How does the brain imagine arm movement after reverse shoulder arthroplasty? A fMRI study.

Reverse shoulder arthroplasty (RSA) a prosthesis technique characterized by the inversion of joint surfaces, usually results in a satisfying recovery of shoulder ROM, but, due the profound biomechanical and proprioceptive modifications, may be associated with incomplete shoulder control. Such considerations led us to ask how the central nervous system behaves in the case of RSA, and consequently to analyze its activity during voluntary RSA shoulder movements. Nine voluntary subjects undergone unilateral RSA (6-24 months post-surgery) participated in the study. Each subject after the assessment of shoulder ROM and sensory-motor performance, underwent aBOLD fMRI session in which he/she performed standardized motor tasks with the RSA and the contralateral healthy shoulder. Although fMRI signal enhancements occurred in the contralateral primary pre- and post-rolandic, associative parietal, insular, temporal and prefrontal areas, and in the ipsilateral cerebellum, we limited the comparisons to the primary sensory motor cortices and to the cerebellum in which, according to the literature, there is a bilateral symmetry of proximal limb representation. The comparison of BOLD enhancements elicited in the primary motor cortex during RSA shoulder movement evidenced a contralateral recruitment, more intense and extended than the one triggered by the movement of the healthy shoulder. The cerebellar activation was mainly localized to the inferior and superior lobules ipsilateral to the moved shoulder, however while no intervention related difference in the inferior lobules was appreciated, a significant asymmetry resulted evident in the upper ones. In fact, during RSA shoulder movement, the ipsilateral upper lobules were less active than their counterparts, intensely activated during healthy shoulder movements These findings suggest crossed reciprocal cerebello-cortical interactions and different functional roles of upper and lower cerebellar lobules. In addition, considering that almost all the subjects evidenced a poor proprioceptive recovery, RSA techniques should aim to preserve proprioceptive afferents and improve sensory feedback.