Volume 88, Issue 9 , Pages 1159-1163, September 2007
Mechanics of Ambulation With Standard and Spring-Loaded Crutches
Article Outline
Abstract
Segura A, Piazza SJ. Mechanics of ambulation with standard and spring-loaded crutches.
Objective
To compare kinetic measures and spatiotemporal variables assessed during walking with standard axillary crutches and spring-loaded crutches.
Design
A repeated-measures design in which healthy subjects walked with both standard and spring-loaded crutches.
Setting
Biomechanics research laboratory.
Participants
Ten healthy young adult volunteers participated. Only female volunteers between 154.9 and 175.3cm in stature were selected to fit the size of the crutches used.
Interventions
Not applicable.
Main Outcome Measures
The main outcome measures were kinetic variables such as ground reaction force, rate of force rise, and impulse and spatiotemporal variables such as stride length, stride time, and percentage of stride spent in stance.
Results
The rate of ground reaction force rise and impulse of the ground reaction force (both P<.001) were reduced by 33% and 13% to 26%, respectively, but the peak ground reaction force was slightly greater (P=.001) with spring-loaded crutches. The stride time was increased with spring-loaded crutches (P=.005), but the stride length did not differ significantly (P=.465).
Conclusions
The use of spring-loaded crutches altered the mechanics of crutch gait in ways that are likely to reduce overuse injury in crutch users. Further study of spring-loaded crutches is warranted, especially with respect to their energetic efficiency.
Key Words: Biomechanics, Crutches, Gait, Rehabilitation
IT IS IMPORTANT FOR PEOPLE with disabilities to be able to use crutches comfortably and effectively. Many long-term crutch users prefer axillary (underarm) crutches to elbow crutches because axillary crutches may offer increased control during gait as well as the ability to stand stably while performing tasks with the arms. However, walking with axillary crutches can be problematic because of complications known to be associated with their long-term use.1, 2 Large forces may be applied to the tips of axillary crutches at initial contact, and these forces may be transmitted to the elbow and shoulder joints, causing irritation and injury.3, 4 Such forces may lead to crutch palsy, aneurysms, and thromboses, causing pain, discomfort, and other serious conditions in some axillary crutch users.2, 5, 6
Spring-loaded crutches are a potentially more comfortable alternative to standard crutches because they may alter the way in which loads are transmitted to the body during crutch gait. Parziale and Daniels7 found a 24% decrease in peak handle load applied and a 22% decrease in the average amplitude of the initial handle force when using spring-loaded axillary crutches in comparison to standard axillary crutches. Shoup8 found that subjects using spring-loaded crutches used a stride length greater than that observed during standard crutch use. The use of spring-like tendons for energy conservation is known to occur in human and animal locomotion,9, 10 and it may be the case that such benefits will also apply to users of spring-loaded crutches. Shoup8 hypothesized that the energy storage and return expected from spring-loaded crutches might represent an energy savings of about 25% of the metabolic energy typically consumed during standard crutch gait, although this hypothesis was not tested.
Previous investigations of the mechanics of spring-loaded crutch gait have been limited in their scope. Parziale and Daniels7 did not measure ground reaction forces during ambulation with the spring-loaded crutches, and Shoup8 evaluated the ground reaction forces in only a single subject and only during the interval just after the initial crutch contact. A recent innovation in spring-loaded crutches was described by Shortell et al,11 who designed a forearm crutch with compliant shafts in S-shapes, but the authors did not measure the ground reaction forces during ambulation with the modified design.
The purpose of this study was to determine if there are differences in ground reaction force, rate of force rise, impulse, and spatiotemporal gait variables between ambulation with spring-loaded axillary crutches and ambulation with standard axillary crutches. A video-based motion analysis system and forceplate were used to record crutch kinematics and ground reaction forces in healthy young adult subjects who were not habitual crutch users as they walked with both types of crutches. It was hypothesized that introducing springs into crutches would reduce peak ground reaction force, rate of force rise, and impulse while increasing stride length.
Methods
Ten healthy women (age range, 21–28y; mean height, 169.4cm; mean body mass, 63.9kg) free of musculoskeletal problems and nonusers of crutches in the previous 6 months volunteered as subjects. Subjects were selected on the basis of their bodily dimensions, such that the stature of each was suited to the size of the modified and standard crutches (167.6–175.3cm) and the body mass of each was suited to the stiffness of the spring introduced into the modified crutches (54.5–86.4kg). This body mass limitation was in accordance with the recommendations of Shortell11 that a spring constant of 21.9kN/m (125lb/in) was suitable for subjects within this range of body masses. All procedures were approved by the Institutional Review Board of The Pennsylvania State University.
