberg DE: 7 Ramana R, Paykel ES, Cooper Z, Hayhurst H, ramine in a controlled multicenter study. Dan- Lithium carbonate augmentation of antide- Saxty M, Surtees PG: Remission and relapse ish University Antidepressant Group. J Affect pressant action: An effective prescription for

in major depression: a two-year prospective Disord 1990; 18: 289–299. treatment-refractory depression. Arch Gen

follow-up study. Psychol Med 1995; 25: 1161– 15 Fava GA, Rafanelli C, Grandi S, Conti S, Bel- Psychiatry 1983; 40: 1335–1342. 1170. luardo P: Prevention of recurrent depression 24 Joffe RT, Schuller DR: An open study of bus- 8 Judd LL, Paulus MJ, Schettler PJ, Akiskal HS, with cognitive therapy: preliminary fi ndings. pirone augmentation of serotonin reuptake in-

Endicott J, Leon AC, Maser JD, Mueller T, Arch Gen Psychiatry 1998; 55: 816–820. hibitors in refractory depression. J Clin Psy-

Solomon DA, Keller MB: Does incomplete re- 16 Keller MB, McCullough JP, Klein DN, Arnow chiatry 1993; 54: 269–271. covery from fi rst lifetime major depressive ep- B, Dunner DL, Gelenberg AJ, Markowitz JC, 25 Kantor D, McNeven S, Leichner P, Harper D, isode herald a chronic course of illness? Am J Nemeroff CB, Russell JM, Thase ME, Trivedi Krenn M: The benefi t of lithium carbonate ad-

Psychiatry 2000; 157: 1501–1504. MH, Zajecka J: A comparison of nefazodone, junct in refractory depression: fact or fi ction?

9 Wells K, Sherbourne C, Schoenbaum M, the cognitive behavioral-analysis system of Can J Psychiatry 1996; 31: 416–418. Ettner S, Duan N, Miranda J, Unutzer J, Ru- psychotherapy, and their combination for the 26 Katona CL, Abou-Saleh M, Harrison DA, Nai- benstein L: Five-year impact of quality im- treatment of chronic depression. N Engl J Med rac BA, Edwards DR, Lock T, Burns RA, Rob-

provement for depression: results of a group- 2000; 342: 1462–1470. ertson MM: Placebo-controlled trial of lithium level randomized controlled trial. Arch Gen 17 Fava M: Augmentation and combination strat- augmentation of fl uoxetine and lofepramine.

Psychiatry 2004; 61: 378–386. egies in treatment-resistant depression. J Clin Br J Psychiatry 1995; 166: 80–86.

10 Unützer J, Katon W, Callahan CM, Williams Psychiatry 2001; 62(suppl 18):4–11. 27 Schopf J, Baumann P, Lemarchand T, Rey M: JW, Hunkeler E, Harpole L, Hoffi ng M, Della 18 Fava M, Alpert J, Nierenberg A, Lagomasino Treatment of endogenous depressions resis- Penna RD, Noel PH, Lin EH, Arean PA, I, Sonawalla S, Tedlow J, Worthington J, Baer tant to tricyclic antidepressants or related Hegel MT, Tang L, Belin TR, Oishi S, Langs- L, Rosenbaum JF: Double-blind study of high- drugs by lithium addition. Pharmacopsychia-

ton C. IMPACT Investigators: Improving dose fl uoxetine versus lithium or desipramine try 1989; 22: 183–187. mood-promoting access to collaborative treat- augmentation of fl uoxetine in partial respond- 28 Stein G, Bernadt M: Double-blind trial of lith- ment: collaborative care management of late- ers and nonresponders to fl uoxetine. J Clin ium carbonate in tricyclic-resistant depression;

life depression in the primary care setting: a Psychopharmacol 2002; 22: 379–387. in NJ Birch (ed): Lithium: Inorganic Pharma-

randomized controlled trial. JAMA 2002; 288: 19 Fava M, Rosenbaum JF, McGrath PJ, Stewart cology and Psychiatric Use. Oxford, IRL Press,

2836–2845. JW, Amsterdam JD, Quitkin FM: Lithium and 1988, p 35. tricyclic augmentation of fl uoxetine treatment 29 Stein G, Bernadt M: Lithium augmentation for resistant major depression: a double-blind, therapy in tricyclic-resistant depression: a con-

controlled study. Am J Psychiatry1994; 151: trolled trial using lithium in low and normal

1372–1374. doses. Br J Psychiatry 1993; 162: 634–640.

