Treating depression with adjunctive antipsychotics
Abstract
Objectives
To evaluate the efficacy and safety of using adjunctive antipsychotics in patients with major depressive disorder.
Method
Studies published since the last Cochrane review conducted in 2010 were identified via a literature search of recognised databases, using the keywords “adjunct*”, “augment*”, “antipsychotic” and “depression”, and systematically evaluated. A targeted review of relevant guidelines was undertaken.
Results
Adjunctive antipsychotics produce a small but significant improvement in depressive symptoms in most studies. Most of the studies focussed on patients with an inadequate response to antidepressants rather than patients with treatment resistant depression. Treatment guidelines were variable but generally supported the use of adjunctive antipsychotics while cautioning about the risk of side effects. Most were non-specific about the length of time adjunctive antipsychotics should be prescribed.
Conclusions
The studies do not support the routine use of adjunctive antipsychotics in patients with an inadequate response to antidepressants. They may be beneficial when used short-term in patients with treatment resistant depression who have specific symptoms (severe ruminations, melancholia, major sleep disturbance) that appear to respond well to adjunctive antipsychotics. There is no support for long-term use. Research should focus on specifying which symptom profiles are responsive and how adjunctive antipsychotics compare to other strategies in treatment resistant depression.
1 INTRODUCTION
Antipsychotics are a class of medication primarily used to treat psychosis. They form a heterogeneous group of molecules that act on a variety of neurotransmitters and neurotransmission systems, in particular those related to dopamine. The use of antipsychotics in treating major depression dates back to the discovery of antipsychotics in the 1950s. The first published trial used thioridazine to treat 29 patients with psychosis and reported that the subgroup which responded best consisted of five patients with psychotic depression.1 In 1982, Robertson and Trimble2 reviewed 34 double blind trials evaluating “major tranquilisers” used for treating depression. They reported that most studies were conducted on patients with mixed anxiety and depression and that, overall, the major tranquilisers “possess some antidepressant properties” (p. 1892). In addition, they noted that major tranquilisers appeared to work rapidly and cause relatively few side effects, while at the same time pointing out that the quality of the studies was generally poor and, importantly, there were no studies of longer term use.
Indeed, few robust studies were undertaken until Shelton et al3 began a revival by using atypical antipsychotics as an adjunctive treatment strategy in patients with treatment resistant, nonpsychotic, unipolar depression. This idea drew on observations at the time that the atypical antipsychotics were effective in treating psychotic depression.4 Shelton et al reported that a combination of fluoxetine and olanzapine was superior to either drug alone in their nonpsychotic depressed sample. In the first decade of the new millennium, a number of other RCTs reported statistically significant improvements in symptoms of depression compared with placebo when patients with treatment resistant depression were given a variety of adjunctive atypical antipsychotics and, by 2010, quetiapine, aripiprazole and the combination of olanzapine and fluoxetine (OFC) were approved by the US Food and Drug Administration (FDA) for adjunctive use in major depression. More recently, in 2015, the new atypical antipsychotic brexpiprazole was also approved as an adjunctive therapy, and gradually, over time, the use of adjunctive antipsychotics as a treatment strategy for treatment resistant depression has become increasingly accepted—and reflected in clinical practice guidelines.
1.1 Treatment resistant depression (TRD)
1.1.1 Defining and managing
Despite the many evidence based pharmacological treatments for depression, in real world practice many patients do not respond adequately to antidepressants and a significant proportion are unable to tolerate their side effects.5 These nonresponders are often described as having “refractory” or “treatment resistant” depression (TRD) but the terms are variably defined and even definitions focusing on medication cannot achieve consensus as to what constitutes TRD. In most of the studies that are reviewed it may be more accurate to consider that the patient had an inadequate response rather than treatment resistance per se (see Table 1 for definitions of treatment resistance used).
