Antidepressant Treatment of Depression in Schizophrenia: A Systematic Review and Meta-Analysis Dr Rachel Upthegrove, Senior Clinical Lecturer, Department of Psychiatry, College of Medical and Dental Sciences. Angharad Gregory, Medical Student University of Birmingham Keywords: Schizophrenia, Psychosis, Depression, Antidepressant, Quality of life and Suicide. Background Co-morbid depression in schizophrenia is common, with prevalence rates reported from 25%(1) to 83%(2). The variation in prevalence is partly due to the challenge in distinguishing symptoms of a mood disorder from core negative symptoms in schizophrenia, for example, blunted affect and difficulty expressing emotion. The discrepancy is further attributed to insufficiency of current classification systems in diagnosing concurrent depression in schizophrenia. However, following results of recent epidemiological studies, the overall consensus is that depressive syndromes are present in 50% of individuals with schizophrenia.(3) Depressive symptoms in schizophrenia can occur in all phases of the illness and cause significant distress. The consequences of depression in schizophrenia are serious and include more frequent relapses(4), increased duration of illness (5) and high rate of substance abuse(6). Suicide is 13 times more likely in people with schizophrenia than in the general population(7). It has also been found that of those that commit suicide, 57% had also been depressed(8). In order to improve quality of life and reduce rates of suicide, effective treatment of depressive symptoms is needed. The optimal treatment approach for depression in schizophrenia is disputed. Guidelines on treatment recommendations from National Institute of Clinical Excellence (NICE)(9) and British Association for Psychopharmacology(10) are brief and without conclusion. NICE recommends Cognitive Behavioural Therapy to all people with schizophrenia or related psychoses but not specifically for concurrent depression(9). The literature on psychological intervention is modest. Most studies on psychological treatments have not investigated depression as a primary outcome or with well-validated scales(11). In terms of pharmacological therapy, in the early 2000’s there was hope that some of the newer atypical antipsychotics may be effective in ameliorating co-morbid depressive symptomatology(12, 13), particularly olanzapine(14, 15) and quetiapine(16, 17). However, a Cochrane review in 2008 (18) found no evidence to support atypical antipsychotics are any better than typical antipsychotics for treating depression in people with schizophrenia. The review was based on three extremely small, poorly reported, short studies, vulnerable to bias. The authors suggested practice at that time should be guided by evidence other than that derived from randomised trials.(18) The potential benefit of antidepressants for co-morbid depression has not received the attention it warrants.(19) Early studies in the 90’s investigating the use of tricyclic antidepressants, mainly imipramine, showed some significance in treating co-morbid depressive features (20, 21). However, there has since been minimal advancement. Whitehead et al (5) conducted a systematic review in 2002 identifying just 11 appropriate trials. The majority used tricyclic antidepressants, with one study investigating a selective serotonin reuptake inhibitor.(22) In general, studies were small and of poor quality, with multiple methodological flaws including small sample size, inadequate description of allocation concealment, absent power calculations and multiple trials did not use an intention-to-treat analysis. (5) Although a subgroup analysis indicated some clinical benefit of antidepressants, drawing results from a small number of small trials is unreliable. Furthermore, it was impossible to investigate the likelihood of publication bias as another consequence of the limited trials in the review. To conclude, Whitehead (5) reported although the evidence suggests some clinical benefit of co-prescribing antidepressants in schizophrenia a fairer conclusion would be evidence at that time remains “unproven”. The review is now out of date and the need for a definitive answer regarding the use of antidepressant therapy persists. A substantial body of evidence suggests the use of antidepressant medication in schizophrenia is increasing.(23) It has been reported, antidepressants are prescribed for 11 to 43% of patients with schizophrenia (4). Notably, alongside depressive symptoms, they are also used to treat anxiety and negative symptoms. However, polypharmacy increases drug interactions and side effects. A number of case reports have shown concurrent use of selective serotonin reuptake inhibitors and new antipsychotics is associated with arrhythmias, prolonged QTc interval on electrocardiogram and orthostatic hypotension in patients with otherwise normal cardiovascular systems (24). Therefore, it is essential to determine if the use of antidepressants for depression is therapeutic or futile in order to prevent unnecessary harm. In summary, the evidence base for the regular co-pharmacy of antidepressants alongside antipsychotics in clinical practice is weak. Considering how frequently depression occurs as a discrete syndrome in schizophrenia, as well as being a significant mediator of disability and potentially leading to suicide, research into the role of antidepressant therapy is essential to establish adequately informed guidelines.(3) Aim The aim of the study is to conduct a systematic review and meta-analysis of the use of antidepressants as an adjunct to antipsychotic medication to treat depression in people with schizophrenia. 1 Objectives To synthesise evidence of the effectiveness of antidepressants for treatment of depression in people with schizophrenia To estimate the aggregate effectiveness of antidepressants for treatment of depression in people with schizophrenia Setting This project is a systematic review therefore will be using published data, not specific to any location. Methods/Design A systematic review and meta-analysis will be conducted with outcomes reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement (25). A PRISMA flow diagram will create a graphical representation of citations reviewed in the course of the systematic review (see Appendix 2). Data will be extrapolated and assessed using Cochrane tools for data extraction (26) and risk of bias (27), respectively. Selection Criteria (See Appendix 3 for inclusion criteria table/screening tool) 1. Types of Participants Participants will include any individual aged 18 years or older with schizophrenia or related psychosis; including schizoaffective disorder and psychotic disorder not otherwise specified (as diagnosed using International Classification Diseases-10) with a depressive episode (as assessed using a standardised rating scale or clinical interview). Studies including participants under 18 years old or with a primary diagnosis of organic brain disorder or bipolar disorder will be excluded from the systematic review. This is justified on the basis that depression experienced in these disorders would have a very different aetiology and would be managed differently to the focus of this review. 2. Types of Intervention Interventions used in included trials will take an antidepressant versus usual care or placebo approach. Any class of antidepressant: tricyclic antidepressant; monoamine oxidase inhibitor; selective serotonin reuptake inhibitor or other type will be valid as per those published in the BNF (section 4.3).(28) Agents with only theoretical antidepressant properties never approved in any country for depression will be excluded. It is anticipated that all people in the study will be receiving antipsychotic medication and possibly an anticholinergic. All medications participants receive throughout the trial duration should be charted, monitored and reported. 3. Types of Outcome Primary Outcome: The primary outcome will be recovery from depression using dichotomous outcomes (depressed/ not depressed) provided by the authors. Continuous outcomes, showing change on standardised scales will also be included. Secondary Outcomes: These will include positive and negative symptoms of schizophrenia and quality of life, using any psychometrically validated scale both self-reported and clinician administered. Timing of Outcome Measures: It is anticipated the time points of measuring outcomes will vary. It may be necessary to group outcome measures into short-term, medium-term and long-term outcomes. For example, as guided by results of Whitehead et al (5), <6 weeks, <12 weeks and >12 weeks, respectively. 4. Types of Studies All appropriate randomised controlled trials or quasi-experimental studies will be included. Observational or qualitative studies will not be included. Search Methods and Identification of Studies The search strategy aims to find both published and unpublished studies to minimise chance of publication bias. 1. Electronic searches The following electronic bibliographic databases will be searched: The Cochrane Central Register of Controlled Trials; MEDLINE (1948 to present); EMBASE (1980 to present); CINAHL (1982 to present); PsycINFO (1967 to present) and Web of Science (1990 to present). The final search update will be 29th February 2015. (See Appendix 1 for MEDLINE search). The following limits will be set: 1. Randomised controlled trials (RCTs) or quasi-experimental studies, 2. English language