Objective
The aim of this study was to develop objective evaluation method which reflect the velocity dependent increase of stretch reflex in spasticity using the pendulum test and dynamic electromyography.

Method
Kinematic analysis for knee angle and angular velocity measurement, and dynamic electromyography were performed simultaneously during pendulum movement of spastic lower extremity in supine position for nineteen spastic hemiplegia patients and ten normal control subjects. Angular relaxation index (ARI), Maximum angular velocity (MAV), Angular threshold (AT) and Angular velocity threshold (AVT) were used for the evaluation of spsticity.

Results
1) ARI was 1.64⁑0.04 in normal, 1.28⁑0.14 in modified Ashworth scale (MAS) I, 1.18⁑0.16 in MAS II, 1.02⁑0.13 in MAS III, 2) MAV was 325.0⁑29.4 in normal, 301.7⁑37.8 in MAS I, 269.2⁑29.7 in MAS II, 232.4⁑28.2 in MAS III, 3) In normal subjects EMG activity did not occur, whereas in spastic hemiplegic patient a stretch reflex appears in thigh muscle. AT and AVT were 57.39⁑4.45, 256.24⁑14.07 in MAS I, 38.59⁑4.26, 184.45⁑12.7 in MAS II, 19.13⁑7.13, 136.06⁑12.88 in MAS III, respectively, 4) Correlation coefficients of the ARI (r=⁣0.786), AT (r=⁣0.960), AVT (r=⁣0.949) showed significantly negative correlations with the MAS.

Conclusion
1) AT and AVT are more sensitive parameters than ARI for documenting spasticity in hemiplegic patients. 2) An evaluation of spasticity using pendulum test and dynamic electromyography would provide consistent results with little error and would not be influenced by the change of examiner or environment. Thus we can get very reliable results using this method.