| Trial | Treatment comparison | Patients | Definition of TRD | Duration | Findings |
|---|---|---|---|---|---|
| Bauer et al15 | Adjunctive quetiapine XR 150 mg/d vs 300 mg/d vs placebo | 493 patients with MDD | History of inadequate response in current episode to ≥6 week antidepressant trial | 6 weeks | Quetiapine significantly improved MADRS scores (2.73-3.05 beyond placebo) |
| El-Khalili et al16 | Adjunctive quetiapine XR 150 mg/d vs 300 mg/d vs placebo | 446 patients with MDD | Continuing depressive symptoms after ≥6 week of antidepressant therapy | 6 weeks | Quetiapine significantly improved MADRS scores (2.4-3.55 beyond placebo) |
| Fava et al17 | Low dose adjunctive aripiprazole 2 mg/d vs adjunctive placebo | 225 patients with MDD | Less than a 50% reduction in depression scores in 1-3 antidepressant trials | 4 weeks | No significant effect of aripiprazole on MADRS score |
| Kamijima et al18 | Fixed (3 mg/d) vs flexible (3-15 mg/d) aripiprazole vs placebo | 586 patients with MDD | Inadequate response to 1-3 antidepressant trials of at least 6 weeks duration | 6 weeks | Aripiprazole significantly improved MADRS score (2.2-3.1 beyond placebo) |
| Thase et al19 | Adjunctive brexpiprazole 2 mg/d vs adjunctive placebo | 379 patients with treatment resistant MDD | Less than 50% reduction to an adequate trial of 1-3 antidepressants | 6 weeks | Brexpiprazole significantly improved MADRS score (3.12 beyond placebo) |
| Thase et al20 | Adjunctive brexpiprazole 1 mg/d vs 3 mg/d vs adjunctive placebo | 379 patients with treatment resistant MDD | Less than 50% reduction to an adequate trial of 1-3 antidepressants | 6 weeks | Brexpiprzole 3mg/d significantly improved MADRS scores (1.95 beyond placebo), but no significant difference for brexpiprazole 1 mg/d |
| Papakostas et al21 | Adjunctive ziprasidone 40-160 mg/d vs escitalopram monotherapy | 139 outpatients with persistent MDD symptoms | Persistence of symptoms (QIDS ˃10) after 8 week trial of escitalopram | 8 weeks | Ziprasidone significantly improved HAM-D scores (3.1 beyond placebo) |
| Durgram et al22 | Adjunctive cariprazine 1-2 mg/d vs adjunctive cariprazine 2-4.5 mg/d vs adjunctive placebo | 819 MDD patients with inadequate antidepressant response | Failure to respond adequately after 6 weeks of antidepressant treatment | 8 weeks | Caripiprazine 2-4.5 mg/d significantly improved MADRS scores (2.2 beyond placebo), but no significant difference for caripiprazine 1-2 mg/d |
Clinical studies in naturalistic settings consistently indicate that barely half of depressed patients respond to initial antidepressant monotherapy (the most common recommendation in practice guidelines) and that only one third eventually achieve satisfactory remission of symptoms (the desired goal in most guidelines).6 Thus, in practice, alternative treatment strategies often have to be employed to achieve suitable outcomes in this large group of patients. These include assessing the adequacy of dose and treatment adherence; reviewing the diagnosis, including possible psychiatric and medical co-morbidity; and considering chronic social difficulties. The next steps are to increase the dose, switch to another antidepressant (possibly from a different class), combine antidepressants or add an adjunctive therapy.7
1.1.2 Adjunctive therapy
The words adjunctive and augmentation are often used interchangeably, but have different meanings. Augmentation implies that the added medication increases the efficacy of the primary medication, while adjunctive is simply the addition of a treatment that has effects independent of the primary medication. There are also synergistic agents, which work together with the primary medication to increase the efficacy of both. With some noted exceptions, most usage of atypical antipsychotics in depression appears to be adjunctive.