Two pairs of standard aluminum axillary crutches,a designed for persons between 154.9cm (61in) and 175.3cm (69in) in height, were used in this study. One of the pairs of crutches was not modified in any way and was used as the standard axillary crutch. The other pair of crutches was modified by the addition of sliding mechanism and a helical compression spring to the bottom of the shaft just above the crutch tip. The springs were 8.9cm long and 1.9cm in diameter, with a stiffness of 22.4kN/m and a small preload of approximately 10N.
A forceplateb was used to measure the ground reaction forces under the crutch tips at a sampling rate of 1000Hz. Subject and crutch kinematics were recorded by using a 6-camera digital video-based motion analysis system.c Five spherical reflective markers, each 12mm in diameter, were placed on the left crutch (fig 1) with 2 of the markers located above and below the spring to monitor spring deflection. Markers were also placed at several locations on the body of the subject, but coordinates of these markers were not used in the present study.
Fig 1.
A subject walking with spring-loaded crutches. Five reflective markers placed on the left crutch were used to track crutch motions from which spatiotemporal variables were calculated. Subjects performed 10 trials for each crutch for which only the left crutch tip struck the forceplate.
Each subject was fitted for both pairs of crutches according to standard guidelines,12 and subjects were instructed in proper axillary crutch walking technique.5 Each subject practiced with each pair of crutches for 15 minutes to become accustomed to crutch gait before data collection. The ordering of the crutch type (springy or standard) was alternated for each subject. For each crutch type, the subject was asked to perform 10 “good” trials by walking over the forceplate with a right-leg-support, swing-through crutch gait. A trial was considered “good” if the left crutch tip struck the forceplate and the subject cleared the plate without striking it again with the crutch tip or with either foot. Each trial was performed at the subject’s self-selected walking speed, with subjects instructed to walk in the way that was most comfortable.
Force and marker coordinate data were low-pass filtered with a cutoff frequency of 45Hz by using a fourth-order Butterworth filter implemented in Matlabd to remove 60Hz noise. Crutch tip strike was identified as occurring at the first frame for which ground reaction force rose above a 10-N threshold. The impulse of the ground reaction force was calculated as the area under force versus time curves. Vertical, anteroposterior (AP), and mediolateral components of impulse as well as total impulse were calculated for the first 50, 100, and 200ms after crutch tip strike. In addition, the following variables were computed for each trial: time of crutch stance (crutch tip strike to tip liftoff), time for 1 stride (crutch tip strike to subsequent tip strike), percentage of stride in crutch stance, stride length (AP distance traveled by the bottom crutch marker during a stride), velocity, and maximum ground reaction force. The average rate of force rise over every 10-ms interval during crutch stance was also calculated, and the largest rate of force rise was noted for each trial.
A 2-way analysis of covariance (ANCOVA) with repeated measures with factors of crutch type (spring-loaded, standard) and trial number (1–10) using velocity (stride length divided by stride time) as a covariant was performed for each variable by using Minitab.e Velocity was chosen as a covariant because it was expected to influence ground reaction force, but it was not controlled to allow the subjects to walk with the most natural gait possible. Tukey post hoc tests were performed to determine if there were significant differences between the standard and springy crutch means only when ANCOVA results were significant. In addition, velocities were compared by using a paired t test. The level of statistical significance was set at α equal to .05 for all tests.
Results
The maximum resultant ground reaction force was slightly but significantly higher (P=.001) when subjects ambulated with spring crutches than when they walked with standard crutches (fig 2). The maximum rate of force rise over any 10-ms interval of crutch stance phase was significantly lower (P<.001) when subjects ambulated with the spring crutches as opposed to when subjects walked with the standard crutches (fig 3).
Fig 2.
The peak ground reaction force (GRF) averaged across trials for each subject. For 7 of 10 subjects, peak ground reaction force was higher for spring-loaded crutches (P=.001). Bars indicate ±1 standard deviation (SD).
Fig 3.
The maximum rate of force rise over any 10-ms interval averaged across trials for each subject. For each of the 10 subjects, the rate of force rise was lower for spring-loaded crutches than for standard crutches (P<.001). Bars indicate ±1 SD.
The average impulse over the first 50ms of crutch stance phase was significantly lower (P<.001) with spring-loaded crutches than with standard crutches. This was true for the total impulse magnitude and for each component of the impulse vector (fig 4). Similarly, total impulses were also significantly lower for spring-loaded crutches when time periods of 100 and 200ms were considered (both P<.001).
Fig 4.
The magnitude of the impulse of the ground reaction force over first 50ms of crutch stance phase averaged across trials for each subject. For each of the 10 subjects, the impulse magnitude was smaller for spring-loaded crutches than for standard crutches (P<.001). Bars indicate ±1 SD.
