Gait & Posture 40 (2014) 640–646
Contents lists available at ScienceDirect
Gait & Posture journal homepage: www.elsevier.com/locate/gaitpost
The relationship between pelvis–trunk coordination and low back pain in individuals with transfemoral amputations§ Elizabeth Russell Esposito *, Jason M. Wilken Center for the Intrepid, Department of Orthopaedics and Rehabilitation, Brooke Army Medical Center, Ft. Sam Houston, TX, USA
A R T I C L E I N F O
A B S T R A C T
Article history: Received 19 April 2013 Received in revised form 14 July 2014 Accepted 17 July 2014
Low back pain (LBP) is common in individuals with transfemoral amputation and may result from altered gait mechanics associated with prosthetic use. Inter-segmental coordination, assessed through continuous relative phase (CRP), has been used to identify specific patterns as risk factors. The purpose of this study was to explore pelvis and trunk inter-segmental coordination across three walking speeds in individuals with transfemoral amputations with and without LBP. Nine individuals with transfemoral amputations with LBP and seven without pain were compared to twelve able-bodied subjects. Subjects underwent a gait analysis while walking at slow, moderate, and fast speeds. CRP and CRP variability were calculated from three-dimensional pelvis and trunk segment angles. A two-way ANOVA and post hoc tests assessed statistical significance. Individuals with transfemoral amputation demonstrated some coordination patterns that were different from able-bodied individuals, but consistent with previous reports on persons with LBP. The patient groups maintained transverse plane CRP consistent with ablebodied participants (p = 0.966), but not sagittal (p < 0.001) and frontal plane CRP (p = 0.001). Sagittal and frontal CRP may have been re-optimized based on new sets of constraints, such as protective rigidity of the segments, muscular strength limitations, or prosthesis limitations. Patients with amputations and without LBP exhibited few differences. Only frontal and transverse CRP shifted toward out-of-phase as speed increased in the patient group with LBP. Although a cause and effect relationship between CRP and future development of back pain has yet to be determined, these results add to the literature characterizing biomechanical parameters of back pain in high-risk populations. Published by Elsevier B.V.
Keywords: Amputee Military Continuous relative phase Walking Variability
1. Introduction Low back pain (LBP) affects 80% of adults at some point during their lifetime [1]. The rates of LBP in individuals with amputations are significantly greater than in the general population [2]. LBP is particularly troublesome for individuals with transfemoral amputation (TFA), compared to trans-tibial amputation. Muscular imbalance, leg length discrepancies and altered gait mechanics may contribute to the increased incidence of pain in this population [3–5]. Abnormal gait patterns, to include altered motion of the trunk and pelvis, may also contribute to the increased incidence of pain in this population. Numerous studies have sought to determine biomechanical risk factors for LBP in the
§ The view(s) expressed herein are those of the author(s) and do not reflect the official policy or position of Brooke Army Medical Center, the U.S. Army Medical Department, the U.S. Army Office of the Surgeon General, the Department of the Army, Department of Defense or the U.S. Government. * Corresponding author. Tel.: +1 2105395824. E-mail address: [email protected] (E. Russell Esposito).
http://dx.doi.org/10.1016/j.gaitpost.2014.07.019 0966-6362/Published by Elsevier B.V.
general population, but few have investigated a relationship between gait and pain in a population with lower extremity amputation. Standard gait analyses, including peak kinematics and ranges of motion (ROM), have been previously used to identify gait deviations associated with LBP. For example, transverse plane trunk rotation has been found to be greater in individuals with TFA who have LBP than in those without LBP or able-bodied individuals [6]. These greater rotations may contribute to disk degeneration [7] and/or soft tissue strain [8,9] which are potential sources of LBP. However, the extremes of lumbar motion are not universally associated with an increased risk in individuals with LBP [10] or amputation [11]. An alternative to the use of discrete measures is the assessment of movement coordination between adjacent segments using continuous data from the entire gait cycle [12]. Continuous measures, such as continuous relative phase (CRP), have been effective in identifying gait deviations associated with LBP. CRP has been used to compare pelvis and trunk coordination in populations with LBP [13,14] and amputations [15]. Coordination can be defined using CRP to categorize movement patterns
E. Russell Esposito, J.M. Wilken / Gait & Posture 40 (2014) 640–646
ranging from ‘‘in-phase’’, where adjacent segments rotate as a unit in the same direction at the same speed, to ‘‘out-of-phase’’, where adjacent segments concurrently rotate at the same speed in the opposite directions to create bending or twisting motions. Seay et al. [13] found that sufferers of LBP had greater synchronous, in-phase pelvis and trunk rotations in the frontal plane during walking than able-bodied control subjects. In the transverse plane, able-bodied patients with LBP [14], as well as patients with TFA [15], never achieved the same degree of axial out-of-phase coordination as healthy, able-bodied subjects. Although no single plane of motion has been identified as the primary contributor to LBP, studies suggest pain is associated with side bending, twisting, and various asymmetrical movements of the trunk and pelvis [16]. Across all three planes, the variability of these phase relationships may provide additional information on LBP. Variability within a biological system allows for a level of flexibility, adaptability and, thus, the ability to overcome perturbations [17]. A movement with low coordination variability may be perceived as safer because of its predictability but its constant repetition may create an overuse situation [18] such as LBP. One study found lower coordination variability between axial trunk– pelvis rotations during walking in individuals with LBP relative to able-bodied controls [19], however another study found no differences between groups in any plane [13]. The differences between these two studies may be due to the subject populations; Selles et al. [19] had patients who could not work full time due to pain and Seay et al. [13] had recreational runners who ran at least 20 miles per week. Changes in walking speed can directly influence coordination and coordination variability. When walking overground, increasing speed shifts axial pelvis–trunk coordination from an in-phase relationship to more out-of-phase in healthy, able-bodied individuals [20]. Sufferers of LBP have less ability to switch coordination patterns as speed increases [13,14,19] but it is unknown how speed may influence coordination patterns in individuals with amputations and if the resulting patterns may be indicative of risk factors for LBP. Although the increased incidence of LBP in individuals with TFA may result from altered gait mechanics from prosthetic use, it is still unknown how coordination and coordination variability may be affected by the development of LBP after TFA. The first objective was to determine how pelvis–trunk kinematics, coordination and coordination variability differed among individuals with TFA who did and did not experience LBP and an able-bodied control group. We hypothesized that, similar to [13–15,19], individuals with TFA would exhibit greater in-phase coordination in the transverse and frontal planes and greater out-of-phase coordination in the sagittal plane than, able-bodied individuals. In agreement with the results of Seay et al. [13], we expected coordination variability to stay consistent across groups [13]. The second objective was to determine if pelvis–trunk kinematics, coordination, and coordination variability modulated with walking speed. We hypothesized that, similar to [13,19,20], CRP would increase with increasing speed and, similar to [13], variability would remain consistent across walking speeds.
