This work presents a protocol to enhance prosthetic function after selective nerve transfer surgery. Rehabilitation interventions include patient information and selection, support of wound healing, cortical re-activation of sensory-motor areas of the upper limb, training of selective muscle activation, prosthetic handling in daily life, and regular follow-up assessments.
Targeted Muscle Reinnervation (TMR) improves the biological control interface for myoelectric prostheses after above-elbow amputation. Selective activation of muscle units is made possible by surgically re-routing nerves, yielding a high number of independent myoelectric control signals. However, this intervention requires careful patient selection and specific rehabilitation therapy. Here a rehabilitation protocol is presented for high-level upper limb amputees undergoing TMR, based on an expert Delphi study. Interventions before surgery include detailed patient assessment and general measures for pain control, muscle endurance and strength, balance, and range of motion of the remaining joints. After surgery, additional therapeutic interventions focus on edema control and scar treatment and the selective activation of cortical areas responsible for upper limb control. Following successful reinnervation of target muscles, surface electromyographic (sEMG) biofeedback is used to train the activation of the novel muscular units. Later on, a table-top prosthesis may provide the first experience of prosthetic control. After fitting the actual prosthesis, training includes repetitive drills without objects, object manipulation, and finally, activities of daily living. Ultimately, regular patient appointments and functional assessments allow tracking prosthetic function and enabling early interventions if malfunctioning.
High amputations of the upper limb provide a challenge for prosthetic replacement1. Aside from elbow joint function, active prosthetic systems should include opening/closing of the prosthetic hand and ideally also pronation/supination and/or wrist extension/flexion. However, the control of standard myoelectric devices usually relies on the input signals from two muscles only2. These are traditionally the biceps and triceps muscles after transhumeral amputations and the latissimus dorsi and pectoralis major muscles after glenohumeral amputations3. To control all prosthetic joints, amputees need to switch between the active joints (e.g., by using a co-contraction of the two muscles)1. While this provides a stable control paradigm, a significant restriction ensues with resulting slow and unintuitive control, which does not allow simultaneous movements of two or more prosthetic joints4. This limits the functionality of the prosthesis and is one of the reasons for high prosthetic abandonment rates after above-elbow amputations5.
To overcome limited and unintuitive control for these types of prosthetic fittings, selective nerve transfers can be utilized. This approach, also known as Targeted Muscle Reinnervation (TMR), consists in surgically establishing myo-control signals by re-routing nerves that initially served the amputated hand and arm to different target muscles within the residual limb6,7. After successful reinnervation, more selective activation of the reinnervated muscle units becomes possible8. The resulting electromyographic (EMG) activity can then be used for prosthetic control and can yield up to six control signals.
While there is a broad agreement that TMR can significantly improve prosthetic function9, selective activation and appropriate control of multiple muscles in the stump pose a challenge to patients, especially in the early post-operative period. This enhanced complexity of prosthetic control paired with the reduced multisensory feedback following amputation requires a specific rehabilitation to fully benefit from the surgical procedure. Here, a step-by-step guideline for the therapy interventions is provided based on recent recommendations10. An overview of the interventions and the estimated time they take in an ideal setting can be found in Figure 1.
Figure 1: Overview of stages within the rehabilitation process, including the milestones that mark the start of a new stage. Please click here to view a larger version of this figure.
The protocol was developed within a European Delphi study10. The assessment of its application on patients was approved by the local research ethics committee of the Medical University of Vienna and carried out according to the Declaration of Helsinki. If not mentioned otherwise, the steps described here should be carried out by an occupational therapist or a physiotherapist.
1. Pre-surgical interventions
2. Early post-surgical interventions
3. Signal training
Figure 2: Setup for surface EMG biofeedback. The therapist places an electrode on the patient's skin where the EMG signal is expected while explaining the needed movement cue (making a fist). The patient and therapist can see the patient's muscular activity (EMG) on the computer screen and use this feedback for finding the best electrode position and movement cue. Please click here to view a larger version of this figure.
