Surface electromyographic activity of five residual limb muscles recorded during isometric contraction in transfemoral amputees with osseointegrated prostheses

Abstract

Background

Femoral osseointegrated implants represent a new development in amputee rehabilitation, eliminating socket pressure discomfort, improving hip range of movement and facilitating prosthetic limb attachment. A clinical aspect that has not previously been reported on is the function of muscles in the residuum with implications concerning energy expenditure, hip-hiking and viability of the electrogram as a myoprocessor. Typically, amputees fitted with osseointegrated fixation have shorter residuums and weaker attachment of cleaved muscles. Function of muscle can be assessed by surface electromyography through changes in amplitude and median frequency of the signal.

Methods

Five male transfemoral amputees with osseointegrated fixations participated together with a control group comprised of ten adult males. Electrodes recorded surface electromyographic activity of five residual limb muscles or left lower limb muscles of control subjects. Isometric contractions were performed against resistance. The increase in mean rectified amplitude from resting to maximally contracting was calculated and median frequencies estimated.

Findings

The amputees were unable to maintain a maximum voluntary contraction of constant amplitude. Amplitude increase was lowest for rectus femoris and adductor magnus. The median frequency of adductor magnus was significantly greater (P = 0.02) for the amputees than intact subjects and there was a significant difference (P < 0.01) between gluteus maximus and adductor magnus for amputee subjects.

Interpretation

High electromyographic amplitude variability suggests that using residuum muscles singly as a myoprocessor might be challenging. Adductor magnus displayed a different sEMG profile compared to intact subjects indicating decreased function and neuromuscular changes. Further work into optimal muscle anchorage is required to ensure maximal mechanical performance.

Introduction

There are estimated to be about 623,000 lower limb amputees currently in the United States of America and this number is predicted to more than double over the next 40 years (Ziegler-Graham et al., 2008). Dysvascularity accounts for the majority of amputations in older individuals whereas trauma is an important cause for younger amputees. This subgroup of amputees, who have sustained trauma, although small, is important as they demand greater performance from their prosthesis over a longer period of time. Prosthetics and amputation surgery have vastly improved over the decades, however the prosthetic limb still remains inferior to its natural counterpart mechanically, aesthetically and comfort wise. Common problems reported by transfemoral amputees include socket discomfort and dissatisfaction with the knee and foot components (Klute et al., 2009, Schmalz et al., 2002).

A promising new development in prosthetics, particularly for traumatic transfemoral amputees, is the direct skeletal fixation. The main advantages of the direct skeletal fixation are that it eliminates discomfort arising from pressure of the socket on the skin and transmits force directly from the external prosthesis to the femur. In the conventional prosthesis, some of the energy will be dissipated through the soft tissue, the socket and the soft tissue–socket interface. Clinically, decreased energy expenditure, improved proprioception and greater range of hip movement have been reported (Hagberg et al., 2004, Hagberg et al., 2005, Jacobs et al., 2000, Sullivan et al., 2003). There are currently three main types of direct skeletal fixations; the Endo–Exo Femur Prosthesis (EEFT), the Intraosseous Transcutaneous Amputation Prosthesis (ITAP) and the Osseointegrated Prosthesis for Rehabilitation of Amputees (OPRA) (Aschoff et al., 2009, Brånemark et al., 2001, Pitkin, 2009, Staubach and Grundei, 2001, Sullivan et al., 2003). This paper will focus on the OPRA as it is the device used in the study. The OPRA, designed by Dr. Per-Ingvar Brånemark, has been implanted into the femurs of amputees in the United Kingdom since 1997 and currently over 100 individuals worldwide have undergone this procedure (Hagberg and Brånemark, 2009). This device involves two stages of surgery; the titanium implant is fitted during the first stage and the abutment is attached approximately six months later (Brånemark et al., 1992, Robinson et al., 2004, Ward and Robinson, 2004). During the first stage, muscle tissue is removed to expose the cut end of the femur and muscles are split to uncover the distal 5–10 cm of the femoral shaft. Suturing of the transected muscles to the periosteum (myodesis), with the adductors being attached by a modified Gottschalk procedure (drill holes are not bored), and suturing of agonist to antagonist muscles (myoplasty) is performed during the second stage (Robinson et al., 2004, Ward and Robinson, 2004).