Augmentation and Combination Psychother Psychosom 2006;75:139–153 149 Treatments for MDD Downloaded by: Llumc.CP 11105 Loma Linda Univ. 151.112.124.130 - 3/13/2014 10:04:42 PM 30 Zusky PM, Biederman J, Rosenbaum JF, Man- 42 Dimitriou EC, Dimitriou CE: Buspirone aug- 56 Berman RM, Anand A, Cappiello A, Miller schreck TC, Gross CC, Weilberg JB, Gast- mentation of antidepressant therapy. J Clin HL, Hu XS, Oren DA, Charney DS: The use

friend DR: Adjunct low dose lithium carbon- Psychopharmacol 1998; 18: 465–469. of pindolol with fl uoxetine in the treatment of ate in treatment-resistant depression: a 43 Jacobsen FM: A possible augmentation of an- major depression: fi nal results from a double- placebo-controlled study. J Clin Psychophar- tidepressant response by buspirone. J Clin Psy- blind, placebo-controlled trial. Biol Psychiatry

macol 1988; 8: 120–124. chiatry 1991; 52: 217–220. 1999; 45: 1170–1177. 31 Nierenberg AA, Papakostas GI, Petersen T, 44 Landen M, Bjorling GB, Agren H, Fahlen T: A 57 Sacristan JA, Gilaberte I, Boto B, Buesching Montoya HD, Worthington JJ, Tedlow J, Al- randomized, double-blind, placebo-controlled DP, Obenchain RL, Demitrack M, Perez Sola pert JE, Fava M: Lithium augmentation of nor- trial of buspirone in combination with an SSRI V, Alvarez E, Artigas F: Cost-effectiveness of triptyline for subjects resistant to multiple an- in patients with treatment-refractory depres- fl uoxetine plus pindolol in patients with major

tidepressants. J Clin Psychopharmacol 2003; sion. J Clin Psychiatry 1998; 59: 664–668. depressive disorder: results from a random-

23: 92–95. 45 Appelberg BG, Syvalahti EK, Koskinen TE, ized, double-blind clinical trial. Int Clin Psy-

32 Januel D, Poirier MF, D’alche-Biree F, Dib M, Mehtonen OP, Muhonen TT, Naukkarinen chopharmacol 2000; 15: 107–113. Olie JP: Multicenter double-blind randomized HH: Patients with severe depression may ben- 58 Maes M, Libbrecht I, van Hunsel F, Campens parallel-group clinical trial of effi cacy of the efi t from buspirone augmentation of selective D, Meltzer HY: Pindolol and mianserin aug- combination clomipramine (150 mg/day) plus serotonin reuptake inhibitors: results from a ment the antidepressant activity of fl uoxetine lithium carbonate (750 mg/day) versus clomip- placebo-controlled, randomized, double-blind, in hospitalized major depressed patients, in-

ramine (150 mg/day) plus placebo in the treat- placebo wash-in study. J Clin Psychiatry 2001; cluding those with treatment resistance. J Clin

ment of unipolar major depression. J Affect 62: 448–452. Psychopharmacol 1999; 19: 177–182.

Disord 2003; 76: 191–200. 46 Onder E, Tural U: Faster response in depres- 59 Moreno F, Gelenberg AJ, Bachar K, Delgado 33 Bauer M, Forsthoff A, Baethge C, Adli M, sive patients treated with fl uoxetine alone than P: Pindolol augmentation of treatment-resis- Berghofer A, Dopfmer S, Bschor T: Lithium in combination with buspirone. J Affect Dis- tant depressed patients. J Clin Psychiatry

augmentation therapy in refractory depres- ord 2003; 76: 223–227. 1997; 58: 437–439. sion-update 2002. Eur Arch Psychiatry Clin 47 Cohn JB, Rickels K: A pooled, double-blind 60 Perez V, Soler J, Puigdemont D, Alvarez E,

Neurosci 2003; 253: 132–139. comparison of the effects of buspirone, diaze- Artigas F: A double-blind, randomized, place- 34 Joffe RT, Singer W: A comparison of triiodo- pam and placebo in women with chronic anxi- bo-controlled trial of pindolol augmentation in

thyronine and thyroxine in the potentiation of ety. Curr Med Res Opin 1989; 11: 304–320. depressive patients resistant to serotonin reup-

tricyclic antidepressants. Psychiatry Res 1990; 48 Fava M, Alpert JE, Carmin CN, Wisniewski take inhibitors. Arch Gen Psychiatry 1999; 56:

32: 241–251. SR, Trivedi MH, Biggs MM, Shores-Wilson K, 375–379. 35 Aronson R, Offman HJ, Joffe T, Naylor CD: Morgan D, Schwartz T, Balasubramani GK, 61 Perry EB, Berman RM, Sanacora G, Anand A, Triiodothyronine augmentation in the treat- Rush AJ: Clinical correlates and symptom pat- Lynch-Colonese K, Charney DS: Pindolol aug- ment of refractory depression: A meta-analy- terns of anxious depression among patients mentation in depressed patients resistant to

sis. Arch Gen Psychiatry 1996; 53: 842–848. with major depressive disorder in STAR*D. selective serotonin reuptake inhibitors: a dou-

36 Agid O, Lerer B: Algorithm-based treatment of Psychol Med 2004; 34: 1299–1308. ble-blind, randomized, controlled trial. J Clin

major depression in an outpatient clinic: clini- 49 Martinez D, Broft A, Laruelle M: Pindolol aug- Psychiatry 2004; 65: 238–243. cal correlates of response to a specifi c serotonin mentation of antidepressant treatment: recent 62 Fava M, Kendler KS: Major depressive disor-

reuptake inhibitor and to triiodothyronine contributions from brain imaging studies. Biol der. Neuron 2000; 28: 335–341.