641
past 4 weeks. TFA-LBP indicated a frequency of LBP of one time per week to all the time or almost all the time. Responses on LBP frequency were collected as part of the Prosthetic Evaluation Questionnaire [21]. Five of the TFA-NP group wore the Otto Bock X2 prosthesis and two wore their standard mechanical prostheses that were not micro-processor controlled. Eight of the TFA-LBP wore the X2 and one wore a mechanical prosthesis. Limb lengths were matched to intact side and subjects were K4 ambulators. Able-bodied individuals had no current or previous orthopedic injuries affecting gait and no injuries to the low back or LBP. All subjects read and signed an informed consent approved by the Institutional Review Board. The experimental setup for overground walking consisted of a 26-camera motion capture system (120 Hz; Motion Analysis, Santa Rosa, CA). Fifty-seven spherical, retro-reflective markers were placed on anatomical landmarks and on segments of the limbs, torso and head [22]. Participants walked at three speeds (Speed 1, 2, and 3) based on dimensionless Froude numbers of 0.10, 0.16, and 0.23 [23] which corresponded to average speeds of 1.0, 1.2 and 1.4 m/s, respectively. Speed was controlled with an auditory tone that sounded as walking speed fell within a 5% range. Participants completed eight trials at each of the speeds during which they walked across approximately 20 m of level ground, the middle portion of which was analyzed. All trials were labeled in EvART (Motion Analysis Corporation, Santa Barbara, CA) and exported to Visual3D (version 4.96.7, C-Motion, Germantown, MD). Raw marker data were interpolated and filtered using a 4th order, low-pass Butterworth filter with a cutoff frequency of 6 Hz. Heel strike was identified according to the methods described by [24]. A successful trial was defined as one containing a complete stride initiated with the prosthetic limb for the TFA groups and initiated with the right leg for able-bodied subjects. CRP was then calculated on each stride using the method of Hamill et al., [18]. Briefly, pelvis and trunk segmental angular velocities were calculated as the first derivative of segment angular position. Angular position and velocity were normalized to 101 data points and scaled between 1 and 1. These normalized position and angular velocity vectors were plotted relative to each other for the trunk segment and for the pelvis segment. Phase angles were calculated as the angle formed between a line from the origin to each data point and the right horizontal. CRP was defined as the difference between the trunk and pelvis phase angles at each percent of the stride. CRP variability was defined as the standard deviation of the CRP at each time point across all eight trials. Individual subject means were calculated for peak segment angles, CRP and CRP variability at each of the three walking speeds. Shapiro–Wilk tests were conducted for normality and a two-way, repeated-measures ANOVA (SPSS Inc, Chicago, IL, USA) examined the main effects. Tukeys post hoc tests were used to compare groups (p < 0.05) and paired t-tests with Bonferroni corrections were used to compare speeds. 3. Results Individual subject characteristics are presented in Table 1. Only age (p = 0.002) and body mass (p = 0.045) were significantly greater in the TFA-LBP group than the able-bodied control group. Self-selected walking velocity (SSWV) was not significantly different from Speed 2. 3.1. Kinematics (Fig. 1 and Supplemental material)
2. Methods Sixteen individuals with unilateral, traumatic TFA and twelve able-bodied control subjects participated in this study. Seven of the TFA subjects were categorized in the no pain group (TFA-NP) and nine were categorized in the LBP group (TFA-LBP). TFA-NP indicated a frequency of LBP never or only once or twice in the
There were no significant interactions for any of the trunk and pelvis kinematics. There were no significant main effects of speed in the sagittal and frontal planes, but there were significant group effects. TFA-LBP maintained significantly less sagittal plane anterior trunk lean than TFA-NP and able-bodied (p < 0.001) and less sagittal plane ROM than the able-bodied group (p = 0.022). Anterior pelvic tilt was significantly greater in both TFA groups relative to the able-bodied (p < 0.001) and sagittal plane pelvis ROM was significantly greater in the TFA-LBP group than the TFA-NP (p < 0.001) and able-bodied controls (p = 0.004).
E. Russell Esposito, J.M. Wilken / Gait & Posture 40 (2014) 640–646
642
Table 1 Subject Characteristics. Patients grouped into the LBP (TFA-LBP) group reported the presence of back pain at least once per week. Patients grouped into the no pain (TFA-NP) group reported no back pain or pain only once or twice in the past month. * Indicates a significant difference from the able-bodied control group. There were no significant differences between patient groups. Self-selected walking velocity (SSWV) was not available for one TFA-NP subject. Subject
Subject characteristics Age (yrs)
Height (m)
Mass (kg)
Leg Length (m)
SSWV (m/s)
Back pain in past month?
Months since first ambulation
Patient (low back pain) TFA-LBP 1 Male TFA-LBP 2 Male TFA-LBP 3 Male TFA-LBP 4 Male TFA-LBP 5 Male TFA-LBP 6 Male TFA-LBP 7 Male TFA-LBP 8 Male TFA-LBP 9 Male
Gender
32 28 39 30 36 26 25 38 35
1.85 1.75 1.83 1.86 1.83 1.80 1.68 1.75 1.86
105.7 81.6 105.0 78.0 98.2 77.3 84.6 94.1 91.1
0.94 0.92 0.99 0.98 0.93 0.94 0.88 0.91 0.95
1.36 1.30 1.27 1.09 1.35 1.28 1.11 1.23 1.11
All the time 4–6/week 1/week 1/week 1/week Every day 2–3/week 1/week 2–3/week
25 19 42 5 71 8 10 23 7
Average (SD)
32.1* (5.2)
1.80 (0.06)
90.6* (10.9)
0.94 (0.03)
1.23 (0.11)
Patient (no pain) TFA-NP 1 Male TFA-NP 2 Male TFA-NP 3 Male TFA-NP 4 Male TFA-NP 5 Male TFA-NP 6 Male TFA-NP 7 Male
25 36 27 33 26 23 29
1.68 1.78 1.85 1.75 1.83 1.88 1.75
72.6 95.3 86.4 85.9 77.1 84.4 74.8
0.87 0.94 0.91 0.93 0.92 0.98 0.90
1.20 1.20 1.11 1.28 1.44 1.45 –
Average (SD)
28.4 (6.4)
1.79 (0.07)
82.4 (8.0)
0.92 (0.03)
1.28 (0.14)
22 27 24 27 22 22 30 30 27 25 22 23
1.79 1.76 1.73 1.75 1.78 1.77 1.82 1.86 1.74 1.93 1.77 1.78
0.93 0.90 0.89 0.90 0.92 0.90 0.99 0.96 0.94 1.01 0.91 0.89
1.38 1.13 1.41 1.16 1.37 1.23 1.21 1.15 1.23 1.34 1.43 1.25
25.1 (3.1)
1.79 (0.06)
0.93 (0.04)
1.27 (0.11)
Control C1 C2 C3 C4 C5 C6 C7 C8 C9 C10 C11 C12 Average (SD)
Male Male Male Male Male Male Male Male Male Male Male Male
68.0 78.0 78.6 85.0 66.5 82.3 83.0 75.2 69.8 103.6 86.8 77.7 79.5 (10.0)
Supplementary material related to this article can be found, in the online version, at http://dx.doi.org/10.1016/j.gaitpost.2014.07.019. Frontal plane peak trunk angles (‘‘shoulder up’’) and ROM were significantly greater in both TFA groups than the able-bodied group (p < 0.001 for all comparisons). Pelvis angle profiles differed between the TFA and able-bodied groups. The patterns displayed by the TFA were more variable than the able-bodied and peak values (‘‘hip up’’) occurred in different phases of the gait cycle. For this reason, ROM was analyzed and compared across groups instead of peak values; ROM was significantly greater in the able-bodied group than the TFA-LBP (p = 0.025) and TFA-NP (p = 0.029) groups. Transverse plane peak trunk angles (‘‘shoulder back’’) were greater in the TFA-NP than LBP group (p = 0.003), which had greater values than able-bodied controls (p = 0.001). Transverse plane trunk ROM was significantly lower in the TFA groups than able-bodied controls (p < 0.001). There was a significant main effect of speed (p = 0.049), and transverse plane trunk ROM decreased with increasing speed. For pelvis motion, the able-bodied group exhibited significantly greater peak motion (‘‘hip back’’) than the TFA-LBP (p < 0.001) and TFA-NP (p = 0.009) groups. The ablebodied control group utilized the greatest pelvis ROM and the TFA-LBP the least.
3.2. Continuous relative phase (Figs. 2 and 3) In the sagittal plane, there were no significant interactions or main effects of speed. Both TFA groups showed greater out-of-phase coordination than able-bodied control subjects across all speeds (p < 0.001 for both comparisons). However, the presence of LBP in the TFA group did not affect sagittal plane CRP. In the frontal plane, there was a significant interaction between group and speed (p = 0.014). Both TFA groups showed more in-phase coordination then the ablebodied group during speeds 1 and 2 (p < 0.001 for both comparisons), and the TFALBP group showed greater in-phase coordination than able-bodied during speed 3. Frontal plane CRP increased from Speed 1 to 3 in the TFA-LBP group. In the transverse plane, there was a significant interaction between group and speed (p = 0.018). The TFA groups experienced greater increases in CRP as speed
23 (21)
Once Once Once No No No Once
or twice or twice or twice
or twice
29 62 52 62 47 6 41 43 (20)
No No No No No No No No No No No No
increased than able-bodied control subjects. The presence of LBP in the TFA group did not affect transverse plane CRP. 3.3. CRP Variability (Figs. 2 and 3) There were no significant interactions or differences among groups or across speeds for CRP variability in the sagittal and transverse planes. In the frontal plane, there was a significant interaction (p = 0.005) such that CRP variability was significantly greater in the able-bodied control subjects than the TFA-NP (p = 0.001) and TFA-LBP (p = 0.008) groups at Speed 1.
