Abstract Background Omega-3 (or n-3) polyunsaturated fatty acids (PUFAs) are promising antidepressant treatments for perinatal depression (PND) because of supporting evidence from clinical trials, the advantage in safety, and their anti-inflammatory and neuroplastic effects. Although several observational studies have showed n-3 PUFA deficits in women with PND, the results of individual PUFAs from different studies were inconsistent. Methods This systematic review and meta-analysis aim to compare the levels of PUFA indices, including eicosapentaenoic acid (EPA), docosahexaenoic acid (DHA), arachidonic acid (AA), total n-3, total n-6, and the n-6/n-3 ratio between women with PND and healthy controls. The meta-analysis included 12 eligible studies available by December 2016. The effect sizes (ESs) were synthesized by using a random effects model. In addition, we performed subgroup analysis for the PUFA levels in patients with prenatal and postnatal depression, both of which were compared with healthy controls. Results There were significant lower levels of total n-3 PUFAs and DHA, and a significant increased n-6/n-3 ratio in PND patients. In the subgroup analyses, there were significant lower levels of n-3 PUFAs, EPA, and DHA in women with prenatal depression. The n-6/n-3 ratio was significantly increased in both prenatal and postnatal depression subgroups. Conclusions Our meta-analysis consolidates the important role of n-3 PUFAs in PND. As nutritional medicine is an important strategy to improve the effectiveness of depression treatment, our findings provide the strong rationale to conduct clinical trials to test the therapeutic and prophylactic effects of n-3 PUFAs in PND.