There were some small but significant differences in spatiotemporal parameters between the 2 types of crutches (table 1). The stride time and the time that the subjects spent in crutch stance phase were significantly higher (P=.005, P=.002, respectively), and the walking velocity was significantly lower (P=.025) when they walked with spring-loaded crutches than when they walked with standard crutches. No significant differences were seen, however, between spring crutches and standard crutches in the duty factor, the percentage of the gait cycle spent in crutch stance (P=.463), or in the stride length (P=.465).
Table 1. Spatiotemporal Variables for Spring-Loaded and Standard Crutches, Taken Across Subjects
| Parameter | Spring-Loaded | Standard | P |
|---|---|---|---|
| Stride time (s) | 1.471±0.174 | 1.433±0.186 | .005 |
| Stance time (s) | 0.793±0.110 | 0.766±0.099 | .002 |
| Duty factor (%)⁎ | 53.9±2.7 | 53.5±3.0 | .463 |
| Stride length (m) | 1.582±0.175 | 1.607±0.165 | .465 |
| Velocity (m/s) | 1.096±0.212 | 1.146±0.225 | .025 |
⁎The percentage of the stride time spent in crutch stance. |
Crutch ground reaction force versus time plots were highly variable between subjects and across trials of the same subject. Trials collected minutes apart for the same subject did not show a consistent ground reaction force profile; it was often the case that some trials exhibited 2 distinct force peaks, some had only a single peak, and still others featured a force plateau (fig 5). Furthermore, there was substantial variability in peak force magnitude seen across trials. ANCOVA revealed no significant effect of trial number on any of the spatiotemporal or kinetic measures.
Fig 5.
Vertical ground reaction force (vGRF) traces for the 10 spring-loaded crutch trials of subject 1. The considerable intertrial variability in loading patterns shown here was evident for all subjects and for both the spring-loaded and standard crutch types.
Discussion
Of the spatiotemporal measures considered in this study, significant differences were found between the spring-loaded and standard crutches only with respect to the period in crutch stance phase, the stride time, and the velocity. Our hypothesis that the stride length would be increased when using the spring crutches, as suggested by Shoup,8 was not supported. Also contrary to our expectation, the maximum ground reaction force was higher when subjects walked with the spring-loaded crutches than when they used standard crutches. In line with our hypotheses, however, were the findings that the maximum rate of force rise and the impulse of the ground reaction force in early crutch stance were lower for spring-loaded crutches.
Shoup8 reported increases in stride length accompanying the use of spring-loaded crutches that were not reproduced in the present study. The reduction of the initial force transient reported by Shoup for a single subject, however, was similar to the reduced rate of force rise and reduced impulse found in 10 subjects in the present study.
Parziale and Daniels7 also added springs to the shafts of standard crutches but measured uniaxial forces at the crutch handles rather than the ground reaction force. The authors reported handle forces that were 24% lower for spring-loaded crutches than for standard crutches, but in the present study a small but significant increase in ground reaction force was noted for spring-loaded crutches. The difference in methodology (forces measured at the handle vs the crutch tip) and transmission of torque at the crutch handle may be the causes of this discrepancy, and future studies should consider simultaneous measurement of these forces because both are likely to have implications for injury.
The unexpected larger maximum ground reaction forces seen when subjects walked with spring-loaded crutches could have been caused by “bottoming out” that might have occurred if the spring became fully compressed during crutch stance. The measurement of crutch deflection from the crutch marker coordinates, however, indicated that this was not the case, and none of the subjects described experiencing this sensation during spring-loaded crutch trials. Alternative explanations are that subjects may have used different techniques for each crutch type and that the increased force was caused by subjects falling through a greater distance before the springs arrested their falls.
The decrease in the rate of force rise when subjects used the spring-loaded crutches indicates that more jarring and quickly changing forces are applied to the crutch user when using standard crutches. Similarly, the findings that the impulses were lower when using the spring-loaded crutches are important because of the implications for overuse injuries. Because injuries, such as aneurysm formation, that occur as a result of axillary crutch use are often caused by repetitive trauma to the upper extremities,6 it is likely that walking with spring-loaded crutches lessens the risk of injury to the crutch user. Repetitive impulsive loading of joints leads to articular stiffness and has been found to lead to injury and osteoarthritis.13 Lower impulses and rates of force rise when using the spring crutches, such as those found in the present study, may indicate that the spring-loaded crutches may decrease skeletal impact loading during crutch stance phase, thus decreasing the likelihood of crutch palsy, aneurysm, and thrombosis that are associated with axillary crutch use. Although it is true that the subjects walked more slowly with spring-loaded crutches, velocity differences cannot explain the differences in impulse and rate of force rise because ANCOVA should have removed the variance accounted for by velocity. When the 10 subjects in this study were asked which crutches they preferred after having used both, 4 chose the spring-loaded crutches, 4 chose standard, and 2 had no preference. One comment made by 2 subjects was that the spring-loaded crutches felt “more comfortable but less stable.” If spring-loaded crutches offer such a trade of stability for comfort, it raises the question whether they are suitable for use by patients with neuromuscular disorders.