augmentation. Int J Neuropsychopharmacol Psychiatry 2000; 48: 844–853. 63 Bouckoms A, Mangini LP: An antidepressant

2003; 6: 41–49. 50 Perez V, Gilaberte I, Faries D, Alvarez E, Ar- adjuvant for mood disorders? Psychopharma-

37 Clayton AH, Shen C: T4 augmentation in tigas F: Randomised, double-blind, placebo- col Bull 1993; 29: 207–211. partial responders with major depression. controlled trial of pindolol in combination 64 Stryjer R, Strous RD, Shaked G, Bar F, Feld- NCDEU Annu Meet, Boca Raton, Fla., USA, with fl uoxetine antidepressant treatment. Lan- man B, Kotler M, Polak L, Rosenzcwaig S,

June 2002. cet 1997; 349: 1594–1597. Weizman A: Amantadine as augmentation 38 Iosifescu DV, Nierenberg AA, Mischoulon D, 51 Tome MB, Isaac MT, Harte R, Holland C: Par- therapy in the management of treatment-resis- Perlis RH, Papakostas GI, Ryan JL, Alpert JE, oxetine and pindolol: a randomized trial of se- tant depression. Int Clin Psychopharmacol

Fava M: An open study of triiodothyronine rotoninergic autoreceptor blockade in the re- 2003; 18: 93–96. augmentation of selective serotonin reuptake duction of antidepressant latency. Int Clin 65 Rogoz Z, Dziedzicka-Wasylewska M, Daniel

inhibitors in treatment-resistant major depres- Psychopharmacol 1997; 12: 81–89. WA, Wojcikowski J, Dudek D, Wrobel A, Zie-

sive disorder. J Clin Psychiatry 2005; 66: 1038– 52 Zanardi R, Artigas F, Franchini L, Sforzini L, ba A: Effects of joint administration of imipra- 1042. Gasperini M, Smeraldi E, Perez J: How long mine and amantadine in patients with drug- 39 Appelhof BC, Brouwer JP, Van Dyck R, Fliers should pindolol be associated with paroxetine resistant unipolar depression. Pol J Pharmacol

E, Hoogendijk WJ, Huyser J, Schene AH, Tijs- to improve the antidepressant response? J Clin 2004; 6: 735–742.

sen JG, Wiersinga WM: Triiodothyronine ad- Psychopharmacol 1997; 17: 446–450. 66 Sporn J, Ghaemi SN, Sambur MR, Rankin dition to paroxetine in the treatment of major 53 Bordet R, Thomas P, Dupuis B: Effect of pin- MA, Recht J, Sachs GS, Rosenbaum JF, Fava depressive disorder. J Clin Endocrinol Metab dolol on onset of action of paroxetine in the M: Pramipexole augmentation in the treat-

2004; 89: 6271–6276. treatment of major depression: intermediate ment of unipolar and bipolar depression: a ret- 40 Altshuler LL, Bauer M, Frye MA, Gitlin MJ, analysis of a double-blind, placebo-controlled rospective chart review. Ann Clin Psychiatry

Mintz J, Szuba MP, Leight KL, Whybrow PC: trial. Am J Psychiatry 1998; 155: 1346–1351. 2000; 12: 137–140. Does thyroid supplementation accelerate tricy- 54 Blier P, Bergeron R: The use of pindolol to po- 67 Lattanzi L, Dell’Osso L, Cassano P, Pini S, Ruc- clic antidepressant response? A review and tentiate antidepressant medication. J Clin Psy- ci P, Houck PR, Gemignani A, Battistini G,

meta-analysis of the literature. Am J Psychia- chiatry 1998; 59(suppl 5):16–23. Bassi A, Abelli M, Cassano GB: Pramipexole in

try 2001; 158: 1617–1622. 55 Isaac MT, Isaac MB, Gallo F, Tournoux A: treatment-resistant depression: a 16-week natu-

41 Bouwer C, Stein DJ: Buspirone is an effective Milnacipran and pindolol: a randomized trial ralistic study. Bipolar Disord 2002; 4: 307–314. augmenting agent of serotonin selective reup- of reduction of antidepressant latency. Hum 68 Cassano P, Lattanzi L, Soldani F, Navari S,

take inhibitors in severe treatment-refractory Psychopharmacol 2003; 18: 595–601. Battistini G, Gemignani A, Cassano GB:

depression. South Afr Med J 1997; 87(suppl Pramipexole in treatment-resistant depres- 4):534–537, 540. sion: an extended follow-up. Depress Anxiety

2004; 20: 131–138.

