Targeted Muscle Reinnervation: Prosthetic Management

Craig Jackman CPO, FAAOP

Brian Monroe CPO

Ryan Sheridan MS, CPO, FAAOP

Craig Jackman or an immediate family member serves as a paid consultant to or is an employee of Hanger Clinic. Brian Monroe or an immediate family member serves as a paid consultant to or is an employee of Hanger Clinic. Ryan Sheridan or an immediate family member serves as a paid consultant to or is an employee of Hanger Clinic.

This chapter is adapted from Lipschutz RD: Targeted muscle reinnervation: prosthetic management, in: Krajbich JI, Pinzur MS, Potter BK, Stevens PM, eds: Atlas of Amputations and Limb Deficiencies: Surgical, Prosthetic, and Rehabilitation Principles, ed 4. American Academy of Orthopaedic Surgeons, 2016, pp 339-350.

ABSTRACT

Targeted muscle reinnervation is a surgical technique that has been developed to improve an individual’s ability to control a myoelectric prosthesis and reduce postamputation pain. Prosthetists who attempt to fit individuals with powered prostheses after target muscle reinnervation must first be knowledgeable in the basic principles of fitting myoelectric devices. In addition, the prosthetist should understand the surgical procedure and the intended outcomes. The application of both traditional and advanced myoelectric control strategies can optimize function for this patient population. The incorporation of emerging technologies will benefit both the user and the prosthetist.

Introduction

For individuals with upper limb amputation, especially at or proximal to the transhumeral level, successful operation of a prosthesis requires the performance of control motions that are rarely analogous to actions performed before the amputation. For bodypowered prostheses, users must incorporate gross body movements such as glenohumeral flexion, scapular protraction, or biscapular protraction to flex the elbow and/or operate the terminal device. In myoelectrically controlled, externally powered systems, the actions for controlling a powered elbow can be more physiologic for individuals with transhumeral amputations if contraction of the biceps is used to control elbow flexion and the triceps is used to control elbow extension. Beyond this exception, for the other powered components of a transhumeral prosthesis (electronic wrists and terminal devices) and all the motors in an externally powered prosthesis for shoulder disarticulation, both the electromyographic (EMG) input signals and other physical body movements that are used as control inputs are nonphysiologic strategies. For example, strategies for using a myoelectrically controlled transhumeral prosthesis may include using the residual biceps and triceps muscles to control elbow flexion and extension, wrist supination and pronation, and terminal device prehension. Various mode selection strategies are required to allow the user to switch between these antagonistic pairs of movements. An example of mode selection may be an intentional co-contraction of the biceps and triceps muscles to switch active control inputs from one component to another (eg, elbow to hand). Other methods of mode selection could include activating a bump switch mounted on the exterior surface of the prosthetic socket or a momentary pull switch incorporated into the harness. Regardless of the method of mode selection, another action is required to direct input signals from one component to another. More importantly, the muscle signals used to control these various motors are rarely consistent with the action being performed.

Targeted muscle reinnervation (TMR) is a means by which users can operate their myoelectrically controlled prostheses in a manner that is more intuitive and physiologically consistent with the actions and thought processes that users had before their amputatio ns.¹^,²^,³^,⁴^,⁵^,⁶^,⁷^,⁸ Adding two to three physiologic electrode sites for the individual with a transhumeral prosthesis, and four to six physiologic sites for the shoulder disarticulation prosthesis, creates the potential for a more natural means of controlling the prosthesis. TMR also allows the user to control multiple motors intuitively and reduce the delays associated with mode selection for most prosthetic actions and movements.

Since its initial application to proximal upper limb amputation levels, multiple surgical centers have performed TMR with the primary goal of improved prosthetic control.⁹ However, subsequent clinical experience and observation suggested that patients who underwent TMR reported less postamputation pain and fewer instances of recurrent painful neuroma formation.⁹ A retrospective multicenter study that examined the role of TMR in postamputation nerve pain confirmed the potential for reducing residual and phantom limb pain.⁹ Further studies have confirmed the potential benefits of performing TMR at both the primary closure of the amputation and as a secondary procedure.¹⁰^,¹¹

The general principles of fitting individuals with upper limb prostheses who have undergone TMR, the challenges encountered during clinical fittings, and technologic advancements associated with TMR are discussed.

