Temporomandibular disorders and neck pain in primary headache patients: a retrospective machine learning study
Martina Ferrillo, Mario Migliario, Nicola Marotta, Francesco Fortunato, Marino Bindi, Federica Pezzotti, Antonio Ammendolia, Amerigo Giudice, Pier Luigi Foglio Bonda, Alessandro de Sire
Abstract
OBJECTIVES: To evaluate the linkage underpinning different clinical conditions as painful TMD and neck pain in patients affected by primary headaches. MATERIALS AND METHODS: In this machine learning study, data from medical records of patients with headaches as migraine, tension-type headache (TTH) and other primary ones, referring to a University Hospital over a 10-year period were analysed. VAS was used to evaluate the intensity of the TMD and neck pain. Moreover, the magnetic resonance imaging was used to supplement the clinical data. RESULTS: A total of 300 patients (72 male, 228 female), mean aged 37.78 ± 5.11 years, were included. Higher TMD and neck pain VAS in migraine patients were reported. The machine learning analysis focussed on type of primary headache demonstrated that a higher TMD VAS was correlated to migraine, whereas a higher neck pain VAS was correlated to TTH or migraine. Concerning the TMD type, arthrogenous and mixed TMD were correlated to mild-moderate TMD pain (depending on neck pain intensity), whereas myogenic TMD was correlated to moderate-severe TMD pain. CONCLUSIONS: Machine-learning approach highlighted the complexity of diagnosis process and demonstrated that neck pain might be an influential variable on the belonging to different group of headaches in TMD patients.