The main options for adjunctive treatment are lithium, atypical antipsychotics and, to a lesser degree, thyroxine. Lithium is perhaps the most versatile in that it can be added to any of the antidepressants currently in use and probably produces benefit both independently (as an antidepressant) and as an augmentation agent that facilitates the primary antidepressant effect. Its onset of action as an augmentation strategy is usually relatively prompt and if a response is not observed within 10-14 days then it is likely that it will be found to be ineffective.8 Despite this broad indication and the promise of a quick response, its use has been surpassed by the administration of atypical antipsychotics—largely because lithium is viewed as complicated to prescribe with the potential for serious side effects. As Nelson recently pointed out,9, 10 the database for adjunctive trials using atypical antipsychotics in major depression is much larger than for any other treatment. As of 2015, there were 22 randomised controlled trials (RCTs) involving 5,531 patients. By comparison, lithium had only 10 trials with 271 patients.
The RCTs, along with the FDA approvals, have had a major impact on clinical practice. Pharmaco-epidemiological studies in North America have shown increased rates of adjunctive antipsychotic treatment in nonpsychotic major depression. For example, outpatient visits in the US where nonpsychotic major depression was diagnosed and adjunctive antipsychotic treatment was included rose from 4.6% (95% CI 2.9-6.3) in 1999-2000 to 12.5% (95% CI 9.7-15.3) in 2009-2010.11 In Canada, prescriptions for quetiapine per 100 Canadians rose from 3.2 in 2007 to 12 in 2012, and mood disorders were the most common diagnosis associated with quetiapine prescriptions.12, 13
This raises a key question: is the large increase in prescribing atypical antipsychotics for depression justified by the quality of the available data? And, perhaps more importantly, do the benefits of treatment outweigh the risks? A Cochrane review that partly addressed these questions was published in 2010.14 The present paper complements this review and provides a timely update. Specifically, it attempts to answer these questions by reviewing recent data concerning the reduction of depressive symptoms and considering the results in the context of the overall quality of life and general functioning of patients.
2 METHOD
Articles published since the Cochrane review and up to 1 March 2018 were identified via a literature search of recognised databases (PubMed, Scopus, PsycINFO, and Google Scholar) using the keywords “adjunct*”, “augment*”, “antipsychotic”, “depression” and excluding studies prior to 2009. To ensure inclusion of all available articles, reference lists of all relevant papers were also checked. Further, we identified and reviewed recently published guidelines which discuss the use of antipsychotics in MDD.
3 RESULTS
3.1 Efficacy data - Quality
The Cochrane review noted a number of issues with the available data. First, as discussed earlier, there is no agreed definition of TRD, with different studies having used varying definitions ranging from nonresponse to one antidepressant to those who have failed more than four antidepressants. In reality, most studies use patients who have not responded to one or two antidepressants, which is probably better regarded as inadequate response. Second, of the 28 RCTs reported in the Cochrane Database,14 only three gave information about the type of randomisation. Seven studies used identical capsules for blinding but no trials examined whether the blinding was effective. Given the significant adverse side effects of the tested drugs, the Cochrane authors felt that there was a high likelihood that patients and researchers would be unblinded. Third, the attrition rate in RCTs was moderate to high. In 10 studies, it was higher than 25%. Fourth, no study was judged to be free of selective reporting. Most trials had incomplete reporting of predefined outcomes and many only reported on adverse events if at least 5%-10% of participants had reported them. Finally, 21 of the studies were industry sponsored and four did not provide adequate information regarding funding. Hence, as noted by the Cochrane authors, “due to potential conflicts of interest a potential bias cannot be excluded.”14
Since the 2010 Cochrane Review, eight RCTs have compared adjunctive atypical antipsychotics to placebo (see Table 1). These recent studies were consistent with the Cochrane review, in that they showed a small but significant benefit from adjunctive atypical antipsychotics in most cases, and were subject to some of the same criticisms. The eight studies used different definitions of TRD, only three provided (partial) details of randomisation; six did not detail blinding procedures and none assessed the effectiveness of their blinding procedure; attrition rates were mixed (14%, 22.9%, 11%, 7.8%, 12.7%, 26.6%, 6.6% and 17.5%); and six studies failed to fully report side-effects. Seven of eight studies were funded by industry.