General Principles

TMR has moved from experimental case studies to the standard of care at many surgical centers in the United States for managing the transected nerves in limb amputation. Although postoperative care for an amputation with nerve treatment is the same as a non-TMR procedure, it is essential that the surgical and rehabilitation teams fully comprehend the surgical procedure, any benefits and risks associated with the process, the postoperative protocol, and prosthetic fitting principles and training.¹²

Muscle Recovery Period

After a successful TMR procedure, the patient will be asked to pay particular attention to their neuromuscular development of reinnervation, most notably what occurs when attempting to perform a muscle contraction that was absent before TMR.¹³ Reinnervation will occur gradually, reaching a consistently measurable EMG level as soon as 3 to 6 months after the procedure. However, EMG signals may continue to increase in magnitude, and optimal electrode locations may reorient as further development occurs in subsequent months and years after the procedure.¹⁴ During this period of reinnervation, the patient should be given a protocol of how to exercise both the reinnervated and natively innervated muscles. In addition to the typical practices of wound care and healing maintenance of the residual limb, overall strength and range of motion should also be pursued.¹³^,¹⁴

One principle often overlooked in this transition period is prosthetic wear after TMR for individuals who had been previously fitted with prostheses. Generally, it is expected that after a brief surgical recovery period, the user will resume prosthetic wear with their legacy device. Prosthesis use may be complicated by the fact that the overall limb volume may change substantially because of the removal or movement of adipose tissue during the TMR procedure as well as transient atrophy of muscles that have been deinnervated and reinnervated. Although the repositioning of adipose tissue, termed debulking, is a benefit for signal acquisition, the removal or movement of subcutaneous fat will also alter the shape of the residual limb to the degree that either major socket modifications or socket replacement becomes necessary.

Myotesting Principles

Fitting individuals with traditional myoelectric control requires that the prosthetist follow the basic strategies for myosite testing. Signal thresholds, antagonistic muscle pairs, the appropriate alignment of the electrodes, and maintenance of electrode contact on the skin surface are essential elements of these fittings. The addition of TMR sites does not alter these strategies. Depending on the level of amputation and surgical technique performed, the patient may have five or more reinnervated muscle sites. The use of pattern recognition systems, discussed later in this chapter, as a diagnostic tool can test up to eight EMG sites simultaneously to assist this pre-prosthetic phase of rehabilitation (Figure 1). The site selection of antagonistic pairs of muscles, including isolation, attaining thresholds, and proportional control, can be identified with either traditional or advanced myotesting systems.

Coactivation of Signals

TMR has created an avenue for discrete EMG signals to control discrete prosthetic movements. For example, elbow flexion and extension can be controlled independently from the opening and closing of a terminal device. These individual signals are comparatively easy to attain when selecting them in isolation and testing the residual limb for control of the desired motions. However, when all the electrodes are in contact with the user’s body, eliminating unintended coactivation of muscles is quite difficult. This is, in part, because following TMR distinct control muscles are part of the same synergistic pattern (elbow flexion and hand close, elbow extension and hand open). In addition to the surgical attempt to separate these muscles by means of adipofascial flaps,⁸ the use of pattern recognition control can eliminate the confusion caused by the unintended coactivation of muscles as such coactivation becomes part of that recognized pattern for a given movement.

