Characteristics of Trunk Compensation Movements during Walking in People with Bilateral Transtibial Amputation: A Cross-sectional Study

ABSTRACT

Objectives: The gait characteristics of people with bilateral transtibial amputation are not well understood. This study aimed to clarify changes in trunk lateral bending and the trunk flexion angle during walking in people with bilateral transtibial amputation.

Methods: In this cross-sectional study, four participants with bilateral transtibial amputation who could walk without assistance (BTTA group) and ten able-bodied participants (control group) were recruited. The range of motion of trunk lateral bending, the trunk flexion angle, and other gait parameters during comfortable-speed and maximum-speed walking were measured using a three-dimensional motion analysis system and force plates. These parameters were compared between the amputees and the controls.

Results: During maximum-speed walking, the BTTA group walked slower with a smaller trunk flexion angle (median, 1.75° vs. 4.79°, P=0.036) and greater double leg support time (0.18 vs. 0.12, P=0.008) when compared with the control group. The other parameters during maximum-speed walking were not significantly different between the two groups. During comfortable-speed walking, none of the parameters were significantly different between the two groups.

Conclusions: Compensatory trunk flexion angle decreases markedly during maximum-speed walking in people with bilateral transtibial amputation. People with bilateral transtibial amputation may be changing the trunk flexion angle to walk faster. When evaluating gait compensation for people with bilateral transtibial amputation, trunk flexion angle may be an important index and maximum-speed walking is needed to detect the change in trunk flexion angle.

INTRODUCTION

People with bilateral transtibial amputation (BTTA) can learn bipedal walking using bilateral transtibial prostheses, although they require more training than those with unilateral transtibial amputation (UTTA), who can compensate for their limb loss with their intact limb. Given that BTTA is rare,¹⁾ there are fewer reports on walking for people with BTTA than for people with UTTA. Therefore, clinicians have limited information about how people with BTTA can achieve stable walking, which hinders systematic intervention for walking after BTTA.

The ability to walk with lower limb prostheses is known to be affected by age, sex, amputation level, skin condition, contracture, muscle strength, and balance ability.^{2,3,4,5,6,7,8)} When walking with prosthetics, the characteristics and performance of prosthetic components also influence walking performance. People with UTTA have reduced stability caused by limited afferent feedback from their lower limb and limited ankle joint mobility.^9,10,11) Previous studies have reported that walking in people with UTTA is characterized by decreased stability in the stance phase, asymmetric walking, and increased energy consumption.^12,13) Therefore, the purpose of physical therapy for most people with UTTA is to reduce asymmetry. However, people with BTTA might walk with a strategy that is different from that used by people with UTTA because those with BTTA do not have a healthy limb. Therefore, physical therapy for people with BTTA may require an approach that is different from therapies used for those with UTTA.

Trunk kinematics is important for walking assessment, and compensation for trunk kinematics and trunk muscle activity occurs in people with lower limb disabilities.^14,15) In previous research focusing on trunk movement in people with UTTA, trunk lateral excursions increased with increasing level of amputation, and in the transfemoral amputees, trunk flexion-extension excursions increased.¹⁶⁾ Many previous studies have examined the relationship between walking speed and trunk activity in people with UTTA.^17,18,19) Hendershot et al. reported that the trunk muscle forces and spinal loads increased with increasing walking speed in persons with lower limb amputation when compared with able-bodied controls.¹⁷⁾ In another study, stride length asymmetry was shown to be dependent on the combination of asymmetries in forward foot placement and trunk progression, with a smaller contribution of trunk-progression asymmetry at slow speed.¹⁹⁾ Altered trunk motion and spinal loading may contribute to increased risk of low-back injury among lower limb amputees.¹⁶⁾ In people with UTTA, an increase in trunk muscle activity and trunk-pelvic movement patterns suggested an increase in the spinal load contributing to elevated risk of low-back pain.¹⁷⁾ Decreased trunk flexion angle and increased trunk lateral bending may cause secondary low-back pain when a prosthetic leg is used over a long period.^17,20)

