Leg muscle activity during walking with assistive devices at varying levels of weight bearing¹

Article Outline

• Abstract
• Methods
  • Participants
  • Experimental design
  • Device fitting and familiarization
  • Testing protocol
  • Muscle activity
  • Forceplate instrumentation
  • Treatment of the data
  • Statistical analysis
• Results
  • Muscle activity
    • Vastus lateralis
    • Biceps femoris
    • Anterior tibialis
    • Soleus
  • Force characteristics during assisted walking
• Discussion
• Conclusions
• References
• Copyright

Abstract 

Clark BC, Manini TM, Ordway NR, Ploutz-Snyder LL. Leg muscle activity during walking with assistive devices at varying levels of weight bearing. Arch Phys Med Rehabil 2004;85:1555–60.

Objective

To evaluate the muscle activation patterns at varying levels of weight-bearing forces during assisted walking with an axillary crutch and a recently designed device that allows weight transfer through the pelvic girdle (ED Walker).

Design

Descriptive, repeated measures.

Setting

University-based research laboratory.

Participants

Twelve healthy volunteers (age, 39.6±13.6y).

Interventions

Not applicable.

Main outcome measures

Electromyographic activity was recorded from the anterior tibialis, soleus, biceps femoris, and vastus lateralis muscles on a test leg during assisted axillary crutch and ED Walker ambulation. Force platform readings measured weight-bearing load (non, light, heavy). These values were normalized to normal walking gait.

Results

In the vastus lateralis and soleus muscles, both devices allowed for approximately 50% and 65% reductions in electromyographic activity during the non-weight-bearing condition. During crutch ambulation, electromyographic activity of the soleus was significantly reduced compared with that required for normal walking at all levels of weight-bearing load. In the vastus lateralis for the weight-bearing conditions, the ED Walker required significantly higher electromyographic activity than crutch ambulation (light: 105.0%±12.3% vs 72.7%±10.1%; heavy: 144.8%±23.5% vs 100.0%±13.5%). Both devices required similar peak vertical ground reaction forces during the heavy weight-bearing conditions (crutch: 75%±1.6%; ED Walker: 73%±1.8%), whereas axillary crutch gait produced less force than the ED Walker in the light condition (32%±2.0% vs 48%±1.6%).

Conclusions

During walking with assistive devices, muscle activation patterns varied with weight-bearing load. The leg extensor muscles appeared to incur a greater reduction in muscle activity when compared with their flexor counterparts. Additionally, the ED Walker and axillary crutch differed with respect to their muscle activity levels and weight-bearing characteristics. Clinically, knowledge of these muscle activity and force characteristics may aid in the decision-making process of prescribing a device type and timeline to follow in restoring weight-bearing loads.

Keywords:  Atrophy, Disuse, Crutches, Rehabilitation, Walkers, Walking

INJURIES AND DISABILITIES associated with the lower extremities frequently result in the impairment of ambulation, for which walking aids are commonly prescribed in an attempt to restore locomotory function.1 In general, these ambulatory devices allow for a reduction in the weight-bearing load placed on the affected or injured limb.2 One of the most commonly prescribed ambulation devices is the axillary crutch, which allows patients to transfer as much as 100% of body weight to the arms and axillary region during swinging gaits.1

In many cases, one of the negative side effects of using an axillary crutch is the inactivity of the associated limb skeletal muscle. The predominantly adaptive response to inactivity is skeletal muscle atrophy (a reduction in muscle size primarily due to reductions in the contractile proteins actin and myosin).3 Experimentally unloading the quadriceps femoris muscle group, using a limb suspension model that involves subjects performing ambulatory activity with crutches, causes reduced cross-sectional area of about 0.4% per day, or 14% to 16% decrements over a 5- to 6-week period.3, 4, 5 Additionally, strength losses in the range of 0.6% per day are usually observed.3 These disuse-induced deficits in skeletal muscle function persist for prolonged periods despite the resumption of normal activity levels and physical rehabilitation.6 Thus, because muscle strength is vital to physical functioning,7 the mitigation of muscle dysfunction during periods of unweighting and disuse are imperative.3

Recently, an ambulation device (ED Walker)^a has been developed that allows for weight transfer through the pelvic girdle (fig 1). The ED Walker differs substantially from the standard axillary crutch: the base of support with the ED Walker is located on the medial side of the affected limb as opposed to the lateral side, and the knee joint is flexed more during ambulation with the ED Walker. Therefore, variations in muscle activity and ground reaction forces between these modes of ambulation would be expected.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 1. 