Visualization of Desired Movement

A primary goal of TMR is to enable the user to provide an intuitive neuromuscular signal that is native to the action desired from the prosthesis. For the individual with a transhumeral amputation, the native neural pattern for elbow flexion and extension signals are present both before and after TMR. However, hand and wrist movements require a rerouted neural signal from a different muscle. For the individual with a shoulder disarticulation, this rerouting occurs for all elbow, wrist, and hand signals. When motor reinnervation occurs, it is unclear as to exactly what neural information reaches its destination in the motor point of the muscle. For example, after reinnervation of the median nerve to a motor point on a targeted muscle, it is expected that the user’s attempt to elicit the EMG
signal for “hand close” will send the desired neural signal to contract the targeted muscle. Generally, this attempt is successful. However, some factors may necessitate slight variations to the user’s visualized movement of their amputated limb to generate the proper corresponding EMG signals. First, the portion of the median nerve that reinnervated the target muscle may have been biased toward thumb movements more so than the second and third digits. In this case, the patient may have to visualize more thumb movement for “hand close” versus visualizing the entire hand closing. In other cases, the patient may think that their phantom limb cannot completely move into the desired position, again necessitating a variation of the desired visualization. Both circumstances require the patient, the prosthetist, and the occupational therapist to be flexible enough to try alternative motions that are normally innervated by the reinnervated peripheral nerve to determine what is most effective.¹⁴^,¹⁵ As in all myoelectric control, consistency in the activation patterns is the key to successful prosthetic operation.

FIGURE 1 Illustration of a graphical user interface screen that allows the simultaneous viewing of multiple electromyographic channels. (Courtesy of Ottobock.)

Electrode Placement and Socket Designs

Prosthetists usually have their own preferred electrodes and socket design for myoelectric fittings that they have found successful. Such components and design concepts should remain within each prosthetist’s repertoire, adding the possibility of minor modifications caused by the surgical method used, the location and number of EMG sites, and significant movement of superficial tissue during muscle contraction.¹⁵

Similar to prosthetists, surgeons have their own ideas of how to achieve successful results. Although TMR has somewhat standardized surgical principles, the results from TMR will depend on what is discovered both before and during the procedure. For example, a muscle transfer from another region of the body may be necessary to provide a viable site for reinnervation of the peripheral nerve. Such transfer may also be necessary to provide coverage over bony prominences, such as within an interscapulothoracic amputation. Alternatively, the originally targeted muscle may be determined as nonviable intraoperatively. In other surgical variations, the surgeon may opt to detach the origin of the muscle tendon to prevent proximal migration of the muscle during contraction. In all of these cases, it is essential for the prosthetist to review the surgical report and/or discuss the case with the surgeon to ensure that both appreciate (and understand) how to best design the socket.

Consideration should be given for the length of the residual limb and the location of usable EMG sites, which need to be contained within the socket interface. Traditional myoelectric prostheses are designed with two electrode sites. After TMR, at least four sites should be available. In these cases, greater surface coverage of the socket may be necessary to capture the reinnervated sites, while ensuring electrodes are placed distal to the remaining major joints.

It has been noted that in both transhumeral and shoulder disarticulation limbs, substantial movement occurs in particular areas of superficial tissue
overlying reinnervated muscle sites¹^,²^,³^,⁴^,⁵^,⁶^,⁷^,⁸ (Figure 2). Such movement is problematic because the electrodes must maintain surface contact with the soft tissue to prevent inadvertent movement of the prosthesis. Sockets may need to be modified for a tighter fit over these regions or, alternatively, use a flexible interface that can expand and still maintain contact with the skin surface. In addition to the movement of superficial tissues, subcutaneous muscle may shift during contraction. This may occur because the newly targeted muscle did not undergo myodesis or myoplasty during the initial amputation surgery. Alternatively, during TMR, the origin of the muscle may not have been detached, allowing for a floating muscle belly. Because this muscle with only a proximal attachment is reinnervated, a contraction makes the muscle move proximally toward its origin.¹² In these instances, it may be necessary to install the electrodes quite proximally on the socket.

FIGURE 2 Clinical photographs of soft-tissue movements of the transhumeral residual limb associated with generating various control signals in a patient after targeted muscle reinnervation. A, Relaxed. B, Hand close. C, Elbow up. (Courtesy of Shirley Ryan AbilityLab, Chicago, IL.)

Only gold members can continue reading. Log In or Register to continue