In a recent study published in Nature Biomedical Engineering, researchers from the United States of America (USA) used commercially available electrodes for closed-loop spinal cord stimulation (SCS) in three individuals with transtibial amputation. They found that the approach helped evoke somatosensations from the missing foot, leading to improved balance, gait stability, and reduced phantom limb pain (PLP) in the patients.
Background
About 0.15 million individuals in the USA undergo lower-limb amputation annually, resulting in significant challenges such as mobility issues and PLP. Although treatments such as prosthetic limbs and pharmaceuticals exist, individuals continue to face falls, gait issues, and persistent PLP. These problems are linked to the disruption of somatosensory feedback from the missing limb. Therefore, restoring somatosensations in the amputated limb may alleviate these issues by addressing sensorimotor mismatches and improving functional outcomes.
Evidence suggests that electrically stimulating the peripheral nerves may help induce sensations in the missing limb and provide tactile feedback. However, the clinical adoption of this approach is limited by its surgical complexity and may be especially challenging in patients with peripheral neuropathy. SCS emerges as a promising alternative to peripheral nerve stimulation.
Previous studies have demonstrated the applications of this well-established technique in the reduction of chronic pain as well as the restoration of sensations in the missing upper limb.
Therefore, researchers in the present study aimed to leverage SCS to restore sensations in the missing foot in patients who underwent below-knee amputation owing to trauma or diabetic peripheral neuropathy.
About the study
The present study included three individuals (ages between 21 and 70 years) who underwent unilateral transtibial amputation about 3–7 years ago. The exclusion criteria were pregnancy/lactation, glycated hemoglobin >8, the presence of metal/medical device implants, and the use of anticoagulants.
In the included patients, commercially available SCS leads were implanted percutaneously in the thoracolumbar epidural region to stimulate the lateral lumbosacral spinal cord. The stimulation pulses were charge-balanced, biphasic, and made of symmetric cathodic and anodic phases. Multiple testing sessions, up to 6 hours each, were conducted to identify electrode contacts that evoked sensations in the missing foot.
A closed-loop system was developed, modulating SCS based on pressure signals recorded from a shoe insole under the prosthetic limb. Real-time somatosensory feedback from the system was used to monitor balance, gait, and PLP over the weeks following implantation.
The location of the evoked sensations (mechanical, movement, tingle, and temperature) was detected, and the quality of sensation was assessed using various descriptors grouped as naturalistic or paresthetic. The change in amplitude required to help the participants identify the more intense stimulus with 75% probability was defined as “just-noticeable difference” (JND).
Linear regression was used to determine the potential relationship between stimulation amplitude and apparent magnitude. A sensory organization test (SOT) was used to evaluate balance, functional gait assessment (FGA) scores were used to quantify gait stability, and the McGill pain questionnaire (MPQ) and visual analog scale (VAS) were used to assess PLP.
Results and discussion
Beyond the first two weeks of the study, all three participants were found to regain sensations in the missing foot (including the toe and heel). Missing limb sensations required higher stimulation amplitudes but were always found to be accompanied by residual limb sensations. The participants experienced a combination of naturalistic and paresthetic descriptors in varying proportions, with no apparent relationship with frequency or intensity of stimulation.
Detection thresholds of the stimulus were found to vary from 0.6 to 4 mA across participants, with multipolar stimulation showing higher thresholds than monopolar stimulation. JNDs were found to be in the range of 0.05 to 0.3 mA. Perceived magnitude of stimulation increased nearly linearly with stimulation amplitude across all the participants and electrodes.
Post-SCS, participant numbers 2 and 3 achieved higher SOT scores and showed a reduction in falls as compared to baseline scores, indicating improved balance. In the gait-stability analysis, a clinically significant improvement (>4 points) was observed in the FGA score of participant number 3.
PLP was found to reduce 50% from the baseline in participant numbers 1 and 3 but was less than one point on VAS in participant number 2. MPQ analysis showed a clinically meaningful decrease in PLP in participant numbers 1 and 2. While participant number 3 showed an initial decrease followed by an increase in pain scores over the weeks, but an overall decrease in PLP episodes was reported.
Conclusion
This is the first study to demonstrate the restoration of sensations in patients with diabetic peripheral neuropathy-related amputations. Thus, independent of the cause of amputation, the findings highlight lumbosacral SCS as a promising, clinically viable intervention for restoring sensations and enhancing function in lower-limb amputees, ultimately improving their quality of life. Further research needs to be conducted in the future to confirm these findings in a larger group of participants, while including additional sham stimulation, assessor blinding, pre-assessment of PLP, and regular reporting of pain scores.