Is an instrumented spasticity assessment an improvement over clinical spasticity scales in assessing and predicting the response to integrated botulinum toxin type a treatment in children with cerebral palsy?

Abstract

To compare responsiveness and predictive ability of clinical and instrumented spasticity assessments after botulinum toxin type A (BTX) treatment combined with casting in the medial hamstrings (MEHs) in children with spastic cerebral palsy (CP). Prospective cohort study. Hospital. Consecutive sample of children (N=31; 40 MEH muscles) with CP requiring BTX injections. Clinical and instrumented spasticity assessments before and on average ± SD 53±14 days after BTX. Clinical spasticity scales included the Modified Ashworth Scale and the Modified Tardieu Scale. The instrumented spasticity assessment integrated biomechanical (position and torque) and electrophysiological (surface electromyography) signals during manually performed low- and high-velocity passive stretches of the MEHs. Signals were compared between both stretch velocities and were examined pre- and post-BTX. Responsiveness of clinical and instrumented assessments was compared by percentage exact agreement. Prediction ability was assessed with a logistic regression and the area under the receiver operating characteristic (ROC) curves of the baseline parameters of responders versus nonresponders. Both clinical and instrumented parameters improved post-BTX (P≤.005); however, they showed a low percentage exact agreement. The baseline Modified Tardieu Scale was the only clinical scale predictive for response (area under the ROC curve=0.7). For the instrumented assessment, baseline values of root mean square (RMS) electromyography and torque were better predictors for a positive response (area under the ROC curve=.82). Baseline RMS electromyography remained an important predictor in the logistic regression. The instrumented spasticity assessment showed higher responsiveness than the clinical scales. The amount of RMS electromyography is considered a promising parameter to predict treatment response.