4. Discussion This study provides a direct comparison of both kinematics and inter-segmental coordination in 16 individuals with TFA with and without LBP and 12 able-bodied controls. In partial support of our first hypothesis, CRP in the sagittal plane was more out-of-phase in individuals with TFA than in healthy control subjects and CRP in the frontal plane was more in-phase, which is consistent with studies on patients with LBP [13]. Contrary to the first hypothesis, transverse plane CRP was not significantly different between TFA with and without LBP and walking with an amputation did not result in different axial CRP values relative to able-bodied. Relatively large pelvis and trunk ROM typically occur in the transverse plane during walking and, therefore, movement in this plane was the primary interest. The peak magnitudes of trunk rotation were greater in individuals with TFA, which may be a risk
[(Fig._1)TD$IG]
E. Russell Esposito, J.M. Wilken / Gait & Posture 40 (2014) 640–646
643
Trunk and Pelvis Segment Angles
Angle (deg)
TFA-NP
TFA-LBP Pelvis Trunk
6 4
Angle (deg)
Frontal
Hip/Shoulder up (+)
Sagial
Anterior lt (+)
Control 20 18 16 14 12 10 8 6 4 2 0 -2
2 0 -2 -4 -6
4
Angle (deg)
Hip/Shoulder back (+)
Transverse
6
2 0 -2 -4 -6
0
20
40
60
80
100
0
20
Gait Cycle
40
60
Gait Cycle
80
100
0
20
40
60
80
100
Gait Cycle
Fig. 1. Ensemble averaged trunk and pelvis segment angles. All data are shown from speed 2. Positive pelvis segment angles indicate anterior tilt, hip up of the prosthetic limb (right limb for able-bodied subjects), and hip posterior rotation (‘‘hip back’’) of the prosthetic limb (right limb for able-bodied subjects) in the sagittal, frontal, and transverse planes, respectively. Positive trunk segment angles indicate anterior lean, shoulder up of the prosthetic side (right limb for able-bodied subjects), and shoulder back of the prosthetic side (right limb for able-bodied subjects) in the sagittal, frontal, and transverse planes, respectively.
factor for LBP [6–9], whereas the able-bodied subjects utilized greater ranges of pelvic motion, which is unlikely to be a source of LBP [10]. The TFA groups did not exhibit significantly different transverse plane coordination than able-bodied controls. At first glance, these results appear to contradict the TFA relative phase data of Goujon-Pillet et al. [15], however, their able-bodied and amputee subjects walked at different speeds (TFA: 1.0, ablebodied: 1.2 m/s). When comparing corresponding speeds between [15] and the present study, both the able-bodied and TFA groups exhibited highly similar CRP values between studies. The ability of the TFA groups to maintain axial pelvis–trunk coordination consistent with able-bodied participants may be accomplished by utilizing sub-optimal coordination in the sagittal and frontal planes. For the purposes of this study, movement coordination in healthy, able-bodied individuals without LBP is considered optimal. It is, however, unclear if deviations identified in the TFA groups suggest re-optimization following TFA to account for impaired limb function. In the sagittal plane, CRP values were consistently greater in the individuals with TFA compared to able-bodied controls. ROM in the sagittal plane is generally not as large as in the other two planes but the counter rotations found in the sagittal plane CRP values may contribute to alternating compressive and tensile forces on soft tissues. The primary concern in this plane is the large pelvis-trunk angle formed between the anterior-tilted pelvis and posteriortilted trunk segments. This posture can result from muscular imbalances and clinical reports relate weak lower abdominal
muscles [25] and tight posterior trunk and hip flexor muscles to LBP. In contrast to the sagittal plane, the frontal plane can have a relatively large ROM during walking and typically demonstrates the greatest out-of-phase motion between the trunk and pelvis segments in healthy, able-bodied individuals [13]. However, the lower frontal plane CRP values in the TFA groups indicate rigidity between the pelvis and trunk during walking by protective ‘‘guarding’’ or splinting of the segments [14,26,27] or a compensatory mechanism to enhance stability. While walking on level ground individuals with TFA may increase rigidity along some planes to allow attention to be diverted to stabilizing the center of mass, hiking the hip to lift the residual limb and prosthesis, and clearing the toe during swing. Constraining the degrees of freedom between the pelvis and trunk may also provide a level of stability for individuals with amputations, however this is speculative and has yet to be studied. Therapeutic interventions may focus on proper frontal plane rotation of the pelvis, particularly in early stance when the greatest load-bearing occurs. Uncoupling the side-to-side sway of the trunk and pelvis and allowing the segments to rotate asynchronously may decrease the muscular efforts needed to maintain their rigid motions [26]. Since ROM in the sagittal plane is generally lowest, therapeutic interventions may focus less on movement patterns and instead on hip flexor and trunk extensor flexibility, and on abdominal strengthening to decrease the lordotic curvature of the spine [10].
[(Fig._2)TD$IG]
E. Russell Esposito, J.M. Wilken / Gait & Posture 40 (2014) 640–646
644
Connuous Relave Phase Variability Control
Sagial
180
CRP Variability
120 CRP
TFA-LBP
90 60
30
0
Frontal
180
Frontal
90
150
In-phase
CRP Variability
120 CRP
Out-of-phase
60
30
90 60
60
30
30 0
0
Transverse
180
Transverse
90
150 CRP Variability
120 CRP
Out-of-phase
TFA-NP
TFA-LBP
0
In-phase
Control
Sagial
90
TFA-NP
150
In-phase
Out-of-phase
Connuous Relave Phase
90 60
60
30
30 0
0
20
40
60
80
100
Gait Cycle
0
0
20
40
60
80
100
Gait Cycle
Fig. 2. Average continuous relative phase (CRP) and CRP variability at each instant of the gait cycle. Data are shown only for speed 2. Note the vertical axes are different for the two variables.
In further support of our first hypothesis, CRP variability was generally similar among groups, with the only exception in the frontal plane at the slowest speed. Although low variability can be interpreted as a system less able to respond to and overcome perturbations [19], individuals with TFA and able-bodied control subjects did not have significantly different levels of variability. Seay et al. [13] reported similar levels of pelvis–trunk transverse plane CRP variability, whereas Selles et al. [19] found less variability in individuals with LBP. The likely explanation for this difference is that Selles et al. [19] used a wide range of thirteen different walking speeds, while Seay et al. [13] and the present study used only three. Additionally, Selles et al. [19] had lowerfunctioning individuals whose SSWV averaged 0.8 m/s while Seay et al. [13] had recreational runners who ran at least 20 miles per week. The individuals with TFA in the present study were highfunctioning with SSWVs that were not significantly different from able-bodied. In partial contrast to our second hypothesis, CRP only shifted to more out-of-phase with increasing speed in the TFA group in the
transverse plane and the TFA-LBP group in the frontal. The lack of other differences may be due to the walking speeds selected relative to other studies. Previous studies that have found changes in CRP with increasing speed have all used wider ranges of walking speeds [13,19,28]. The restricted range of speeds was chosen to match the abilities of the TFA group and is therefore, a possible reason that a phase transition with speed was not observed in all conditions. Additional limitations of the present study require consideration. It has been reported that individuals with resolved LBP, who have not experienced pain during the past six months, still exhibit CRP and CRP variability values that differ from controls [13,29,30]. Only three of the participants in the current study had no back pain in the past month and it is unknown if they experienced pain prior to that time point. Given the prevalence of LBP in individuals with amputations, it was not possible to find a distinct subset of TFA with no history of LBP. It is, therefore, unknown whether amputation or the presence of any pain in recent months plays a larger role in pelvis and trunk coordination. Other factors not
[(Fig._3)TD$IG]
E. Russell Esposito, J.M. Wilken / Gait & Posture 40 (2014) 640–646
Connuous Relave Phase
CRP
120
*
*
*
90 60
Sagial
90
Variability
150
Connuous Relave Phase Variability Control TFA - NP TFA - LBP
Sagial
180
645
Control TFA - NP TFA - LBP
60
30
30 0
0
Frontal
180
CRP
90
*
*
*
60
Variability
†
150 120
Frontal
90
60
* 30
30 0
0
Transverse †
180 150
Variability
120
CRP
Transverse
90
90 60
60
30
30
0
0
Speed 1
Speed 2
Speed 3
Speed 1
Speed 2
Speed 3
Fig. 3. Ensemble averages and standard deviation bars of continuous relative phase (CRP) (left) and continuous relative phase variability (right). Note the vertical axes are different for the two variables. Averages are taken across the entire gait cycle. Standard deviation bars of the CRP indicate the variability across subjects, not the CRP variability. Similarly, the standard deviation bars of the CRP variability are the variability across subjects. *Significant differences between groups. ySignificant differences between speeds.
