Prediction of voluntary movements of the upper extremities by resting state‐brain regional glucose metabolism in patients with chronic severe brain injury: A pilot study

Confirmation of the exact voluntary movements of patients with disorder of consciousness following severe traumatic brain injury (TBI) is difficult because of the associated communication disturbances. In this pilot study, we investigated whether regional brain glucose metabolism assessed by 18F‐fluorodeoxyglucose positron emission tomography (FDG‐PET) at rest could predict voluntary movement in severe TBI patients, particularly those with sufficient upper limb capacity to use communication devices. We visually and verbally instructed patients to clasp or open their hands. After video capture, three independent rehabilitation therapists determined whether the patients' movements were voluntary or involuntary. The results were compared with the standardized uptake value in the primary motor cortex, referring to the Penfield's homunculus, by resting state by FDG‐PET imaged 1 year prior. Results showed that glucose uptake in the left (p = 0.0015) and right (p = 0.0121) proximal limb of the primary motor cortex, based on Penfield's homunculus on cerebral cartography, may reflect contralateral voluntary movement. Receiver operating characteristic curve analysis showed that a mean cutoff standardized uptake value of 5.47 ± 0.08 provided the best sensitivity and specificity for differentiating between voluntary and involuntary movements in each area. FDG‐PET may be a useful and robust biomarker for predicting long‐term recovery of motor function in severe TBI patients with disorders of consciousness. Resting FDG‐PET of the primary motor cortex may predict upper limb movements. Glucose uptake differs between voluntary and involuntary movement at resting FDG. Glucose uptake cutoff in the primary motor cortex may help detect voluntary movements.