Previous research has reported that people with BTTA have a symmetrical walk, decreased walking speed, increased step width, and increased energy consumption during walking when compared with able-bodied controls.^21,22) In a comparison of people with BTTA and able-bodied participants walking at similar speeds, the BTTA group demonstrated decreased ankle dorsiflexion and knee flexion in the stance phase, decreased peak ankle internal plantar flexion moment, decreased ankle power generation, and increased bilateral hip hiking when compared with able-bodied participants.²¹⁾ It is necessary to understand the compensatory movements unique to people with BTTA and to carry out rehabilitation with consideration for the prevention of low back pain and secondary disorders. In a previous study focusing on trunk movement in people with BTTA, it was reported that the trunk lateral flexion angle and forward trunk lean increased during walking when compared with able-bodied participants at similar walking speed and the degree of lean was directly proportional to speed.²²⁾ However, there are only a few studies that have examined the influence of trunk angle change and walking speed.²²⁾ As with people with UTTA, it is important to understand the compensatory motion used by people with BTTA when walking. Understanding trunk compensation during walking in people with BTTA may contribute to the planning of optimal physical therapy and prevention of secondary disorders. Therefore, this study aimed to clarify the changes in trunk lateral bending and trunk flexion angle during walking in people with BTTA.

MATERIALS AND METHODS

Ethical Considerations

All participants gave their informed consent. This study was approved by the Ethical Review Board of the National Rehabilitation Center for Persons with Disabilities (Approval number: 28-149, 1 March 2017).

Participants

In this cross-sectional study, four participants with BTTA (BTTA group) who could walk without upper-limb walking aids (mean [standard deviation, SD]; age, 31.5 [17.1] years; height, 175.4 [2.8] cm; mass, 69.8 [5.3] kg) and ten able-bodied participants (control group) (age, 23.6 [3.3] years; height, 173.1 [4.0] cm; mass, 65.2 [10.4] kg) were recruited by convenient sampling. The characteristics of the BTTA group are presented in Table 1. None of the participants had cognitive dysfunction, skin disorders (e.g., wounds, pain, sensory disorders), or an extremely short residual limb that would make it difficult to wear a prosthesis. The data were measured from June 2017 to November 2017.

Table 1. Profile of the bilateral transtibial amputees (BTTA) group

Subject	Age (years)	Time since amputation (days)	Height (m)	Mass (kg)	Residual length right/left (mm)	Sex	Cause	Foot type	Socket type	Suspension system	Complication
BTTA1	21	1382	1.77	69.4	100/130	Male	Trauma	Echelon (Blatchford)	TSB	Pin-lock liner	-
BTTA2	16	427	1.71	66.4	130/105	Male	Trauma	1C60 (Otto Bock)	TSB	Pin-lock liner	-
BTTA3	29	276	1.77	78.6	140/130	Male	Trauma	1C61 (Otto Bock)	TSB	Pin-lock liner	Unilateral below-elbow amputation
BTTA4	60	1229	1.77	64.8	190/165	Male	Disease	FS3000 (Freedom)	TSB	Pin-lock liner	Bilateral below-elbow amputation

TSB, total surface bearing.

Experimental Protocol

All participants wore shoes during the measurements. Retroreflective markers were attached to the participant according to a modified Helen Hayes marker set.^23,24) On the shoes, a marker was affixed to a position corresponding to an anatomical landmark, and prostheses were affixed to positions corresponding to healthy participants. All participants walked along the 10-m walkway at a comfortable speed and at maximum speed. For the monitoring of comfortable-speed walking and maximum-speed walking, the following respective instructions were given: “Please walk at a normal comfortable speed,” and “Please walk at the fastest speed.” After one practice trial, ten walking trials were recorded for each participant. Participants were first instructed to walk at a comfortable speed and then at maximum speed. Participants could rest at any time during the experiments. Ten gait cycles with clean heel strikes on one of the eight force plates without apparent acceleration and deceleration were used for the kinematic analysis.

Data Collection

Walking analysis was performed using a 12-camera three-dimensional motion analysis system (MAC3D; Motion Analysis, Natick, MA, USA) with eight Kistler force plates, using Cortex software (version 3.6, Motion Analysis). The sampling rate was set at 100 Hz for the cameras and the force plates. The collected data were processed using Visual3D software (version 4.96, C-Motion, Germantown, MD, USA). The marker trajectories were filtered using a fourth-order, zero-lag Butterworth filter with cutoff frequencies of 6 Hz. Trunk flexion/side bending angles (with respect to the ground) and temporospatial parameters were calculated for each gait cycle using the trunk and foot markers and the force data.