investigated, such as strength and prosthesis type could also have influenced walking kinematics in the TFA population. Thirteen of the subjects, however, wore the same make and model of prosthesis and all TFA had the same prosthetist. Additionally, it is unknown if coordination patterns can be viewed as a causal factor for developing LBP since both the painful and pain-free subjects experienced similar findings. Prospective studies investigating the long-term effects of inphase trunk–pelvis coordination on the development of LBP would be needed to establish a direct link. Lastly, intersegmental coordination and coordination variability are robust features of locomotion with various quantification methods (CRP, discrete relative phase, vector coding, cross correlations, principle components analysis, etc.). CRP was selected because it was allowed the researchers to address the questions posed and compare to previous studies.
5. Conclusion Individuals with TFA with and without LBP exhibited similar transverse plane movement coordination to able bodied. This was achieved potentially at the expense of optimal coordination in the sagittal and frontal planes. The presence of LBP in individuals with TFA had few effects on coordination or coordination variability. Only the individuals with TFA may have found the increases in speed challenging enough to elicit a transition in transverse plane coordination. With few exceptions, coordination variability was not different across groups or across speeds. Although the cause and effect relationship between CRP and future development of LBP has yet to be determined, the results of this study add to the literature attempting to characterize biomechanical parameters of LBP in high-risk populations. In the future, clinicians may further emphasize proper rotations of the pelvis and trunk during walking
E. Russell Esposito, J.M. Wilken / Gait & Posture 40 (2014) 640–646
646
as a potential means for preventing LBP in individuals with unilateral TFA. Acknowledgements This study was supported by the Telemedicine and Advanced Technology Research Center and the Center for Rehabilitation Sciences Research, Uniformed Services University of the Health Sciences, Bethesda, MD. Conflict of interest statement The authors declare that they have no conflict of interest. References [1] Pai S, Sundaram LJ. Low back pain: an economic assessment in the United States. Orthop Clin N Am 2004;35:1–5. [2] Smith DG, Ehde DM, Legro MW, Reiber GE, del Aguila M, Boone DA. Phantom limb, residual limb, and back pain after lower extremity amputations. Clin Orthop 1999;29–38. [3] Friberg O. Biomechanical significance of the correct length of lower-limb prostheses – a clinical and radiological study. Prosthet Orthot Int 1984;8:124–9. [4] Kulkarni J, Gaine WJ, Buckley JG, Rankine JJ, Adams J. Chronic low back pain in traumatic lower limb amputees. Clin Rehabil 2005;19:81–6. [5] Hendershot BD, Nussbaum MA. Persons with lower-limb amputation have impaired trunk postural control while maintaining seated balance. Gait Posture 2013;38:438–43. [6] Morgenroth DC, Orendurff MS, Shakir A, Segal A, Shofer J, Czerniecki JM. The relationship between lumbar spine kinematics during gait and low-back pain in transfemoral amputees. Am J Phys Med Rehabil 2010;89:635–43. [7] Fujiwara A, Lim TH, An HS, Tanaka N, Jeon CH, Andersson GB, et al. The effect of disc degeneration and facet joint osteoarthritis on the segmental flexibility of the lumbar spine. Spine (Phila Pa 1976) 2000;25:3036–44. [8] Krismer M, Haid C, Rabl W. The contribution of anulus fibers to torque resistance. Spine 1996;21:2551–7. [9] Schmidt H, Kettler A, Heuer F, Simon U, Claes L, Wilke HJ. Intradiscal pressure, shear strain, and fiber strain in the intervertebral disc under combined loading. Spine 2007;32:748–55. [10] McGregor AH, Hukins DWL. Lower limb involvement in spinal function and low back pain. J Back Musculoskelet 2009;22:219–22. [11] Shakir A, Barr K, Segal A, Orendurff M, Pecoraro J, Czerniecki J. Poster 152. Sagittal plane biomechanics and low back pain in transfemoral amputees. Arch Phys Med Rehabil 2003;84:E31. [12] Hamill J, Van Emmerik REA, Heiderscheit BC, Li L. Variability and stability: a dynamical systems perspective. J Appl Biomech 2000;16:392–3.
[13] Seay JF, Van Emmerik REA, Hamill J. Low back pain status affects pelvis–trunk coordination and variability during walking and running. Clin Biomech 2011;26:572–8. [14] Lamoth CJ, Meijer OG, Wuisman PI, van Dieen JH, Levin MF, Beek PJ. Pelvis– thorax coordination in the transverse plane during walking in persons with nonspecific low back pain. Spine (Phila Pa 1976) 2002;27:E92–9. [15] Goujon-Pillet H, Sapin E, Fode P, Lavaste F. Three-dimensional motions of trunk and pelvis during transfemoral amputee gait. Arch Phys Med Rehabil 2008;89:87–94. [16] Marras WS, Lavender SA, Leurgans SE, Rajulu SL, Allread WG, Fathallah FA, et al. The role of dynamic three-dimensional trunk motion in occupationallyrelated low back disorders. The effects of workplace factors, trunk position, and trunk motion characteristics on risk of injury. Spine (Phila Pa 1976) 1993;18:617–28. [17] Glass L. Synchronization and rhythmic processes in physiology. Nature 2001;410:277–84. [18] Hamill J, van Emmerik RE, Heiderscheit BC, Li L. A dynamical systems approach to lower extremity running injuries. Clin Biomech (Bristol Avon) 1999;14:297–308. [19] Selles RW, Wagenaar RC, Smith TH, Wuisman PIJM. Disorders in trunk rotation during walking in patients with low back pain: a dynamical systems approach. Clin Biomech 2001;16:175–81. [20] vanEmmerik REA, Wagenaar RC. Effect of walking velocity on relative phase dynamics in the trunk in human walking. J Biomech 1996;29:1175–84. [21] Legro MW, Reiber GD, Smith DG, del Aguila M, Larsen J, Boone D. Prosthesis evaluation questionnaire for persons with lower limb amputations: assessing prosthesis-related quality of life. Arch Phys Med Rehabil 1998;79:931–8. [22] Wilken JM, Rodriguez KM, Brawner M, Darter BJ. Reliability and minimal detectible change values for gait kinematics and kinetics in healthy adults. Gait Posture 2012;35:301–7. [23] Vaughan CL, O’Malley MJ. Froude and the contribution of naval architecture to our understanding of bipedal locomotion. Gait Posture 2005;21:350–62. [24] Zeni JA, Richards JG, Higginson JS. Two simple methods for determining gait events during treadmill and overground walking using kinematic data. Gait Posture 2008;27:710–4. [25] Hodges PW, Richardson CA. Inefficient muscular stabilization of the lumbar spine associated with low back pain – a motor control evaluation of transversus abdominis. Spine 1996;21:2640–50. [26] Arendt-Nielsen L, Graven-Nielsen T, Svarrer H. The influence of low back pain on muscle activity and coordination during gait: a clinical experimental study. Pain 1995;64:231–40. [27] Ahern DK, Hannon DJ, Goreczny AJ, Follick MJ, Parziale JR. Correlation of chronic low-back pain behavior and muscle function examination of the flexion–relaxation response. Spine (Phila Pa 1976) 1990;15:92–5. [28] Van Emmerik REA, McDermott WJ, Haddad JM, Van Wegen EEH. Age-related changes in upper body adaptation to walking speed in human locomotion. Gait Posture 2005;22:233–9. [29] Seay JF, Van Emmerik REA, Hamill J. Influence of low back pain status on pelvis–trunk coordination during walking and running. Spine 2011;36: E1070–79. [30] Taylor NF, Evans OM, Goldie PA. The effect of walking faster on people with acute low back pain. Eur Spine J 2003;12:166–72.
