Gait & Posture 33 (2011) 511–526

Contents lists available at ScienceDirect

Gait & Posture journal homepage: www.elsevier.com/locate/gaitpost

Review

Biomechanics and physiological parameters during gait in lower-limb amputees: A systematic review Yoshimasa Sagawa Jr.a,b,c,*, Katia Turcot f, Ste´phane Armand f, Andre Thevenon b,d,e, Nicolas Vuillerme g,h, Eric Watelain a,b,c,i a

UVHC, LAMIH, F-59313 Valenciennes, France Univ Lille Nord de France, F-59000 Lille, France CNRS, FRE 3304, F-59313 Valenciennes, France d Physical Medicine and Rehabilitation Department, Lille University Hospital, Lille, France e Laboratory of Human Movement Studies, Faculty of Sports Sciences and Physical Education, Lille, France f Willy Taillard Laboratory of Kinesiology, Geneva University Hospitals and Geneva University, Geneva, Switzerland g Laboratoire TIMC-IMAG UMR UJF CNRS 5525. Equipes ARFIRM/AGIM3, Faculte´ de Me´decine, La Tronche, France h CIC-IT 805, INSERM/AP-HP, Hoˆpital Raymond Poincare´, EA 4497, Garches, France i HandiBio, EA 4322, Univ du Sud Toulon Var, La Garde, France b c

A R T I C L E I N F O

A B S T R A C T

Article history: Received 13 July 2010 Received in revised form 3 February 2011 Accepted 6 February 2011

Objective: : The purpose of this systematic review was to identify which biomechanical and physiological parameters are the most relevant, commonly used, able to discriminate and/or have specific clinical relevance for the gait analysis of lower-limb amputees (LLA). Methods: : We performed an electronic search via the PubMed, EMBASE and ISI Web of Knowledge databases from 1979 to May 2009. Two independent reviewers assessed the title and abstract of each identified study. The quality assessment of the full text was undertaken using a 13-item checklist divided into three levels: A, B, and C. Results: : The literature search identified 584 abstracts to be considered. After applying the inclusion criteria, we reviewed the full text of a total of 89 articles. The mean article quality was 8  2. No A-level article was found; the primary reason was a negative score in blinded outcome assessment. Sixty-six articles (74%) corresponded to a B-level, and two articles (2%) corresponded to a C-level. Twenty-one articles (24%) did not acquire enough points to be assigned to any level. In this study, we present and discuss the most commonly used and most relevant 32 parameters. Many of the parameters found were not reported in enough studies or in enough detail to allow a useful evaluation. Conclusion: : This systematic review can help researchers compare, choose and develop the most appropriate gait evaluation protocol for their field of study, based on the articles with best scores on the criteria list and the relevance of specific biomechanical and physiological parameters. ß 2011 Elsevier B.V. All rights reserved.

Keywords: Amputee Biomechanics Physiological Gait Lower limb Prosthesis

1. Introduction In 2007, approximately 1.7 million people experienced limb loss in the United States [1]. In this country, more than 185,000

Abbreviations: AB, able-bodied; EMG, electromyography; FS, fast speed; GRF, ground reaction force; LLA, lower-limb amputees; NS, normal speed; ROM, range of motion; SACH, solid ankle cushion heel; TF, transfemoral amputation; TT, transtibial amputation. * Corresponding author at: Laboratoire d’Automatique et de Me´canique d’Informatique Industrielles et Humaines (FRE CNRS 3304) Universite´ le Mont Houy, Baˆtiment Malvache, 59313 Valenciennes cedex 9, France. Tel.: +33 (0) 6 65177936/327511349; fax: +33 (0) 3 27511317. E-mail address: [email protected] (Y. Sagawa Jr.). 0966-6362/$ – see front matter ß 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.gaitpost.2011.02.003

new amputations are performed each year [1,2]. The prevalence rate in 1996 was 4.9 per 1000 persons. The prevalence rate was higher for people aged 65 years and older: 19.4 per 1000 [1]. The incidence rate was 46.2 per 100,000 for people with vascular disease, 5.86 per 100,000 for people with secondary trauma and .35 per 100,000 for people with bone or joint malignancy. Annual acute and post-acute medical care costs associated with caring for vascular amputees exceed $4.3 billion in the United States [3]. After lower-limb amputation, a person is routinely prescribed a prosthesis that may include a prosthetic foot, pylon, knee and socket, depending on the level of amputation and the cause. There are a number of cost-effective components presently available [4,5], but until now, there has been no consensus among the different professionals (e.g., doctors, physiotherapists, prosthetists) in terms of the main criteria used to select an appropriate

512

Y. Sagawa Jr. et al. / Gait & Posture 33 (2011) 511–526

lower-limb prosthesis that corresponds to the patient abilities and needs. Therefore, lower-limb prostheses are usually prescribed based on empirical knowledge. On the other hand, the healthcare policies require objective and reliable criteria for prosthetic prescriptions. Depending on the prosthesis (e.g., mechanical knees versus microprocessor knees), costs may vary widely [6]. Some countries have already begun to clarify these healthcare policies, for example, the Health Care Financing Administration’s Common Procedure Coding System in the USA [7] and the Dutch Health Care Insurance Board in Netherlands [8]. In the current evidence-based medicine context, many studies have been conducted to assess the relationship between patient characteristics, the prosthesis and the environment (see references below). Most of these studies used gait analysis to assess biomechanical and physiological aspects of gait. In fact, walking is considered to be one of the most important aspects of independence [9]. Moreover, biomechanical and physiological analyses allow a precise quantification of both body movements and energy expenditure. This systematic review aimed at identifying which biomechanical and physiological parameters are the most relevant, commonly used, able to discriminate and/or have specific clinical relevance for the gait analysis of lower-limb amputees (LLA). 2. Methods 2.1. Method used to identify the studies to include We performed an electronic search via the PubMed, EMBASE and ISI Web of Knowledge databases from 1979 to May 2009. The search strategy was based on a combination of the following six keywords: amputee*, lower limb*, lower extremity*, gait*, locomotion* and walking*. We further narrowed the field to include only published articles (i.e., conventional articles, comparative studies, evaluations) written in English or French with an entirely adult study population (i.e., +18 years). 2.2. A preliminary selection based on the abstract Two independent reviewers assessed the title and abstract of each study identified. Based on the abstract, studies were included in the full text review when they satisfied the following three set of criteria: (1) patient characteristics, (2) descriptive or intervention studies and (3) outcome measures. These criteria are described below. Participant characteristics: The studies had to include a population with lower-limb amputation(s) (1) with a traumatic, vascular or other cause; (2) with hip or knee disarticulation, transfemoral (TF) or transtibial (TT) amputation level; (3) with unilateral or bilateral involvement; (4) with different types of foot, knee or socket prosthesis; and/or (5) walking with or without assistive devices. Descriptive or intervention studies: The studies had to involve a particular subpopulation (e.g., above-knee amputations, vascular causes, athletes), an effect of the prosthesis components (e.g., sockets, knees, feet), or an effect of fitness, rehabilitation or other methodological aspects (e.g., influence of walking speed). Outcome measures: The studies had to report gait-related biomechanical parameters (e.g., spatio-temporal, kinematic and/or kinetic; transducers based measurements and accelerometry) or physiological parameters (e.g., energy consumption, energy cost, heart rate, electromyography (EMG)). The exclusion criteria of this abstract-based selection were theoretical studies, studies validating a model/protocol, and

studies about osseo integrated prostheses, gait under conditions other than on level surface (e.g., uphill, downhill, stair ambulation, obstacle crossing), and the effect of prosthesis mass or prosthesis settings. In addition, the references of the full texts selected were examined to extend our search. 2.3. Method used to assess the quality of the selected articles The quality of the articles selected was assessed using the 13item checklist developed by van der Linde et al. [8]. This checklist was adapted to evaluate non-randomized controlled trials using two other randomized controlled trial checklists [10,11]. Each criterion was scored ‘‘0’’ if it was invalid or the answer was ‘‘no’’ and ‘‘1’’ if it was valid or the answer was ‘‘yes’’. If a criterion was not applicable, it was scored ‘‘0’’. The focus of the article review process was not the intervention approaches per se, but rather the main parameters used during these interventions. Four independent reviewers piloted the adapted quality checklist on three randomly chosen articles in order to assess the content and to certify reliable data extraction. The reviewers’ results were compared and the differences were resolved through discussion. After completing this pilot session, we standardized the item descriptions to guarantee good inter-rater reliability. The final quality checklist involved 13 items with a theoretical maximum score of 13 points. The checklist covered three different domains: (a) adequacy of the description of inclusion and exclusion criteria (four items, maximum four points), (b) intervention and assessment (five items, maximum five points) and (c) statistical validity (four items, maximum four points). The checklist was converted into an electronic data extraction sheet, and then two independent reviewers performed the data extraction. 2.4. Analysis In order to ensure agreement of the quality assessments, we performed Kappa statistics and bootstrap confidence intervals. All studies included in this systematic review were required to appropriately control for selection and measurement bias, similarly to in the review by van der Linde et al. [8]. Studies were classified as:  A-level studies – Studies with a total score of at least 11 of 13 points, including 6 points in criteria sets a and b described above; a positive score for blinded outcome assessment (criterion b7) and for measurement timing (criterion b8). This last criterion measured the time that the subjects were given to adapt to the change in prosthesis. In fact, an adequate adaptation period is required. According to English et al. [12], transfemoral (TF) amputees need at least 3 weeks of walking with a new knee mechanism to ensure that their gait parameters are stable, LLA need a period of at least 1 week to adapt to a new prosthetic foot or to a change in prosthetic mass [12].  B-level studies – Studies with a score between 6 and 12 points, including a positive score for measurement timing (criterion b8).  C-level studies – Studies with a score of at least six points out for the criteria sets a and b, with an invalid score on criteria b7 and b8. 3. Results 3.1. A preliminary selection based on the abstract The preliminary literature search identified 584 abstracts. None of the articles was excluded on the basis of language. After applying

Y. Sagawa Jr. et al. / Gait & Posture 33 (2011) 511–526

513

parameters related to spatio-temporal, kinematic and kinetic parameters. Additionally, 98 specific parameters were found and added to create an exhaustive list of 220 parameters. Of the 122 parameters proposed by Benedetti et al. [16], 78 (64%) have never been used to analyze the gait of LLA. Eight of the studies examined (9%) [17–24] had a combination of biomechanical, physiological and EMG outcomes; 12 (13%) [15,25–35] had a combination of biomechanical and clinical/ functional outcomes; 2 (2%) [36,37] had a combination of physiological and clinical/functional outcomes; and 17 (19%) [26,28,31,33–35,37–47] had a combination of biomechanical or physiological outcomes with various questionnaires about levels of activity, prosthesis comfort and/or functionality. Table 3 shows the frequency distribution of the 89 articles selected from 23 biomedical journals. The mean impact factor of these journals was 1.67  0.77 and ranged from .34 to 2.78. Thirtythree percent of these studies were published in a specialized journal concerning LLA (impact factor .37). 4. Discussion 4.1. The main objective of this study and its relevance Fig. 1. Procedure for the study selections with utilized databases and for the literature search-and-selection criteria.