A feature of the OPRA procedure and transfemoral amputations in general, that has not been adequately investigated is the function of muscles following amputation, although it is acknowledged by orthopaedic surgeons that retention of muscle function is an important determinant in achieving optimal gait patterns (Gottschalk, 1999, Ward and Robinson, 2004). Following a transfemoral amputation, the ability of transected muscles to generate force and consequentially moments acting on the hip joint, are radically reduced due to biomechanical and neurological changes. Clinically, these changes are reflected in the altered gait of amputees compared to non-amputees, major differences reported being a slower speed, smaller stride length, asymmetry in stance phase, increased hip extension at the end of stance, and increased lateral flexion of the trunk and greater energy consumption (Bae et al., 2007, Boonstra et al., 1994, Jaegers et al., 1995b). Muscle changes include reduction in cross-sectional area, decreased length and lever arm, altered site and tissue type anchoring cleaved muscle, unbalanced impact on agonists and antagonists, decreased compartment pressure, reduction in muscle–tendon receptors and denervation of muscle fibres. A biomechanical factor unique to direct skeletal fixations is that the absence of a socket compressing the residuum may result in greater expansion of contracting muscles radially from the femoral shaft, thereby reducing the force component along the femoral longitudinal axis. Significant atrophy of transected muscles of up to 73% has been reported in transfemoral amputees with conventional prostheses and greater atrophy was observed in the absence of adequate fixation (Jaegers et al., 1995a). There is no published study detailing changes in muscle morphology following the OPRA procedure. Clinically, decreased femoral abduction and side lurch in individuals was observed where a myodesis of adductor magnus (AM) was performed (Gottschalk, 1999). AM is an important but underestimated gait muscle, with the adductor group accounting for over 20% of the cross sectional area of the thigh (Fig. 1) (Brand et al., 1986). Neurologically, denervation of a muscle in high level transfemoral amputations with complete or partial reinnervation may result in a greater proportion of type II (fast twitch) muscle fibres, with implications for earlier onset of muscle fatigue during physical activity (Schmalbruch and Lewis, 2000).

Another recent development in prosthetics is the myoprocessor controlled prosthesis, in which the activity of the muscle, as measured by surface electromyography (sEMG), is utilised as a sensor to regulate movement of the prosthetic limb (Englehart and Hudgins, 2003). Currently, this has not been used commercially for lower limb prosthetics due in part to high levels of noise and low reproducibility of the sEMG. The portion of sEMG noise arising from movement between the electrodes and the socket is eliminated in amputees fitted with a direct skeletal fixation, thereby rendering this approach better suited for combination with a myoprocessor controlled knee component.

Specific questions this study aimed to address regarding muscle function in the residual limb were:

• How is the function of transected muscles affected in terms of magnitude of contraction and variability?

• Is there a change in motor units in the cleaved muscles?

The aims of the study were to investigate muscle activity in the residuum of amputees with direct skeletal fixations through EMG analysis. Change in magnitude of sEMG in an isometric contraction is related to force of muscle contraction and spectral analysis of the signal may provide information regarding types of muscle fibres present, with type II fibres displaying a greater median frequency than type I fibres (Merletti and Lo Conte, 1997). Same-day changes in amplitude and frequency of sEMG of five residual limb muscles were assessed in transfemoral amputees with direct skeletal fixations and a control group following an isometric maximum voluntary contraction. Differences in the sEMG amplitude and frequency between the five muscles were determined for both groups. No previous reports have been published on sEMG in individuals who have been fitted with a direct skeletal fixation. The hypotheses are that transected muscles show less increase in sEMG amplitude than intact muscles, median frequency of AM is greater in the amputee group and amputees display increased variability of sEMG amplitude compared to control subjects. This is a preliminary study designed to provide baseline information on sEMG under controlled conditions (isometric contractions only) with further work on sEMG patterns during gait in progress.