Study Limitations
There were a number of limitations to the present study. One of the most obvious was that young, healthy subjects were tested rather than habitual crutch users. The reason these subjects were used was that they were easy to recruit, healthy, and therefore likely to be able to adequately use crutches with minimal risk of falling or injury. However, subjects practiced for only 15 minutes with each crutch type before data collection. This practice was enough for subjects to become accustomed to using the crutches but probably not enough for subjects to master crutch walking with each crutch type at the level of a habitual crutch user. The lack of extended practice was likely to have been a major determinant of the variability within subjects that was seen in the ground reaction force traces, but more studies of habitual crutch users’ gait are required to confirm this. Testing of the crutches with habitual crutch users would most likely reduce the within-subject variability because subjects would be more comfortable using crutches but, because people with a variety of medical conditions and varying degrees of disability use crutches regularly, the between-subject variability may be greater than that seen in this study. The spring constant used in this study was based on the qualitative rather than quantitative recommendations of Shortell et al,11 and unfortunately only 1 spring constant was tested. It would have been helpful to evaluate walking with spring crutches with a variety of spring constants to detect differences between spring constants in subjects with different body weights.
Conclusions
The results of this study have important implications for the design of axillary crutches and suggest avenues for future research. The reduced impulses and rates of force rise imply that spring-loaded crutches may allow crutch users to ambulate more comfortably with axillary crutches and with fewer complications. Further work is needed to quantify how loads applied to the body (as opposed to ground reaction forces) are altered and what loading differences exist for habitual crutch users and for varied spring constants. Although oxygen consumption during crutch ambulation has been measured previously, there have not been studies of the oxygen consumption while walking with spring-loaded crutches compared with standard crutches. LeBlanc et al,14 however, failed to find differences between spring-loaded and standard crutches in a heart rate–based energy expenditure index. If using spring crutches could substantially reduce the high-energy demands of crutch walking, many people with disabilities might opt to use crutches rather than wheelchairs.
Suppliers
Acknowledgment
We thank Laura Nigro, BSc, for assistance with the data collection.
References
- . Electrophysiological findings in crutch palsy. Electromyogr Clin Neurophysiol. 1989;29:281–285
- . Bilateral brachial plexus compressive neuropathy (crutch palsy). J Orthop Trauma. 1997;11:136–138
- . Upper limb loadings of gait with crutches. J Biomech Eng. 1987;109:285–290
- . Forces and impulses during aided gait. Arch Phys Med Rehabil. 1987;68:715–722
- . Bilateral predominant radial nerve crutch palsy (A case report). Clin Orthop Relat Res. 1993;245–246Dec(297)
- . Crutch-induced axillary artery injury. Cardiovasc Intervent Radiol. 1995;18:296–299
- . The mechanical performance of ambulation using spring-loaded axillary crutches (A preliminary report). Am J Phys Med Rehabil. 1989;68:192–195
- . Design and testing of a child’s crutch with conservative energy storage. J Mech Des N Y. 1980;102:672–676
- . Compliant leg behaviour explains basic dynamics of walking and running. Proc Biol Sci. 2006;273:2861–2867
- . Walking and running. Am Sci. 1984;72:348–354
- . The design of a compliant composite crutch. J Rehabil Res Dev. 2001;38:23–32
- . Canes, crutches and walkers. Am Fam Physician. 1991;43:535–542
- . Effect of prolonged walking on concrete on the knees of sheep. J Biomech. 1982;15:487–492
- . A quantitative comparison of four experimental axillary crutches. J Prosthet Orthot. 1993;5:20–28
- a Guardian Red Dot; Sunrise Medical, 2382 Faraday Ave, Ste 200, Carlsbad, CA 92008-7220.
- b Model 9287; Kistler Instrument, 75 John Glenn Dr, Amherst, NY 14228-2171.
- c Eagle 500RT; Motion Analysis Corp, 3617 Westwind Blvd, Santa Rosa, CA 95403-1067.
- d The MathWorks Inc, 3 Apple Hill Dr, Natick, MA 01760-2098.
- e Minitab Inc, Quality Plaza, 1829 Pine Hall Rd, State College, PA 16801-3008.
Supported by the National Science Foundation (grant no. BES-0134217).
No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the author(s) or upon any organization with which the author(s) is/are associated.
PII: S0003-9993(07)00422-4
doi:10.1016/j.apmr.2007.05.026
© 2007 American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation. Published by Elsevier Inc. All rights reserved.
Volume 88, Issue 9 , Pages 1159-1163, September 2007