  (A) The ED Walker is a newly designed assisted walking device that allows for weight transfer through the pelvic girdle. (B) A subject using the ED Walker device.

Despite the widespread use of axillary crutches and other ambulation devices, muscle activation patterns during walking with assistive devices have received little scientific attention.8 To our knowledge no study has evaluated the muscle activity of the lower-limb muscles during axillary crutch ambulation. Therefore, the aims of our study were to characterize the muscle activation patterns in healthy adults during assisted walking using the axillary crutch and to compare these characteristics with assisted walking with the ED Walker. Additionally, we wanted to determine how these devices differed with respect to musculoskeletal loading during self-selected variations of weight-bearing (non, light, heavy).

Back to Article Outline

Methods 

Participants 

Twelve subjects (7 men, 5 women) were recruited from a university and surrounding community to participate in this study. To ensure a wide age range of participants, we recruited 3 subjects each from the third, fourth, fifth, and sixth decades of life (23.3±2.5y, 33.0±4.4y, 46.6±1.5y, 56.7±4.9y, respectively) (total mean age, 39.6±13.6y). All subjects were healthy and recreationally active adults, with no current orthopedic limitations (mean weight, 79.7±20kg; mean height, 165±11.6cm). The experimental protocol was approved by the university’s institutional review board, and before testing all subjects provided written informed consent.

Experimental design 

Subjects participated in an orientation to become familiarized with assisted walking using the crutches and the ED Walker. Subjects then performed assisted walking with both devices at 3 varying weight-bearing levels (non, light, heavy). Additionally, subjects performed normal (unassisted) walking. Surface electromyographic activity was recorded from the vastus lateralis, biceps femoris, soleus, and anterior tibialis muscles on the test leg during trials.

Device fitting and familiarization 

When the subjects reported to the laboratory, axillary crutch and ED Walker length were adjusted to match each subject’s height. For the crutches, approximately 1 hand width was allowed between the axilla and top of the crutches.9 For the ED Walker, approximately 3 to 5cm were allowed between the device seat and pelvic girdle when standing erect (manufacturer’s instructions). Next, subjects followed a familiarization protocol for assisted walking on the axillary crutches and the ED Walker; subjects were taught how to remove 100% of their body weight from a test leg with both devices (swing through gait pattern) and how to perform partial-weight-bearing gaits (crutches: 3-point gait; ED Walker: 2-point gait).1 Subjects were allowed to practice walking with light and heavy weight-bearing loads, with forceplate feedback provided. Once the subjects felt comfortable performing the various types of ambulation, they practiced performing the tasks at a set cadence of 1 stride per 1.5 seconds. During all testing, this stride pace was kept constant via a metronome and investigator feedback. Again, subjects were allowed unlimited time to practice the tasks, and testing did not proceed until subjects felt comfortable with all types of ambulation.

Testing protocol 

During testing, subjects were first asked to perform 2 trials of normal walking gait (at least 10 stride cycles) while electromyographic activity was collected from vastus lateralis, anterior tibialis, soleus, and biceps femoris muscles over strides 4 to 8 and used to normalize the assisted walking electromyography. Next, subjects were asked to perform a similar amount of ambulation with the axillary crutch and the ED Walker (the order was counterbalanced among subjects). During these trials, subjects were randomly asked to perform 3 trials of swing-through gait (non-weight-bearing) and gaits with a self-selected “light” and “heavy” weight-bearing load (10-stride cycles). Two trials were performed for each ambulatory condition. During the partial weight-bearing loads, subjects were instructed to ambulate placing a light and a heavy load on the test leg. We did not ask subjects to place a specific amount of weight on the test leg, because one of our interests was to determine to what extent the devices varied with respect to self-selected loads. During the orientation trials, we provided feedback to subjects if the conditions (light vs heavy) did not differ with respect to their ground reaction forces. During all trials, the peak vertical ground reaction force was recorded on the seventh stride cycle by force platforms that were embedded in the ground.