the inclusion criteria, we included a total of 89 articles (i.e., 15%) in the full text review (Fig. 1). 3.2. Data quality The agreement on data quality between the two reviewers was high [13]. The estimated mean Kappa value was .92 (SD .17) and the 95% confidence intervals ranged from .82 to 1. The mean quality score of studies was 8 (SD 2) and these scores ranged from 2 to 12. However, no articles corresponded to the A-level, due mainly to a negative score in blinded outcome assessment. Sixty-six articles (74%) corresponded to the B-level, and two articles (2%) corresponded to the C-level. Twenty-one articles (24%) did not have enough points to be assigned a level. 3.3. Participant characteristics The number of participants ranged from 2 (intra-individual analyses) [14] to 94 [15] (Table 1). The participants from the studies reviewed were heterogeneous, and a blend of different amputation levels and causes of amputation was often found (Table 1). 3.4. Parameters used for gait analysis The frequency distribution of biomechanical and physiological parameters used in gait analysis is illustrated in Fig. 2. A few studies also associated the psychological and cost parameters with the biomechanical and physiological parameters (Fig. 2, others). The main biomechanical parameters used were walking speed (43 times), knee angles (31 times), vertical ground reaction force (30 times), knee moments (27 times), hip power (26 times) and ankle angles (22 times). The main physiological parameters used were VO2 (ml/min/kg) (30 times), EMG of lower-limb muscle activity (17 times) and VO2 cost (ml/m/kg) (13 times). In Table 2, the reference parameters are distributed in four areas: foot (26 studies: 29%), knee (13 studies: 15%), socket (6 studies: 7%) and other (i.e., descriptive, rehabilitation, fitness studies) (44 studies: 49%) are represented. This list was based on the one proposed by Benedetti et al. [16], which involves 122

The objective of this systematic review was to identify the most relevant biomechanical and physiological parameters used to analyze the gait of LLA. To our knowledge, there are currently no studies with this objective. This review can help researchers compare, choose and develop the most appropriate gait evaluation protocol for their field of study (Table 1), based on the articles with best scores on the criteria list (Tables 1 and 3) and the relevance of specific biomechanical and physiological parameters (Fig. 2 and Table 2). The review also offers important information on research fields and gait analysis parameters used for LLA (Fig. 2 and Tables 1 and 2). 4.2. Article quality Two independent reviewers used a checklist adapted to LLA studies to score the quality of the articles that satisfied the criteria [8]. Similarly to the systematic review by van der Linde et al. [8] on the contribution of different prosthesis components, our investigation obtained limited unbiased information. In our study, no article received an A-level score, and only one study had a blind assessor [48]. In the study with a blind assessor (10 points, B-level), Postema et al. [48] compared four different prosthetic feet and performed a double-blind randomized trial. This was possible because the prosthetic feet were covered by a cosmetic overlay that mimicked a normal foot. In addition, the prostheses were aligned by the same orthopedic technician, who was not involved in the trials. A double-blind experimental design is more difficult when comparing other prosthetic components, such as the knees or the socket, because these components are often apparent. This limitation in performing double-blind trials, which prevents more evidencebased results in non-pharmacological experiments, has been already discussed in the literature [49–51]. For the sample size, van der Linde et al. [8] suggested that the number of independent variables (K) was adequate if the ratio K:n exceeded 1:10. According to this ratio, 74% of the articles that we reviewed did not have an adequate sample size (n = 17.2  14.2 subjects, range 2–94 subjects). A reduced sample size reduces the ability to validate the hypothesis and increases the risk of type-II error. It also makes it difficult to illustrate the discriminating capacity of a parameter. On the whole, due to the methodological limitations of the studies evaluated, it is recommended to use caution when interpreting the parameters and determining their relevance.

Y. Sagawa Jr. et al. / Gait & Posture 33 (2011) 511–526

514

Table 1 Methodological aspects of reviewed articles: authors, participants, main objective(s) and level of evidence. Ref

Articles

Participants (N)

Sex and age

Main objective(s)

L

[75]

Arya et al. (1995)

3 Uni, Tt, ?

3 M, 45  ?

5-U

[76]

Bae et al. (2007)

[29]

? M,? F, 40  8 ? M, ? F, 24  2 15 M, 5 F, 61

[77]

Baker and Hewison (1990) Barnett et al. (2009)

[33]

Berge et al. (2005)

8 Uni, Tf, ? 10 Ab 20 Uni, Tf, Tt, Tr, Va, Ca 8 Uni, Tt, Va, ? 7 Uni, Tt, Va, ? 15 Uni, Tt, Tr, Va, In

7 M, 1 F, 50  16 5 M, 2 F, 58  11 15 M, 51  9

[30]

Board et al. (2001)

11 Uni, Tt, Tr

? M, ? F, 45  ?

[45]

Boonstra et al. (1993)

9 Uni, Kd, Tr, Va, Ca

6 M, 3 F, 41  18

[18] [78]

Boonstra et al. (1994) Buckley et al. (1997)

29 Uni, Tf, ? 3 Uni, Tf, Tr

24 M, 5 F, 41  13 3 M, 48  10

[37]

Casillas et al. (1995)

12 Uni, Tt, Va 12 Uni, Tt, Tr

10 M, 2 F, 73  7 12 M, 50  14

[36]

Chin et al. (2002)

9 Uni, Tf, Va 8 Uni, Tt, Va

? M,? F, 63  2 ? M,? F, 72  2

[79]

Chin et al. (2006)

49 Uni, Hd, Tf, Va, Tr, Ca, In

34 M, 15 F, 67  6

[80]

Chin et al. (2006)

4 Uni, Tf, Tr, Ca 14 Ab

4 M, 24  8 10 M, 4 F, 25  4

[81]

Chin et al. (2009)

7 Uni, Hd, Ca, In

6 M, 1 F, 68  4

[82]

Cortes et al. (1997)

8 Uni, Tt, Tr 7 Ab

8 M, 35  12 7 M, 32  10

[83]

Culham et al. (1986)

10 Uni, Tt, Tr, Va

8 M, 2 F, 61  ?

[39]

Datta et al. (2004)

21 Uni, Tt, Tr, Va

19 M, 2 F, 51.7  15

[43]

Datta et al. (2005)

10 Uni, Tf

? M, ? F, 38  ?

[24]

Detrembleur et al. (2005)

6 Uni, Tf, Tr 6 Uni, Tt, Va

? M, ? F, 38  12 ? M, ? F, 50  11

[84] [85]

Doane and Holt (1983) Gailey et al. (1994)

[42]

Gard and Konz (2003)

8 Uni, Tt, ? 39 Uni, Tt, ? 21 Ab 10 Uni Tt, Tr, Vc

8 M, ?  ? 39 M, 46  16 21 M, 31  6 9 M, 1 F, 54  17

[86]

Genin et al. (2008)

[73]

Gitter et al. (1991)

10 Uni, Tf, Tr 9 Uni, Tt, Tr 13 Ab 5 Uni, Tt, ? 5 Ab

10 M, 35  5 9 M, 35  7 10 M, 3 F, 28  5 5 M, 20–50  ? ? M, ? F, ?  ?

To assess the performance characteristics of the Jaipur foot by comparing its shock absorption capacity and influence on gait style with that of SACH and Seattle feet, using the GRF To evaluate the muscle condition by acquiring the root mean square electromyogram To determine the rate at which gait recovers as measured by temporal distance factors (velocity and symmetry) To investigate the gait patterns of amputees, using either the amputee mobility aid or pneumatic post-amputation aid To determine if a shock-absorbing pylons affects gait mechanics To compare the volume changes associated with normal and vacuum conditions using a total surface-bearing suction socket To investigate the gait patterns when wearing prostheses fitted with either the Multiflex or Quantum foot To describe the gait quantitatively To quantify the physiological energy cost of using the so-called ‘‘Intelligent Prosthesis’’ compared to the cost of using a more conventional pneumatic swing-phase controlled device To use bioenergetic parameters to assess a new energy-storing foot prothesis (Proteor foot) by comparing it with the SACH foot in different walking situations To investigate whether or not %VO2max as an indicator of physical fitness is useful in predicting the outcome for a prosthetic rehabilitation after dysvascular amputation To evaluate physical fitness and prosthetic ambulatory ability and to investigate the level of fitness required for successful prosthetic ambulation To examine the impact of the characteristic differences between the Intelligent Knee Prosthesis and C-Leg on walking speed and energy expenditure during walking To investigate the differences in energy consumption between prosthetic walking and wheelchair locomotion To present an objective quantitative method for studying prosthetic gait, which allows gait patterns to be compared To evaluate the effect of the terminal prosthetic component on the electromyography activity of the quadriceps and hamstring muscle groups during gait To evaluate gait characteristics, cost and time analysis, and subjective opinion when subjects changed from a PTB to an ICEX1 fitting technique To test the effect of switching to an intelligent prosthesis on oxygen consumption and gait for users of pneumatic swingphase control knee joints To assess the influence of self-selected gait speed, efficiency of the pendulum mechanism and smoothness of center of body mass (CMb) displacement on metabolic energy costs To compare the SACH and uni-axis foot during the gait To compare the metabolic cost, heart rate, and self-selected speed of ambulation To investigate the effect that shock-absorbing pylon has on walking To investigate the effect of speed on the energy expenditure rate

[19]

Gitter et al. (1995)

[59] [87]

Goh et al. (1984) Goh et al. (2004)

8 Uni, Tf, Tr, Ca 8 Ab 11 Uni, Tf, Tt, ? 4 Uni, Tt, Tr, Va

7 M, 1 F, 37  7 8 M, 32  ? 11 M, 48  11 4 M, 40  10

[63]

Goujon-Pillet et al. (2008)

? M, ? F, 51  14 ? M,? F, 44.3  ?