150 Psychother Psychosom 2006;75:139–153 Fava /Rush

Downloaded by: Llumc.CP 11105 Loma Linda Univ. 151.112.124.130 - 3/13/2014 10:04:42 PM 69 Perugi G, Toni C, Ruffolo G, Frare F, Akiskal 83 Ninan PT, Hassman HA, Glass SJ, McManus 95 Cheng-Shannon J, McGough JJ, Pataki C, H: Adjunctive dopamine agonists in treat- FC: Adjunctive modafi nil at initiation of treat- McCracken JT: Second-generation antipsy- ment-resistant bipolar II depression: an open ment with a selective serotonin reuptake in- chotic medications in children and adoles-

case series. Pharmacopsychiatry 2001; 34: 137– hibitor enhances the degree and onset of thera- cents. J Child Adolesc Psychopharmacol

141. peutic effects in patients with major depressive 2004; 14: 372–394.

70 Zarate CA, Payne JL, Singh J, Quiroz JA, disorder and fatigue. J Clin Psychiatry 2004; 96 Nemets B, Mishory A, Levine J, Belmaker

Luckenbaugh DA, Denicoff KD, Charney DS, 65: 414–420. RH: Inositol addition does not improve de- Manji HK: Pramipexole for bipolar II depres- 84 Ostroff RB, Nelson JC: Risperidone augmen- pression in SSRI treatment failures. J Neural

sion: a placebo-controlled proof of concept tation of selective serotonin reuptake inhibi- Transm 1999; 106: 795–798.

study. Biol Psychiatry 2004; 56: 54–60. tors in major depression. J Clin Psychiatry 97 Levine J, Mishori A, Susnosky M, Martin M,

71 Fawcett J, Kravitz HM, Zajecka JM, Schaff 1999; 60: 256–259. Belmaker RH: Combination of inositol and MR: CNS stimulant potentiation of mono- 85 Shelton RC, Tollefson GD, Tohen M, Stahl S, serotonin reuptake inhibitors in the treat-

amine oxidase inhibitors in treatment-refrac- Gannon KS, Jacobs TG, Buras WR, Bymaster ment of depression. Biol Psychiatry 1999; 45:

tory depression. J Clin Psychopharmacol 1991; FP, Zhang W, Spencer KA, Feldman PD, Melt- 270–273.

11: 127–132. zer HY: A novel augmentation strategy for 98 Stoll AL, Rueter S: Treatment augmentation 72 Feighner JP, Herbstein J, Damlouji N: Com- treating resistant major depression. Am J Psy- with opiates in severe and refractory major

bined MAOI, TCA, and direct stimulant ther- chiatry 2001; 158: 131–134. depression. Am J Psychiatry 1999; 156: 2017. apy of treatment-resistant depression. J Clin 86 Adson DE, Kushner MG, Eiben KM, Schulz 99 Bodkin JA, Zornberg GL, Lukas SE, Cole JO:

Psychiatry 1985; 46: 206–209. SC: Preliminary experience with adjunctive Buprenorphine treatment of refractory de-

73 Linet LS: Treatment of a refractory depression quetiapine in patients receiving selective sero- pression. J Clin Psychopharmacol 1995; 15:

with a combination of fl uoxetine and d-am- tonin reuptake inhibitors. Depress Anxiety 49–57.

phetamine. Am J Psychiatry 1989; 146: 803– 2004; 19: 121–126. 100 Stahl SM: Basic psychopharmacology of anti- 804. 87 Worthington JJ 3rd, Kinrys G, Wygant LE, depressants. Part 2: estrogen as an adjunct to 74 Masand PS, Anand VS, Tanquary JF: Psycho- Pollack MH: Aripiprazole as an augmentor of antidepressant treatment. J Clin Psychiatry

stimulant augmentation of second generation selective serotonin reuptake inhibitors in de- 1998; 59(suppl 4):15–24. antidepressants: a case series. Depress Anxiety pression and anxiety disorder patients. Int Clin 101 Shapira B, Oppenheim G, Zohar J, Segal M,

1998; 7: 89–91. Psychopharmacol 2005; 20: 9–11. Malach D, Belmaker RH: Lack of effi cacy of 75 Metz A, Shader RI: Combination of fl uoxetine 88 Papakostas GI, Petersen TJ, Kinrys G, Burns estrogen supplementation to imipramine in with pemoline in the treatment of major de- AM, Worthington JJ, Alpert JE, Fava M, Nie- resistant female depressives. Biol Psychiatry

pressive disorder. Int Clin Psychopharmacol renberg AA: Aripiprazole augmentation of se- 1985; 20: 576–579.

