This review provides a critical perspective on recent meta-analyses suggesting that several anti-inflammatory modalities, including nonsteroidal anti-inflammatory drugs (NSAIDs), omega-3 fatty acids, and cytokine antagonist, possess generalizable antidepressant properties. By examining confounds and limitations in the available literature it is suggested that current data suggest that only a sub-group of individuals with major depressive disorder (MDD) have evidence of increased inflammatory biomarkers and it is in these individuals that anti-inflammatory agents show promise for reducing depressive symptoms. The treatment implications of this cautionary perspective are discussed.