[41]

Goujoun et al. (2006)

[46]

Graham et al. (2007)

27 Uni, Tf, Tr, Co, Ca 33 Ab 4 Uni, Tf, Tr 6 Uni, Tt, Tr 35 Ab 6 Uni, Tf, Tr

4 M, 49  13 5 M, 1 F, 45  13 ? M, ? F, 33  ? 6 M, 40  6

To determine the biomechanical adaptations necessary to walk while wearing a conventional prosthetic foot-ankle assembly and subsequently to evaluate the effects of energy-storing feet in the restoration of normal gait characteristics To define the relationships between mechanical and metabolic factors in pathological gait To evaluate SACH and uniaxial feet biomechanically To compare the pressure distribution of the pressure cast (PCast) socket to that of the patellar-tendon-bearing (PTB) socket To identify specific 3-D motion patterns for the pelvic and scapular girdles during gait To evaluate prosthetic feet with an original protocol that records fore-foot and ankle kinematics together with the global body kinematics and GRF during gait To explore the differences of using an energy-storing foot through gait analysis and a timed walking test to produce objective measurements and a comfort score for the patients’ subjective opinion

7-B 6-B 9-B 9-B 8-B 8-B 8-U 8-U

9-B

7-B

8-B

6-B

9-B 10-B

7-B

8-B

3-U

8-B

7-U 8-B 10-B 12-B

8-B

8-B 6-U 8-U

8-B

3-U

9-B

Y. Sagawa Jr. et al. / Gait & Posture 33 (2011) 511–526

515

Table 1 (Continued ) Ref

Articles

Participants (N)

Sex and age

Main objective(s)

L

[65]

Han et al. (2003)

6 Uni, Tt, Tr, Va, Ca

4 M, 2 F, 41  10

9-B

[71] [22]

Hansen et al. (2006) Hoffman et al. (1997)

14 Uni, Tt 5 Bi, Tf, Tr, Co 5 Ab

9 M, 5 F, 46  11 4 M, 1 F, 22  3 4 M, 1 F, 22  6

[20]

Houdijk et al. (2009)

11 Uni, Tt, Tr, Va 11 Ab

? M, ? F, 46  9 ? M, ? F, 47  11

[88]

Hsu et al. (1999)

5 Uni, Tt, ?

5 M, 32  4

[89]

Hurley et al. (1990)

[90]

Isacov et al. (1885)

7 Uni, Tt, Tr, Va, Co 4 Ab 3 Uni, Tf, Tr 14 Uni, Tf, Va

? M, ? F, 35  6 ? M, ? F, 25  2 3 M, 35  11 13 M, 1 F 60  7

[56]

Isacov et al. (1996)

14 Uni, Tt, Tr, Va

11 M, 3 F, 40  13

[66]

Isakov et al. (2000)

14 Uni, Tt, Tr

14 M, 45  7

[91]

Isakov et al. (2001)

11 Uni, Tt, Tr

11 M, 37  8

[92]

Jaeger et al. (1996)

[25]

Jones et al. (1997)

11 Uni, Tf, Tr, Ca 3 Ab 10 Uni, Tt, Va

11 M, 36  ? 3 M, 38  12 10 M, 67  6

[26]

Kahle et al. (2008)

? M, ? F, 51  19

[93]

Lacroix et al. (1992)

21 Uni, Tf, Tr, Va, Ca, Co 5 Uni, Tf, Tr, Ca 3 Uni, Tf, Tr

4 M, 1 F, 39  4 3 M, 26  ?

[94]

Lee and Hong (2009)

5 Uni, Tf, Tr

5 M, 48  2

[34]

Lee et al. (2006)

4 Uni, Tt, Tr, Co

4 M, 41  16

[17]

Lehmann et al. (1993)

10 Uni, Tt, ?

? M, ? F

[27] [95] [14]

Lemaire and Fisher (1994) Lemaire et al. (1993) Linden et al. (1999)

12 Uni, Tt, Tr 12 Ab 8 Uni, Tt, Tr 2 Uni, Tf, ?

12 M, 72  4 12 M, 70  3 8 M, 69  2 2 M, 34  1

[40]

McNealy and Gard (2008)

[96]

Michaud et al. (2000)

4 Bi, Tf, Tr, Co 9 Ab 9 Uni, Tf, Tt, Tr

4 M, 41  23 8 M, 1 F, 28  4 9 M, 45  16

[32]

Mizuno et al. (1992)

[97]

Murray (1980)

10 Uni, Tt, ? 5 Ab 10 Uni, Tf, Tr 30 Ab

10 M, 51  14 5 M, 41  ? 10 M, 41  ? ? M, ? F, ?  ?

[98]

Murray et al. (1983)

10 Uni, Tf, Tr ? Ab

10 M, 41  ? ? M, ? F, ?  ?

[31]

Nadollek et al. (2002)

22 Uni, Tt, Va

? M, ? F, 72  10

[68]

Nolan and Lee (2000)

[99]

Nyska et al. (2002)

4 Uni, Tf, Tr 4 Uni, Tt, Tr 10 Ab 3 Uni, Tt, Tr

? M, ? F, 28  8 ? M, ? F, 41  6 ? M, ? F, 29  10 3 M, 50  ?

To evaluate the gait patterns during walking with and without shoes and to identify the differences in barefoot gait patterns when using different prosthetic feet To examine the effect of roll-over shape arc length on gait To examine the aerobic demands and cardio-respiratory responses during walking for a range of speeds in addition to the subjects’ chosen walking speeds To investigate whether the increased energy cost of amputee gait could account for an increase in the mechanical work dissipated during the step-to-step transition in walking To investigate and compare the differences in energy cost, gait efficiency, and relative exercise intensity of multiple-speed walking and running with three different types of prosthetic feet: the SACH foot, the Flex foot, and the Re-Flex Vertical Shock Pylon To investigate the role of the contralateral limb in gait by determining lower limb’s joint reaction forces and symmetry To compare a prosthesis with an open knee mechanism versus a locked knee mechanism in terms of performance and physiological responses To investigate gait characteristics at two different speeds and the influence of speed on symmetry of selected gait parameters obtained To outline differences between both legs in terms of the kinematic parameters and the activity of the muscles controlling the knees To investigate the activity of the vastus medialis and biceps femoris muscles during ambulation To study the electromyographic activity of the superficial hip muscles of both legs during walking To compare standing prosthetic weight-bearing tolerance to the forces experienced during walking To compare subject performance using a non-microprocessor knee mechanism versus a C-Leg To describe the energy cost of gait in young traumatic transfemoral amputees using a contact socket for different knee prosthesis To investigate the effect of artificial ankle mobility in the saggital plane on the gait of amputees wearing a stance- and swing-controlled knee prosthesis To evaluate the gait performance and perception of amputees while using a flexible elliptican-shank monolimb as compared to a thicker circular-shank monolimb and a conventional modular prosthesis To quantify metabolic rate and efficiency, biomechanical gait parameters, and prosthesis comfort of the Seattle Ankle/Lite Foot compared to the SACH foot To assess the incidence of OA and relate these findings to the kinematic walking gait characteristics To examine the kinematic and kinetic gait parameters To describe a methodology for investigating the effects of various prosthetic feet on amputee gait To determine if adding prosthetic ankle motion would improve gait To assess the qualitative and quantitative differences in pelvic obliquity To investigate the functional features of various prostheses to facilitate the task of prescribing them To document several previously unreported motion patterns in the lower limbs and the trunk using prostheses with constantfriction knee components To measure the multiple displacement patterns in order to compare the stride dimensions and temporal components during slow, free-speed, and fast walking, using a constantfriction knee component to a hydraulic swing-control knee component To establish the relationship between weight distribution, the anterior–posterior and medio-lateral center of exerted pressure, the strength of the hip abductor muscle and gait parameters To quantify the sagittal plane kinematic characteristics and the joint moment and power demands placed on the intact limb during walking To compare three prostheses – the SACH, the energy-storing Seattle prosthesis and the Indian Jaipur prosthesis, which is more prevalent in eastern countries

5-U 10-B

9-B

11-B

6-B 7-U

10-B

11-B

11-B 8-B 11-B 9-B 3-U

6-U

9-C

8-B

11-B 8-U 7-U 7-B 6-B 3-U 8-B

8-B

10-B

9-B

6-U

Y. Sagawa Jr. et al. / Gait & Posture 33 (2011) 511–526

516 Table 1 (Continued ) Ref

Articles

Participants (N)

Sex and age

Main objective(s)

L

[100]

Pagliarulo et al. (1979)

15 Uni, Tt, Tr, Co

12 M, 3 F, 29  11

8-U

[21]

Paysant et al. (2006)

[101]

Pinzur et al. (1991)

[58]

Pinzur et al. (1991)

10 Uni, Tt, Tr 10 Ab 7 Uni, Tt, Va 7 Ab 7 Uni, Tt, Va 5 Uni, Tt, Va

10 M, 39  14 10 M, 39  ? ? M, ? F, 64  ? ? M, ? F, ?  ? ? M, ? F, ?  ? ? M, ? F, ?  ?

[102]

Pinzur et al. (1992)

[48]

Postema et al. (1997)

25 Uni, Tf, Tt, Va 5 Ab, Va 10 Uni, Tt, Tr, Va, Co

? M, ? F, 58  ? ? M, ? F, 54  ? 9 M, 1 F, 49  11

To determine the metabolic energy cost of walking with crutches and with prostheses To investigate the influence of ground surface on walking in real-world situations To evaluate the pressures applied at several ‘‘at-risk’’ locations on the plantar surface of the intact foot To evaluate the phasic myoelectric activity of the quadriceps and the hamstring muscles in both the sound and amputated limbs of active amputees that do a limited amount of walking To evaluate the metabolic demands during gait

[67]

Powers et al. (1994)

10 Uni, Tt, Tr

10 M, 45  15

[15]

Powers et al. (1996)

[54]

Powers et al. (1998)

22 72 10 10

15 M, 7 F, 60  11 ? M, ? F, ?  ? 10 M, 62  7 5 M, 5 F, 51  9

[60]

Prince et al. (1998)

5 Uni, Tt, Tr, Va

? M, ? F, 42  14

[61]

Rabuffetti et al. (2005)

[103]

Royer and Wasilewski (2006) Sadeghi et al. (2001)

11 Uni, Tf 7 Ab 10 Uni, Tt, Tr, Va, Co 5 Uni, Tt, Tr, Va

10 M, 1 F, 36  25 7 M, 38  19 9 M, 1 F, 41  10 3 M, 27  13

6 Uni, 6 Ab 5 Uni, 6 Uni, 23 Ab 7 Uni, 8 Uni, 6 Uni, 6 Uni, 8 Uni, 9 Ab

6 M, 40  7 6 M, 33  7 5 M, 53  14 6 M, 58  11 ? M, ? F, 51  ? ? M, ? F, 49  17 ? M, ? F, 44  17 ? M, ? F, 33  6 ? M, ? F, 36  9 7 M, 1 F, 47  13 6 M, 3 F, 29  8

[72]

[55] [28]

Sanderson and Martin (1997) Sapin et al. (2008)

[23]

Schmalz et al. (2002)

[64]

Segal et al. (2006)

[104]

Segal et al. (2009)

[70]

Seroussi et al. (1996)

[74]

Silverman et al. (2008)

[62]