1991; 6: 93–96. lective serotonin reuptake inhibitors for treat- 102 Zohar J, Shapira B, Oppenheim G, Ayd FJ, 76 Stoll AL, Pillay SS, Diamond L, Workum SB, ment-resistant major depressive disorder. J Belmaker RH: Addition of estrogen to imip-

Cole JO: Methylphenidate augmentation of se- Clin Psychiatry 2005; 66: 1326–1330. ramine in female-resistant depressives. Psy-

rotonin selective reuptake inhibitors: a case se- 89 Papakostas GI, Petersen TJ, Nierenberg AA, chopharmacol Bull 1985; 21: 705–706.

ries. J Clin Psychiatry 1996; 57: 72–76. Murakami JL, Alpert JE, Rosenbaum JF, Fava 103 Amsterdam J, Garcia-Espana F, Fawcett J, 77 Wharton RN, Perel JM, Dayton PG, Malitz S: M: Ziprasidone augmentation of selective se- Quitkin F, Reimherr F, Rosenbaum J, Beas- A potential clinical use for methylphenidate rotonin reuptake inhibitors (SSRIs) for SSRI- ley C: Fluoxetine effi cacy in menopausal with tricyclic antidepressants. Am J Psychiatry resistant major depressive disorder. J Clin Psy- women with and without estrogen replace-

1971; 127: 1619–1625. chiatry 2004; 65: 217–221. ment. J Affect Disord 1999; 55: 11–17. 78 Lavretsky H, Kim MD, Kumar A, Reynolds 90 Dunner DL, Amsterdam JD, Shelton RC, Ro- 104 Rasgon NL, Altshuler LL, Fairbanks LA, CF: Combined treatment with methylpheni- mano SJ, Loebel A: Adjunctive ziprasidone in Dunkin JJ, Davtyan C, Elman S, Rapkin AJ: date and citalopram for accelerated response treatment-resistant depression: randomized, Estrogen replacement therapy in the treat- in the elderly: an open trial. J Clin Psychiatry double-blind, 8-week pilot study. 44th Annu ment of major depressive disorder in peri-

2003; 64: 1410–1414. Meet Am Coll Neuropsychopharmacol, San menopausal women. J Clin Psychiatry 2002;

79 Menza MA, Kaufman KR, Castellanos AM: Juan, P.R., USA, 2004. 63(suppl 7):45–48. Modafi nil augmentation of antidepressant 91 Nemeroff CB, Gharabawi GM, Canuso CM, 105 Schneider LS, Small GW, Hamilton SH, Bys- treatment in depression. J Clin Psychiatry Mahmoud R, Loescher A, Turkoz I, Rapaport tritsky A, Nemeroff CB, Meyers BS: Estrogen

2000; 61: 378–381. MH, Gharabawi GM: Augmentation with ris- replacement and response to fl uoxetine in a 80 DeBattista C, Solvason HB, Flores BH, et al: peridone in chronic resistant depression: A multicenter geriatric depression trial. Am J

Modafi nil as an adjunctive agent in the treat- double-blind placebo-controlled maintenance Geriatr Psychiatry 1997; 5: 97–106. ment of fatigue and hypersomnia associated trial. 44th Annu Meet Am Coll Neuropsycho- 106 Schneider LS, Small GW, Clary CM: Estro- with depression. 39th Annu Meet Am Coll pharmacol, San Juan, P.R., USA, 2004. gen replacement therapy and antidepressant Neuropsychopharmacol, San Juan, P.R., USA, 92 Dube S, Andersen S, Paul S: Meta-analysis response to sertraline in older depressed

2000. of olanzapine-fl uoxetine use in treatment- women. Am J Geriatr Psychiatry 2001; 9:

81 DeBattista C, Doghramji K, Menza MA, resistant depression (abstract P.1.021). J Eur 393–399.

Rosenthal MH, Fieve RR: Adjunct modafi nil Coll Neuropsychopharmacol 2002; 12(suppl 107 Stahl SM: Augmentation of antidepressants

for the short-term treatment of fatigue and 3):S182. by estrogen. Psychopharmacol Bull 1998b;34:

sleepiness in patients with major depressive 93 Casey DE: Dyslipidemia and atypical antipsy- 319–321.

disorder: a preliminary double-blind, placebo- chotic drugs. J Clin Psychiatry. 2004; 6(suppl 108 Wolkowitz OM, Reus VI, Keebler A, Nelson

controlled study. J Clin Psychiatry 2003; 64: 18):27–35. N, Friedland M, Brizendine L, Roberts E:

1057–1064. 94 Newcomer JW: Abnormalities of glucose me- Double-blind treatment of major depression 82 Fava M, Thase ME, DeBattista C: A multi- tabolism associated with atypical antipsychot- with dehydroepiandrosterone. Am J Psychia-

center, placebo-controlled study of modafi nil ic drugs. J Clin Psychiatry 2004; 65(suppl try 1999; 156: 646–649. augmentation in partial responders to selective 18):36–46. 109 Pope HG Jr, Cohane GH, Kanayama G, Sie- serotonin reuptake inhibitors with persistent gel AJ, Hudson JI: Testosterone gel supple-

fatigue and sleepiness. J Clin Psychiatry 2000; mentation for men with refractory depres-

66: 85–93. sion: a randomized, placebo-controlled trial.

Am J Psychiatry 2003; 160: 105–111.

