Botulinum Toxin Modulates Posterior Parietal Cortex Activation in Post-stroke Spasticity of the Upper Limb

Abstract

Post-stroke spasticity (PSS) is effectively treated with intramuscular botulinum toxin type A (BoNT-A), although the clinical improvement is likely mediated by changes at the central nervous system level. Using functional magnetic resonance imaging (fMRI) of the brain, this study aims to confirm and locate BoNT-A-related changes during motor imagery with the impaired hand in severe PSS. Temporary alterations in primary and secondary sensorimotor representation of the impaired upper limb were expected. Thirty chronic stroke patients with upper limb PSS undergoing comprehensive treatment including physiotherapy and indicated for BoNT treatment were investigated. A change in PSS of the upper limb was assessed with the modified Ashworth scale (MAS). fMRI and clinical assessments were performed before (W0) and 4 weeks (W4) and 11 weeks (W11) after BoNT-A application. fMRI data were acquired using 1.5-Tesla scanners during imagery of finger-thumb opposition sequences with the impaired hand. At the group level, we separately modeled (1) average activation at each time point with the MAS score and age at W0 as covariates; and (2) within-subject effect of BoNT-A and the effect of time since W0 as independent variables. Comprehensive treatment of PSS with BoNT-A significantly decreased PSS of the upper limb with a maximal effect at W4. Task-related fMRI prior to treatment (W0) showed extensive activation of bilateral frontoparietal sensorimotor cortical areas, bilateral cerebellum, and contralesional basal ganglia and thalamus. After BoNT-A application (W4), the activation extent decreased globally, mostly in the bilateral parietal cortices and cerebellum, but returned close to baseline at W11. The intra-subject contrast revealed a significant BoNT-A effect, manifesting as a transient decrease in the activation of the ipsilesional intraparietal sulcus and superior parietal lobule. We demonstrate that BoNT-A treatment of PSS of the upper limb is associated with transient changes in the ipsilesional posterior parietal cortex, possibly resulting from temporarily altered sensorimotor upper limb representations.

Keywords: botulinum toxin; functional magnetic resonance imaging; motor imagery; neuronal plasticity; spasticity; stroke.

Figure 1

Group-wise stroke lesion map. The color-spectrum overlay represents an unthresholded sum of normalized individual stroke lesion masks on top of an MNI 152 standard space T₁-weighted template. Cold colors (blue) indicate low overlap (~1–2 subjects), hot colors (yellow-red-white) indicate frequent overlap (up to 16 subjects). Right is right according to neurological convention.

Figure 2

Effect of BoNT-A treatment on global MAS scores. The edges of the box represent the 25th and 75th percentiles, the horizontal thick line inside the box represents the median, and the whiskers represent the maximum and minimum values.

Figure 3

Mean activation at each time point. The red-yellow color overlays represent Z statistical maps thresholded cluster-wise at Z > 3.0 and cluster-wise corrected p < 0.05. An averaged T₁-weighted image was used as a background. The top panel shows the mean activation at week 0 (W0), the middle panel shows activation at 4 weeks after BoNT-A injection (W4), whereas the bottom panel displays activation at 11 weeks after BoNT-A injection (W11). Right is right according to neurological convention.

Figure 4

Treatment-related activation change. The red-yellow color overlay represents a Z statistical map of the negative BoNT effect (i.e., transient activation decrease at W4), thresholded cluster-wise at Z > 2.3 and cluster-wise corrected p < 0.05. On the right, the statistical map has been superimposed onto a reconstructed average brain surface. sup., superior view; post., posterior view. Remaining conventions (see Figure 2).