Contents lists available at ScienceDirect
Gait & Posture journal homepage: www.elsevier.com/locate/gaitpost
The relationship between pelvis–trunk coordination and low back pain in individuals with transfemoral amputations§ Elizabeth Russell Esposito *, Jason M. Wilken Center for the Intrepid, Department of Orthopaedics and Rehabilitation, Brooke Army Medical Center, Ft. Sam Houston, TX, USA
A R T I C L E I N F O
A B S T R A C T
Article history: Received 19 April 2013 Received in revised form 14 July 2014 Accepted 17 July 2014
Low back pain (LBP) is common in individuals with transfemoral amputation and may result from altered gait mechanics associated with prosthetic use. Inter-segmental coordination, assessed through continuous relative phase (CRP), has been used to identify specific patterns as risk factors. The purpose of this study was to explore pelvis and trunk inter-segmental coordination across three walking speeds in individuals with transfemoral amputations with and without LBP. Nine individuals with transfemoral amputations with LBP and seven without pain were compared to twelve able-bodied subjects. Subjects underwent a gait analysis while walking at slow, moderate, and fast speeds. CRP and CRP variability were calculated from three-dimensional pelvis and trunk segment angles. A two-way ANOVA and post hoc tests assessed statistical significance. Individuals with transfemoral amputation demonstrated some coordination patterns that were different from able-bodied individuals, but consistent with previous reports on persons with LBP. The patient groups maintained transverse plane CRP consistent with ablebodied participants (p = 0.966), but not sagittal (p < 0.001) and frontal plane CRP (p = 0.001). Sagittal and frontal CRP may have been re-optimized based on new sets of constraints, such as protective rigidity of the segments, muscular strength limitations, or prosthesis limitations. Patients with amputations and without LBP exhibited few differences. Only frontal and transverse CRP shifted toward out-of-phase as speed increased in the patient group with LBP. Although a cause and effect relationship between CRP and future development of back pain has yet to be determined, these results add to the literature characterizing biomechanical parameters of back pain in high-risk populations. Published by Elsevier B.V.
Keywords: Amputee Military Continuous relative phase Walking Variability
1. Introduction Low back pain (LBP) affects 80% of adults at some point during their lifetime [1]. The rates of LBP in individuals with amputations are significantly greater than in the general population [2]. LBP is particularly troublesome for individuals with transfemoral amputation (TFA), compared to trans-tibial amputation. Muscular imbalance, leg length discrepancies and altered gait mechanics may contribute to the increased incidence of pain in this population [3–5]. Abnormal gait patterns, to include altered motion of the trunk and pelvis, may also contribute to the increased incidence of pain in this population. Numerous studies have sought to determine biomechanical risk factors for LBP in the
§ The view(s) expressed herein are those of the author(s) and do not reflect the official policy or position of Brooke Army Medical Center, the U.S. Army Medical Department, the U.S. Army Office of the Surgeon General, the Department of the Army, Department of Defense or the U.S. Government. * Corresponding author. Tel.: +1 2105395824. E-mail address: [email protected] (E. Russell Esposito).
http://dx.doi.org/10.1016/j.gaitpost.2014.07.019 0966-6362/Published by Elsevier B.V.
general population, but few have investigated a relationship between gait and pain in a population with lower extremity amputation. Standard gait analyses, including peak kinematics and ranges of motion (ROM), have been previously used to identify gait deviations associated with LBP. For example, transverse plane trunk rotation has been found to be greater in individuals with TFA who have LBP than in those without LBP or able-bodied individuals [6]. These greater rotations may contribute to disk degeneration [7] and/or soft tissue strain [8,9] which are potential sources of LBP. However, the extremes of lumbar motion are not universally associated with an increased risk in individuals with LBP [10] or amputation [11]. An alternative to the use of discrete measures is the assessment of movement coordination between adjacent segments using continuous data from the entire gait cycle [12]. Continuous measures, such as continuous relative phase (CRP), have been effective in identifying gait deviations associated with LBP. CRP has been used to compare pelvis and trunk coordination in populations with LBP [13,14] and amputations [15]. Coordination can be defined using CRP to categorize movement patterns
E. Russell Esposito, J.M. Wilken / Gait & Posture 40 (2014) 640–646
ranging from ‘‘in-phase’’, where adjacent segments rotate as a unit in the same direction at the same speed, to ‘‘out-of-phase’’, where adjacent segments concurrently rotate at the same speed in the opposite directions to create bending or twisting motions. Seay et al. [13] found that sufferers of LBP had greater synchronous, in-phase pelvis and trunk rotations in the frontal plane during walking than able-bodied control subjects. In the transverse plane, able-bodied patients with LBP [14], as well as patients with TFA [15], never achieved the same degree of axial out-of-phase coordination as healthy, able-bodied subjects. Although no single plane of motion has been identified as the primary contributor to LBP, studies suggest pain is associated with side bending, twisting, and various asymmetrical movements of the trunk and pelvis [16]. Across all three planes, the variability of these phase relationships may provide additional information on LBP. Variability within a biological system allows for a level of flexibility, adaptability and, thus, the ability to overcome perturbations [17]. A movement with low coordination variability may be perceived as safer because of its predictability but its constant repetition may create an overuse situation [18] such as LBP. One study found lower coordination variability between axial trunk– pelvis rotations during walking in individuals with LBP relative to able-bodied controls [19], however another study found no differences between groups in any plane [13]. The differences between these two studies may be due to the subject populations; Selles et al. [19] had patients who could not work full time due to pain and Seay et al. [13] had recreational runners who ran at least 20 miles per week. Changes in walking speed can directly influence coordination and coordination variability. When walking overground, increasing speed shifts axial pelvis–trunk coordination from an in-phase relationship to more out-of-phase in healthy, able-bodied individuals [20]. Sufferers of LBP have less ability to switch coordination patterns as speed increases [13,14,19] but it is unknown how speed may influence coordination patterns in individuals with amputations and if the resulting patterns may be indicative of risk factors for LBP. Although the increased incidence of LBP in individuals with TFA may result from altered gait mechanics from prosthetic use, it is still unknown how coordination and coordination variability may be affected by the development of LBP after TFA. The first objective was to determine how pelvis–trunk kinematics, coordination and coordination variability differed among individuals with TFA who did and did not experience LBP and an able-bodied control group. We hypothesized that, similar to [13–15,19], individuals with TFA would exhibit greater in-phase coordination in the transverse and frontal planes and greater out-of-phase coordination in the sagittal plane than, able-bodied individuals. In agreement with the results of Seay et al. [13], we expected coordination variability to stay consistent across groups [13]. The second objective was to determine if pelvis–trunk kinematics, coordination, and coordination variability modulated with walking speed. We hypothesized that, similar to [13,19,20], CRP would increase with increasing speed and, similar to [13], variability would remain consistent across walking speeds.