Figure 3: Schematic drawing of the EMG signals displayed via biofeedback. Every channel (with a different color) is mapped to a specific muscle part and will later be responsible for a particular prosthetic movement. Good separation, as depicted here, ensures that the prosthesis only performs intended movements. Please click here to view a larger version of this figure.
Figure 4: Patient controlling a table-top prosthesis with surface electrodes mounted on his residual limb. Please click here to view a larger version of this figure.
4. Prosthetic training
5. Follow-up assessments
The described rehabilitation protocol was implemented in a clinical setting at the Medical University of Vienna, and its feasibility and outcomes were assessed in a clinical study, which was recently published9. As reported9, 30 patients participated in the trial to evaluate the feasibility of TMR surgery and subsequent rehabilitation. Figure 5 displays that out of these 30 patients, 11 underwent TMR as a pain treatment rather than a means to improve function via prosthetic fitting. Out of the remaining 19 patients originally aiming for a prosthetic fitting, five decided against it due to the high costs of the fitting (estimated between 75,000-150,000 €), insufficient time for rehabilitation, or high weight of the prosthesis. In one patient, intra-operative exploration revealed a global brachial plexus injury, making further nerve transfers impossible. This patient kept using his body-powered device. Of the remaining 13 patients undergoing prosthetic rehabilitation, 10 were available for a follow-up assessment.
Figure 5: Flowchart showing the patients included in the feasibility study. Please click here to view a larger version of this figure.
Outcomes were assessed using the Southampton Hand Assessment Procedure (SHAP)19, the Action Research Arm Test (ARAT)20,21, and the Clothespin-Relocation Test (CPRT)6,26. These assessments are commonly used tests to evaluate prosthetic function. The evaluation took place at least 6 months after the final prosthetic fitting. Additionally, patients were asked about their prosthetic wearing habits.
As described by Salminger et al.9, assessment of the 10 patients after TMR surgery revealed a SHAP score of 40.5 ± 8.1 (with a healthy upper extremity having a score of about 100) and ARAT score of 20.4 ± 1.9 (with 57 being the maximum score and 0 representing no upper extremity function) (Table 1). In the CPRT, patients were able to complete the tasks within 34.3 ± 14.4 s. They reported wearing their prosthesis daily with a wearing time ranging from 3-10 hours per day.
Outcome assesment | Score | Expected score for healthy upper extremity |
SHAP | 40.5 ± 8.1 | 100 |
ARAT | 20.4 ± 1.9 | 57 |
CPRT | 34.3 ± 14.4 s | – |
Table 1: Prosthetic function of patients following TMR surgery and rehabilitation. In the SHAP and ARAT, higher scores mean a better function, which is also indicated by less time needed in the CPRT. Total patient assessed: n = 10. Adapted with permission from Reference9.
In recent years, selective nerve transfers have been increasingly used to enhance prosthetic function27. Experienced clinicians in this field have come to appreciate that rehabilitation is essential to enable amputees to use a prosthesis after the surgical procedure skilfully27. However, there is a lack of structured therapy programs. The current protocol aimed to provide the occupational and physical therapists with the tools and structure to guide the patients throughout the long TMR process. In contrast to previous suggestions for therapy (developed for less complex nerve transfers)28, there is a stronger focus on pre-prosthetic training and the use of EMG biofeedback to allow selective muscular control.
As shown in the feasibility study9, discussing the patient's expectations is essential for post-operative success. The inclusion of highly motivated patients certainly helped to achieve the described excellent outcomes. Less compliance to the described protocol might result in reduced prosthetic function. Additionally, not all patients wish to receive a prosthetic fitting (or can afford to get one). However, TMR may still be feasible to improve neuroma or phantom limb pain since recent studies have shown the potential of nerve transfers to alleviate these conditions29,30,31. For such cases, the rehabilitation program is foreshortened. Still, we have experienced that regular training of controlled activation of the reinnervated muscles and a prosthesis can further improve the pain situation32. Here, shared decision-making is essential as some patients might wear a prosthesis for its potential to reduce pain in the long term32, while others might not be interested.
