Texture analysis on muscle T2 maps can reveal changes in upper trapezius muscles related to primary headache disorders.

Migraine and tension-type headache (TTH) are highly prevalent primary headache disorders. Neck pain is common among patients with migraine or TTH, but objective biomarkers for the neck's myofascial involvement within the trigemino-cervical complex (TCC) are lacking. This study aimed (1) to apply texture analysis (TA) to magnetic resonance imaging (MRI) covering the upper trapezius muscles with T2 maps, which are particularly sensitive to subtle edematous changes such as those following inflammation; and (2) to associate demographics, characteristics of sleep and sports, headache, and neck pain with muscle T2 values and derived texture features. Seventy-one participants (mean age 24.0 ± 3.0 years, 59 females; 22 healthy controls [HC], 21 migraine, 16 TTH, 12 migraine with TTH) underwent MRI of the upper trapezius muscles, followed by bilateral muscle segmentations and extraction of muscle T2 and first- and second-order texture features. Comparisons between HC and participants with primary headache disorders were performed following removal of outliers with the robust regression followed by outlier identification (ROUT) method (Mann-Whitney U tests, adjustment for multiple comparisons). Principal component analysis (PCA), linear, and logistic regression analyses (using bootstrapping with 5,000 permutations) were conducted based on image-derived parameters and headache and neck pain characteristics. Significant group differences were found between HC and participants with primary headache disorders for muscle T2 (30.14 ± 1.05 vs. 31.10 ± 0.92 ms, p < 0.001), sumaverage (0.007163 ± 0.0011 vs. 0.006366 ± 0.0012, p = 0.017), kurtosisglobal (1.15 ± 0.32 vs. 1.63 ± 0.81, p = 0.024), and energy (0.003514 ± 0.0006 vs. 0.004176 ± 0.0012, p = 0.043). According to PCA, a cluster including age, days with headache over the last three months, neck pain, muscle T2, kurtosisglobal, and skewnessglobal was identified, with a cluster close to the opposite direction including average hours of sports per week, average hours of sleep per week, sex, and varianceglobal. Muscle T2 (p = 0.015) and sumaverage (p = 0.022) were significantly associated with days with headache over the last three months. Furthermore, sumaverage was significantly associated with the presence of neck pain (p = 0.024). T2 mapping with TA may provide objective and quantitative parameters for myofascial involvement and peripheral sensitization within the TCC, partially associated with clinical parameters such as the number of days with headache.