Polypoidal choroidal vasculopathy (PCV), the predominant subtype of neovascular age-related macular degeneration in the Asian population, is associated with genetic polymorphism of lipid metabolism. In this study, we performed the untargeted lipidomics approach of ultra-performance liquid chromatography coupled with mass spectrometry (UPLC-MS) to reveal the potential discriminating lipid profile of PCV patients in serum (21 PCV patients and 19 age-matched controls). Unsupervised principal component, supervised orthogonal partial least squares analysis, correlation analysis, and heatmap analysis were performed with the data obtained by UPLC-MS. Forty-one discriminating metabolites were identified. Receiver operating characteristic curve analysis, pathway analysis and functional analysis were performed subsequently, and platelet-activating factor (PAF) was further selected as the key indicator of the distinct lipid metabolism in PCV patients. Finally, the serum level of PAF was validated by enzyme-linked immunosorbent assay, which is significantly higher in PCV patients compared to controls (65 PCV patients and 63 age-matched controls, p < 0.0001), consistent with the UPLC-MS analysis. Our results suggested that PAF is considered as the major indicator of the distinct lipid metabolism in PCV patients.