Uni, Tt, Va Ab Uni, Tt, Va Ab

Tt, Tr Tf, Tr, Co Tf, Tr Tt, Tr Tt, Tr Tf, Tr Tf, Tr Tf, Tr

9 M, 1 F, 56  12 6 M, 4 F, 44  14

10 Uni, Tt, Tr, Va, In, Ca 10 Ab 8 Uni, Tf, ? 8 Ab 14 Uni, Tt, Tr, Va 10 Ab

? M, ? F, 37  ? ? M, ? F, 32  ? 13 M, 1 F, 45  9 7 M, 3 F, 33  12

Sjodahl et al. (2002)

9 Uni, Tf, Tr, Ca 18 Ab

5 M, 4 F, 33  33 9 M, 9 F, 36  8

[105] [57]

Snyder et al. (1995) Su et al. (2007)

[106]

Torburn et al. (1995)

[107]

Traballesi et al. (2008)

7 Uni, Tf, Va 19 Bi, Tt 14 Ab 7 Uni, Tt, Va 7 Uni, Tt, Tr 16 Uni, Tf, Va 8 Uni, Tt, Va

7 M, 62, 8 ? M, ? F, 53  18 ? M, ? F, 26  ? 7 M, 62  8 7 M, 51  16 11 M, 5 F, 61  11 6 M, 2 F, 56  17

[44]

Underwood et al. (2004)

11 Uni, Tt, Tr

8 M, 3 F, 42  12

[108]

Van Jaarsveld et al. (1990)

5 Uni, Tt, ?

5 M, 39  15

[35]

Wirta et al. (1991)

19 Uni, Tt, ?

15 M, 4 F, 48  16

[38]

Wright et al. (2008)

10 Bi, Tt, Tf, Tr, Co 10 Ab

10 M, 40  12 10 M, ? F  ?

To obtain a better understanding of the user benefits of the energy storing and release behavior of some prosthetic feet that are regularly used in patient care To examine the joint motion and GRF characteristics of five different prosthetic feet To establish a relationship between muscular torque capability and stride characteristics To use an ‘‘integrated approach’’ (i.e. EMG, kinematics, kinetics) to evaluate the knee mechanics and to identify factors contributing to abnormal knees To evaluate the net energy stored or dissipated and then recovered, as well as spring efficiency, in order to distinguish among three prosthetic feet To determine the effects of the body/socket interface on amputee motor strategies To examine frontal plane moments

12-B 6-B 5-U

5-U 10-B

10-B 9-B 11-B

8-U

7-B 10-B

To gain insight into how hip muscle powers can generate or absorb muscle power activity on the amputated side to compensate for the lack of normal ankle muscle power function, and how these compensatory mechanisms can influence muscle power activities in the sound limb To quantify the adaptation of the joint kinetics in the ankle, knee and hip of both prosthetic and intact limbs To describe amputee gait patterns using two different uni-axis knee joints with a hydraulic swing-phase control and in particular to study the effect of a mechanical knee/ankle link To define more clearly the influence of prosthetic alignment on metabolic energy consumption during walking

9-B

To compare the differences in gait biomechanics of subjects wearing the C-Leg versus a non-computerized prosthesis (Mauch SNS) using a intra-subject design To determine if a commercially available torsion adapter can reduce transtibial amputee joint torques compared to a rigid adapter during straight-line walking and turning gait To estimate the compensatory strategies of ankle, knee, and hip muscles in amputated and intact limbs To better understand compensatory mechanisms by examining the anterior/posterior GRF impulses and joint kinetics, across a wide range of steady-state walking speeds To describe the effect of a training program on temporal parameters and on movements, moments and power in the sagittal plane in the pelvis, hip, knee and ankle joints simultaneously To study the loading patterns for five different prosthetic feet To characterize walking patterns

6-B

To compare the effects of five different commercially-available prosthetic feet on energy expenditure To verify whether or not the energy cost of treadmill walking tests and during free walking are really equivalent, and if not, see where there are measurement differences To examine the effects of two prosthetic feet (the conventional semi-rigid foot versus the dynamic elastic response Flex foot) on the 3-D kinetic patterns during steady-state gait To evaluate the differences in absorption of high-level accelerations among commercially-available prosthetic feet and the influence of the shoe type on these differences To analyze the effect of five commonly prescribed devices on gait and to propose additional guidelines for selecting and prescribing them To determine the physiological cost of walking

10-B 8-B

8-U

7-B

7-B 7-B

8-B

10-B 8-B 9-B 9-B

9-C

9-B

6-B

6-B

Y. Sagawa Jr. et al. / Gait & Posture 33 (2011) 511–526

517

Table 1 (Continued ) Ref

Articles

Participants (N)

Sex and age

Main objective(s)

L

[109]

Zhang and Lee (2006)

12 Uni, Tt

12 M, 50  9

9-B

[47]

Zmitrewicz et al. (2006)

15 Uni, Tt, Tr

14 M, 1 F, 58  7

To evaluate and compare the load-tolerance of different regions of the stump To examine the influence of energy storage and return (ESAR) feet and multi-axis ankles on the ability to generate GRFs and the corresponding impulses during walking at the subjects’ self-selected speeds

10-B

Abbreviations: Ref, references; L, level of evidence [8]; Ab, abled-body; Uni, unilateral; Bi, bilateral; Hd, Hip disarticulation; Tf, transfemoral; Kd, knee disarticulation; Tt, transtibial; Tr, traumatic; Va, vascular; Ca, cancer; Co, congenital; In, infection; ‘‘?’’, not given; U, unassigned.

4.3. The most common parameters for LLA gait analysis The article results often present spatio-temporal parameters first. We found 17 parameters (Table 2), which were mostly compared among AB groups or among other LLA studies. Unsurprisingly, self-selected gait speed was the most common parameter (cited 43 times in 39.5% of the articles), followed by cadence, step length (19% of the articles) and stride length (16% of the articles). These parameters can be obtained with relative precision using a variety of instruments (e.g., accelerometers, optoelectronic cameras, time-sensor cells, footswitches) and they represent a global gait predictor [52,53]. The most commonly used kinematics and kinetic parameters are discussed in Sections 4.4.2– 4.4.4 respectively. Thirty-three parameters were based on ground reaction force (GRF) and impulse provided by force plates, mostly on the vertical and anteroposterior axes. The articles on prosthetic foot components employed these parameters most often (42.3%) because changes in the absorption and propulsive characteristics of the prosthetic feet can be directly reflected in the GRF and impulses (Table 2). There were seven physiological parameters found in the articles, with oxygen consumption and stump EMG signals being

the most frequent. Oxygen consumption was expressed in milliliters per minute per kilogram of body weight (ml/min/kg) or in milliliters per meter per kilogram of body weight (ml/m/kg). These parameters were most often employed to describe gait in function of speed and to assess prostheses aiming to reduce energy expenditure or to achieve energy expenditure close to that of ablebodied (AB) subjects (e.g., mechanical versus microprocessor knees, SACH versus dynamic feet) (Table 2). The EMG intensity and duration were employed to describe and quantify stump muscle activity. This cannot be quantified using biomechanical parameters since muscle contractions do not occur to produce movement distally but to compensate for the absence of adjacent lower-limb structures and to maintain the prosthesis stability [54]. 4.4. Relevant gait parameters for LLA 4.4.1. Spatio-temporal parameters Additional spatio-temporal parameters that may be relevant for LLA gait analysis include stance time, stance time ratio and step time ratio. During the gait cycle, the stance phase on the sound limb is slightly longer than on the prosthetic side. This contributes to a more asymmetrical gait [14,30,40,55,56]. Subjects with

Fig. 2. The list and the frequency of main parameters used in the gait analysis of lower-limb amputees. Abbreviations: t, time; sup, support; CV, coefficient of variation; A, articular; C, center; GRF, ground reaction force; GRI, ground reaction impulse; COP, center of pressure; Acc, acceleration; M, manual; EMG, electromyography; RPE, rating of perceived exertion; n, numbers; diag, diagnosis.

518 Table 2 The list of references for all parameters in biomechanics, physiology and other domains used in the gait analysis of lower-limb amputees for different themes. The number with a star means parameters with a significant difference. The repeated reference number indicates the number of times that a parameter was used. Parameters

N Art.

Feet

Knee

Spatio-temporal – 17 parameters Stance time (s)

14

47*/83

98*/43

Swing time (s) Stride length (m) Cycle time (s) Cadence (step/min) Velocity (m/s)

5 15 3 17 35

Step time (s) Single support time (s) Double support time (s) Foot-flat time (s) Step length (m)

3 2 4 3 17

45*/45, 83 14*, 54*, 105*, 67*/106, 83, 100 47, 83 40, 47, 54, 106, 106, 83, 100 46*, 71*, 41*, 14*, 105*/40, 47, 106, 106, 45, 83, 100 45 83 41*/83 54*, 60*, 59* 47*, 41*, 17*, 32*/40, 106, 83

1 4 1 6 1 3

Ground reaction forces and impulses – 23 parameters Vert. max. F. LR (N/kg)

9

61, 39

64*, 19*, 37*, 98*, 26*, 90*/43

61, 39 61, 39

61*

14*

64*/64

30*

46*, 32*/71, 71

43

30*

46*, 14*, 105*, 105*, 67*, 67*, 17*/46 46* 47*, 75* 47*/46 75*, 75* 75*, 75* 71*, 71*, 71*, 71*/71 41*, 32* 75* 32* 47*, 75*/47, 48, 75 47*, 47*, 75*, 17*/48 47*, 47* 32*

64*/64

Vert. max. F. TS (N/kg) 3 Fore-aft. max. F. LR (N/kg) 4 Fore-aft. max. F. TS (N/kg) 5 28* Vert. F. IC (Impact F. peak) (N/kg) 2 Vert. F. rate IC (Impact F. rate) (N/s/kg) 1 Vert. F. diff. sound/prosthetic (%) 3 Vert. F. excursion (N/kg) 2 Vert. Imp. (N s/kg) 2 Fore-aft. F. excursion (N/kg) 1 Fore-aft. + Imp. (N s/kg) 4 Fore-aft. – Imp. (N s/kg) 5 Fore-aft. Imp. Ratio (%) 2 Fore-aft. F. pattern 1 Med-lat. F. pattern 1 Ground reaction moment (N m/kg) 1 COP excursion (m) 1 Effective foot length COP (m) 1 71*, 71*, 71*, 71* Vert min. F. MS (N/kg), Fore-aft. Min. F. MS (N/kg), Med-lat. min. F. LR (N/kg), Med-lat max. F. MS (N/kg), Med-lat. max. F. TS (N/kg)

Others 63*, 63*, 57*, 72*, 66*, 73*, 76*, 55*, 77*/103, 56 66*/103, 56 102*, 76*, 15*, 27*/31, 23 76* 57*, 62*, 72*, 102*, 76*, 21*, 15*, 77*/103, 21 63*, 62*, 72*, 102*, 102*, 20*, 81*, 107*, 76*, 21*, 15*, 27*, 29*/23, 85 66*/56 66* 31, 56 63*, 57*, 62*, 72*, 66*, 21*/57, 103, 31, 56, 21 57* 72* 31 21 56 72*, 68*/95