The primary outcomes were the trunk lateral bending range and the average trunk flexion angle during one gait cycle relative to the ground. The trunk lateral bending range was defined as the range of left–right side bending movement values within a gait cycle. Trunk flexion angle was defined as the average value of the trunk flexion angle during a gait cycle.

The secondary outcomes were walking speed, step width, cadence, stride length, and double support time. The stride length was defined as the sum of the left and right stride lengths. For other gait parameters, the average value of each measurement item was extracted as a representative value for each participant. This study compared the average value of each variable between the BTTA group and the control group.

Statistical Analysis

All statistical analyses were conducted using SPSS Statistics version 24.0 (IBM, Armonk, NY, USA). Each measurement item between the BTTA group and the control group were compared using the Mann–Whitney U test. The threshold for significance was set at P<0.05.

RESULTS

In the maximum-speed walking test, the median (minimum–maximum) walking speed was significantly lower in the BTTA group than in the control group (1.78 [1.63–1.95] vs. 2.14 [2.00–2.60] m/s, P=0.002). The double leg support time was significantly greater in the BTTA group than in the control group (0.18 [0.15–0.21] vs. 0.12 [0.06–0.16] s, P=0.008). There were no significant differences between the BTTA group and the control group regarding stride length (1.59 [1.36–1.73] vs. 1.76 [1.60–1.95] m), cadence (142.03 [112.99–151.13] vs. 150.77 [129.03–172.41] steps/min), and step width (0.13 [0.12–0.15] vs. 0.12 [0.09–0.15] m) (Table 2).

Table 2. Maximum-speed walking parameters

Outcome measures	Control group (n=10)	BTTA group (n=4)	P-value	BTTA 1	BTTA 2	BTTA 3	BTTA 4
Walking speed (m/s)	2.14 (2.00–2.60)	1.78 (1.63–1.95)	0.002	1.63	1.85	1.95	1.71
Step width (m)	0.12 (0.09–0.15)	0.13 (0.12–0.15)	0.304	0.15	0.12	0.12	0.14
Stride length (m)	1.76 (1.60–1.95)	1.59 (1.36–1.73)	0.054	1.73	1.63	1.56	1.36
Cadence (steps/min)	150.77 (129.03–172.41)	142.03 (112.99–151.13)	0.188	112.99	134.98	149.07	151.13
Double leg support time (s)	0.12 (0.06–0.16)	0.18 (0.15–0.21)	0.008	0.21	0.17	0.15	0.19
Trunk lateral bending (°)	4.73 (3.08–6.47)	5.71 (4.50–7.91)	0.304	7.91	5.37	6.05	4.50
Trunk flexion angle (°)	4.79 (0.02–8.01)	1.75 (0.18–2.64)	0.036	2.44	2.64	1.07	0.18

Data shown as median (minimum–maximum) for groups and as mean values for BTTA participants.

In the comfortable-speed walking test, there were no significant differences between the two groups for all parameters. The median (minimum–maximum) walking speed (BTTA vs. control: 1.29 [1.22–1.48] vs. 1.35 [1.19–1.59] m/s), stride length (1.43 [1.25–1.54] vs. 1.44 [1.28–1.66] m), cadence (114.12 [100.59–121.46] vs. 113.42 [107.05–121.09] steps/min), step width (0.12 [0.09–0.15] vs. 0.12 [0.07–0.14] m), and double leg support time (0.25 [0.23–0.28] vs. 0.22 [0.19–0.28] s) were not significantly different between the two groups (Table 3).

Table 3. Comfortable-speed walking parameters

Outcome measures	Control group (n=10)	BTTA group (n=4)	P-value	BTTA 1	BTTA 2	BTTA 3	BTTA 4
Walking speed (m/s)	1.35 (1.19–1.59)	1.29 (1.22–1.48)	0.839	1.29	1.48	1.29	1.22
Step width (m)	0.12 (0.07–0.14)	0.12 (0.09–0.15)	0.635	0.14	0.09	0.15	0.11
Stride length (m)	1.44 (1.28–1.66)	1.43 (1.25–1.54)	0.839	1.54	1.47	1.40	1.25
Cadence (steps/min)	113.42 (107.05–121.09)	114.12 (100.59–121.46)	0.945	100.59	121.46	110.60	117.65
Double leg support time (s)	0.22 (0.19–0.28)	0.25 (0.23–0.28)	0.106	0.26	0.23	0.28	0.24
Trunk lateral bending (°)	3.93 (3.02–6.15)	4.28 (3.42–8.07)	0.839	8.07	3.47	5.09	3.42
Trunk flexion angle (°)	7.53 (6.60–10.87)	6.55 (4.82–8.60)	0.240	4.82	7.32	8.60	5.77

Data shown as median (minimum–maximum) for groups and as mean values for BTTA participants.