Augmentation and Combination Psychother Psychosom 2006;75:139–153 151 Treatments for MDD Downloaded by: Llumc.CP 11105 Loma Linda Univ. 151.112.124.130 - 3/13/2014 10:04:42 PM 110 Orengo CA, Fullerton L, Kunik ME: Safety 124 Normann C, Hummel B, Scharer LO, Horn 138 Kennedy SH, McCann SM, Masellis M, Mc- and effi cacy of testosterone gel 1% augmenta- M, Grunze H, Walden J: Lamotrigine as ad- Intyre RS, Raskin J, McKay G, Baker GB: tion in depressed men with partial response junct to paroxetine in acute depression: a pla- Combining bupropion SR with venlafaxine, to antidepressant therapy. J Geriatr Psychia- cebo-controlled, double-blind study. J Clin paroxetine, or fl uoxetine: a preliminary re-

try Neurol 2005; 18: 20–24. Psychiatry 2002; 63: 337–344. port on pharmacokinetic, therapeutic, and 111 Alpert JE, Mischoulon D, Nierenberg AA, et 125 Barbosa L, Berk M, Vorster M: A double- sexual dysfunction effects. J Clin Psychiatry

al: Nutrition and depression: focus on folate. blind, randomized, placebo-controlled trial 2002; 63: 181–186.

Nutrition 2000; 16: 544–546. of augmentation with lamotrigine or placebo 139 DeBattista C, Solvason HB, Poirier J, Kend- 112 Spillmann M, Fava M: S-adenosylmethio- in patients concomitantly treated with fl uox- rick E, Schatzberg AF: A prospective trial of nine (Ademetionine) in psychiatric disorders: etine for resistant major depressive episodes. bupropion SR augmentation of partial and

historical perspective and current status. J Clin Psychiatry 2003; 64: 403–407. non-responders to serotonergic antidepres-

CNS Drugs 1996; 6: 416–425. 126 Wong IC, Lhatoo SD: Adverse reactions to sants. J Clin Psychopharmacol 2003; 23: 27–

113 Alpert JE, Mischoulon D, Rubenstein GE, new anticonvulsant drugs. Drug Saf 2000; 23: 30.

Bottonari K, Nierenberg AA, Fava M: Folin- 35–56. 140 Lam RW, Hossie H, Solomons K, Yatham ic acid (Leucovorin) as an adjunctive treat- 127 Cates M, Powers R: Concomitant rash and LN: Citalopram and bupropion-SR: combin- ment for SSRI-refractory depression. Ann blood dyscrasias in geriatric psychiatry pa- ing versus switching in patients with treat-

Clin Psychiatry 2002; 14: 33–38. tients treated with carbamazepine. Ann Phar- ment-resistant depression. J Clin Psychiatry

114 Coppen A, Bailey J: Enhancement of the an- macother 1998; 32: 884–887. 2004; 65: 337–340. tidepressant action of fl uoxetine by folic acid: 128 Nolen WA, Haffmans PM, Bouvy PF, Duiv- 141 Carpenter LL, Jocic Z, Hall JM, Rasmussen a randomised, placebo-controlled trial. J Af- envoorden HJ: Hypnotics as concurrent med- SA, Price LH: Mirtazapine augmentation in

fect Disord 2000; 60: 121–130. ication in depression. A placebo-controlled, the treatment of refractory depression. J Clin

115 Alpert JE, Papakostas G, Mischoulon D, double-blind comparison of fl unitrazepam Psychiatry 1999; 60: 45–49. Worthington JJ 3rd, Petersen T, Mahal Y, and lormetazepam in patients with major de- 142 Carpenter LL, Yasmin S, Price LH: A double- Burns A, Bottiglieri T, Nierenberg AA, Fava pression, treated with a (tri)cyclic antidepres- blind, placebo-controlled study of antidepres-

M: S-adenosyl-L-methionine (SAMe) as an sant. J Affect Disord 1993; 28: 179–188. sant augmentation with mirtazapine. Biol

adjunct for resistant major depressive disor- 129 Smith WT, Londborg PD, Glaudin V, Paint- Psychiatry 2002; 51: 183–188. der: an open trial following partial or nonre- er JR: Is extended clonazepam cotherapy of 143 Debonnel G, Gobbi G, Turcotte J, DeMon- sponse to selective serotonin reuptake inhibi- fl uoxetine effective for outpatients with ma- tigny C, Blier P: The alpha-2 antagonist mir-

tors or venlafaxine. J Clin Psychopharmacol jor depression? J Affect Disord 2002; 70: 251– tazapine combined with the SSRI paroxetine

2004; 24: 661–664. 259. induces a greater antidepressant response: a 116 Berlanga C, Ortega-Soto HA, Ontiveros M, 130 Asnis GM, Chakraburtty A, DuBoff EA, double-blind controlled study. 39th Annu Senties H: Effi cacy of S-adenosyl-L-methio- Krystal A, Londborg PD, Rosenberg R, Roth- Meet Am Coll Neuropsychopharmacol, San nine in speeding the onset of action of imip- Schechter B, Scharf MB, Walsh JK: Zolpidem Juan, P.R., USA, 2000.

ramine. Psychiatry Res 1992; 44: 257–262. for persistent insomnia in SSRI-treated de- 144 Lauritzen L, Clemmesen L, Klysner R,

117 Yasmin S, Carpenter LL, Leon Z, Siniscalchi pressed patients. J Clin Psychiatry 1999; 60: Loldrup D, Lunde M, Schaumburg E, Waarst

JM, Price LH: Adjunctive gabapentin in 668–676. S, Bech P: Combined treatment with imipra- treatment-resistant depression: a retrospec- 131 Fava M, Buysse DJ, Rubens R, Wessel T, mine and mianserin. A controlled pilot study.

tive chart review. J Affect Disord 2001; 63: Caron J, Roth T: Eszopiclone coadministered Pharmacopsychiatry 1992; 25: 182–186.