3.2 Efficacy data - Outcome
The Cochrane review authors were cautious about efficacy, concluding that the evidence of positive treatment effects of olanzapine and risperidone augmentation has to be considered as rather low, and that tolerability was worse in terms of weight gain and sedation (olanzapine) and prolactin (risperidone). Somewhat more data suggest a benefit of quetiapine and aripiprazole augmentation in some aspect of efficacy, but this again must be weighed against worse overall tolerability, such as more sedation (quetiapine) and more weight gain (quetiapine, aripiprazole) compared to single-treatment regimens with antidepressants.
The eight new RCTs reviewed mostly reported significant but small improvement in MADRS scores using adjunctive antipsychotics. The quetiapine studies reported a small but significant change beyond placebo (2.4-3.55 change in MADRS scores). The low dose aripiprazole study (2 mg/d) reported no significant effect, while the higher dose trial (3 mg-15 mg/d) reported a significant change (2.2-3.1 improvement in MADRS score). Similarly, adjunctive brexpiprazole significantly improved MADRS scores (1.95-3.12) as did ziprasidone (3.1) and cariprazine (2.2), all with small effect sizes.
A number of reviews and meta-analyses examining adjunctive antipsychotic treatment for major depression have been published with varying conclusions.23-26 All report that adjunctive atypical antipsychotics were more effective than adjunctive placebo in reducing measures of depressive symptoms and improving response and remission. All acknowledged that the effect sizes were small or moderate to small and that the drugs demonstrated substantial risks of several adverse side effects. However, it is important to note that where an atypical antipsychotic is added to an antidepressant and the effect is compared with the effects of an antidepressant alone, the potential effect size is likely to be smaller than when comparing the effects of an active agent to placebo—a supposedly inactive comparator. Furthermore, whether these small effect sizes were deemed to be clinically significant was inconsistent, partly because there is ongoing debate as to what is a clinically significant effect size in drug studies. The National Institute for Health and Clinical Excellence (NICE) in the United Kingdom deems an effect size of 0.5 clinically significant,27 while others claim 0.3 may be an appropriate minimum effect size.28 The two most recent comprehensive meta-analyses23, 24 give an overall effect size of –0.3424 and a range of –0.27 to −0.43.23 Spielmans et al24 interpreted their finding to be of questionable clinical relevance while Zhou et al23 did not comment. Another way of quantifying the relative effects is to express the findings as number needed to treat (NNT). The best estimates of the NNT for remission were nine for aripiprazole, quetiapine and risperidone and 19 for OFC.24
Returning to the depression severity scores, it is important to consider what these differences in depression severity scores actually measure. Spielmans et al24 reported that change on the MADRS score was the most commonly reported outcome. The MADRS consists of 10 items each rated on a 0-6 scale giving a potential total of 60 points. The pooled difference for placebo vs adjunctive antipsychotics in mean change in MADRS score was 2.69 points in the Spielmans et al24 meta-analysis. Because of the sedating effects of most atypical antipsychotics, they speculated that it was possible the effect may be confined to a few symptoms rather than the overall depression score on the MADRS. While it is rare for trials to publish individual MADRS item scores, the pooled analysis of two quetiapine RCTs did so.29 From their figures, it appears that the most prominent effect of quetiapine at 300 mg and 150 mg was on the MADRS item “reduced sleep,” both at 1 week and 6 weeks. The effects on sadness, inner tension, pessimistic thoughts and suicidal thoughts were less marked and effects on other MADRS items were not significant. In contrast, the pooled figures for adjunctive brexpiprazole reported that the medication had most effects on sadness, inner tension, lassitude, inability to feel and pessimistic thoughts and no influence on sleep.30
On other outcome measures such as quality of life or functioning, atypical antipsychotic adjunctive therapy has been shown to have minimal or no benefits. Spielmans et al24 report that olanzapine and quetiapine had no benefits on quality of life, while the effects of aripiprazole were significant but small. Risperidone appeared most useful in improving quality of life measures but, as they pointed out, risperidone had the smallest sample size and the largest risperidone trial published did not report on quality of life at the primary endpoint but 2 weeks later.
