Evaluating large saccades in patients with brain-stem or cerebellar disorders.

Clinicians conventionally test saccades at the bedside by noting the accuracy, initiation time, and speed of large movements, with the patient's head stationary. Partly for methodological reasons, laboratory analysis of saccades has mainly focused on movements of 20 degrees or less. By measuring the velocity waveform of large saccades, it is possible to examine more closely the way in which brain stem and cerebellum guide the eye to the target. Large saccades made by healthy humans show a positively skewed velocity profile. Slow saccades made by patients with brain-stem disorders show a prolonged plateau of low velocity. Some patients with cerebellar disorders may show increased acceleration and deceleration of saccades. Each of these velocity waveforms can be modeled by changing the parameters that describe medium-lead burst neuron firing. In certain other brain-stem and cerebellar disorders, transient decelerations or premature terminations of saccades occur; such velocity waveforms cannot be modeled solely by changing the parameters that describe burst neuron firing. Instead, it is necessary to postulate dysfunction of the mechanism that normally inhibits pontine omnipause neurons, thereby permitting burst neurons to discharge until the saccade is completed. Analysis of large, abnormal saccades calls for application of novel techniques to identify the beginning and end of the saccadic pulse command.