Muscle activity 

We evaluated the electromyographic activity of the vastus lateralis, biceps femoris, soleus, and anterior tibialis because of their role in locomotion and physical functioning.10 To assess muscle activity, we used surface electromyography, which is commonly used to assess muscle fiber action potential activity in skeletal muscle.11 Expression of the electromyographic signal in the time domain allows for evaluation of neuromuscular activation patterns, because a greater amplitude appears to be primarily due to an increase in the number of motor units recruited and increased motor unit discharge rate.12

After familiarization, 1 leg was prepped as the test leg. Before electrode application, the skin was shaved, abraded, and cleaned with alcohol to minimize skin impedance. Electromyographic signals were recorded with Ag/AgCl bipolar surface electrodes (diameter, 4cm; interelectrode distance, 25mm) from the vastus lateralis, anterior tibialis, soleus, and biceps femoris muscles. Reference electrodes were placed with respect to the differential electrodes on bony prominences. Electrode placement was chosen based on Cram and Kasman’s standardized electrode placement atlas.13

The analog signal was preamplified 100 times with a BioAmp 100^b and then amplified 10 times with the use of a Cyber Amp 380^b (total gain, 1000). The signal was band-pass filtered between 10 and 600Hz. The analog signal was digitized at 1000Hz with an analog-to-digital board via a data acquisition card.^c The raw electromyographic signal was saved for subsequent analysis. To eliminate artifacts in the electromyographic signal, lead wires were secured to the subject to minimize movement.

Forceplate instrumentation 

Vertical ground reaction forces during ambulation were collected with a force platform.^d Analog signals were collected with the same data acquisition system and sampling rate as the electromyographic activity.

Treatment of the data 

Interference electromyographic data from each trial were full-wave rectified, and total electromyographic activity over 5 stride cycles was determined (strides 4–8) using LabView software.^e Average electromyographic activity from the 2 trials was calculated. The peak vertical ground reaction force was calculated and averaged using LabView. These values were used for all statistical analyses.

Statistical analysis 

To evaluate differences in electromyographic activity between the axillary crutch and ED Walker at each respective loading condition, a repeated-measures analysis of variance (ANOVA) was conducted (dependent variable: electromyographic activity; within-subject independent variable: ambulation condition [unassisted walking, axillary crutches, ED Walker]). Additionally, a repeated-measures ANOVA was used to evaluate differences in weight-bearing characteristics between the 2 ambulation devices (dependent variable: force; within-subject independent variable: weight-bearing condition [unassisted walking, axillary crutches, ED Walker]). For graphic representation, the electromyographic activity and forces recorded during assisted walking were normalized to that of unassisted walking (expressed as a percentage of unassisted walking). Significant main effects were investigated with Sidak post hoc tests, with a preset α level of significance of .05. All data are reported as mean ± standard error (SE), unless otherwise noted. The SPSS statistical package^f was used for data analysis.

Back to Article Outline

Results 

Muscle activity 

Both devices allowed for significant reductions in vastus lateralis electromyographic activity during the swing-through (non-weight-bearing) gait when compared with normal walking (P≤.05) (fig 2A, table 1). Additionally, during the light weight-bearing condition, vastus lateralis electromyographic activation was reduced as compared with normal walking (P≤.05) (fig 2A, table 1). In both the light and heavy weight-bearing conditions, the ED Walker yielded 45% higher levels of vastus lateralis electromyographic activation than the axillary crutch (P≤.05) (fig 2A, table 1).

• 
  • View Large Image
  • Download to PowerPoint

• Fig 2. 

