Correlation between parameters of spinal cord impact and resultant injury.

It is difficult and expensive to produce and maintain animals with experimental spinal cord injuries. In order to reduce the expenditure of time, money, and animals, a method of predicting the final neurologic deficit from the mechanical parameters of the initial injury is needed. An attempt was made to ascertain the mechanical parameter(s) of a spinal cord injuring impact which best predict(s) the extent of the subsequent injury. Twelve rats were laminectomized and the spinal cords contused with an impactor which recorded force and cord surface displacement. Spinal cord lesion volume was measured after killing at 21 days. The records of displacement and force were used to generate velocity, momentum, power, and energy. The maximum values of the six descriptors of the impact were checked for linear statistical correlation with lesion volume. The nonparametric correlations of the impact descriptors with gait scores from other work were also examined. All descriptors correlated at the 1% level many times; force and displacement correlated at the 1% level most of the time. The displacement of one cord surface with respect to the other was judged to be the most useful parameter because it correlated very nearly as well as force with the subsequent measures of trauma and better than the others (but perhaps not significantly better), and because it is technically easier to measure and control.