The SCATS produced a valid measure of 3 distinct types of spastic motor behaviors in SCI and may provide a complementary tool for measuring spastic hypertonia. Such a measure is valuable because current assessment tools do not differentiate between the different types of spastic motor behaviors that manifest after SCI. Distinguishing the 3 spastic reactions using an efficient and valid clinical tool may help guide management of spastic hypertonia in SCI.

The physiological basis of flexion spasms in individuals after spinal cord injury (SCI) may involve alterations in the properties of spinal neurons in the flexion reflex pathways. We hypothesize that these changes would be manifested as progressive increases in reflex response with repetitive stimulus application (i.e., "windup") of the flexion reflexes. We investigated the windup of flexion reflex responses in 12 individuals with complete chronic SCI. Flexion reflexes were triggered using trains of electrical stimulation of plantar skin at variable intensities and inter-stimulus intervals. For threshold and suprathreshold stimulation, windup of both peak ankle and hip flexion torques and of integrated tibialis anterior electromyographic activity was observed consistently in all patients at inter-stimulus intervals < or =3 s. For subthreshold stimuli, facilitation of reflexes occurred only at intervals < or =1 s. Similarly, the latency of flexion reflexes decreased significantly at intervals < or =1 s. Patients that were receiving anti-spasticity medications (e.g., baclofen) had surprisingly larger windup of reflex responses than those who did not take such medications, although this difference may be related to differences of spasm frequency between the groups of subjects. The results indicate that the increase in spinal neuronal excitability following a train of electrical stimuli lasts for < or =3 s, similar to previous studies of nociceptive processing. Such long-lasting increases in flexion reflex responses suggest that cellular mechanisms such as plateau potentials in spinal motoneurons, interneurons, or both, may partially mediate spinal cord hyperexcitability in the absence of descending modulatory input.

Extensor spasms, which are a significant component of spasticity in spinal cord injury (SCI), were investigated in an attempt to identify the role that hip proprioceptors play in triggering an extensor reflex response. In ten SCI subjects, a controlled hip extension movement was imposed on one leg while the knee and ankle were held in an isometric position using an instrumented leg brace. Isometric joint torques of the hip, knee, and ankle were measured following a constant velocity (30 degrees /s), 45 degrees -75 degrees extension movement of the hip that was applied using the motor of a Biodex rehabilitation/testing system. Electromyograms (EMGs) from four to eight muscles were also recorded during the ten movement trials. The stereotypical torque response to an imposed hip extension consisted of hip flexion, knee extension, and ankle plantarflexion, although all components were not observed in every subject. EMGs indicated coactivation at the knee and ankle joints, with extensor activity generally outlasting flexor activity. These observations are consistent with clinical descriptions of extensor spasms. In contrast, the response to imposed hip flexion, which was observed in six of the ten subjects, comprised hip extension, knee flexion and ankle extension. This difference between the response to hip flexion and the response to extension indicates a specificity of the reflex, suggesting that organized pathways for coordinating leg movements are involved.

We have reported earlier that externally imposed ankle movements trigger ankle and hip flexion reflexes in individuals with spinal cord injury (SCI). In order to examine the afferent mechanisms underlying these movement-triggered reflexes, controlled ankle movements were imposed in 17 SCI subjects. In 13 of these subjects, reflex torques were recorded at the hip, knee and ankle in response to 5 ankle movement ranges, and 4 movement speeds. Subjects were tested using both ankle plantarflexion and dorsiflexion movements. The principal outcome measure, peak hip flexion torque of the induced reflexes, was used for comparing the effects of movement range and speed on the reflex response. We found that movement-triggered reflexes were sensitive to the angular range of ankle deflection, but insensitive to the velocity of the movement. Movement amplitudes sufficient to trigger hip and ankle flexion were routinely associated with increases in ankle passive force, suggesting that force-sensitive receptors participated in the reflex response. However, increases in angular range also corresponded to increases in muscle length, making it difficult to distinguish whether the response was triggered by a load-sensitive receptor (e.g., Golgi tendon organ or muscle free nerve ending) or a position-sensitive receptor responsive to absolute ankle angle (e.g., muscle spindle secondary afferent). The absence of velocity dependence of the reflex suggested that spindle Ia afferents were not major contributors. These results suggest movement-triggered reflexes originate in muscle receptors that are sensitive to either absolute muscle length, to muscle force or to both. Although receptors that are sensitive to absolute muscle length cannot be excluded with certainty, the finding that reflex responses require that ankle movements elicit an increase in passive force argues for a prominent role of nonspindle mechanoreceptors, such as group III/IV muscle afferents. These afferents are activated preferentially as muscles are stretched to near maximum length, and they appear to have potent reflex effects in spinal cord injury.

Local sign withdrawal, a reflex to direct the limb away from noxious cutaneous stimuli, is thought to be indicative of a modular organization of the spinal cord. To assess the integrity of such an organization of the spinal cord in chronic human spinal cord injury (SCI), we tested the electromyogram (EMG) and joint torque responses to cutaneous stimuli applied to 6 locations of the leg in 10 SCI volunteers and 3 spinal-intact controls. The 6 locations included the medial arch of the foot, the second metatarsal, the dorsum, the region over the sural nerve at the lateral malleolus, and the anterior and posterior aspects of the lower leg. Although spinal-intact subjects demonstrated local sign withdrawal, the data from SCI subjects indicated that an invariant flexion response pattern was produced regardless of stimulus location. Ankle dorsiflexion and hip flexion were produced in all subjects at all locations and no difference in the ratio of hip:ankle torques could be detected for the 6 test locations. A windup-crossover test, employing a sequence of 6 stimuli at 1-s intervals was used to assess whether common neuronal pathways were responsible for the loss of modular organization. An additional 10 SCI volunteers were tested using stimuli in which the stimulus location was switched between the 2nd and 3rd stimulus of the test sequence. The response to the crossover stimulus more closely resembled the response to the 3rd stimulus of a windup sequence than a response without conditioning stimuli. These results indicate that increased excitability produced by windup at one stimulus site is maintained at the 2nd site. This observation suggests that deep dorsal horn neurons, typically associated with musculotopic mapping, may be reorganized in chronic spinal cord injury.