During maximum-speed walking, the trunk flexion angle was significantly lower in the BTTA group than in the control group (1.75 [0.18–2.64]° vs. 4.79 [0.02–8.01]°, P=0.036) (Table 2). The trunk lateral bending range was not significantly different between the two groups (5.71 [4.50–7.91]° vs. 4.73 [3.08–6.47]°, P=0.304). The trunk flexion angle was lower in the BTTA group than in the control group throughout the gait cycle at both walking speeds (Fig. 1). The difference in trunk flexion angle between the two walking conditions was greater in the BTTA group than in the control group. During comfortable-speed walking, the trunk flexion angle was lower in the BTTA group than in the control group, although the difference was not statistically significant (6.55 [4.82–8.60]° vs. 7.53 [6.60–10.87]°, P=0.24) (Table 3). The trunk lateral bending range during comfortable-speed walking was not significantly different between the two groups (BTTA group: 4.28 [3.42–8.07]° vs. 3.93 [3.02–6.15]°, P=0.839).

Fig. 1.

Trunk flexion angle during one gait cycle. Data shown as median of each group.

DISCUSSION

This study examined the walking characteristics of people with BTTA, such as trunk movement, according to the walking speed. The BTTA group tended to walk slower with an increased double leg support time and a smaller trunk flexion during maximum-speed walking when compared with the able-bodied controls. The difference between the two groups was greater during maximum-speed walking than during comfortable-speed walking. The results showed that the median values of the trunk flexion angle decreased during maximum-speed walking.

The decrease in the trunk flexion angle in the BTTA group was observed throughout one gait cycle, which indicates that the BTTA group walks with their trunk upright. Theoretically, people with BTTA using passive prosthetic feet have to rely more on hip flexors for swing initiation than able-bodied people who can rely on plantar flexors and hip flexors.²⁵⁾ Posterior pelvic tilt relative to the femur, induced by an upright trunk, lengthens hip flexor muscles, which would facilitate force generation according to the force–length relationship of the muscles. This mechanism would contribute to better gait performance in people with BTTA.

Previous studies have reported that trunk lateral bending increased during the prosthetic stance phase in people with UTTA.^16,26) However, in the current study, there was no left–right asymmetry in the BTTA group. A previous study of people with UTTA found that increased trunk lateral bending toward the amputation side compensated for reduced hip abductor moment.²⁶⁾ People with BTTA have been reported to walk symmetrically with a slower walking speed, greater step width, and higher energy consumption than able-bodied controls.^21,22) In the present study, the BTTA group had a slower walking speed than the control group during maximum-speed walking but the step width was not significantly different from the control group. The trunk lateral bending results in this study were not significantly different between the two groups. Our participants and those in the previous studies differ in age (mean [SD]: 31.5 [17.1] years vs. 52.8 [17.6] years²¹⁾ and 50 [18] years²²⁾), which might have contributed to the difference in the results regarding step width. In addition, the absence of a healthy side in people with BTTA may have prevented the observation of trunk asymmetry in the frontal plane. This study also measured the maximum-speed walking conditions to make it easier to observe the compensatory movement of the BTTA group.

For people with UTTA, a goal is to walk with less left–right asymmetry, but this is not the case for people with BTTA. In this study, the BTTA group showed a decrease in trunk flexion angle without a significant difference in trunk lateral bending. The results of this study differed from those of the previous study.²²⁾ Differences in the age of the participants, causes of amputation, and faster walking speeds may explain the different results. Additionally, no significant difference was observed between people with BTTA and able-bodied controls in comfortable-speed walking. This result may be due to recent developments in prosthetic rehabilitation, or because the participants in the BTTA group in our study were younger and lacked vascular amputees, in contrast with previous studies.^21,22) Trunk flexion angle may be an important indicator when evaluating the gait compensation for people with BTTA, and maximum-speed walking is required to detect any changes in their trunk flexion angle. The results of this study may contribute to the design of rehabilitation programs and goal setting for people with BTTA.