641
past 4 weeks. TFA-LBP indicated a frequency of LBP of one time per week to all the time or almost all the time. Responses on LBP frequency were collected as part of the Prosthetic Evaluation Questionnaire [21]. Five of the TFA-NP group wore the Otto Bock X2 prosthesis and two wore their standard mechanical prostheses that were not micro-processor controlled. Eight of the TFA-LBP wore the X2 and one wore a mechanical prosthesis. Limb lengths were matched to intact side and subjects were K4 ambulators. Able-bodied individuals had no current or previous orthopedic injuries affecting gait and no injuries to the low back or LBP. All subjects read and signed an informed consent approved by the Institutional Review Board. The experimental setup for overground walking consisted of a 26-camera motion capture system (120 Hz; Motion Analysis, Santa Rosa, CA). Fifty-seven spherical, retro-reflective markers were placed on anatomical landmarks and on segments of the limbs, torso and head [22]. Participants walked at three speeds (Speed 1, 2, and 3) based on dimensionless Froude numbers of 0.10, 0.16, and 0.23 [23] which corresponded to average speeds of 1.0, 1.2 and 1.4 m/s, respectively. Speed was controlled with an auditory tone that sounded as walking speed fell within a 5% range. Participants completed eight trials at each of the speeds during which they walked across approximately 20 m of level ground, the middle portion of which was analyzed. All trials were labeled in EvART (Motion Analysis Corporation, Santa Barbara, CA) and exported to Visual3D (version 4.96.7, C-Motion, Germantown, MD). Raw marker data were interpolated and filtered using a 4th order, low-pass Butterworth filter with a cutoff frequency of 6 Hz. Heel strike was identified according to the methods described by [24]. A successful trial was defined as one containing a complete stride initiated with the prosthetic limb for the TFA groups and initiated with the right leg for able-bodied subjects. CRP was then calculated on each stride using the method of Hamill et al., [18]. Briefly, pelvis and trunk segmental angular velocities were calculated as the first derivative of segment angular position. Angular position and velocity were normalized to 101 data points and scaled between 1 and 1. These normalized position and angular velocity vectors were plotted relative to each other for the trunk segment and for the pelvis segment. Phase angles were calculated as the angle formed between a line from the origin to each data point and the right horizontal. CRP was defined as the difference between the trunk and pelvis phase angles at each percent of the stride. CRP variability was defined as the standard deviation of the CRP at each time point across all eight trials. Individual subject means were calculated for peak segment angles, CRP and CRP variability at each of the three walking speeds. Shapiro–Wilk tests were conducted for normality and a two-way, repeated-measures ANOVA (SPSS Inc, Chicago, IL, USA) examined the main effects. Tukeys post hoc tests were used to compare groups (p < 0.05) and paired t-tests with Bonferroni corrections were used to compare speeds. 3. Results Individual subject characteristics are presented in Table 1. Only age (p = 0.002) and body mass (p = 0.045) were significantly greater in the TFA-LBP group than the able-bodied control group. Self-selected walking velocity (SSWV) was not significantly different from Speed 2. 3.1. Kinematics (Fig. 1 and Supplemental material)
2. Methods Sixteen individuals with unilateral, traumatic TFA and twelve able-bodied control subjects participated in this study. Seven of the TFA subjects were categorized in the no pain group (TFA-NP) and nine were categorized in the LBP group (TFA-LBP). TFA-NP indicated a frequency of LBP never or only once or twice in the
There were no significant interactions for any of the trunk and pelvis kinematics. There were no significant main effects of speed in the sagittal and frontal planes, but there were significant group effects. TFA-LBP maintained significantly less sagittal plane anterior trunk lean than TFA-NP and able-bodied (p < 0.001) and less sagittal plane ROM than the able-bodied group (p = 0.022). Anterior pelvic tilt was significantly greater in both TFA groups relative to the able-bodied (p < 0.001) and sagittal plane pelvis ROM was significantly greater in the TFA-LBP group than the TFA-NP (p < 0.001) and able-bodied controls (p = 0.004).
E. Russell Esposito, J.M. Wilken / Gait & Posture 40 (2014) 640–646
642
Table 1 Subject Characteristics. Patients grouped into the LBP (TFA-LBP) group reported the presence of back pain at least once per week. Patients grouped into the no pain (TFA-NP) group reported no back pain or pain only once or twice in the past month. * Indicates a significant difference from the able-bodied control group. There were no significant differences between patient groups. Self-selected walking velocity (SSWV) was not available for one TFA-NP subject. Subject
Subject characteristics Age (yrs)
Height (m)
Mass (kg)
Leg Length (m)
SSWV (m/s)
Back pain in past month?
Months since first ambulation
Patient (low back pain) TFA-LBP 1 Male TFA-LBP 2 Male TFA-LBP 3 Male TFA-LBP 4 Male TFA-LBP 5 Male TFA-LBP 6 Male TFA-LBP 7 Male TFA-LBP 8 Male TFA-LBP 9 Male
Gender
32 28 39 30 36 26 25 38 35
1.85 1.75 1.83 1.86 1.83 1.80 1.68 1.75 1.86
105.7 81.6 105.0 78.0 98.2 77.3 84.6 94.1 91.1
0.94 0.92 0.99 0.98 0.93 0.94 0.88 0.91 0.95
1.36 1.30 1.27 1.09 1.35 1.28 1.11 1.23 1.11
All the time 4–6/week 1/week 1/week 1/week Every day 2–3/week 1/week 2–3/week
25 19 42 5 71 8 10 23 7
Average (SD)
32.1* (5.2)
1.80 (0.06)
90.6* (10.9)
0.94 (0.03)
1.23 (0.11)
Patient (no pain) TFA-NP 1 Male TFA-NP 2 Male TFA-NP 3 Male TFA-NP 4 Male TFA-NP 5 Male TFA-NP 6 Male TFA-NP 7 Male
25 36 27 33 26 23 29
1.68 1.78 1.85 1.75 1.83 1.88 1.75
72.6 95.3 86.4 85.9 77.1 84.4 74.8
0.87 0.94 0.91 0.93 0.92 0.98 0.90
1.20 1.20 1.11 1.28 1.44 1.45 –
Average (SD)
28.4 (6.4)
1.79 (0.07)
82.4 (8.0)
0.92 (0.03)
1.28 (0.14)
22 27 24 27 22 22 30 30 27 25 22 23
1.79 1.76 1.73 1.75 1.78 1.77 1.82 1.86 1.74 1.93 1.77 1.78
0.93 0.90 0.89 0.90 0.92 0.90 0.99 0.96 0.94 1.01 0.91 0.89
1.38 1.13 1.41 1.16 1.37 1.23 1.21 1.15 1.23 1.34 1.43 1.25
25.1 (3.1)
1.79 (0.06)
0.93 (0.04)
1.27 (0.11)
Control C1 C2 C3 C4 C5 C6 C7 C8 C9 C10 C11 C12 Average (SD)
Male Male Male Male Male Male Male Male Male Male Male Male
68.0 78.0 78.6 85.0 66.5 82.3 83.0 75.2 69.8 103.6 86.8 77.7 79.5 (10.0)
Supplementary material related to this article can be found, in the online version, at http://dx.doi.org/10.1016/j.gaitpost.2014.07.019. Frontal plane peak trunk angles (‘‘shoulder up’’) and ROM were significantly greater in both TFA groups than the able-bodied group (p < 0.001 for all comparisons). Pelvis angle profiles differed between the TFA and able-bodied groups. The patterns displayed by the TFA were more variable than the able-bodied and peak values (‘‘hip up’’) occurred in different phases of the gait cycle. For this reason, ROM was analyzed and compared across groups instead of peak values; ROM was significantly greater in the able-bodied group than the TFA-LBP (p = 0.025) and TFA-NP (p = 0.029) groups. Transverse plane peak trunk angles (‘‘shoulder back’’) were greater in the TFA-NP than LBP group (p = 0.003), which had greater values than able-bodied controls (p = 0.001). Transverse plane trunk ROM was significantly lower in the TFA groups than able-bodied controls (p < 0.001). There was a significant main effect of speed (p = 0.049), and transverse plane trunk ROM decreased with increasing speed. For pelvis motion, the able-bodied group exhibited significantly greater peak motion (‘‘hip back’’) than the TFA-LBP (p < 0.001) and TFA-NP (p = 0.009) groups. The ablebodied control group utilized the greatest pelvis ROM and the TFA-LBP the least.
3.2. Continuous relative phase (Figs. 2 and 3) In the sagittal plane, there were no significant interactions or main effects of speed. Both TFA groups showed greater out-of-phase coordination than able-bodied control subjects across all speeds (p < 0.001 for both comparisons). However, the presence of LBP in the TFA group did not affect sagittal plane CRP. In the frontal plane, there was a significant interaction between group and speed (p = 0.014). Both TFA groups showed more in-phase coordination then the ablebodied group during speeds 1 and 2 (p < 0.001 for both comparisons), and the TFALBP group showed greater in-phase coordination than able-bodied during speed 3. Frontal plane CRP increased from Speed 1 to 3 in the TFA-LBP group. In the transverse plane, there was a significant interaction between group and speed (p = 0.018). The TFA groups experienced greater increases in CRP as speed
23 (21)
Once Once Once No No No Once
or twice or twice or twice
or twice
29 62 52 62 47 6 41 43 (20)
No No No No No No No No No No No No
increased than able-bodied control subjects. The presence of LBP in the TFA group did not affect transverse plane CRP. 3.3. CRP Variability (Figs. 2 and 3) There were no significant interactions or differences among groups or across speeds for CRP variability in the sagittal and transverse planes. In the frontal plane, there was a significant interaction (p = 0.005) such that CRP variability was significantly greater in the able-bodied control subjects than the TFA-NP (p = 0.001) and TFA-LBP (p = 0.008) groups at Speed 1.