3.3 Adverse effects
3.3.1 Short term
All studies in the Cochrane review reported significantly increased adverse events using atypical antipsychotics. These were akathisia (aripiprazole), sedation (quetiapine, OFC and aripiprazole), abnormal metabolic laboratory results (quetiapine and OFC) and weight gain (all four drugs, particularly OFC). While most trials used a structured instrument to collect data on adverse events, many limited the items to EPS and akathisia, weight and sometimes sexual functioning. The more recent studies listed in Table 1 appear to have more carefully collected data on adverse events. Those that were more common were weight gain (aripiprazole, ziprasidone, brexpiprazole), akathisia and tremor (aripiprazole, quetiapine, cariprazine, brexpiprazole) and insomnia (cariprazine). However, the akathisia was predominantly mild and weight gain over a 6-8 week period was usually less than 1 kg.
3.3.2 Long term
To date, there are no data on the long-term adverse events associated with adjunctive atypical antipsychotics in double-blind studies of major depression. The longest study is an industry study on the efficacy and safety of OFC for up to 47 weeks.31 Clinically significant weight gain (defined as ≥7% increase in weight) was observed in 55.7% of the patients who remained on OFC throughout the study. In addition, adverse changes in glucose, triglyceride and cholesterol were greater in OFC treated patients and more OFC treated patients experienced high prolactin values. There was no change in ECGs and no difference in induction of mania.
There are multiple studies reporting metabolic changes, including dyslipidaemia, diabetes, prolactin increase, thyroid disorders, obesity, cardiovascular, respiratory and renal diseases, as well as movement and seizure disorders in patients taking atypical antipsychotics for other conditions such as schizophrenia.32, 33 These findings are of significant concern as data from the US National Health and Nutrition Examination Surveys34 has reported that both moderate to severe depressive symptoms and antidepressant use are independently associated with increased obesity and metabolic changes. Therefore, the adjunctive use of atypical antipsychotics for long periods of time needs to be administered with caution.
3.4 What do treatment guidelines tell us?
We identified and reviewed a number of guidelines which discuss the use of adjunctive antipsychotics in major depression. The American Psychiatric Association (APA) most recent 3rd Edition Practice Guidelines was published in 2010.35 They reported that adjunctive treatment with an atypical antipsychotic agent was significantly more effective than placebo in terms of response and remission. They considered that the same level of confidence (Level 2) was applicable to adjunctive treatment with atypical antipsychotics, lithium, thyroid hormone or other antidepressants. The Canadian Network of Mood and Anxiety treatment guidelines were updated in 2016.36 This Guideline recommends adjunctive treatment with atypical antipsychotics as follows; aripiprazole, quetiapine and risperidone are first line, brexpiprazole and olanzapine second line, and ziprasidone third line. The NICE Guidelines for Treatment of Major Depression37 recommend adjunctive treatment as a treatment option and consider atypical antipsychotics such as aripiprazole, quetiapine, risperidone, olanzapine, as well as lithium, to be appropriate adjunctive treatment agents. They noted that lithium and olanzapine are supported by a Category A recommendation while aripiprazole, quetiapine, and risperidone carry a Category B recommendation. They consider the level of evidence for atypical antipsychotics to be Level 1. German National Clinical Practice Guidelines38 are more conservative, not recommending atypical antipsychotics as an adjunctive treatment except in psychotic depression. The World Federation of Societies of Biological Psychiatry Guidelines39 stated that the use of adjunctive treatments was Strategy 5 in managing treatment resistant depression, following (a) maximising the dose of the initial antidepressant; (b) switching to another antidepressant from a different class; (c) switching to another antidepressant from the same class and (d) combining two antidepressants from different classes. Their summary box recommended quetiapine or aripiprazole, but at the same time pointed out the potential unwanted effects.39
The Royal Australian and New Zealand College of Psychiatrists’ Mood Disorder Guidelines, published in 2015, was similarly cautious.40 While noting that placebo controlled studies have found that aripiprazole, olanzapine, quetiapine and risperidone can be effective as adjunctive treatments, they also noted that, excepting quetiapine, they are “off-label”—meaning that they are not formally indicated for adjunctive treatment in Australia or New Zealand. Their advice was that adjunctive antipsychotic treatment may be used in TRD but that, once a stable response has been achieved, the gradual withdrawing of the agent should be considered since there is no evidence of long-term efficacy. The Guidelines also recommended that adverse effects be closely monitored since they are of great concern, especially in the context of long-term therapy. In summary, all the guidelines acknowledge that there is an evidence base to support adjunctive treatment with antipsychotics in patients with major depression, but all guidelines caution about the potential side effects and most do not comment on how long these agents should be used.
