The aims of this study were to identify the critical value of relaxation index of pendulum test according to the Modified Ashworth Scale and to create new measurement scale for evaluation of spasticity using pendulum test in the 127 hemiplegic patients with varing degrees of spasticity.

The results were :

1) Modified Ashworth Scale was reversely correlated with the relaxation indices of the isokinetic dynamometer pendulum test and the electrogoniometric ones, but correlation coefficients were not so high.

2) The relaxation indices of the pendulum test were not significantly different among the Brunnstrom stages.

3) The relaxation indices of the isokinetic dynamometer pendulum test were gene rally higher than those of the electrogoniometric ones according to Modified Ashworth Scale and Brunnstrom Stage, but statistically not significant.

4) When Modified Ashworth Scale was grade 1, the distribution of relaxation indices was mostly distributed above 1.0. But we could not identify the ranges of the relaxation indices when Modified Ashworth Scale was 2 and 3 because the distribution of those relaxation indices was greatly wide.

5) Concordance rates of relaxation indices and Modified Ashworth Scale in 17 follow up patients were also so low (24%).

In the above study, when Modified Ashworth Scale was grade 1, the distribution of relaxation indices was mostly distributed above 1.0, but we could not identify the ranges of the relaxation indices when Modified Ashworth Scale was 2 and 3 because the distribution of those relaxation indices was greatly wide. Also we could not create new measurement scale for evaluation of spasticity using pendulum test.

So, when measuring spasticity using pendulum test - especially Modified Ashworth Scale is high grade (ie, 2 or 3) - we would better be cautious about other patient's clinical status (synergic pattern, etc), but further study for contributing factors to pendulum test was also needed.