34*/57, 55 34*, 55* 55*/57 55*/57 42* 39

31* 27*, 27* 74*, 74*/74, 74 74*, 74* 74* 82* 63* 31, 31

Ground reaction force and impulse times (% stride) – 11 parameters Time at Fore-aft. max. F. LR 2 47*, 67* Time at Fore-aft. max. F. TS 2 47*, 67*, 67* Time at Vert. max. F. LR, Time at Vert min. F. MS, Time at Vert. max. F. TS, Time at Fore-aft. min. F. MS, Time at Med-lat. min. F. LR, Time at Med-lat max. F. MS, Time at Med-lat. max. F. TS Trunk angles (Deg) – 3 parameters Total sagittal plane excursion Total coronal plane excursion Total transversal plane excursion Pelvis angles (Deg) – 9 parameters

1 1 1

63* 63* 63*

Y. Sagawa Jr. et al. / Gait & Posture 33 (2011) 511–526

Step width (m) Stance time ratio (%) Swing time ratio (%) Step length ratio (%) Timing gait events (%) CV (%)

Socket

Total sagittal plane excursion 4 46 Total coronal plane excursion 2 Total tranversal plane excursion 1 Pelvis/scapular girdles r. phase 1 Pelvis saggital pattern 1 Min. rot. sagittal plane, max. rot. coronal plane, max. rot. coronal plane, max. rot. transverse plane

61*

63*, 57* 63*, 96* 63* 63* 62

Pelvis angle times (% stride) – 4 parameters Time at Min. rot. sagittal plane, time at max. rot. coronal plane, time at max. rot. coronal plane, time at max. rot. transverse plane Hip angles (Deg) – 13 parameters Max. flex. at LR 2 62*/62, 56 Max. ext. in stance phase 2 46 77* Total sagittal plane excursion 5 48*/40 61* 57*, 92* Total coronal plane excursion 1 45* Hip sagittal pattern 1 95 Flexion at IC, Flexion at toe off, max. flex. in swing phase, max. add. in stance phase, max. abd. in swing phase, total transverse plane excursion, max. int. rot. in stance phase, max. exte. rot. in swing phase Hip angle times (% stride) – 7 parameters Time at max. flex. at LR, time at max. ext. in stance phase, time at max. flex. in swing phase, time at max. add. in stance phase, time at max. abd. in swing phase, time at max. int. rot. in stance phase, time at max. exte. rot. in swing phase

Knee angle times (% stride) – 7 parameters Time at max. flex. at LR, time at max. ext. in stance phase, time at max. flex. in swing phase, time at max. add. in stance phase, time at max. add. in swing phase, time at max. int. rot. in stance phase, time at max. exte. rot. in swing phase Ankle angles (Deg) – 10 parameters Max. plant. flex. at LR Max. dorsiflex. in stance phase Flexion at toe off Max. dorsiflex. in swing phase Total sagittal plane excursion Total coronal plane excursion Max. eversion in stance phase Ankle saggital pattern Flexion at IC, max. inversion in swing phase

5 4 1 1 4 1 1 2

14*, 48*, 59*, 84* 46*, 105*, 67*, 67*

77* 57* 70* 28*

48*, 17* 102* 44

68*, 82*/68, 68

Y. Sagawa Jr. et al. / Gait & Posture 33 (2011) 511–526

Knee angles (Deg) – 13 parameters Max. flex. at LR 10 40*, 54* 64, 64 57*, 24*, 65*, 62*, 66*, 56*, 56*, 33* Max. ext. in stance phase 2 70*, 62 Flexion at toe off 2 64, 64 56* Max. flex. in swing phase 4 46*/40 64*, 64*/64 56* Total sagittal plane excursion 3 17*, 17* 39 68* Flexion at IC, total coronal plane excursion, max. add. in stance phase, max. add. in swing phase, total transverse plane excursion, max. int. rot. in stance phase, max. exte. rot. in swing phase Knee sagittal pattern 3 94 55*/95

95*, 55*

Ankle angle times (% stride) – 5 parameters Time at Max. plant. flex. at LR, time at max. dorsiflex. in stance phase, time at max. dorsiflex. in swing phase, time at max. eversion in stance phase, time max. inversion in swing phase Lower-limb moments (N m/kg) – 2 parameters Max. lower-limb moment Lower-limb moment pattern Hip joint moments (N m/kg) – 8 parameters Max. flex. moment at IC Max. ext. moment at TS First max. add. moment Max. exte. rot. moment Max. flex. moment at MS Hip sagittal pattern Max. int. rot. moment, second max. add. moment

1 1

5 3 2 1 1 1

70* 55

40*, 14* 40

68*, 70*/15 68*, 70* 103*/15 104*

40 55

519

Hip joint moment times (% stride) – 6 parameters Time at max. flex. moment at IC, time at max. ext. moment at TS, time at first max. add. moment, time at second max. add. moment, time at max. exte. rot. moment, time at max. int. rot. moment

520

Table 2 (Continued ) Parameters Knee joint moments (N m/kg) – 10 parameters First max. ext. moment First max. flex. moment Second max. ext. Moment TS First max. add. moment Max. int. rot. Moment Second max. flex. moment Knee sagittal moment pattern Max. abd. Moment, second max. add. Moment, Max. exte. rot. moment

N Art.

5 5 2 3 1 2 2

Feet

Knee

40*, 44*, 14* 54* 40 44*, 44*

64

44* 54

94

64*, 64*, 64*

Socket

Others

39, 39

55*/55 62*, 70*, 55* 55 103* 104* 68*, 68*

Knee joint moment times (% stride) – 8 parameters Time at first max. flex. moment 1 54* Time at first max. ext. moment, time at second max. ext. moment at TS, time at Max. abd. moment, time at first max. add. moment, time at second max. add. moment, time at max. exte. rot. moment, time at max. int. rot. moment

4 8 1

44*, 14* 71*, 71*, 14*, 17*/40 71*, 71*

68*/62 57*, 62*, 70*, 70*, 55*

Ankle joint moment times (% stride) – 4 parameters Time at max. plant. flex. moment 2 40* Time at max. dorsiflex. moment at TS, time at max. eversion moment, time at max. inversion moment

55*

Hip joint powers (W/kg) – 8 parameters Max. ext. generate LR (H1S) Max. flex. absorb (H2S) Max. flex. generate PS (H3S) Max. abd. absorb (H1F) Max. abd. generate (H3F) Max. ext. rot. generate (H2T)

9 6 9 1 1 1

40*, 14*, 73*/44 40, 44 40*, 46*/44

64 64 64

74*, 74*, 57*, 62*, 70* 57*, 62*/72 57*, 62*/72, 68, 70 72* 72* 72*

Knee Max. Max. Max. Max. Max.

7 7 5 1 2

40*, 14*, 73*, 73* 54*, 73* 44*/40

64 64 64 64

62*, 72*, 27* 74*, 74*, 62*, 72*, 27* 72*, 68*, 68*

joint powers (W/kg) – 6 parameters ext. absorb LR (K1S) ext. generate (K2S) ext. absorb TS (K3S) ext. absorb (K4S) add, abd. generate (K2F)

44*

72*

Ankle joint powers (W/kg) – 3 parameters Max. dorsiflex. absorb (A1S) Max. plant. flex. generate at PS (A2S) Recovery energy (plantarflex generate/dosiflex absorb (%))

3 12 1

44*, 44* 40*, 46*, 74*, 73*/44, 48 60*

64 64

74* 74*, 57*, 62*, 72*, 70*

Physiological – 5 parameters Heart rate (bpm) Blood pressure (mm Hg) VO2 (ml/min/kg)

7 1 15

100*, 88* 100 17, 106

90*, 90*

22*, 85*/102, 107

VO2 cost (ml/m/kg) Respiratory quotient (%)

12 1

100*, 88*

79*, 43*, 78*, 37*/43, 78, 37, 37, 80 19*, 37*, 80*/37

86*, 23*, 23*, 23*, 23*, 22*, 85*, 107*, 21*, 21*, 36*/86, 23, 23, 85, 20, 107, 21 86*, 22*, 102*, 81*, 107*/24, 20 86*

EMG – 2 parameters Stump muscle activity intensity Stump muscle activity time

7 3

54*, 83*, 83*/95 54*

Psychological – 4 parameters Prosthesis comfort Prosthesis preference

1 2

46 45*

66*, 92*, 76*/66, 66, 58 91*, 91*, 91*, 91*, 92*, 92*, 92*, 92*/91, 91, 91, 91, 91, 92, 92

64*

Y. Sagawa Jr. et al. / Gait & Posture 33 (2011) 511–526

Ankle joint moments (N m/kg) – 5 parameters Max. plant. flex. moment Max. dorsiflex. Moment at TS Max. dosiflex. moment diff. sound/prosthetic (%) Max. eversion moment, max. inversion moment

5 3

Others – 26 parameters Hip muscle force Weight-bearing distal end ability Socket pressure peak Socket pressure contact time Socket pressure contact area Foot store/dissipate energy Shock factor (accelerometer) Trunk center of mass vertical excursion COM vertical excursion COM ext. mechanical energy COM energy recovery COM negative external mechanical work of the leading limb COM positive external mechanical work of the trailing limb COM negative external mechanical work to step-to-step transition Cost of components (total) Cost number of visits Cost prosthesis process time Timed walking test Steps/week Falls PEQ score Comorbidity number Socket volume Socket displacement Osteoarthritis clinical diagnosis

1 1 3 1 1 1 1 1 1 2 1 1 1 1 1 1 1 1 1 2 1 1 1 1 1

44

43*, 37*, 26*

39* 42*, 22*/33

31 109 109*, 109*

99* 99* 99* 60* 17*

101*

19 24* 24*, 20* 24 20* 20* 20* 39* 39 39* 46 33 36*

26*, 26* 26

36* 30*, 30* 30*, 30* 27*, 27*

Number: reference. Number*: reference and significant difference using such parameter. Others: descriptive, rehabilitation, fitness and pylon studies. Abbreviations: N Art, number of articles, which utilized such parameter; F, force; Imp, impulse; Vert, vertical; Med-lat, mediolateral; IC, initial contact; LR, loading response; MS, mid stance; TS, terminal stance; diff, difference; flex, flexion; ext, extension; add, adduction; abd, abduction; int. internal; rot, rotation; exte. external; plant, plantar; dorsiflex, dorsiflexion; r, relative; COM center of body mass.