4 RECOMMENDATIONS
4.1 Where does this leave us?
This review highlights the potential concerns regarding the use of atypical antipsychotics as an adjunctive treatment in people with major depression. The first concern is who is being treated? While treatment resistant depression is considered the clinical indication for adjunctive treatment in the majority of studies reviewed, the patients more accurately may be described as having an inadequate response rather than being treatment resistant. This is a potential problem for two reasons. It does not test the medication in the population where clinicians may wish to use adjunctive antipsychotics, and it may reduce the perceived efficacy of the drugs, since a significant proportion of those with inadequate response rather than “true” treatment resistance will go on to recover regardless of specific treatments.
The second concern is the quality of the efficacy data. Most studies were industry sponsored. None assessed whether the blinding was effective despite the prominent side effects of the study medication, and few were judged to be free of selective reporting, with adverse events, for instance, not always reported clearly.
The third concern is that, despite these inadequacies, the effect sizes of adjunctive antipsychotics in reducing depressive symptoms were small.
Fourth is the significant side effect burden of atypical antipsychotics reported in all studies and a lack of data on the long-term adverse effects associated with adjunctive antipsychotic use in major depression.
Finally, we are concerned that there are few trials comparing atypical antipsychotics with other strategies in patients with an inadequate response to antidepressants. A recent, large, well-conducted trial attempted to answer these concerns.41 Over 1500 patients who were unresponsive to at least one antidepressant treatment were randomised to switching to bupropion, continuing their antidepressant and adding bupropion, or continuing their antidepressant and adding aripiprazole. The aripiprazole augmentation group exceeded the switch group in remission but other response comparisons were not significant. Remission rates were 23.3% for the switch group, 26.9% for the bupropion augmentation group and 28.9% for the aripiprazole augmentation group. The authors concluded that augmentation with aripiprazole resulted in statistically significant but only modestly increased likelihood of remission during 12 weeks of treatment compared with switching to bupropion monotherapy. They considered that, given the small effect size and adverse effects associated with aripiprazole, further analysis including cost effectiveness was needed to understand the utility of this approach. More pragmatic research of this kind is needed to fully understand the place for atypical antipsychotics in the treatment of MDD.
4.2 Clinical recommendations
The studies reviewed do not support routine use of atypical antidepressants as augmentation in TRD. The benefits are small and the risks moderate. However, this finding does not rule out the possibility that adjunctive antipsychotics may be useful in some patients with TRD. Clinical practice suggests that clinicians find them useful. The question becomes who are the patients where their improvement outweighs the side effects associated with treatment.
The reviewed studies varied in their definitions of TRD for inclusion but the majority used a suboptimal response to a patient's first or second antidepressant, which is better defined as an inadequate response to treatment. This in turn leads to a heterogeneous group whose treatment resistance may arise as a result of inadequate treatment or poor compliance as well as TRD. A more stringent definition of TRD, focussing for example on the melancholic subtype,42 might lead to finding patients who receive substantial benefit from adjunctive antipsychotics. It is likely that the placebo response rate in such patients will be less than most of the trials conducted in those who are simply unresponsive to antidepressant treatment.