243–247. with fl uoxetine for insomnia-associated with 145 Medhus A, Heskestad S, Thue JF: Mianserin 118 Schmidt do Prado-Lima PA, Bacaltchuck J: major depressive disorder (MDD): Effects on added to tricyclic antidepressants in de- Topiramate in treatment-resistant depres- sleep and depression. 45th Annu NCDEU pressed patients not responding to a tricyclic sion and binge-eating disorder. Bipolar Dis- Meet, Boca Raton, Fla., USA, 2005. antidepressant alone. A randomized, place-

ord 2002; 4: 271–273. 132 Terzano MG, Rossi M, Palomba V, Smerieri bo-controlled, double blind study. Nord J

119 Otani K, Yasui N, Kaneko S, Ohkubo T, Osa- A, Parrino L: New drugs for insomnia: com- Psychiatry 1994; 48: 355–358. nai T, Sugawara K: Carbamazepine augmen- parative tolerability of zopiclone, zolpidem 146 Ferreri M, Lavergne F, Berlin I, Payan C, Pu-

tation therapy in three patients with trazo- and zaleplon. Drug Saf 2003; 26: 261–282. ech AJ: Benefi ts from mianserin augmenta- done-resistant unipolar depression. Int Clin 133 Roehrs T, Roth T: Hypnotics: an update. tion of fl uoxetine in patients with major de-

Psychopharmacol 1996; 11: 55–57. Curr Neurol Neurosci Rep 2003; 3: 181–184. pression non-responders to fl uoxetine alone.

120 Varney NR, Garvey MJ, Cook BL, Campbell 134 Moller HJ: Effectiveness and safety of benzo- Acta Psychiatr Scand 2001; 103: 66–72.

DA, Roberts RJ: Identifi cation of treatment- diazepines. J Clin Psychopharmacol 1999; 147 Licht RW, Qvitzau S: Treatment strategies in

resistant depressives who respond favorably 19(suppl 2):S2–S11. patients with major depression not respond-

to carbamazepine. Ann Clin Psychiatry 1993; 135 Fredman SJ, Fava M, Kienke AS, White CN, ing to fi rst-line sertraline treatment. A ran-

5: 117–122. Nierenberg AA, Rosenbaum JF: Partial re- domised study of extended duration of treat- 121 De la Fuente JM, Mendlewicz J: Carbamaze- sponse, non-response, and relapse on SSRIs ment, dose increase or mianserin augmen-

pine addition in tricyclic antidepressant-re- in major depression: a survey of current ‘next- tation. Psychopharmacology (Berl) 2002;

sistant unipolar depression. Biol Psychiatry step’ practices. J Clin Psychiatry 2000; 61: 161: 143–151.

1992; 32: 369–374. 403–407. 148 Nelson JC, Mazure CM, Bowers MB, Jatlow 122 Hantouche EG, Akiskal HS, Lancrenon S, 136 Bodkin JA, Lasser RA, Wines JD Jr, Gardner P: A preliminary open study of the combina- Chatenet-Duchene L: Mood stabilizer aug- DM, Baldessarini RJ: Combining serotonin tion of fl uoxetine and desipramine for rapid mentation in apparently ‘unipolar’ MDD: reuptake inhibitors and bupropion in partial treatment of major depression. Arch Gen

predictors of response in the naturalistic responders to antidepressant monotherapy. J Psychiatry 1991; 48: 303–307.

French national EPIDEP study. J Affect Dis- Clin Psychiatry 1997; 58: 137–145. 149 Nelson JC, Mazure CM, Jatlow PI, Bowers

ord 2005; 84: 243–249. 137 Ramasubbu R: Treatment of resistant de- MB Jr, Price LH: Combining norepinephrine 123 Rybakowski JK, Suwalska A, Chlopocka- pression by adding noradrenergic agents to and serotonin reuptake inhibition mecha- Wozniak M: Potentiation of antidepressants lithium augmentation of SSRIs. Ann Phar- nisms for treatment of depression: a double-

with lithium or carbamazepine in treatment- macother 2002; 36: 634–640. blind, randomized study. Biol Psychiatry

resistant depression. Neuropsychobiology 2004; 55: 296–300.

1999; 40: 134–139.

