Quantifying muscle tone in spinal cord injury patients using isokinetic dynamometric techniques

The torque generated during a passive movement of the knee joint was used to quantify muscle tone in normal able-bodied subjects and spastic and flaccid spinal cord injury (SCI) subjects using a computerized isokinetic dynamometer. Maximum peak (Tmax) and the sum of four consecutive peaks (Tsum) were calculated for each velocity (30, 60, 120 degrees/s) and for each phase (flexing or extending) separately and compared statistically using a one-way ANOVA. Statistical significance between groups was found in Tmax FLEXION (FLX) at 60 and 120 degrees/s. Scheffe's tests revealed that the spastic group was significantly less than both the flaccid and normal groups, although the flaccid and normal groups were not significantly different from each other. The slopes of the linear regression curve of the torque-velocity data were found and compared statistically using a t-test for parallelism. In all parameters, the data increased in a linear fashion with increasing velocity of knee motion. The slope of the regression curve for the spastic group was significantly lower than that of the normal group for Tmax and was significantly lower than that of the flaccid group for Tsum while the slopes for the flaccid and normal groups were not significantly different. The ability of the entire set of variables to classify subjects into three groups (normal, spastic, and flaccid) was tested using discriminant analysis. By taking into account 7 of the 12 original variables, this multivariate technique correctly classified 100% of the spastic, 90% of the normal, but only 67% of the flaccid subjects. Separation of observations was between spastic and normal subjects was good, except for only one case. This feature could be useful when dealing with assessment of individual responses to therapeutic interventions aimed at modification of spasticity.