• major depressive disorder
• antidepressants
• clinical trials outcome
• alcohol use disorders
• comorbidity

Abstract
Major depressive disorder (MDD) and dysthymic disorder (DD) are often complicated by the co-occurrence of alcohol use disorders (AUDs), each condition often aggravating the course and outcome of the other in terms of severity, functional impairment, risk of suicide, relapse of depression and drinking behaviors, and chronicity. Recently, there has been some evidence that antidepressants may improve depressive symptoms in patients with concurrent AUDs, even if they have only a limited effect in decreasing alcohol use in these subjects. To date, however, there is a paucity of randomized, double-blind, placebo-controlled trials that have been conducted to evaluate the efficacy, safety, and tolerability of antidepressants as monotherapy for patients with MDD/DD and co-occurring AUDs. Although previous meta-analyses have found antidepressants to be more effective than placebo in the treatment of this specific patient population, efficacy for newer agents (i.e. the SSRIs) was questionable. The aim of this study is to examine the efficacy of antidepressants in patients with unipolar depression (MDD and/or DD) with co-morbid AUDs, and to compare compare study design characteristics, patient characteristics, and drug-/placebo- outcomes between depressed patients with or without co-morbid AUDs.
Method
Medline/Pubmed publication databases were searched for randomized, double-blind, placebo-controlled trials of antidepressants used as monotherapy for the acute-phase treatment of MDD and/or DD in patients with or without AUDs. The search was limited to articles published between January 1980 and March 2010 (inclusive). We selected for studies which focused on the treatment of adult patients, were at least 4 weeks of duration, used antidepressants in their oral formulation, and used either the Hamilton Depression Rating Scale (HDRS), the Montgomery-Asperg Depression Rating Scale (MADRS), or the Clinical Global Impression-Improvement Scale (CGI) as one of their outcome measures. The primary outcome for our meta-analysis was clinical response, defined as a 50% or greater reduction in HDRS or MADRS scores, baseline to endpoint, or a CGI-I<3 at the final visit.
Results
A total of 195 manuscripts were found eligible for inclusion in our analysis (n=46,820 patients), 11 of which focused on the treatment of MDD/DD in patients with co-morbid AUDs.
We found that antidepressant therapy was more effective than placebo in patients with co-morbid AUDs (RR=1.336; p=0.021), but not when SSRIs were examined alone (RR= 1.145; p=0.316). There was no difference in the relative efficacy of antidepressants (versus placebo) when comparing studies in MDD/DD patients with or without AUDs (p=0.973). Meta-regression analyses yielded no significant differences in the RR of responding to antidepressants versus placebo in trials with co-morbid AUDs whether antidepressants were used alone or adjunctively to psychotherapy, whether used in patients actively drinking or recently sober, or whether used in pure MDD or in combined MDD and DD populations. Moreover, the baseline severity of drinking (assessed as the number of heavy drinking days in the week before randomization) did not predict a significantly difference in the response rate to antidepressants (coefficient= -0.644, p=0.467). Finally, there was no statistically significant difference in the percentage of heavy drinking days at the end of the study between antidepressant- and placebo- treated patients (RR = 0.691, p=0.275)
Conclusions
Our results suggests that antidepressants are more effective than placebo in treating depression in patients with co-morbid AUDs, and lend further support to the argument that antidepressants should represent first-line therapy for targeting depressive symptoms in patients with alcohol use disorders, a condition which is highly prevalent and is associated with high rates of medical and psychiatric comorbidity, disability, and increased use of general medical health services as well as psychiatric hospitalizations. However, the use of SSRIs for treating depression in such patients is not convincingly supported by the evidence. More data on the use of newer antidepressants, including SNRIs and SSRIs, for this select patient population are needed.
Moreover, our work showed that the efficacy of antidepressants was not influenced whether patients were actively drinking or recently sober, and we did not find any relationship between the severity of baseline drinking and treatment outcome. This suggests that the decision whether to recommend antidepressants in this patient population should not be determined by these variables, at least as far as the potential efficacy of treatment is concerned.