Knowledge of blood triacylglycerol (TAG) species is essential to clarify the physiological functions of individual TAG molecules and also to develop potential biomarkers for related diseases. Commonly, lipid samples prepared by organic liquid-liquid extraction contain complex components, thus cannot be directly characterized by mass spectrometry (MS) and often require an additional purification step. Here, we described a laser desorption ionization - mass spectrometry (LDI-MS) method that utilized aggregated graphene oxide (AGO) as both lipid extractant and MS matrix (AGOLDI-MS), to characterize and quantify plasma TAG species without the use of harmful solvent or complex separation step. We first designed and synthesized the AGO material with a multi-layered sheet structure, which could efficiently break up the structure of lipoproteins, and extract plasma TAGs as solid-phase extraction material. Furthermore, in AGOLDI-MS procedure, the AGO could directly act as matrix and selectively produce the MS signals of TAGs without the interferences of phospholipids, which was hardly achieved by using the routine LDI-MS method based on liquid-liquid extraction and small molecular matrix. We confirmed the suitability of AGOLDI-MS as characterization and quantitative tool for TAG species through studying the analysis performances in TAG standards and real plasma samples. To establish potential utility of our method, we characterized 42 human plasmas from healthy and hyperlipemic donators, indicating that the AGOLDI-MS could not only generate comparable quantitative results of total TAGs to current clinical technology, but also monitor the changes of TAG species between different sample groups. This approach could further characterize the compositions of the fatty acid moieties in even low abundant TAGs by the assistance of tandem MS-MS. This concise, specific, and high-throughput approach will facilitate the rapid and precise characterizations of plasma TAGs, and make the MS approach for TAGs more adaptable for clinical uses. (c) 2018 Elsevier B.V. All rights reserved.