In this study, we used Chaihu Shugan San (CSS), a traditional Chinese herbal formula, as a probe to investigate the involvement of brain functional network connectivity and hippocampus energy metabolism in perimenopausal depression. A network pharmacology approach was performed to discover the underlying mechanisms of CSS in improving perimenopausal depression, which were verified in perimenopausal depression rat models. Network pharmacology analysis indicated that complex mechanisms of energy metabolism, neurotransmitter metabolism, inflammation, and hormone metabolic processes were closely associated with the anti-depressive effects of CSS. Thus, the serum concentrations of estradiol (E), glutamate (Glu), and 5-hydroxytryptamine (5-HT) were detected by ELISA. The brain functional network connectivity between the hippocampus and adjacent brain regions was evaluated using resting-state functional magnetic resonance imaging (fMRI). A targeted metabolomic analysis of the hippocampal tricarboxylic acid cycle was also performed to measure the changes in hippocampal energy metabolism using liquid chromatography-tandem mass spectrometry (LC-MS/MS). CSS treatment significantly improved the behavioral performance, decreased the serum Glu levels, and increased the serum 5-HT levels of PMS + CUMS rats. The brain functional connectivity between the hippocampus and other brain regions was significantly changed by PMS + CUMS processes but improved by CSS treatment. Moreover, among the metabolites in the hippocampal tricarboxylic acid cycle, the concentrations of citrate and the upregulation of isocitrate and downregulation of guanosine triphosphate (GTP) in PMS + CUMS rats could be significantly improved by CSS treatment. A brain functional network connectivity mechanism may be involved in perimenopausal depression, wherein the hippocampal tricarboxylic acid cycle plays a vital role.