  Electromyographic activity of the (A) vastus lateralis, (B) biceps femoris, (C) tibialis anterior, and (D) soleus muscles during assisted walking with the axillary crutch and ED Walker at 3 different weight-bearing loads. The black horizontal lines at 100% represent electromyographic activity required for normal walking. Abbreviation: EMG, electromyographic activity. ∗ED Walker greater than axillary crutch (P≤.05); ^†walking with assisted device significantly different from that of normal walking (P≤.05).

Table 1. Electromyographic Activity During Varying Levels of Weight-Bearing Load During Assisted Walking With Axillary Crutches and the ED Walker

NOTE. Values are mean ± SE and are expressed as a percentage of unassisted walking.

Abbreviations: AT, anterior tibialis; BF, biceps femoris; SOL, soleus; VL, vastus lateralis.

During the light weight-bearing gait using the ED Walker, a 32% greater electromyographic activity was recorded when compared with that using the crutches (P<.05) (fig 2B, table 1). Large increases in biceps femoris muscle activity were observed during the non-weight-bearing condition using the crutches and the ED Walker when compared with normal walking (P≤.05) (fig 2B, table 1). During the heavy weight-bearing condition, the ED Walker required increased biceps femoris electromyographic activity when compared with normal walking (P≤.05) (fig 2B, table 1).

For the non-weight-bearing condition (swing-through gait), electromyographic activity during crutch ambulation was greater than that required for normal walking (P≤.05) (fig 2C, table 1). The light weight-bearing gaits resulted in a 34% decrease in electromyographic activity when compared with normal walking for both devices (P≤.05) (fig 2C, table 1). Additionally, during the heavy weight-bearing condition, the crutches resulted in a 17% decrease in electromyographic activity when compared with normal walking (P≤.05) (fig 2C, table 1). For the anterior tibialis muscle, the only activation differences between the 2 devices was during the heavy weight-bearing condition, with the ED Walker electromyographic activity being 24.5% higher (P≤.05) (fig 2C, table 1).

Ambulation with the crutches resulted in significant decreases in soleus electromyographic activity at all levels of weight-bearing load when compared with normal walking (P≤.05) (fig 2D, table 1). The ED Walker resulted in decreased soleus electromyographic activity versus that of normal walking only during non-weight bearing gait (P≤.05) (fig 2D, table 1). During the light weight-bearing condition, the ED Walker required a greater amount of electromyographic activity than the axillary crutches (P≤.05) (fig 2D, table 1).

Force characteristics during assisted walking 

The self-selected heavy weight-bearing condition required 75.6%±1.7% and 72.7%±1.9% of the peak vertical force produced during normal ambulation with the axillary crutch and ED Walker gaits, respectively (not significant). During the light condition, subjects transferred a greater amount of their body weight off the test limb with crutches when compared with the ED Walker (32.0%±2.0% vs 48.7%±1.6%; P≤.05) (fig 3). All conditions resulted in a significant decrease in force when compared with that of normal walking (P≤.05) (fig 3).

• 
  • View Large Image
  • Download to PowerPoint

• Fig 3. 

  Normalized peak vertical ground reaction force when subjects were instructed to self-select a light and heavy weight-bearing load while ambulating with the assistance of axillary crutches and the ED Walker. The black horizontal lines at 100% represent force recorded during normal walking. ∗ED Walker greater than axillary crutch (P≤.05); ^†walking with assisted device significantly different from that of normal walking (P≤.05).

Back to Article Outline

Discussion 

The most novel finding of this study is that muscle activation patterns varied during axillary crutch ambulation depending on the level of weight-bearing load. In general, these findings suggest that non-weight-bearing crutch ambulation resulted in reduced muscle activity of the antigravity extensor muscles (ie, vastus lateralis, soleus) (fig 1). Conversely, the leg flexors (ie, anterior tibialis, biceps femoris) exhibited an increased activity level (fig 1). During partial weight-bearing ambulation with the crutches, the majority of muscle activation levels return toward that required for normal ambulation, although decreases in soleus activity are observed even during the heavy weight-bearing condition. To our knowledge, this is the first study reporting leg muscle activity during crutch ambulation.