4. Discussion This study provides a direct comparison of both kinematics and inter-segmental coordination in 16 individuals with TFA with and without LBP and 12 able-bodied controls. In partial support of our first hypothesis, CRP in the sagittal plane was more out-of-phase in individuals with TFA than in healthy control subjects and CRP in the frontal plane was more in-phase, which is consistent with studies on patients with LBP [13]. Contrary to the first hypothesis, transverse plane CRP was not significantly different between TFA with and without LBP and walking with an amputation did not result in different axial CRP values relative to able-bodied. Relatively large pelvis and trunk ROM typically occur in the transverse plane during walking and, therefore, movement in this plane was the primary interest. The peak magnitudes of trunk rotation were greater in individuals with TFA, which may be a risk
[(Fig._1)TD$IG]
E. Russell Esposito, J.M. Wilken / Gait & Posture 40 (2014) 640–646
643
Trunk and Pelvis Segment Angles
Angle (deg)
TFA-NP
TFA-LBP Pelvis Trunk
6 4
Angle (deg)
Frontal
Hip/Shoulder up (+)
Sagial
Anterior lt (+)
Control 20 18 16 14 12 10 8 6 4 2 0 -2
2 0 -2 -4 -6
4
Angle (deg)
Hip/Shoulder back (+)
Transverse
6
2 0 -2 -4 -6
0
20
40
60
80
100
0
20
Gait Cycle
40
60
Gait Cycle
80
100
0
20
40
60
80
100
Gait Cycle
Fig. 1. Ensemble averaged trunk and pelvis segment angles. All data are shown from speed 2. Positive pelvis segment angles indicate anterior tilt, hip up of the prosthetic limb (right limb for able-bodied subjects), and hip posterior rotation (‘‘hip back’’) of the prosthetic limb (right limb for able-bodied subjects) in the sagittal, frontal, and transverse planes, respectively. Positive trunk segment angles indicate anterior lean, shoulder up of the prosthetic side (right limb for able-bodied subjects), and shoulder back of the prosthetic side (right limb for able-bodied subjects) in the sagittal, frontal, and transverse planes, respectively.
factor for LBP [6–9], whereas the able-bodied subjects utilized greater ranges of pelvic motion, which is unlikely to be a source of LBP [10]. The TFA groups did not exhibit significantly different transverse plane coordination than able-bodied controls. At first glance, these results appear to contradict the TFA relative phase data of Goujon-Pillet et al. [15], however, their able-bodied and amputee subjects walked at different speeds (TFA: 1.0, ablebodied: 1.2 m/s). When comparing corresponding speeds between [15] and the present study, both the able-bodied and TFA groups exhibited highly similar CRP values between studies. The ability of the TFA groups to maintain axial pelvis–trunk coordination consistent with able-bodied participants may be accomplished by utilizing sub-optimal coordination in the sagittal and frontal planes. For the purposes of this study, movement coordination in healthy, able-bodied individuals without LBP is considered optimal. It is, however, unclear if deviations identified in the TFA groups suggest re-optimization following TFA to account for impaired limb function. In the sagittal plane, CRP values were consistently greater in the individuals with TFA compared to able-bodied controls. ROM in the sagittal plane is generally not as large as in the other two planes but the counter rotations found in the sagittal plane CRP values may contribute to alternating compressive and tensile forces on soft tissues. The primary concern in this plane is the large pelvis-trunk angle formed between the anterior-tilted pelvis and posteriortilted trunk segments. This posture can result from muscular imbalances and clinical reports relate weak lower abdominal
muscles [25] and tight posterior trunk and hip flexor muscles to LBP. In contrast to the sagittal plane, the frontal plane can have a relatively large ROM during walking and typically demonstrates the greatest out-of-phase motion between the trunk and pelvis segments in healthy, able-bodied individuals [13]. However, the lower frontal plane CRP values in the TFA groups indicate rigidity between the pelvis and trunk during walking by protective ‘‘guarding’’ or splinting of the segments [14,26,27] or a compensatory mechanism to enhance stability. While walking on level ground individuals with TFA may increase rigidity along some planes to allow attention to be diverted to stabilizing the center of mass, hiking the hip to lift the residual limb and prosthesis, and clearing the toe during swing. Constraining the degrees of freedom between the pelvis and trunk may also provide a level of stability for individuals with amputations, however this is speculative and has yet to be studied. Therapeutic interventions may focus on proper frontal plane rotation of the pelvis, particularly in early stance when the greatest load-bearing occurs. Uncoupling the side-to-side sway of the trunk and pelvis and allowing the segments to rotate asynchronously may decrease the muscular efforts needed to maintain their rigid motions [26]. Since ROM in the sagittal plane is generally lowest, therapeutic interventions may focus less on movement patterns and instead on hip flexor and trunk extensor flexibility, and on abdominal strengthening to decrease the lordotic curvature of the spine [10].
[(Fig._2)TD$IG]
E. Russell Esposito, J.M. Wilken / Gait & Posture 40 (2014) 640–646
644
Connuous Relave Phase Variability Control
Sagial
180
CRP Variability
120 CRP
TFA-LBP
90 60
30
0
Frontal
180
Frontal
90
150
In-phase
CRP Variability
120 CRP
Out-of-phase
60
30
90 60
60
30
30 0
0
Transverse
180
Transverse
90
150 CRP Variability
120 CRP
Out-of-phase
TFA-NP
TFA-LBP
0
In-phase
Control
Sagial
90
TFA-NP
150
In-phase
Out-of-phase
Connuous Relave Phase
90 60
60
30
30 0
0
20
40
60
80
100
Gait Cycle
0
0
20
40
60
80
100
Gait Cycle
Fig. 2. Average continuous relative phase (CRP) and CRP variability at each instant of the gait cycle. Data are shown only for speed 2. Note the vertical axes are different for the two variables.
In further support of our first hypothesis, CRP variability was generally similar among groups, with the only exception in the frontal plane at the slowest speed. Although low variability can be interpreted as a system less able to respond to and overcome perturbations [19], individuals with TFA and able-bodied control subjects did not have significantly different levels of variability. Seay et al. [13] reported similar levels of pelvis–trunk transverse plane CRP variability, whereas Selles et al. [19] found less variability in individuals with LBP. The likely explanation for this difference is that Selles et al. [19] used a wide range of thirteen different walking speeds, while Seay et al. [13] and the present study used only three. Additionally, Selles et al. [19] had lowerfunctioning individuals whose SSWV averaged 0.8 m/s while Seay et al. [13] had recreational runners who ran at least 20 miles per week. The individuals with TFA in the present study were highfunctioning with SSWVs that were not significantly different from able-bodied. In partial contrast to our second hypothesis, CRP only shifted to more out-of-phase with increasing speed in the TFA group in the
transverse plane and the TFA-LBP group in the frontal. The lack of other differences may be due to the walking speeds selected relative to other studies. Previous studies that have found changes in CRP with increasing speed have all used wider ranges of walking speeds [13,19,28]. The restricted range of speeds was chosen to match the abilities of the TFA group and is therefore, a possible reason that a phase transition with speed was not observed in all conditions. Additional limitations of the present study require consideration. It has been reported that individuals with resolved LBP, who have not experienced pain during the past six months, still exhibit CRP and CRP variability values that differ from controls [13,29,30]. Only three of the participants in the current study had no back pain in the past month and it is unknown if they experienced pain prior to that time point. Given the prevalence of LBP in individuals with amputations, it was not possible to find a distinct subset of TFA with no history of LBP. It is, therefore, unknown whether amputation or the presence of any pain in recent months plays a larger role in pelvis and trunk coordination. Other factors not
[(Fig._3)TD$IG]
E. Russell Esposito, J.M. Wilken / Gait & Posture 40 (2014) 640–646
Connuous Relave Phase
CRP
120
*
*
*
90 60
Sagial
90
Variability
150
Connuous Relave Phase Variability Control TFA - NP TFA - LBP
Sagial
180
645
Control TFA - NP TFA - LBP
60
30
30 0
0
Frontal
180
CRP
90
*
*
*
60
Variability
†
150 120
Frontal
90
60
* 30
30 0
0
Transverse †
180 150
Variability
120
CRP
Transverse
90
90 60
60
30
30
0
0
Speed 1
Speed 2
Speed 3
Speed 1
Speed 2
Speed 3
Fig. 3. Ensemble averages and standard deviation bars of continuous relative phase (CRP) (left) and continuous relative phase variability (right). Note the vertical axes are different for the two variables. Averages are taken across the entire gait cycle. Standard deviation bars of the CRP indicate the variability across subjects, not the CRP variability. Similarly, the standard deviation bars of the CRP variability are the variability across subjects. *Significant differences between groups. ySignificant differences between speeds.
