Plasticity in Spinal Locomotor Circuits

Dorn Spinal Therapy

Spine Healing Therapy

Get Instant Access

The cat's deafferented spinal cord below a low thoracic transection can generate alternating flexor and extensor muscle activity a few hours after surgery when DOPA or clonidine are administered intravenously or when the dorsal columns or dorsal roots are continuously stimulated. This is called fictive locomotion. Several weeks after a complete lower thoracic spinal cord transection without deafferenta-tion, adult cats and other mammals have been trained on a treadmill so that their paralyzed hindlimbs fully support their weight, rhythmically step, and adjust their walking speed to that of the treadmill belt in a manner that is similar to normal locomotion.375,376 Postural support alone is detrimental to subsequent locomotion, whereas rhythmic alternating movements of the limbs with joint loading seems critical to the recovery of locomotor output.377 Serotonergic and noradrenergic drugs enhance the stepping pattern378 and strychnine, through a glycinergic path, quickly induces it in spinalized animals who have trained only been to stand.379

The transected animal also transiently learns locomotor-related tasks such as stepping over an ob-ject.186 The cat retains its learned, stereotyped locomotor patterns for at least 6 weeks after treadmill training and rapidly relearns to walk on the treadmill after another period of training.380 Step training in the spinal transected cat and rat also affects the firing thresholds of motoneurons in the cord below the lesion. Changes in excitability are related to an increase in the GABA-synthesizing enzyme, GAD67, in the cord after spinal transection and glycine-mediated inhibition.381 The trained spinalized cat's lumbar cord can make other adaptations. After the nerve to the lateral gastrocnemius and soleus was cut, the lumbar locomotor circuits compensated for the induced gait deficit, a yield at the ankle during stance that produced a more forward placement of the foot and shortened the stance phase, by 8 days postneurectomy.382 The data show that compensatory changes can be attributed to spinal rather than to supraspinal pathways and that some combination of inputs from cutaneous and group I afferents led to the change in gait pattern.

Sensory feedback from cutaneous and proprioceptor inputs during stepping has a powerful affect on locomotor rhythm and muscle activation. The step phase transitions from stance to swing are triggered by afferent feedback related to extension at the hip and the unloading of leg extensor muscles.161 One possible explanation is that group Ib Golgi tendon input during early stance inhibits the generation of flexor activity. As this input wanes near the end of stance, it releases the flexor burst generating system and enables the initiation of the swing phase.

SCI.180 The movements were evoked when the subject was supine with the hips and knees in extension and when the subject was suspended over a treadmill belt. Noxious input from one hip appeared to initiate the rhythmical locomotor activity. In parallel to this human example, cats after a low thoracic spinal transection perform hindlimb stepping on a treadmill that is enabled by noxious stimulation below the lesion and hip extension caused passively by the posterior movement of the treadmill belt (see Experimental Case Studies 1-2).

Other evidence for a CPG in humans includes the occurrence of rhythmic myoclonic activity generated by a patient's transected spinal cord. Peripheral stimulation of flexor reflex afferents induced, slowed, or interrupted a subject's symmetrical 0.3-0.6 Hz rhythmic activity in extensor muscles.181 Dimitrijevic and colleagues induced step-like locomotor activity in subjects with chronic complete spinal cord injuries by epidural electrical stimulation of the posterior spinal cord below the level of the lesion.182 They placed quadripolar electrodes at vertebral levels T-11 through L-1 and measured surface EMG activity in five muscles of each leg. Nonpatterned stimulation with 6-9 volts at 25-50 Hz at the L-2 level of the spinal cord produced the most rhythmic unilateral, but occasionally bilateral, alternating flexor-extensor muscle activity. Bilateral activity was found only when the electrodes happened to be placed in the midline. The L-2 level is especially important for hip flexion and, as noted above, has also been a key level for activation of CPGs in mammals. Stimulation at T-10 produced rhythmic irregular flexor withdrawal movements. Stimulation at 100 Hz at L-2 changed locomotor-like activity to tonic muscle firing. The current probably stimulated dorsal root fibers and, perhaps, dorsal column fibers. Neuromodulation by spinal electrical stimula tion is being evaluated as a potential neuro-prosthesis for locomotion (see Chapter 4).