In our experience, a detailed discussion with the patient is essential to evaluate future compliance. Depending on the reinnervation time, motor learning capacity, and the patient's availability, the rehabilitation process is likely to take between 9-15 months. Suppose a patient does not strive toward the improvement of upper limb function or might make better use of another device (e.g., body-powered prosthetics). In that case, one might not consider the time (and possibly financial) commitment worth it. To save resources, we strongly recommend only including patients who express a strong interest in the procedure and only perform the surgery for functional purposes when the full rehabilitation procedure is anticipated. Finally, the costs for the surgery, therapy, and fitting should likely be covered at that point.
The described study protocol needs to be adapted for each individual based on clinical reasoning to meet their specific needs. Physical and psychological co-morbidities need to be considered and adequate treatment (e.g., psychotherapy) offered in addition to the interventions described here. In patients receiving TMR immediately after amputation, a closer screening for psychological conditions developing overtime may be needed. Apart from this, no change in the protocol is required for this group of patients. They might even progress faster in motor learning as they might still be used to bimanual activities. Within this protocol, the nerve transfers operated by the surgeon define, which motor commands need to be trained and are expected for which muscle parts. The choice of the prosthetic end device influences prosthetic training. For multi-articulated prostheses, switching between different grasp types and how to use them needs to be included in therapy, if necessary.
For patients living far away from the clinical center or those who cannot attend in-person rehabilitation regularly, adoptions in the rehabilitation protocol are needed. They include a stronger focus on home training, the possible involvement of a therapist near the patient's home, and telerehabilitation sessions via online video calls. Solutions for telerehabilitation need to provide a stable video and audio connection while fulfilling all data protection requirements. In these patients, a first visit to the clinical center should be planned at 6-9 months after surgery for signal training. The visit is usually for 1 week, with therapy sessions twice a day. In a majority of cases, good signal separation can be achieved at this time. Otherwise, another stay for signal training is needed, and the patient may get a simple sEMG biofeedback device for home training. When good signal separation is established, the prosthetist can fabricate a test socket, and the signal positions can be defined during the stay. This allows the prosthetist to create the final fitting when the patient returns home. The final prosthesis can be fitted in a second 1-week visit 1-2 months later, and prosthetic training can be initiated. Advanced prosthetic training and further follow-up visits can either happen in a remote setting or during a further visit to the center, depending on the patient's needs.
Furthermore, other surgical interventions, such as osseointegration33 to improve the mechanical interface for the prosthesis, can be combined with TMR34. If this is the case, specific interventions must be included (such as the graded weight-bearing training after osseointegration35). Additionally, while the described protocol is intended for direct prosthetic control systems (where one electrode corresponds to one movement), its principles remain the same if a pattern recognition control system is planned. The main difference in rehabilitation is that the selective activation of single muscles becomes less relevant, while particular and repeatable activation patterns of several muscles need to be trained36.
The authors have nothing to disclose.
This study has received funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation program (grant agreement No. 810346). The authors thank Aron Cserveny for preparing the illustrations used in this publication.
Dynamic Arm Plus® system with a Variplus Speed prosthetic hand | Ottobock Healthcare, Duderstadt, Germany | This prosthetic system was used together with a computer (and Bluetooth connection) for sEMG Biofeedback. Later, it was used for table top prosthetic training and as the patient's prosthetic fitting. | |
ElbowSoft TMR | Ottobock Healthcare, Duderstadt, Germany | In combination with the Dynamic Arm Plus system and a standard computer (with Windows 7, 8 or 10), this software allows the visualisation of EMG signals as well as changing settings in the prosthetic system. | |
EMG electrodes | Ottobock Healthcare, Duderstadt, Germany | electrodes 13E202 = 50 | The EMG electrodes used in this study were bipolar and included a ground and a 50 Hz filter. They were used with the Dynamic Arm Plus®. |
Folding Mirror Therapy Box (Arm/Foot/Ankle) | Reflex Pain Management Therapy Store | This box was used for mirror therapy. |