Based on our findings of reduced muscle activity in the soleus and vastus lateralis muscles during assisted ambulation, it appears that they would be the most susceptible to dysfunction associated with unloading. The finding of decreased vastus lateralis activation agrees with previous studies reporting marked decreases in quadriceps femoris strength and cross-sectional area after short-term unweighting of the limb using axillary crutches.3, 4, 5, 14, 15 Because the vastus lateralis and soleus muscles showed a wide range of muscle activity depending on weight-bearing load during ED Walker gait, ambulation with this device may allow selective muscle activation and serve to reduce the negative side effects of unloading on the musculoskeletal system. However, the self-selected light weight-bearing force is greater with the ED Walker (vs the axillary crutches); ergo, if the purpose of using a walking device is to reduce the load placed on an injured limb, then the potential benefit of having increased muscle activity may be negated. Additionally, although increased muscle activity could attenuate the disuse-induced muscle atrophy, one must weigh the potential for increased fatigue when deciding on the device prescription and load-bearing condition.

Our study also suggests that when subjects are instructed to self-select a light partial weight-bearing load, the ground reaction force during crutch gait could be 68% less than that of normal walking and 16% less than that of ED Walker gait. Interestingly, subjects self-selected the same heavy weight-bearing loads (around 25% less than that of normal walking) with both devices. Therefore, if the goal of a rehabilitative device is to allow for a large variation in selective weight-bearing loads, the axillary crutch appears to allow this through a modest range (30%–75% of body weight).

In addition to the aforementioned issues (degree of muscle activity and unweighting), other variables should be considered when deciding on the appropriate ambulation device for a given subject. For example, the 2 devices we tested differ in their upper-body and spinal mechanics: crutches are known to cause brachial plexus injuries,16, 17 whereas the ED Walker places the spine in a more flexed position. Thus, evaluating one’s medical history and predisposition to specific injuries should be taken into account in this decision-making process.

There are several limitations to this study that should be acknowledged. First, we did not simulate a cast or brace. Thus, it is possible that with the addition of a cast or brace, the associated muscle activation patterns could change. Additionally, because we chose to control for ambulation speed (1 stride/1.5s), it is possible that allowing for self-selected speeds could alter our findings. Future research is needed to determine the effectiveness of the ED Walker as an ambulatory device. It is suggested that future studies investigating ambulation devices evaluate differences in ambulatory speed and efficiency and consider biomechanical differences.

Back to Article Outline

Conclusions 

These findings provide considerable detail of the leg muscle activation patterns and force characteristics during assisted axillary crutch and ED Walker ambulation. In general, walking with assisted devices resulted in altered muscle activation patterns in the test limb that were dependent on weight-bearing load and device type. During swing-through gait patterns using axillary crutches and the ED Walker, the leg extensor muscles displayed a drastic reduction in muscle activity. As weight-bearing load increased, the activity of the vastus lateralis increased, with the ED Walker allowing for a greater range of voluntary activation during assisted walking. Regarding ground reaction forces, axillary crutch walking allowed for a large range of weight transfer off the affected limb when subjects self-selected light and heavy weight-bearing loads. Clinically, knowledge of these muscle activities and force characteristics may aid in the decision-making process of prescribing a device type following a timeline for restoring weight-bearing loads.