A parallel can be drawn between the the results of step training in the cat and in humans after a complete SCI. After a low thoracic spinal cord transection, the segmental sensory inputs discussed above have been used to train cats183-186 and rats187 to step independently on a moving treadmill belt over a range of speeds. In people with a complete thoracic SCI who are suspended with body weight support over a moving treadmill belt while therapists move their legs reciprocally through the step cycle, sensory inputs such as levels of loading on the stance leg and the degree of hip extension prior to the swing phase lead to step-like, alternating EMG activity.174,179,188-191 Some subjects swing their legs without physical assistance and take a few steps that require onlyplacement of the foot at the end of swing.174,179

Figure 1-7A shows the rhythmical EMG activity elicited from a subject with a clinically complete thoracic SCI. If considered in parallel to the spinal transected cat experiments in Experimental Case Studies 1-2, this rhythmic activity suggests the possiblity that spinal auto-maticity in humans can be driven by locomo-tor-related sensory inputs that are recognized by CPGs. Repetitive step training of a subject with a complete SCI may lead to greater amplitude of the elicited EMG bursts and improved organization, as shown in Figure 1-7B. This training-induced change supports the potential for activity-dependent plasticity in the motor pools and their motor units and functional stepping despite partial disconnection from supraspinal influences.

Thus, the lumbosacral spinal cord in humans recognizes patterned afferent input related to the step cycle and produces basic locomotor

Assisted stepping before training Minimally assisted stepping following training

Assisted stepping before training Minimally assisted stepping following training

0 seconds 7.3 0 seconds 6.9

Figure 1-7. Electromyographic (EMG) activity from the flexor and extensor muscles of the legs in a subject with a complete thoracic spinal cord injury obtained during fully assisted treadmill stepping with 40% body weight support early (A) and late (B) after training. The level of weightbearing is shown at the bottom, highest during the phase of single and double-limb stance. The EMG about the ankle and knee muscles increased in amplitude, including the medial hamstrings (MH) and vastus lateralis (VL) at the knee and the soleus (SOL) and medial gastrocnemius (MG) at the ankle over the time of training, which suggests the recruitment of more motor units. The double burst that evolved in the tibialis anterior (TA) is typical of its normal pattern of firing. The rectus femoris (RF) came on only during stance and the iliopsoas (IL) fired at onset of swing (see Chapter 6 for details about normal firing patterns). Source: UCLA Locomotor Laboratory. S. Harkema, V.R. Edgerton, B. Dobkin.

0 seconds 7.3 0 seconds 6.9

Figure 1-7. Electromyographic (EMG) activity from the flexor and extensor muscles of the legs in a subject with a complete thoracic spinal cord injury obtained during fully assisted treadmill stepping with 40% body weight support early (A) and late (B) after training. The level of weightbearing is shown at the bottom, highest during the phase of single and double-limb stance. The EMG about the ankle and knee muscles increased in amplitude, including the medial hamstrings (MH) and vastus lateralis (VL) at the knee and the soleus (SOL) and medial gastrocnemius (MG) at the ankle over the time of training, which suggests the recruitment of more motor units. The double burst that evolved in the tibialis anterior (TA) is typical of its normal pattern of firing. The rectus femoris (RF) came on only during stance and the iliopsoas (IL) fired at onset of swing (see Chapter 6 for details about normal firing patterns). Source: UCLA Locomotor Laboratory. S. Harkema, V.R. Edgerton, B. Dobkin.

synergies. The spinal cord, then, can learn.192 This training approach has been operational-ized using body weight-supported treadmill training BWSTT) in people with incomplete SCI and hemiplegic stroke (see Chapter 6).179,193-196 Although a network of CPGs may be less useful in humans than for quadripeds who need coupling between the step cycles of the forelimbs and hindlimbs, it seems unlikely that evolution would dismiss the computational flexibility offered by the interaction of spinal reflexes, CPGs, and, as discussed in the next section, motor primitives.

Was this article helpful?

0 0
The Donts of Treadmill Buying

The Donts of Treadmill Buying

Though competitive runners are advised to run on the road, there are several reasons why you should buy treadmills anyway. You might have a family which means that your schedule does not have the flexibility it once had.

Get My Free Ebook


Post a comment