This study has some limitations. This study recruited only four participants with heterogeneous characteristics: the age and the cause of amputation could not be controlled. It should be noted that these results are generalized, and this study only considers the tendencies for BTTA in medically stable people. However, the data are valuable because the BTTA group was composed of participants from the general population, rather than military participants, and amputations in some participants were due to diseases. Further gait analysis studies with larger sample sizes or with electromyographic data collection would help confirm the relationship between the walking kinematics and secondary musculoskeletal disorders in people with BTTA. In the future, it will be necessary to examine how physical therapy interventions for people with BTTA affect trunk movement and the frequency of secondary disorders caused by long-term use of prostheses.

CONCLUSION

There were no significant differences in gait parameters between the BTTA group and the control group during comfortable-speed walking, but differences in walking speed, double leg support time, and trunk flexion angle were observed during maximum-speed walking. It is suggested that when the BTTA group walks fast, the trunk stands upright more than when walking at comfortable speed. When evaluating the gait compensation for people with BTTA, trunk flexion angle may be an important index and maximum-speed walking is needed to detect their trunk flexion angle change.

ACKNOWLEDGMENTS

The authors thank the study participants for their time and dedication.

CONFLICTS OF INTEREST

The authors have no conflicts of interest to declare.

REFERENCES

• 1.  Ohmine S, Kimura Y, Saeki S, Hachisuka K: Community-based survey of amputation derived from the physically disabled person’s certification in Kitakyushu City, Japan. Prosthet Orthot Int 2012;36:196–202. PMID:22314214, DOI:10.1177/0309364611433443
• 2.  Nolan L: A training programme to improve hip strength in persons with lower limb amputation. J Rehabil Med 2012;44:241–248. PMID:22367416, DOI:10.2340/16501977-0921
• 3.  Sansam K, O’Connor RJ, Neumann V, Bhakta B: Can simple clinical tests predict walking ability after prosthetic rehabilitation? J Rehabil Med 2012;44:968–974. PMID:23037865, DOI:10.2340/16501977-1046
• 4.  Raya MA, Gailey RS, Fiebert IM, Roach KE: Impairment variables predicting activity limitation in individuals with lower limb amputation. Prosthet Orthot Int 2010;34:73–84. PMID:20196689, DOI:10.3109/03093640903585008
• 5.  Molina-Rueda F, Alguacil-Diego IM, Cuesta-Gómez A, Iglesias-Giménez J, Martín-Vivaldi A, Miangolarra-Page JC: Thorax, pelvis and hip pattern in the frontal plane during walking in unilateral transtibial amputees: biomechanical analysis. Braz J Phys Ther 2014;18:252–258. PMID:25003278, DOI:10.1590/bjpt-rbf.2014.0032
• 6.  Visser JM, McCarthy I, Marks L, Davis RC: Is hip muscle strength the key to walking as a bilateral amputee, whatever the level of the amputations? Prosthet Orthot Int 2011;35:451–458. PMID:21983043, DOI:10.1177/0309364611422268
• 7.  Isakov E, Burger H, Gregorič M, Marinčcek Č: Stump length as related to atrophy and strength of the thigh muscles in trans-tibial amputees. Prosthet Orthot Int 1996;20:96–100. PMID:8876002, DOI:10.3109/03093649609164425
• 8.  Wakasa M, Fukuda A: The time-dependent changes in EMG recorded from the medial gastrocnemius of young and elderly amputees. J Phys Ther Sci 2013;25:1511–1513. PMID:24396222, DOI:10.1589/jpts.25.1511
• 9.  Viton JM, Mouchnino L, Mille ML, Cincera M, Delarque A, Pedotti A, Bardot A, Massion J: Equilibrium and movement control strategies in transtibial amputees. Prosthet Orthot Int 2000;24:108–116. PMID:11061197, DOI:10.1080/03093640008726533