Y. Sagawa Jr. et al. / Gait & Posture 33 (2011) 511–526

Satisfaction rate Rate of perceived exertion

521

522

Y. Sagawa Jr. et al. / Gait & Posture 33 (2011) 511–526

Table 3 The main scientific journals pertaining to gait analyses and the distribution of articles. Journals

IF 2008

N

Articles

L (mean  SD, min–max)

J Biomech Gait Posture IEEE Trans Neural Syst Rehabil Eng Phys Ther Arch Phys Med Rehabil Scand J Rehabil Med Clin Biomech J Rehabil Med Eur J Appl Physiol Clin Orthop Relat Res Clin Rehabil Am J Phys Med Rehabil J Rehabil Res Dev Bull Prosthet Res Foot Ankle Orthopedics Physiother Res Int Prosthet Orthot Int

2.784 2.743 2.7 2.190 2.159 215 2 1.983 1.931 1.893 1.840 1.695 1.446 1.29 1.061 0.588 0.561 0.377

1 8 1 2 11 2 2 1 1 1 1 6 13 1 1 2 1 30

7 9  1.3, 7–11 9 8.5  0.7, 8–9 8.9  1.2, 7–11 7.5  0.7, 7–8 8  1.4, 7–9 10 12 8 3 8.3  0.5, 8–9 8.3  1.8, 6–12 8 6 5 10 7.5  2, 3–11

0.343 – – –

1 1 1 1

14 20, 34 15, 17, 18, 44, 91 86 92 43 19, 21, 97 58 58, 31 25, 66, 99 93 88 94

Int J Rehabil Res Ann Phys Rehabil Med J Orthop Sports Phys Ther Proc Inst of Mech Eng Part H, J Eng Med

23, 24, 54, 55, 74, 103, 107 100 22, 27, 37, 46, 47, 63, 67, 70, 78, 98 90 76

65, 72, 73, 80, 81 26, 33, 35, 42, 57, 60, 64, 87, 96, 104–106

102 28–30, 32, 36, 38–41, 45, 48, 56, 59, 61, 62, 68, 71, 75, 77, 80, 82–85, 89, 95, 108, 109

6 3 11 6

Abbreviations: IF, impact factor; N, number; L, level of evidence obtained from the van der Linde et al. [8] criteria list with a maximum score of 13 points.

unilateral amputations rely more on their sound leg to compensate for some of the deficiencies associated with prostheses [57]. However, the increased loading period may explain the development of complications in the remaining limb [58]. Some studies have demonstrated positive effects from specific prosthetic components. Subjects felt more confident with such a prosthetic foot and therefore compensated less on the sound side [14]. The use an appropriated socket, and through appropriate fitting, the degree of gait asymmetry can be reduced [30]. Appropriated fitting may be offering LLA better control over the prosthesis position, possibly by improving proprioception and force transfer to the prosthesis. This probably improves the symmetry of their gait [30]. Foot flat time is another spatio-temporal parameter that seems appropriate for LLA gait analysis. AB subjects reached foot flat phase at 12–17% of the gait cycle [54,59] compared to subjects using a solid ankle cushion heel (SACH) foot, who were found to make heel contact only for the first 20% [60] or 44.5% [59]. The inability to achieve foot flat contact during loading phase could be attributed to the compromised plantar flexion of some prosthetic feet. For example, Seattle Light foot has been reported to provide only 2–38 of motion [54]. Reduced motion during weight acceptance may result in a period of instability during which balance relies on the rear foot. It is therefore likely that the contraction of the adjacent muscles (e.g., quadriceps and hamstrings) could represent an attempt to provide joint stability during this phase of the gait cycle. Powers et al. [54] suggest that individuals with transtibial (TT) amputation need to stabilize the knee during weight acceptance due to the prolonged heel-only contact caused by reduced prosthetic foot mobility. Nevertheless, other prosthetic feet, such as the Golden-Ankle one, attained the foot flat phase significantly earlier in the gait cycle (14%) compared with those fitted with either the Seattle Light foot (21%), or the SACH foot (20%) [60]. Golden-Ankle foot users showed a more natural gait pattern, probably because its ankle articulation approached natural ankle function [60]. 4.4.2. Joint angle parameters Reduced hip range of motion (ROM) in the sagittal plane is associated with ischium–socket interference in TF amputees [61]. Reduced hip ROM, mainly during hip extension, also employs

compensatory mechanisms on the pelvis and on the sound side in order to maintain adequate speed. Rabuffetti et al. [61] suggested that, when the hip on the prosthetic side is extended the ischium– socket interference limits the physiological ROM. In order to maintain functional step length, the pelvic range of motion (ROM) in the sagittal plane and the hip flexion of the sound side increased compared to AB subjects. In accordance with previous studies [61,62], Goujon-Pillet et al. [63] found that, for TF subjects (88  58), the pelvis ROM is twice that of AB subjects (48  18). In the long-term, this motor strategy may cause lower back pain, which is often reported by TF amputees. The literature reviewed suggests, that prosthetic feet have no influence [40] or a minor influence [48] on the hip motion in the sagittal plane in TT and TF amputees. Pelvic ROM in the frontal plane is increased, in LLA compared to AB subjects. Su et al. [57] found a significant difference when comparing LLA at self-selected speeds (8.48  2.88) to AB subjects at slow speeds (6.38  2.18). LLA lifted the pelvis on the swing side while walking. This compensatory motion, known as hip hiking, is often observed in individuals with unilateral TT or TF amputations and is believed to compensate for the inability to dorsiflex the prosthetic ankle. Hip hiking increases the prosthetic foot clearance [57]; however, it may also require additional metabolic energy to lift body mass against gravity, thus reducing gait efficiency. Individuals with bilateral TT amputations display bilateral hip hiking, which probably requires increased energy expenditure during walking compared to unilateral amputees [57]. Knee flexion during the loading phase has a ‘‘shock-absorbing’’ effect which is important in the prevention of wear and tear of weight-bearing joints [56]. For AB subjects or the LLA’s sound side, this parameter value is about 15–188 [54,57,64]. For TT amputees, it is limited to 9–128 [54,56,57]; however, for TF amputees, it is often absent or negative [64]. During gait speed changes, this parameter is less variable on the prosthetic side than on the sound side [56,57]. It can be influenced by many conditions, such shoed or barefoot walking [65], the type of socket [66] and rehabilitation [62]. However, Segal et al. [64] demonstrated that TF patients who used microprocessor controlled knee (e.g., C-Leg) designed to allow controlled stance phase knee flexion, this did not normalize. They suggested that, although stance phase knee flexion is possible, it is difficult for TF amputees to achieve it, possibly because they associate this action with buckling and falling.

Y. Sagawa Jr. et al. / Gait & Posture 33 (2011) 511–526

Plantar flexion in the early stance phase is an important parameter, which quantifies the prosthetic foot’s capabilities to be flat on the ground, allowing more floor contact, which permits better stability during the stance phase. This parameter changes considerably with the prosthetic foot design. Most of the dynamic prosthetic feet are composed by a blade without an articulated ankle joint, so that the measured ankle plantar flexion motion during the early stance phase is primarily due to heel compression. This feature has been reported by Postema et al. [48], who found that the Multiflex articulated foot has better plantar flexion at normal speeds (NS) ( 8.38  .48) and fast speeds (FS) ( 6.58  .78) than other so-called ‘‘flexible’’ feet: Springlite II (NS: 6.18  3.28; FS: 4.98  .28); Carbon copy II (NS: 4.58  .98; FS: 4.68  .78); and Seattle light foot (NS: 4.68  .68; FS: 4.88  .78). The next important parameter is the dorsiflexion motion during the mid-to-late stance phase. Prosthetic feet provide less ankle motion during the stance phase than the natural ankle motion in AB subjects: 12.58  3.18 at self-selected speeds for LLA versus 20.28  3.58 at slow speeds for AB subjects [48,67]. This parameter also seems to be greatly influenced by prosthetic foot designs. As explained above, for dynamic prosthetic feet the dorsiflexion motion also depends of the blade capacity to bend. Power et al. [67] found better results for Flex foot (23.28) and Quantum (19.58) feet than for Seattle (15.18), Carbon copy II (12.18) or SACH (128) feet. The last parameter is the total ankle ROM in the sagittal plane. Nolan and Lee [68] reported a 218 ROM for AB subjects, a 208 ROM for the LLA’s prosthetic side and a 268 ROM for their sound side. The increased ROM for the sound limb was attributed to the limited ankle movement on the prosthetic limb: LLA need to increase sound limb length in order to clear the prosthetic limb during the swing phase, a compensation method that has been previously reported for TF amputee gait. Nolan and Lee [68] also found that the type of prosthesis used can affect lower-limb kinematics. They observed in their study that TT amputees using a SACH foot had 108 ROM less in the prosthetic ankle and 158 ROM greater in the sound ankle than when they used a multi-axis foot. 4.4.3. Joint moment parameters McNealy and Gard [40] found that the mean hip joint moment in the sagittal plane at initial contact in LLA was +.8 Nm/kg, which was more than twice the one observed in AB subjects (+.3 Nm/kg). It appeared that the hip joint in the LLA group was critical in generating power during the early stance phase. Since the TF amputees investigated did not have any active control of the ankle and knee joints, the moment generated at the hip probably assisted forward progression [40]. Nolan and Lee [68] also found similar results for the same moment on the sound hip side. Seroussi et al. [70] found that the hip moment became a flexor moment (external) substantially earlier for the sound limb than the prosthetic limb of LLA and both limbs of AB subjects. The greatest difference between LLA and AB subjects for the knee moment in the sagittal plane occurs during the external flexion moment in the loading-response phase. According to Powers et al. [54], AB subjects rely on the knee extensors, which are acting to control knee flexion during weight acceptance. This was confirmed through an EMG of the vastus lateralis, which demonstrated an average intensity of 29% of a maximal muscle contraction test and lasted until 22% of the gait cycle [54]. In contrast, the moment data obtained from the TT amputees suggested negligible extensor demand as the knee flexion moment was significantly smaller during the stance phase. However, the EMG results showed that activity of the vastus lateralis in the TT group was significantly greater than that of the AB group. On average, the TT amputees demonstrated a 25% increase in the mean intensity of the vastus lateralis and significantly longer duration of activity (lasting until 33% of the gait cycle) compared to normal values [54].

