The aetiology and pathophysiology of depression have long been associated with inflammation, at least in a proportion of patients. Altered cytokine activity in the periphery and in the brain has brought support to a concept of depression-associated inflammation. However, these immunological changes - and inflammation in particular - in depression have only been recently targeted for treatment. Non-steroidal anti-inflammatory drugs (NSAIDs) have been proposed to be of clinical use in the treatment of depression either as monotherapy or as adjuncts in combination with antidepressants. Specifically, selective cyclooxygenase (COX)-2 and non-selective COX inhibitor NSAIDs as adjuncts or monotherapy have been trialled clinically. A limited body of clinical research has been conducted with mixed results so far. Although meta-analyses appear to support the use of NSAIDs in acute depression, the overall effect is mainly biased by the effects of celecoxib for which the best evidence exists to date. Efficacy data of non-selective COX inhibitor NSAIDs on depressive symptoms is limited and out of six studies, only a retrospective analysis shows positive results for non-selective COX inhibitor. Clinical data on aspirin, an irreversible inhibitor of both COX-1 and COX-2, are mainly experimental and hypothetical at this stage, but may be promising in depressed patients with concomitant inflammatory conditions. The main problematic factor is that current evidence rests on trials in acute depression. Because of the dynamic nature of depression, it is important exploring if NSAIDs and other anti-inflammatory treatments may have a preventive role in early stages of depression and for relapse prevention. The possible impact of anti-inflammatory treatments on immune changes in different phases of depression warrants caution for a wide and preventive use of anti-inflammatory agents in depression-associated inflammation.