Suppliers

Back to Article Outline

References 

1. Joyce BM, Kirby RL. Canes, crutches and walkers. Am Fam Physician. 1991;43:535–542
   • View In Article
   • MEDLINE
2. Williamson VC. Management of lower extremity fractures. Orthop Nurs. 1998;17:84–87
   • View In Article
   • MEDLINE
3. Dudley GA, Ploutz-Snyder LL. Deconditioning and bed rest (musculoskeletal response). In:  Roitman JL editors. ACSM’s resource manual for guidelines for exercise testing and prescription. 3rd ed.. Baltimore: Williams & Wilkins; 1998;p. 200–205
   • View In Article
4. Dudley GA, Duvoisin MR, Adams GR, Meyer RA, Belew AH, Buchanan P. Adaptations to unilateral lower limb suspension in humans. Aviat Space Environ Med. 1992;63:678–683
   • View In Article
   • MEDLINE
5. Ploutz-Snyder LL, Tesch PA, Crittenden DJ, Dudley GA. Effect of unweighting on skeletal muscle use during exercise. J Appl Physiol. 1995;79:168–175
   • View In Article
   • MEDLINE
6. Reardon K, Galea M, Dennett X, Choong P, Byrne E. Quadriceps muscle wasting persists 5 months after total hip arthroplasty for osteoarthritis of the hip (a pilot study). Intern Med J. 2001;31:7–14
   • View In Article
   • MEDLINE
   • CrossRef
7. Ploutz-Snyder LL, Manini T, Ploutz-Snyder RJ, Wolf DA. Functionally relevant thresholds of quadriceps femoris strength. J Gerontol A Biol Sci Med Sci. 2002;57:B144–B152
   • View In Article
   • MEDLINE
   • CrossRef
8. Peacock B. A myographic and photographic study of walking with crutches. Physiotherapy. 1966;52:264–268
   • View In Article
   • MEDLINE
9. Harkess JW, Ramsey WC, Harkess JW. Principles of fracture and dislocation. In:  Rockwood CA,  Green DP,  Bucholz RW editor. Rockwood and Green’s fractures in adults. 4th ed.. Philadelphia: Lippincott-Raven; 1996;p. 102–107
   • View In Article
10. Barr AE, Backus SI. Biomechanics of gait. In:  Nordin M,  Frankel VH editor. Basic biomechanics of the musculoskeletal system. 3rd ed.. Baltimore: Lippincott, Williams & Wilkins; 2001;p. 438–457
    • View In Article
11. Enoka RM, Fuglevand AJ. Motor unit physiology (some unresolved issues). Muscle Nerve. 2001;24:4–17
    • View In Article
    • MEDLINE
    • CrossRef
12. Basmajian JV, DeLuca CL. Muscles alive (their functions revealed by electromyography). Baltimore: Williams & Wilkins; 1985;
    • View In Article
13. Cram JR, Kasman GS. Introduction to surface electromyography. Gaithersburg (MD): Aspen; 1998;
    • View In Article
14. Hather BM, Adams GR, Tesch PA, Dudley GA. Skeletal muscle responses to lower limb suspension in humans. J Appl Physiol. 1992;72:1493–1498
    • View In Article
    • MEDLINE
15. Ploutz-Snyder LL, Tesch PA, Hather BM, Dudley GA. Vulnerability to dysfunction and muscle injury after unloading. Arch Phys Med Rehabil. 1996;77:773–777
    • View In Article
    • Abstract
    • Full-Text PDF (728 KB)
    • CrossRef
16. Poddar SB, Gitelis S, Heydemann PT, Piasecki P. Bilateral predominant radial nerve crutch palsy. A case report. Clin Orthop. 1993;Dec(297):245–246
    • View In Article
17. Raikin S, Froimson MI. Bilateral brachial plexus compressive neuropathy (crutch palsy). J Orthop Trauma. 1997;11:136–138
    • View In Article
    • MEDLINE
    • CrossRef

• 1 No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit on the authors or on any organization with which the authors are associated.
• a Hartford Walking Systems, 22 Pearl St, New Hartford, NY 13413.
• b Axon Instruments Inc, 1101 Chess Dr, Foster City, CA 94404.
• c Compute Boards Inc, 16 Commerce Blvd, Middleboro, MA 02346.
• d Kistler Instrument Corp, 75 John Glenn Dr, Amherst, NY 14228.
• e National Instruments, 6504 Bridge Point Pkwy, Austin, TX 78730.
• f SPSS Inc, 233 S Wacker Dr, 11th Fl, Chicago, IL 60606.