523

The inconsistency between the moment data and the EMG results for this population indicates a discrepancy between the mechanical measurements of knee extensor demand and the physiological response. Hence, caution is needed while interpreting the results [54]. For TF amputees, the knee flexion moment is even smaller because the prosthetic knee cannot replace quadriceps muscle activity. However, Segal et al. [64] found a small but significant flexor knee moment when TF subjects used the microprocessor knee C-Leg (.14  .05 N m/kg), compared to subjects using a mechanical knee (.06  .07 N m/kg). These values are still modest when compared with those of AB subjects (.47  .1 N m/kg). LLA develop an external plantar ankle moment over the first 20% of the gait cycle, whereas AB subjects display this moment only for the first 9% of the gait cycle [40]. AB subjects are able to rapidly achieve the foot flat phase and to transfer the load onto the leading leg during the loading-response phase in preparation for single support. In doing so, they advance the center of pressure under the foot to a position anterior to the ankle joint axis [40]. Additionally, the stance phase knee flexion in AB subjects probably helps to reduce the duration of the negative ankle moment [40]. LLA spend more time, compared to AB subjects, rotating their prosthetic legs forward until the foot flat phase is achieved. This is due to reduced movement of the prosthetic ankle, combined with the absence of stance phase knee flexion [40]. Sjodahl et al. [62] found that, both before and after a rehabilitation program, there was a bump in the external ankle dorsiflexion moment at 26% of the gait cycle on the sound side, indicating that LLA were using a vaulting movement. This adaptation was probably used to facilitate toe clearance on the prosthetic side [55,68]. Su et al. [57] reported that the external ankle dorsiflexion moment at the end of the stance phase of unilateral TT amputees was only 60–70% of the moment found in AB subjects. They suggested that this reduction was due to the absence of the ankle plantar flexors. Another possibility is that the keel of the prosthetic foot was functionally shorter than that of the biological foot, reducing the moment arm between the GRFs and the ankle joint center during stance [71]. However, the ankle moment on both the prosthetic and sound sides may vary according to the rehabilitation program [62] or prosthetic foot characteristics [14,17,44]. As Undewood et al. [44] have shown, LLA using a dynamic Flex foot have 15% greater ankle dorsiflexion moment when compared to a non-dynamic foot prosthesis. 4.4.4. Joint power parameters On both the prosthetic and sound sides, LLA have a greater amplitude and duration of hip joint power lasting throughout the first half of the stance phase (55–60% of gait cycle) than AB subjects (20% of gait cycle) [72]. Sadeghi et al. [72] called this hip joint power ‘‘the first limb propeller parameter’’. The increase in hip extensor power output in TT amputees corresponds to an increase in the gluteus maximus and hamstring activity, as shown by EMG. Increased hip extensor use in the early stance phase appears both to control knee flexion during limb loading and to ‘‘pull’’ the trunk forward after heel strike as the foot makes contact with the floor. Using hip extensors as a source of power represents a compensation for the lack of appropriate ankle function during push-off [72]. The excessive work produced at the hip may further contribute to the increased energy expenditure in LLA walking [57]. The increased hip power has been confirmed by other studies [44,57,62,73,74]. Seroussi et al. [70] found an increase of 270% in the concentric hip extensor work in the early stance phase on the sound side of LLA when the prosthetic limb was pushing off. Gitter et al. [73] have shown an increased use of concentric hip extensor energy generation during the early and mid stance (H1S) in

524

Y. Sagawa Jr. et al. / Gait & Posture 33 (2011) 511–526

ambulation with a non-dynamic foot, such as the SACH. In addition, Underwood et al. [44] confirmed the influence of the type of prosthetic foot. They showed, that the H1S power generation tended to be smaller on both the prosthetic limb (a 66% decrease) and the sound limb (a 75% decrease) when LLA were wearing the dynamic Flex foot, compared to a SACH foot. This finding suggests that the dynamic Flex foot requires less passive and active stability, compared to a conventional prosthetic foot. Compared to AB subjects, LLA exhibit another large source of power generation at the hip joint prior to toe off (H3S) [40,70,72]. The H3S power burst represents the action of the hip flexors pulling the lower-limb upward and forward. As shown in AB subjects, the ankle power generation and the H3S hip power bursts occur simultaneously during the gait cycle as the leading limb prepares for the swing phase. Thus, LLA apparently compensate for the lack of energy produced in the prosthetic ankle (A2S) by increasing power generation at the hip (H3S) immediately prior to toe off [40,70,72]. Sadeghi et al. [72] reported a reduction of 63% in the knee power absorption during the loading-response phase (K1S) on the prosthetic side of TT amputees, compared to the sound side. The problem was worse for TF amputees because the negative external knee moment did not allow prosthetic knee motion in the stance phase. Thus there was no energy storage (K1S) or energy return (K2S) in the knee power curves during this phase of the gait cycle [40]. Consequently, the knee joint did not contribute to shock absorption, energy storage or energy returns for most of the gait cycle [40]. Both the TT and TF amputees appeared to use the hip joint as an alternative means of shock absorption, power absorption and power generation [40,72], or they overloaded their sound knees [62]. There are two important parameters for ankle power: the ankle power absorption (A1S) at the loading-response phase and the mid stance phase and the ankle power generation (A2S) at the end of the stance phase. The deformation properties of dynamic feet allow for a greater power absorption (A1S) during weight acceptance and, consequently, a trend towards a greater external dorsiflexion moment; power generation (A2S) takes place at push-off. Although the use of the dynamic foot increased the dorsiflexion moment and push-off power of the prosthetic ankle, these values were still well below those for the ankles of AB subjects [44]. Seroussi et al. [70] found that the LLA’s prosthetic ankle at push-off reached only about 20% of the ankle work generated by the AB subjects’ ankle. Su et al. [57] found that the prosthetic side of TT amputees, walking at a self-selected speed (.38  .18 W/kg), generated four times less ankle power than the AB subjects, walking at a slow speed (1.26  .38 W/kg). For the same ankle parameter, Sadeghi et al. [72] found a reduction of 76% for the TT prosthetic side compared to the sound side, which was expected because of the limited deformation capability of prosthetic feet [72]. Ankle power parameters seem to be greatly influenced by the type of prosthetic foot [14,44,46,48,70] or by the rehabilitation program [62]. It has been shown that the ankle plantarflexors at push-off are a major source of energy generation when walking [70]. Therefore, the decrease in prosthetic ankle push-off represents a substantial loss of the mechanical work generated by the lower extremity during walking, resulting in a number of possible compensatory mechanisms [70,72]. Seroussi et al. [70] described three possible compensatory mechanisms in their study. First, the sound ankle of LLA generates approximately one third more work than the ankles of the AB subjects during push-off. Second, the decrease in the prosthetic ankle push-off creates an increase in the concentric hip extensor work in the early stance phase on the sound side of LLA. Third, there is a relative increase in the concentric hip pull-off in the prosthetic limb (H3S) [70].

4.5. The main limitations of studies and parameters discussed There is a lack of studies in the literature on biomechanical models adapted for LLA, which would take into account the characteristics of prosthetic components (i.e., mass, center of articulation, center of mass, moment of inertia). All results for ankle motion in the sagittal plane described above were obtained using a method in which the joint position of the prosthetic ankle was assumed to be in the same position as that of an intact ankle. However, a recent study [69] has shown that the motion of the prosthetic feet is different from that of the intact ankle. Thus, the use of this method is subject to systematic errors as it could not reflect the real motion of the prosthetic foot. The same is probably the case for other prosthetic components, such as the polycentric knee mechanisms, and the same errors should be expected. Moreover, many of the studies investigated stated that the prosthetic components were adjusted and aligned by an experienced prosthetist, without providing any more information. However, individual fitting characteristics may affect motion analysis and lead to error. The manufacturers of some dynamic feet, for example, recommend that for an optimal function, these feet require to be set in slight plantar flexion. Another example is the socket of TF subjects, which is often manufactured with an initial flexion to facilitate support during gait. Thus, if this information is not explicit, it could bias interpretations of LLA gait analysis. Indeed, in our opinion, further work is required to establish a methodological consensus or a guideline before clinically meaningful measurements can be confidently based on LLA gait analysis. 5. Conclusion In this study, we presented and discussed the 32 most common parameters published until now. Many of the parameters found were not reported in enough studies or in enough detail to allow a useful discussion. The diversity in the outcomes selected to describe the LLA gait cannot be explained by differences in research objectives only. This parameter diversity suggests that there is a lack of consensus among researchers about the aspects of gait that are important when assessing LLA outcomes. Finally, due to the methodological inconsistency of the studies and the parameter diversity, it has been difficult to identify the main parameters that should be used in gait analysis for LLA. Although this systematic review cannot correct the biases and methodological flaws observed in the original studies, it could help guide future studies for choosing parameters, thus bringing about a more evidence-based compromise. Further research emphasizing the clinical usefulness of LLA gait analysis may help determine which gait parameters provides the most useful information. Acknowledgments The present research work has been supported by OSEO project number A0607009N, International Campus on Safety and Intermodality in Transportation, the Nord-Pas-de-Calais Region, the European Community, the Regional Delegation for Research and Technology, the Ministry of Higher Education and Research, and the National Center for Scientific Research. The authors gratefully acknowledge the support of these institutions. Conflict of interest The authors state that no conflicts of interest are present in this research.

