Iron‐Sensitive MRI Reveals Individual Differences in Pain Modulation Rather Than Categorical Markers in Chronic Pain and Migraine

Abstract

Objective To determine whether regional iron levels are associated with endogenous pain‐inhibitory function and whether iron‐sensitive MRI techniques, quantitative magnetic susceptibility (QSM) and R2∗ mapping capture differences in midbrain iron content in the globus pallidus (GP), periaqueductal grey (PAG), and red nucleus (RN) across chronic pain (CP), migraineurs (M), and healthy volunteers (HV). Methods Sixty participants (20 CP, 20 M, and 20 HV) underwent a 3T MRI examination. Pain modulatory and autonomic function was assessed using conditioned pain modulation (CPM) and pupil reflex dilation (PRD) across all participants (n = 60), whereas quantitative sensory testing (QST) and pain sensitivity questionnaire (PSQ) were assessed in CP and M only (n = 40). Group differences and iron–pain relationships were evaluated using ANOVA or Kruskal–Wallis tests and linear regression models. Results Distinct region‐ and modality‐specific relationships were observed between iron‐sensitive MRI measures and pain‐related outcomes. R2∗ RN was associated with absolute and relative CPM effects. QSM PAG was associated with temporal summation and showed a trend‐level association with subjective pain sensitivity (PSQ‐minor). Autonomic reactivity PRD was associated with R2∗ measures in both regions. R2∗ RN was related to PRD during heat stimulation, whereas R2∗ PAG was related to PRD during heat and cold‐water conditioning. QSM/R2∗ RN were further associated with mechanical pain threshold (MPT) and mechanical pain sensitivity (MPS). Group‐level differences were, however, minimal between CP, migraine, and HV cohorts. Conclusion Modality‐specific associations between iron‐sensitive MRI measures and descending inhibition, nociceptive gain, and autonomic reactivity were observed. These findings indicate that QSM and R2∗ mapping together capture individual variation in pain‐modulatory physiology, supporting the concept of an iron‐related biotype of pain modulation rather than a categorical marker. In contrast, regional iron content did not differentiate CP, M, and HV groups.