Effect of intrathecal baclofen bolus injection on temporospatial gait characteristics in patients with acquired brain injury.

OBJECTIVE: To evaluate changes in temporospatial gait parameters after intrathecal baclofen (ITB) bolus administration in patients with spasticity resulting from acquired brain injury by using computerized gait analysis.
DESIGN: Case series; before-after intervention. Walking performance and spasticity in lower-extremity muscles were assessed before and at 2, 4, and 6 hours after ITB bolus injection.
SETTING: Tertiary care free-standing rehabilitation hospital. PARTICIPANTS: Consecutive sample of 28 adults with acquired brain injury due to stroke, trauma, or anoxia. INTERVENTION: A 50-microg ITB bolus injection (2 subjects received 75 microg and 100 microg, respectively). MAIN OUTCOME MEASURES: Ashworth Scale scores, self-selected gait velocity, stride length, cadence, step length symmetry, step width, and percentage of single support on the more involved side.
RESULTS: Ashworth score decreased from 2.0+/-0.5 at baseline to 1.3+/-0.3 at peak response ( P <.001), whereas gait velocity increased from 41+/-26 to 47+/-31 cm/s ( P <.001). Significant improvements also occurred in stride length ( P <.05) and step width ( P <.001). Gait velocity was the most sensitive temporospatial outcome measure for differentiating functional response to ITB bolus injection. Sixteen patients increased velocity by an average of 12 cm/s, representing a mean gain of 33% over their baseline walking speed, while 5 decreased (mean loss, -6 cm/s [52% of baseline]) and 7 were unchanged. There was a significant correlation between baseline velocity and peak change in velocity after ITB bolus ( r =.39, P <.05). Baseline Ashworth score correlated inversely with velocity, stride length, cadence, and percentage single support (r range, -.46 to -.57). No significant correlations were found between change in Ashworth score and change in any temporospatial outcome measure.
CONCLUSIONS: ITB bolus injection consistently reduces spasticity in ambulatory patients with acquired brain injury but may result in a range of changes in walking performance that can be reliably detected using computerized motion analysis. Velocity appears to be the most sensitive parameter with which to classify individual patient response. The relation between baseline gait velocity and peak change in velocity after bolus administration may have application in predicting the outcome of screening trials for pump implantation; it warrants further investigation.