Surface-based cortical morphometry in fibromyalgia reveals left temporal lobe thinning and dissociated thickness-area patterns: a FreeSurfer replication and extension study

Background Fibromyalgia (FM) is a nociplastic pain disorder associated with structural brain changes. Surface-based cortical thickness analysis using FreeSurfer, enabling independent measurement of cortical thickness and surface area, has been applied to FM in only a small number of studies. Methods We applied FreeSurfer 7.4.1 to structural MRI from 20 women with FM and 20 age-matched healthy controls from OpenNeuro dataset ds001928 (GE 3T SPGR), previously analysed by Agoalikum et al. using source-based morphometry (SBM). Group comparisons of 68 Desikan-Killiany cortical parcels and 16 subcortical structures used Benjamini-Hochberg FDR correction, covarying for age (and eTIV for volumes). Results Three cortical regions survived FDR correction (q < 0.05), all left temporal: inferior temporal cortex (d = −1.77), fusiform gyrus (d = −1.35), and middle temporal gyrus (d = −1.22); a whole-surface vertex-wise GLM independently confirmed this left temporal thinning and additionally resolved entorhinal involvement. A thickness-area dissociation was identified: middle temporal gyrus showed laminar thinning (thickness reduced, area preserved), whereas inferior temporal cortex showed concurrent thickness and area reduction. Subcortical findings included bilateral cerebellar cortex reduction (d ∼ −1.0), right amygdala reduction (d = −1.01), and bilateral pallidum enlargement (d≈+0.85); all FDR q < 0.05. Thalamic volumes did not differ between groups (p > .54), directly contradicting the main finding of the prior SBM analysis on the same subjects. Left ACC cortical thickness correlated inversely with depression severity in this FM cohort (age-adjusted GLM: partial r = −0.65, p = .003, FDR q = 0.022). Conclusions SBM and surface-based morphometry capture distinct structural properties in FM even when applied to identical subjects, with implications for methodological standardisation. Temporal association cortex may represent a candidate structural substrate for the cognitive symptoms of FM, though this interpretation is provisional as no cognitive measures were available in this cohort.