Brain perfusion in fibromyalgia patients and its differences between responders and poor responders to gabapentin

INTRODUCTION: The aim of the present study was to determine the brain areas associated with fibromyalgia, and whether pretreatment regional cerebral blood flow (rCBF) can predict response to gabapentin treatment. METHODS: A total of 29 women with fibromyalgia and 10 healthy women (without pain) matched for age were finally enrolled in the study. Technetium-99m ethyl cysteinate dimer single photon emission computed tomography ((99m)Tc-ECD SPECT) was performed in the fibromyalgia patients and controls. A voxel-by-voxel group analysis was performed using Statistic Parametric Mapping 5 (SPM5). After treatment with gabapentin, 16 patients were considered 'responders', with decrease in pain of greater than 50% as evaluated by visual analogue scale (VAS). The remaining 13 patients were considered 'poor responders'. RESULTS: We observed rCBF abnormalities, compared to control subjects, in fibromyalgia including hypoperfusion in the left culmen and hyperperfusion in the right precentral gyrus, right posterior cingulate, right superior occipital gyrus, right cuneus, left inferior parietal lobule, right middle temporal gyrus, left postcentral gyrus, and left superior parietal lobule. Compared to responders, poor responders exhibited hyperperfusion in the right middle temporal gyrus, left middle frontal gyrus, left superior frontal gyrus, right postcentral gyrus, right precuneus, right cingulate, left middle occipital gyrus, and left declive. The right middle temporal gyrus, left superior frontal gyrus, right precuneus, left middle occipital gyrus, and left declive exhibited high positive likelihood ratios. CONCLUSIONS: The present study revealed brain regions with significant hyperperfusion associated with the default-mode network, in addition to abnormalities in the sensory dimension of pain processing and affective-attentional areas in fibromyalgia patients. Furthermore, hyperperfusion in these areas was strongly predictive of poor response to gabapentin.