Conditioning Reflex Protocols

Traditional tenets about the hard-wired nervous system have long been dispelled with mounting evidence for activity-dependent plasticity throughout the CNS. Fascinating results from animal, and more recently, human studies have shown that even the “simplest” spinal cord reflex, the stretch reflex pathway or its electrical analog, the H-reflex, can be altered to increase or decrease in size through operant conditioning (Wolpaw 2010). In animals, a reward is provided whenever the H-reflex amplitude is above or below a threshold value. Through modulation of descending influence, the animal can gradually learn to maintain its H-reflex amplitude at a certain level. Humans can also learn to increase or decrease the size of the soleus H-reflex (Thompson et al. 2009). Some gait impairments following SCI could be associated with hyperreflexia and abnormal reflex responses in the ankle plantar flexors (Dietz & Sinkjaer 2007). The possibility that H-reflex amplitude could be down-conditioned raises the compelling question of whether such protocols may benefit individuals with SCI who present with spastic gait disorder.

This idea was recently tested in a group of 13 individuals with chronic (>8 months) motor-incomplete SCI who all were ambulatory and presented with spasticity (e.g. ³1 on Modified Ashworth Scale) and weak ankle dorsiflexion (Thompson et al. 2013). Participants were randomly assigned at a 2:1 ratio to the down-conditioning (DC) group (n=9) or the unconditioned (UC) group (n=4). Each participant completed 6 baseline sessions followed by 30 sessions (3 sessions/week) of control (UC group) or conditioning (DC group). Visual feedback was provided to the DC group to inform them of whether they were successful in reducing their H-reflex amplitude to within the target range. In the UC group, each session involved H-reflex recordings without any visual feedback or instructions about H-reflex amplitude. Note that in this study, no locomotor training was provided; training sessions consisted of only the H-reflex down-conditioning (or control protocol).

Author Year; Country
Score
Research Design
Sample Size
Methods Outcome

Thompson et al. 2013; USA
PEDro=7
RCT
Level 1
N=13

Population: 13 ambulatory participants with SCI (9M 4F); mean(SD) age: 48.4 (13.9) yrs; DOI ranging from 8 months to 50 yrs.
Treatment: Participants randomly assigned at a 2:1 ratio to the down-conditioning (DC) group (6M 3F) or the unconditioned (UC) group (3M 1F). Each participant completed 6 baseline sessions and 30 control (UC participants) or conditioning (DC participants) sessions at a rate of 3 sessions/week. Electrical stimulation of the soleus H-reflex was elicited by a 1ms square pulse stimulus.
Outcome Measures: Locomotion (participant asked to walk 10 m at comfortable speed 3 times; average walking time determined); locomotor symmetry; EMG activity; H-reflex modulation.
  1. Success (average conditioned H-reflexes significantly less for session 25-30 than baseline) rate for participants with SCI = 67%, which is slightly, but not significantly, less than that for neurologically normal participants (89%).
  2. Conditioned H-reflex for UC group as a whole showed a slight by significant increase (to (mean[SE]) 116(7)%). Down-conditioning was achieved in 6 of 9 participants.
  3. Over the 30 conditioning or control sessions, the participants’ 10m walking speeds increased by 0-123%. The increase was significant in the 6 DC participants in whom the H-reflex decreased. For the 7 participants in whom H-reflex did not decrease, walking speed increased less and not significantly.
  4. For DC participants with decreased H-reflex (n=6), locomotion became faster and more symmetrical and the modulation of EMG activity across the step cycle increased bilaterally.

Discussion

Among the 9 participants in the DC group, 6 were able to successfully down-condition their H-reflex amplitude by the last 5 training sessions. There was no reduction in H-reflex amplitude in the UC group. Across the 6 participants who could successfully down-condition their soleus H-reflex amplitude, there was a significant increase in their 10MWT speeds of 59% (range: 0-123%) along with a significant improvement in gait symmetry. For the 7 participants in whom H-reflex did not decrease, walking speed increased less and not significantly.

Conditioning reflex protocols have been published many years ago for the upper extremity in SCI (Segal & Wolf 1994) to reduce spasticity and are an important neuroscience observation. However, they have not been accepted into practice likely due to the variable results and laborious number of sessions to get a small effect. This one small RCT for the lower extremity shows similar findings as the upper extremity – that soleus spinal reflexes can be down-conditioned in about 2/3 of the participants, although a few of these participants did demonstrate large improvements in gait speed. The success rate of down-conditioning in the SCI participants was comparable to previous studies in able-bodied participants. Unfortunately, absolute values were not reported here, making the clinical significance of these results difficult to ascertain. Furthermore, the complexity of this approach may make it inaccessible for most clinicians. Nevertheless, these results are very intriguing and point towards another potential approach of directly manipulating spinal cord plasticity to enhance functional recovery.

Conclusion

There is level 1b evidence from one RCT (Thompson et al. 2013) that down-conditioning reflex protocols of the soleus could facilitate gait outcomes.