• 10.  Quai TM, Brauer SG, Nitz JC: Somatosensation, circulation and stance balance in elderly dysvascular transtibial amputees. Clin Rehabil 2005;19:668–676. PMID:16180604, DOI:10.1191/0269215505cr857oa
• 11.  Kavounoudias A, Tremblay C, Gravel D, Iancu A, Forget R: Bilateral changes in somatosensory sensibility after unilateral below-knee amputation. Arch Phys Med Rehabil 2005;86:633–640. PMID:15827911, DOI:10.1016/j.apmr.2004.10.030
• 12.  Breakey J: Gait of unilateral below knee amputees. Orthot Prosthet 1976;30:17–24.
• 13.  Vllasolli T, Orovcanec N, Zafirova B, Krasniqi B, Murtezani A, Krasniqi V, Rama B: Physiological cost index and comfort walking speed in two level lower limb amputees having no vascular disease. Acta Inform Med 2015;23:12–17. PMID:25870485, DOI:10.5455/aim.2015.23.12-17
• 14.  Saha D, Gard S, Fatone S: The effect of trunk flexion on able-bodied gait. Gait Posture 2008;27:653–660. PMID:17920272, DOI:10.1016/j.gaitpost.2007.08.009
• 15.  Li K, Ackland DC, McClelland JA, Webster KE, Feller JA, de Steiger R, Pandy MG: Trunk muscle action compensates for reduced quadriceps force during walking after total knee arthroplasty. Gait Posture 2013;38:79–85. PMID:23206902, DOI:10.1016/j.gaitpost.2012.10.018
• 16.  Hendershot BD, Wolf EJ: Three-dimensional joint reaction forces and moments at the low back during over-ground walking in persons with unilateral lower-extremity amputation. Clin Biomech 2014;29:235–242. PMID:24393361, DOI:10.1016/j.clinbiomech.2013.12.005
• 17.  Hendershot BD, Shojaei I, Acasio JC, Dearth CL, Bazrgari B: Walking speed differentially alters spinal loads in persons with traumatic lower limb amputation. J Biomech 2018;70:249–254. PMID:29217090, DOI:10.1016/j.jbiomech.2017.11.026
• 18.  Butowicz CM, Acasio JC, Dearth CL, Hendershot BD: Trunk muscle activation patterns during walking among persons with lower limb loss: influences of walking speed. J Electromyogr Kinesiol 2018;40:48–55. PMID:29625268, DOI:10.1016/j.jelekin.2018.03.006
• 19.  Roerdink M, Roeles S, van der Pas SC, Bosboom O, Beek PJ: Evaluating asymmetry in prosthetic gait with step-length asymmetry alone is flawed. Gait Posture 2012;35:446–451. PMID:22153771, DOI:10.1016/j.gaitpost.2011.11.005
• 20.  Hendershot BD, Nussbaum MA: Altered flexion-relaxation responses exist during asymmetric trunk flexion movements among persons with unilateral lower-limb amputation. J Electromyogr Kinesiol 2014;24:120–125. PMID:24332679, DOI:10.1016/j.jelekin.2013.11.005
• 21.  Su PF, Gard SA, Lipschutz RD, Kuiken TA: Gait characteristics of persons with bilateral transtibial amputations. J Rehabil Res Dev 2007;44:491–501. PMID:18247246, DOI:10.1682/JRRD.2006.10.0135
• 22.  Major MJ, Stine RL, Gard SA: The effects of walking speed and prosthetic ankle adapters on upper extremity dynamics and stability-related parameters in bilateral transtibial amputee gait. Gait Posture 2013;38:858–863. PMID:23680427, DOI:10.1016/j.gaitpost.2013.04.012
• 23.  Kadaba MP, Ramakrishnan HK, Wootten ME: Measurement of lower extremity kinematics during level walking. J Orthop Res 1990;8:383–392. PMID:2324857, DOI:10.1002/jor.1100080310
• 24.  Burnett DR, Campbell-Kyureghyan NH, Topp RV, Quesada PM: Biomechanics of lower limbs during walking among candidates for total knee arthroplasty with and without low back pain. BioMed Res Int 2015;2015:1–8. PMID:26171387, DOI:10.1155/2015/142562
• 25.  Lewis CL, Ferris DP: Walking with increased ankle pushoff decreases hip muscle moments. J Biomech 2008;41:2082–2089. PMID:18606419, DOI:10.1016/j.jbiomech.2008.05.013
• 26.  Rueda FM, Diego IM, Sánchez AM, Tejada MC, Montero FM, Page JC: Knee and hip internal moments and upper-body kinematics in the frontal plane in unilateral transtibial amputees. Gait Posture 2013;37:436–439. PMID:23021254, DOI:10.1016/j.gaitpost.2012.08.019