Osteoarthritis (OA) is a leading cause of chronic joint pain in the older human population. Diagnosis of OA at an earlier stage may enable the development of new treatments to one day effectively modify the progression and prognosis of the disease. In this work, we explore whether an integrated metabolomics approach could be utilized for the diagnosis of OA. Synovial fluid (SF) samples were collected from symptomatic chronic knee OA patients and normal human cadaveric knee joints. The samples were analyzed using 1 H nuclear magnetic resonance (NMR) spectroscopy and gas chromatography-mass spectrometry (GC-MS) followed by multivariate statistical analysis. Based on the metabolic profiles, we were able to distinguish OA patients from the controls and validate the statistical models. Moreover, we have integrated the 1 H NMR and GC-MS results and we found that 11 metabolites were statistically important for the separation between OA and normal SF. Additionally, statistical analysis showed an excellent predictive ability of the constructed metabolomics model (area under the receiver operating characteristic curve ¼ 1.0). Our findings indicate that metabolomics might serve as a promising approach for the diagnosis and prognosis of degenerative changes in the knee joint and should be further validated in clinical settings.