investigated, such as strength and prosthesis type could also have influenced walking kinematics in the TFA population. Thirteen of the subjects, however, wore the same make and model of prosthesis and all TFA had the same prosthetist. Additionally, it is unknown if coordination patterns can be viewed as a causal factor for developing LBP since both the painful and pain-free subjects experienced similar findings. Prospective studies investigating the long-term effects of inphase trunk–pelvis coordination on the development of LBP would be needed to establish a direct link. Lastly, intersegmental coordination and coordination variability are robust features of locomotion with various quantification methods (CRP, discrete relative phase, vector coding, cross correlations, principle components analysis, etc.). CRP was selected because it was allowed the researchers to address the questions posed and compare to previous studies.
5. Conclusion Individuals with TFA with and without LBP exhibited similar transverse plane movement coordination to able bodied. This was achieved potentially at the expense of optimal coordination in the sagittal and frontal planes. The presence of LBP in individuals with TFA had few effects on coordination or coordination variability. Only the individuals with TFA may have found the increases in speed challenging enough to elicit a transition in transverse plane coordination. With few exceptions, coordination variability was not different across groups or across speeds. Although the cause and effect relationship between CRP and future development of LBP has yet to be determined, the results of this study add to the literature attempting to characterize biomechanical parameters of LBP in high-risk populations. In the future, clinicians may further emphasize proper rotations of the pelvis and trunk during walking
E. Russell Esposito, J.M. Wilken / Gait & Posture 40 (2014) 640–646
646
as a potential means for preventing LBP in individuals with unilateral TFA. Acknowledgements This study was supported by the Telemedicine and Advanced Technology Research Center and the Center for Rehabilitation Sciences Research, Uniformed Services University of the Health Sciences, Bethesda, MD. Conflict of interest statement The authors declare that they have no conflict of interest. References [1] Pai S, Sundaram LJ. Low back pain: an economic assessment in the United States. Orthop Clin N Am 2004;35:1–5. [2] Smith DG, Ehde DM, Legro MW, Reiber GE, del Aguila M, Boone DA. Phantom limb, residual limb, and back pain after lower extremity amputations. Clin Orthop 1999;29–38. [3] Friberg O. Biomechanical significance of the correct length of lower-limb prostheses – a clinical and radiological study. Prosthet Orthot Int 1984;8:124–9. [4] Kulkarni J, Gaine WJ, Buckley JG, Rankine JJ, Adams J. Chronic low back pain in traumatic lower limb amputees. Clin Rehabil 2005;19:81–6. [5] Hendershot BD, Nussbaum MA. Persons with lower-limb amputation have impaired trunk postural control while maintaining seated balance. Gait Posture 2013;38:438–43. [6] Morgenroth DC, Orendurff MS, Shakir A, Segal A, Shofer J, Czerniecki JM. The relationship between lumbar spine kinematics during gait and low-back pain in transfemoral amputees. Am J Phys Med Rehabil 2010;89:635–43. [7] Fujiwara A, Lim TH, An HS, Tanaka N, Jeon CH, Andersson GB, et al. The effect of disc degeneration and facet joint osteoarthritis on the segmental flexibility of the lumbar spine. Spine (Phila Pa 1976) 2000;25:3036–44. [8] Krismer M, Haid C, Rabl W. The contribution of anulus fibers to torque resistance. Spine 1996;21:2551–7. [9] Schmidt H, Kettler A, Heuer F, Simon U, Claes L, Wilke HJ. Intradiscal pressure, shear strain, and fiber strain in the intervertebral disc under combined loading. Spine 2007;32:748–55. [10] McGregor AH, Hukins DWL. Lower limb involvement in spinal function and low back pain. J Back Musculoskelet 2009;22:219–22. [11] Shakir A, Barr K, Segal A, Orendurff M, Pecoraro J, Czerniecki J. Poster 152. Sagittal plane biomechanics and low back pain in transfemoral amputees. Arch Phys Med Rehabil 2003;84:E31. [12] Hamill J, Van Emmerik REA, Heiderscheit BC, Li L. Variability and stability: a dynamical systems perspective. J Appl Biomech 2000;16:392–3.
[13] Seay JF, Van Emmerik REA, Hamill J. Low back pain status affects pelvis–trunk coordination and variability during walking and running. Clin Biomech 2011;26:572–8. [14] Lamoth CJ, Meijer OG, Wuisman PI, van Dieen JH, Levin MF, Beek PJ. Pelvis– thorax coordination in the transverse plane during walking in persons with nonspecific low back pain. Spine (Phila Pa 1976) 2002;27:E92–9. [15] Goujon-Pillet H, Sapin E, Fode P, Lavaste F. Three-dimensional motions of trunk and pelvis during transfemoral amputee gait. Arch Phys Med Rehabil 2008;89:87–94. [16] Marras WS, Lavender SA, Leurgans SE, Rajulu SL, Allread WG, Fathallah FA, et al. The role of dynamic three-dimensional trunk motion in occupationallyrelated low back disorders. The effects of workplace factors, trunk position, and trunk motion characteristics on risk of injury. Spine (Phila Pa 1976) 1993;18:617–28. [17] Glass L. Synchronization and rhythmic processes in physiology. Nature 2001;410:277–84. [18] Hamill J, van Emmerik RE, Heiderscheit BC, Li L. A dynamical systems approach to lower extremity running injuries. Clin Biomech (Bristol Avon) 1999;14:297–308. [19] Selles RW, Wagenaar RC, Smith TH, Wuisman PIJM. Disorders in trunk rotation during walking in patients with low back pain: a dynamical systems approach. Clin Biomech 2001;16:175–81. [20] vanEmmerik REA, Wagenaar RC. Effect of walking velocity on relative phase dynamics in the trunk in human walking. J Biomech 1996;29:1175–84. [21] Legro MW, Reiber GD, Smith DG, del Aguila M, Larsen J, Boone D. Prosthesis evaluation questionnaire for persons with lower limb amputations: assessing prosthesis-related quality of life. Arch Phys Med Rehabil 1998;79:931–8. [22] Wilken JM, Rodriguez KM, Brawner M, Darter BJ. Reliability and minimal detectible change values for gait kinematics and kinetics in healthy adults. Gait Posture 2012;35:301–7. [23] Vaughan CL, O’Malley MJ. Froude and the contribution of naval architecture to our understanding of bipedal locomotion. Gait Posture 2005;21:350–62. [24] Zeni JA, Richards JG, Higginson JS. Two simple methods for determining gait events during treadmill and overground walking using kinematic data. Gait Posture 2008;27:710–4. [25] Hodges PW, Richardson CA. Inefficient muscular stabilization of the lumbar spine associated with low back pain – a motor control evaluation of transversus abdominis. Spine 1996;21:2640–50. [26] Arendt-Nielsen L, Graven-Nielsen T, Svarrer H. The influence of low back pain on muscle activity and coordination during gait: a clinical experimental study. Pain 1995;64:231–40. [27] Ahern DK, Hannon DJ, Goreczny AJ, Follick MJ, Parziale JR. Correlation of chronic low-back pain behavior and muscle function examination of the flexion–relaxation response. Spine (Phila Pa 1976) 1990;15:92–5. [28] Van Emmerik REA, McDermott WJ, Haddad JM, Van Wegen EEH. Age-related changes in upper body adaptation to walking speed in human locomotion. Gait Posture 2005;22:233–9. [29] Seay JF, Van Emmerik REA, Hamill J. Influence of low back pain status on pelvis–trunk coordination during walking and running. Spine 2011;36: E1070–79. [30] Taylor NF, Evans OM, Goldie PA. The effect of walking faster on people with acute low back pain. Eur Spine J 2003;12:166–72.
