Tendon reflexes for predicting movement recovery after acute spinal cord injury in humans.

Objective: Use the tendon reflex to examine spinal cord excitability after acute spinal cord injury (SCI), relating excitability findings to prognosis.

Methods: We conducted repeated measures of reflex responses to mechanical taps at the patellar and Achilles tendons of the lower limbs, and the wrist flexor tendons of the upper limbs in persons with acute SCI, beginning as early as the day of injury. The single largest EMG response (peak-to-peak) for each site was recorded. Subjects were compared based on level of injury and final neurologic status of lower limb motor function (i.e. absence of any voluntary recruitment in a lower limb muscle: motor-complete; voluntary recruitment in 1 or more lower-limb muscles: motor-incomplete).

Results: We studied 229 subjects with acute SCI. Persons with injury to the cervical or thoracic spinal cord and who were (or became) motor-incomplete showed large tendon responses, even at the time of initial evaluation. In combination with larger tendon response amplitudes, the presence of the 'crossed-adductor' response to patellar tendon taps at the acute stage was highly predictive of functional motor recovery following SCI. In marked contrast, tendon responses were small (e.g. < 0.1 mV) or absent in persons with acute, motor-complete injury (and which remained motor-complete), and the crossed-adductor response was never seen. Reflex amplitudes and the incidence of the crossed-adductor response increased somewhat over time in persons with motor-complete SCI, but did not approach the values seen in motor-incomplete subjects.

Conclusions: Taken together, tendon response amplitude and reflex spread were sensitive and specific indicators of preserved supraspinal control over lower limb musculature in subjects with acute SCI. A simple algorithm using these outcome measures predicted a 'motor-complete' status with 100% accuracy, and a motor-incomplete status with accuracy exceeding 91%.