Early diagnosis of oral squamous cell carcinoma (OSCC) is an attractive strategy to increase survival rate of patient. An integrated separation approach of reversed phase liquid chromatography and hydrophilic interaction chromatography combining with time of flight mass spectrometer has been firstly developed for performing global saliva metabonomics analysis for early diagnosis of OSCC. This approach was designed to overcome the limitations of a single chromatographic method due to different polarity of endogenous metabolites. As a result, 14 potential salivary metabolites were identified. Eight biomarkers up-regulated in OSCC patients are compared with control and six down-regulated groups. Receiver operating characteristic analysis was exploited to evaluate the diagnostic power of the candidate biomarkers, and related metabolic pathways have also been studied. Five salivary biomarkers (propionylcholine, N-Acetyl-L-phenylalanine, sphinganine, phytosphingosine, and S-carboxymethyl-L-cysteine) in combination yielded satisfactory accuracy (AUC = 0.997), sensitivity (100%), and specificity (96.7%) in distinguishing early stage of OSCC from the control. In this study, a comprehensive saliva metabonomics analysis for identifying potential biomarkers to early diagnose OSCC is successfully demonstrated, which has the advantages of non-invasive, simple, reliable, and low-cost. These novel metabolic biomarkers have obvious clinical utility that will help to diagnose OSCC at its early stage.