Clinical accounts report that the principal benefit of using antipsychotics is found in patients with extensive ruminations, which are almost delusional in intensity, as well as those with marked psychomotor changes, major sleep problems, and guilt and remorse, rather than simply those who have failed to respond to one or two antidepressants. In such cases, the short-term use of adjunctive antipsychotics may be justified. We would also argue that these are the type of patients that need to be included in future clinical trials in order to obtain a better idea of the cost-effectiveness of adjunctive antipsychotics.
In summary, adjunctive antipsychotics may be useful in selected patients with major depression who are treatment resistant. One advantage of adding an antipsychotic is that its clinical effect is usually apparent within the first week and the medication can then be tapered and stopped after a relatively brief treatment trial. The current evidence only supports relatively short-term use of these treatments, which then only need to be maintained for a further short period of time if they are felt to be of benefit. This positions the use of adjunctive antipsychotics as “facilitators” of an acute response to treatment, and it is important to note that their effect is likely to be most helpful in patients with severe and disabling depression.
4.3 Research recommendations
Ideally, research needs to systematically examine the whole class of antipsychotics and their effect on individual clinical profiles and symptoms. Such research, however, would require very large numbers to have the power to study subgroups. A more practical goal might be for studies to focus on subgroups of TRD, such as melancholic or severely agitated depression, since the risk/benefit ratio is likely to be better in such patients. More studies directly comparing adjunctive antipsychotics with other treatment strategies for specifically defined TRD, such as switching or combining antidepressants, or adding lithium or psychotherapy, would also be useful. Finally, exploring the mechanisms by which antipsychotics combined with antidepressants work would be ideal. Are they augmenting (ie increasing the efficacy of the antidepressant) or adjunctive (ie added but not essential to the antidepressant) or synergistic (working together with antidepressants)?
5 CONCLUSION
There is a moderately large database suggesting that adjunctive antipsychotics may be useful in treating TRD. However, the effect size is small, there are issues with the quality of the efficacy data, the side effects are moderate to severe, and the varying definitions of TRD have led to the assessment of differing samples. Therefore, we do not recommend the routine use of adjunctive antipsychotics in patients who have failed to improve with one or two trials of an antidepressant. Nonetheless, adjunctive antipsychotics are useful in psychotic depression and it remains possible that some patients with TRD, such as the melancholic subtype or narrowly defined severe cases, might receive significant benefit from short-term adjunctive antipsychotics. Studies focussing on specific depression profiles and directly comparing adjunctive antipsychotics with other treatment strategies are urgently needed.
DISCLOSURES
This work has been made possible in part through funding from the NHMRC Program Grant APP1073041. RJP has received grants from Health Research Council of New Zealand, Canterbury Medical Research Foundation, and Lotteries Health (NZ); and is provided software for research at no cost from SBT-pro. PB has received grant or research support from the NHMRC, Western Sydney LHD, Eli Lilly & Co, Ferring Pharmaceuticals and the Brian Resource Company; has been a speaker for AstraZeneca, Eli Lilly & Co., Janssen Cilag, Lundbeck, Servier; has been a consultant for Servier and served on advisory boards for AstraZeneca, Eli Lilly & Co., Janssen Cilag, Lundbeck, and Servier. RM has recent grant funding from Health Research Council of New Zealand. GSM has received grant or research support from National Health and Medical Research Council, Australian Rotary Health, NSW Health, Ramsay Health, American Foundation for Suicide Prevention, Ramsay Research and Teaching Fund, Elsevier, AstraZeneca and Servier; has been a speaker for AstraZeneca, Janssen-Cilag, Lundbeck, and Servier; and has been a consultant for AstraZeneca, Janssen Cilag, Lundbeck and Servier. AH, GM, GSM, PB, RJP, RM have all received travel and/or accommodation support from Servier. The SMART Group was supported logistically by Lundbeck who provided financial assistance with travel and accommodation for those members travelling interstate or overseas to attend the meeting in Sydney (held on 3rd March 2018). None of the committee were paid to participate in this project and Lundbeck have not had any input into the content, format or outputs from this project.