152 Psychother Psychosom 2006;75:139–153 Fava /Rush

Downloaded by: Llumc.CP 11105 Loma Linda Univ. 151.112.124.130 - 3/13/2014 10:04:42 PM 150 Weilburg JB, Rosenbaum JF, Meltzer-Brody 156 Taylor FB, Prather MR: The effi cacy of ne- 164 Rush AJ, Fava M, Wisniewski SR, Lavori S, Shushtari J: Tricyclic augmentation of fazodone augmentation for treatment-resis- PW, Trivedi MH, Sackeim HA, Thase ME,

fl uoxetine. Ann Clin Psychiatry 1991; 3: 209– tant depression with anxiety symptoms or Nierenberg AA, Quitkin FM, Kashner TM,

214. anxiety disorder. Depress Anxiety 2003; 18: Kupfer DJ, Rosenbaum JF, Alpert J, Stewart

151 Zajecka JM, Jeffries H, Fawcett J: The effi - 83–88. JW, McGrath PJ, Biggs MM, Shores-Wilson cacy of fl uoxetine combined with heterocyclic 157 Nierenberg AA, Cole JO, Glass L: Possible K, Lebowitz BD, Ritz L, Niederehe G; antidepressant in treatment-resistant depres- trazodone potentiation of fl uoxetine: a case STAR*D Investigators Group: Sequenced

sion: a retrospective analysis. J Clin Psychia- series. J Clin Psychiatry 1992; 53: 83–85. Treatment Alternatives to Relieve Depres-

try 1995; 56: 338–343. 158 Maes M, Vandoolaeghe E, Desnyder R: Effi - sion (STAR*D): rationale and design. Con-

152 Hawley CJ, Sivakumaran T, Ochocki M, Be- cacy of treatment with trazodone in combina- trol Clin Trials 2004; 25: 119–142. van J: Co-administration of reboxetine and tion with pindolol or fl uoxetine in major de- 165 Thase ME, Rush AJ, Howland RH, Korn-

serotonin selective reuptake inhibitors in pression. J Affect Disord 1996; 41: 201–210. stein SG, Kocsis JH, Gelenberg AJ, Schatz- treatment-resistant patients with major de- 159 Nierenberg AA, Adler LA, Peselow E, Zorn- berg AF, Koran LM, Keller MB, Russell JM, pression. 39th Annu Meet Am Coll Neuro- berg G, Rosenthal M: Trazodone for antide- Hirschfeld RM, La Vange LM, Klein DN, psychopharmacol, San Juan, P.R., USA, pressant-associated insomnia. Am J Psychia- Fawcett J, Harrison W: Double-blind switch

2000. try 1994; 151: 1069–1072. study of imipramine or sertraline treatment 153 Dursun SM, Devarajan S: The effi cacy and 160 Smith DL, Wenegrat BG: A case report of se- of antidepressant-resistant chronic depres-

safety of reboxetine plus citalopram in treat- rotonin syndrome associated with combined sion. Arch Gen Psychiatry 2002; 59: 233– ment-resistant depression – An open, natural- nefazodone and fl uoxetine. J Clin Psychiatry 239.

istic case series. 39th Annu Meet Am Coll 2000; 61: 146. 166 McGrath PJ, Stewart JW, Nunes EV, Oce- Neuropsychopharmacol, San Juan, P.R., 161 Stewart DE: Hepatic adverse reactions asso- pek-Welikson K, Rabkin JG, Quitkin FM, USA, 2000. ciated with nefazodone. Can J Psychiatry Klein DF: A double-blind crossover trial of

154 Lucca A, Serretti A, Smeraldi E: Effect of re- 2002; 47: 375–377. imipramine and phenelizine for outpatients boxetine augmentation in SSRI resistant pa- 162 Cassano P, Fava M: Tolerability issues during with treatment-refractory depression. Am J

tients. Hum Psychopharmacol 2000; 15: 143– long-term treatment with antidepressants. Psychiatry 1993; 150: 118–123.

145. Ann Clin Psychiatry 2004; 16: 15–25. 167 Fava GA, Ruini C: What is the optimal treat- 155 Milosavljevic N, Schecter JM, Price LH, Car- 163 Fava M, Rush AJ, Trivedi MH, Nierenberg ment of mood and anxiety disorders? Clin

penter LL: Antidepressant augmentation AA, Thase ME, Sackeim HA, Quitkin FM, Psychol Sci Pract 2005; 12: 92–96. with open-label atomoxetine. APA Annu Wisniewski S, Lavori PW, Rosenbaum JF, 168 Moncrieff J, Cohen D: Rethinking models of Meet, New York, N.Y., USA, 2004. Kupfer DJ; for the STAR*D Investigators psychotropic drug action. Psychother Psy-

Group: Background and rationale for the Se- chosom 2005; 74: 145–153. quenced Treatment Alternatives to Relieve Depression (STAR*D) study. Psych Clin

North Am 2003; 26: 1–38.

Augmentation and Combination Psychother Psychosom 2006;75:139–153 153 Treatments for MDD Downloaded by: Llumc.CP 11105 Loma Linda Univ. 151.112.124.130 - 3/13/2014 10:04:42 PM