Profiling migraine patients according to clinical and psychophysical characteristics: clinical validity of distinct migraine clusters.

Aims: Investigate if different clinical and psychophysical bedside tools can differentiate between district migraine phenotypes in ictal/perictal (cohort 1) and interictal (cohort 2) phases.

Method: This observational study included two independent samples in which patients were subgrouped into distinct clusters using standardized bedside assessment tools (headache frequency, disability, cervical active range of motion, pressure pain threshold in different areas): (A) cohort 1-ictal/perictal migraine patients were subgrouped, based on previous studies, into two clusters, i.e., Cluster-1.1 No Psychophysical Impairments (NPI) and Cluster-1.2 Increased Pain Sensitivity and Cervical Musculoskeletal Dysfunction (IPS-CMD); (B) cohort 2-interictal migraine patients were subgrouped into three clusters, i.e., Cluster-2.1 NPI, Cluster-2.2 IPS, and Cluster-2.3 IPS-CMD. Clinical characteristics (multiple questionnaires), somatosensory function (comprehensive quantitative sensory testing (QST)), and cervical musculoskeletal impairments (cervical musculoskeletal assessment) were assessed and compared across headache clusters and a group of 56 healthy controls matched for sex and age.

Results: Cohort 1: A total of 156 subjects were included. Cluster-1.2 (IPS-CMD) had higher headache intensity (p = 0.048), worse headache-related (p = 0.003) and neck-related disability (p = 0.005), worse quality of life (p = 0.003), and higher symptoms related to sensitization (p = 0.001) and psychological burden (p = 0.005) vs. Cluster-1.1(NPI). Furthermore, Cluster-1.2 (IPS-CMD) had (1) reduced cervical active and passive range of motion (p < 0.023), reduced functionality of deep cervical flexors (p < 0.001), and reduced values in all QST(p < 0.001) vs. controls, and (2) reduced active mobility in flexion, left/right lateral flexion (p < 0.045), and reduced values in QST (p < 0.001) vs. Cluster-1.1 (NPI). Cohort 2: A total of 154 subjects were included. Cluster-2.3 (IPS-CMD) had (1) longer disease duration (p = 0.006), higher headache frequency (p = 0.006), disability (p < 0.001), and psychological burden (p = 0.027) vs. Cluster-2.2 (IPS) and (2) higher headache-related disability (p = 0.010), neck-related disability (p = 0.009), and higher symptoms of sensitization (p = 0.018) vs. Cluster-2.1 (NPI). Cluster-2.3(IPS-CMD) had reduced cervical active and passive range of motion (p < 0.034), and reduced functionality of deep cervical flexors (p < 0.001), vs. controls, Custer-2.1 (NPI), and Cluster-2.2 (IPS). Cluster-2.2 (IPS) and 2.3 (IPS-CMD) had reduced QST values vs. controls (p < 0.001) and Cluster-2.1 (p < 0.039).

Conclusion: A battery of patient-related outcome measures (PROMs) and quantitative bedside tools can separate migraine clusters with different clinical characteristics, somatosensory functions, and cervical musculoskeletal impairments. This confirms the existence of distinct migraine phenotypes and emphasizes the importance of migraine phases of which the characteristics are assessed. This may have implications for responders and non-responders to anti-migraine medications.