Although temporomandibular disorder (TMD) is the second most common musculoskeletal disorder in the general population, the disease is multifactorial and presents symptoms common to other conditions which misdiagnosis can lead to treatment failure. In this case-control study, we performed, for the first time, a high-resolution 1H-nuclear magnetic resonance spectroscopy metabolomic analysis of the saliva of 26 women with TMD of muscular origin (experimental group [EG]) at the beginning (EG-pre) and at the end (EG-post) of a conservative treatment, and of 27 normal women (control group [CG]) to identify a metabolic signature for TMD. One-way analysis of variance showed changes in the concentration of phenylacetate, dimethylamine, maltose, acetoin, and isovalerate. Partial least-square discriminant analysis showed that metabolite signals did not overlap in CG X EG-pre and EG-pre X EG-post, but overlapped in CG X EG-post. The area under the receiver operating characteristic curve was 1 in CG X EG-pre (95% CI, 1.000-1.000; p < 0.002), 0.993 in EG-pre X EG-post (95% CI, 0.963-1.000), and 0.832 in CG X EG-post (95% CI, 0.699-0.961). These results suggest that the metabolomic profiles of women with and without TMD differ, while after treatment there is a lower distinction and slight tendency towards overlapping between CG and EG-post compared to pre treatment. We also found that phenylacetate, dimethylamine, maltose, acetoin, and isovalerate are potential biomarkers for TMD of muscular origin.
Keywords: Biomarkers; Conservative treatment; Counseling; Facial pain; Metabolomics; Proton magnetic resonance spectroscopy; Saliva; Self-management; Temporomandibular joint dysfunction syndrome.
Copyright © 2020 Elsevier B.V. All rights reserved.