Y. Sagawa Jr. et al. / Gait & Posture 33 (2011) 511–526

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.gaitpost.2011.02.003. References [1] Amputation coalition of America [cited 2010 March, 23]; Available from: www.amputee-coalition.org. [2] Dillingham TR, Pezzin LE, MacKenzie EJ. Limb amputation and limb deficiency: epidemiology and recent trends in the United States. South Med J 2002;95:875–83. [3] Dillingham TR, Pezzin LE, Shore AD. Reamputation, mortality, and health care costs among persons with dysvascular lower-limb amputations. Arch Phys Med Rehabil 2005;86:480–6. [4] Hafner BJ, Sanders JE, Czerniecki J, Fergason J. Energy storage and return prostheses: does patient perception correlate with biomechanical analysis? Clin Biomech 2002;17(5):325–44. [5] Michael JW. Modern prosthetic knee mechanisms. Clin Orthop Relat Res 1999;361:39–47. [6] Gerzeli S, Torbica A, Fattore G. Cost utility analysis of knee prosthesis with complete microprocessor control (C-leg) compared with mechanical technology in trans-femoral amputees. Eur J Health Econ 2009;10:47–55. [7] Gailey RS, Roach KE, Applegate EB, Cho B, Cunniffe B, Licht S, et al. The amputee mobility predictor: an instrument to assess determiants of the lower-limb amputee’s ability to ambulate. Arch Phys Med Rehabil 2002;83:613–27. [8] van der Linden ML, Hofstad CJ, Geurts AC, Postema K, Geertzen JH, van Limbeek J. A systematic literature review of the effect of different prosthetic componets on human functioning with a lower-limb prosthesis. J Rehabil Res Dev 2004;41:555–70. [9] Sansam K, Neumann V, O’Cornnor R, Bhakta B. Predicting walking ability following lower limb amputation: a systematic review of the literature. J Rehabil Med 2009;41:593–603. [10] van Tulder MW, Assendelft WJ, Koes BW, Bouter LM. Method guidelines for systematic reviews in the Cochrane Collaboration Back Review Group for spinal disorders. Spine 1997;22:2323–30. [11] Verhagen AP, de Vet HC, de Bie RA, Kessels AG, Boers M, Bouter LM, et al. The Delphi list: a criteria list for quality assessment of randomized clinical trials for condcting systematic reviews developed by Delphi consensus. J Clin epidemiol 1998;51:1235–41. [12] English RD, Hubbard WA, McElroy GK. Establishment of consistent gait after fitting of new components. J Rehabil Res Dev 1995;32:32–5. [13] Landis JR, Koch GG. An application of hierarchical kappa-type statistics in the assessment of majority agreement among multiple observers. Biometrics 1977;33:363–74. [14] van der Linden ML, Solomonidis SE, Spence WD, Li N, Paul JP. A methodology for studying the effects of various types of prosthetic feet on the biomechanics of trans-femoral amputee gait. J Biomech 1999;32:877–89. [15] Powers CM, Boyd LA, Fontaine CA, Perry J. The influence of lower-extremity muscle force on gait characteristics in individuals with below-knee amputations secondary to vascular disease. Phys Ther 1996;76:369–77. [16] Benedetti MG, Catani F, Leardini A, Pignotti E, Giannini S. Data management in gait analysis for clinical applications. Clin Biomech 1998;13:204–15. [17] Lehmann JF, Price R, Boswell-Bessette S, Dralle A, Questad K. Comprehensive analysis of dynamic elastic response feet: Seattle Ankle/Lite Foot versus SACH foot. Arch Phys Med Rehabil 1993;74:853–61. [18] Boonstra AM, Schrama J, Fidler V, Eisma WH. The gait of unilateral transfemoral amputees. Scand J Rehabil Med 1994;26:217–23. [19] Gitter A, Czerniecki J, Weaver K. A reassessment of center-of-mass dynamics as a determinate of the metabolic inefficiency of above-knee amputee ambulation. Am J Phys Med Rehabil 1995;74:332–8. [20] Houdijk H, Pollmann E, Groenewold M, Wiggerts H, Polomski W. The energy cost for the step-to-step transition in amputee walking. Gait Posture 2009;30:35–40. [21] Paysant J, Beyaert C, Datie AM, Martinet N, Andre JM. Influence of terrain on metabolic and temporal gait characteristics of unilateral transtibial amputees. J Rehabil Res Dev 2006;43:153–60. [22] Hoffman MD, Sheldahl LM, Buley KJ, Sandford PR. Physiological comparison of walking among bilateral above-knee amputee and able-bodied subjects, and a model to account for the differences in metabolic cost. Arch Phys Med Rehabil 1997;78:385–92. [23] Schmalz T, Blumentritt S, Jarasch R. Energy expenditure and biomechanical characteristics of lower limb amputee gait: the influence of prosthetic alignment and different prosthetic components. Gait Posture 2002;16: 255–63. [24] Detrembleur C, Vanmarsenille JM, De Cuyper F, Dierick F. Relationship between energy cost, gait speed, vertical displacement of centre of body mass and efficiency of pendulum-like mechanism in unilateral amputee gait. Gait Posture 2005;21:333–40. [25] Jones ME, Steel JR, Bashford GM, Davidson IR. Static versus dynamic prosthetic weight bearing in elderly trans-tibial amputees. Prosthet Orthot Int 1997;21:100–6. [26] Kahle JT, Highsmith MJ, Hubbard SL. Comparison of nonmicroprocessor knee mechanism versus C-Leg on prosthesis evaluation questionnaire, stumbles,

[27] [28]

[29] [30] [31]

[32]

[33]

[34]

[35] [36]

[37]

[38] [39]

[40] [41]

[42] [43]

[44]

[45]

[46]

[47]

[48]

[49]

[50]

[51]

[52]

[53]

[54] [55]

[56]

525

falls, walking tests, stair descent and knee preference. J Rehabil Res Dev 2008;45:1–13. Lemaire ED, Fisher FR. Osteoarthritis and elderly amputee gait. Arch Phys Med Rehabil 1994;75:1094–9. Sapin E, Goujon H, de Almeida F, Fode P, Lavaste F. Functional gait analysis of trans-femoral amputees using two different single-axis prosthetic knees with hydraulic swing-phase control: kinematic and kinetic comparison of two prosthetic knees. Prosthet Orthot Int 2008;32:201–18. Baker PA, Hewison SR. Gait recovery pattern of unilateral lower-limb amputees during rehabilitation. Prosthet Orthot Int 1990;14:80–4. Board WJ, Street GM, Caspers C. A comparison of trans-tibial amputee suction and vacuum socket conditions. Prosthet Orthot Int 2001;25:202–9. Nadollek H, Brauer S, Isles R. Outcomes after trans-tibial amputation: the relationship between quiet stance ability, strength of hip abductor muscles and gait. Physiother Res Int 2002;7:203–14. Mizuno N, Aoyama T, Nakajima A, Kasahara T, Takami K. Functional evaluation by gait analysis of various ankle-foot assemblies used by below-knee amputees. Prosthet Orthot Int 1992;16:174–82. Berge JS, Czerniecki JM, Klute GK. Efficacy of shock-absorbing versus rigid pylons for impact reduction in transtibial amputees based on laboratory, field, and outcome metrics. J Rehabil Res Dev 2005;42:795–807. Lee WCC, Zhang M, Chan PPY, Boone DA. Gait analysis of low-cost flexibleshank transtibial prostheses. IEEE Trans Neural Syst Rehabil Eng 2006;14:370– 7. Wirta RW, Mason R, Calvo K, Golbranson FL. Effect on gait using various prosthetic ankle-foot devices. J Rehabil Res Dev 1991;28:13–24. Chin T, Sawamura S, Fujita H, Ojima I, Oyabu H, Nagakura Y, et al. %VO2max as an indicator of prosthetic rehabilitation outcome after dysvascular amputation. Prosthet Orthot Int 2002;26:44–9. Casillas JM, Dulieu V, Cohen M, Marcer I, Didier JP. Bioenergetic comparison of a new energy-storing foot and SACH foot in traumatic below-knee vascular amputations. Arch Phys Med Rehabil 1995;76:39–44. Wright DA, Marks L, Payne RC. A comparative study of the physiological costs of walking in ten bilateral amputees. Prosthet Orthot Int 2008;32:57–67. Datta D, Harris I, Heller B, Howitt J, Martin R. Gait, cost and time implications for changing from PTB to ICEX sockets. Prosthet Orthot Int 2004;28:115–20. McNealy LL, Gard SA. Effect of prosthetic ankle units on the gait of persons with bilateral trans-femoral amputations. Prosthet Orthot Int 2008;32:111–26. Goujon H, Bonnet X, Sautreuil P, Maurisset M, Darmon L, Fode P, et al. A functional evaluation of prosthetic foot kinematics during lower-limb amputee gait. Prosthet Orthot Int 2006;30:213–23. Gard SA, Konz RJ. The effect of a shock-absorbing pylon on the gait of persons with unilateral transtibial amputation. J Rehabil Res Dev 2003;40:109–24. Datta D, Heller B, Howitt J. A comparative evaluation of oxygen consumption and gait pattern in amputees using Intelligent Prostheses and conventionally damped knee swing-phase control. Clin Rehabil 2005;19:398–403. Underwood HA, Tokuno CD, Eng JJ. A comparison of two prosthetic feet on the multi-joint and multi-plane kinetic gait compensations in individuals with a unilateral trans-tibial amputation. Clin Biomech 2004;19:609–16. Boonstra AM, Fidler V, Spits GM, Tuil P, Hof AL. Comparison of gait using a Multiflex foot versus a Quantum foot in knee disarticulation amputees. Prosthet Orthot Int 1993;17:90–4. Graham LE, Datta D, Heller B, Howitt J, Pros D. A comparative study of conventional and energy-storing prosthetic feet in high-functioning transfemoral amputees. Arch Phys Med Rehabil 2007;88:801–6. Zmitrewicz RJ, Neptune RR, Walden JG, Rogers WE, Bosker GW. The effect of foot and ankle prosthetic components on braking and propulsive impulses during transtibial amputee gait. Arch Phys Med Rehabil 2006;87:1334–9. Postema K, Hermens HJ, de Vries J, Koopman HF, Eisma WH. Energy storage and release of prosthetic feet. Part 1: biomechanical analysis related to user benefits. Prosthet Orthot Int 1997;21:17–27. Boutron I, Tubach F, Giraudeau B, Ravaud P. Methodological differences in clinical trials evaluating nonpharmacological and pharmacological treatments of hip and knee osteoarthritis. JAMA 2003;290:1062–70. Boutron I, Tubach F, Giraudeau B, Ravaud P. Blinding was judged more difficult to achieve and maintain in nonpharmacologic than pharmacologic trials. J Clin Epidemiol 2004;57:543–50. Boutron I, Guittet L, Estellat C, Moher D, Hrobjartsson A, Ravaud P. Reporting methods of blinding in randomized trials assessing nonpharmacological treatments. PLoS Med 2007;4:e61. Montero-Odasso M, Schapira M, Varela C, Pitteri C, Soriano ER, Kaplan R, et al. Gait velocity in senior people. An easy test for detecting mobility impairment in community elderly. J Nutr Health Aging 2004;8:340–3. Montero-Odasso M, Schapira M, Soriano ER, Varela M, Kaplan R, Camera LA, et al. Gait velocity as a single predictor of adverse events in healthy seniors aged 75 years and older. J Gerontol A Biol Sci Med Sci 2005;60:1304–9. Powers CM, Rao S, Perry J. Knee kinetics in trans-tibial amputee gait. Gait Posture 1998;8:1–7. Sanderson DJ, Martin PE. Lower extremity kinematic and kinetic adaptations in unilateral below-knee amputees during walking. Gait Posture 1997;6: 126–36. Isakov E, Burger H, Krajnik J, Gregoric M, Marincek C. Influence of speed on gait parameters and on symmetry in trans-tibial amputees. Prosthet Orthot Int 1996;20:153–8.

526

Y. Sagawa Jr. et al. / Gait & Posture 33 (2011) 511–526

[57] Su PF, Gard SA, Lipschutz RD, Kuiken TA. Gait characteristics of persons with bilateral transtibial amputations. J Rehabil Res Dev 2007;44:491–501. [58] Pinzur MS, Asselmeier M, Smith D. Dynamic electromyography in active and limited walking below-knee amputees. Orthopedics 1991;14:535–7. [59] Goh JCH, Solomonidis SE, Spence WD, Paul JP. Biomechanical evaluation of SACH and uniaxial feet. Prosthet Orthot Int 1984;8:147–54. [60] Prince F, Winter DA, Sjonnensen G, Powell C, Wheeldon RK. Mechanical efficiency during gait of adults with transtibial amputation: a pilot study comparing the SACH, Seattle, and Golden-Ankle prosthetic feet. J Rehabil Res Dev 1998;35:177–85. [61] Rabuffetti M, Recalcati M, Ferrarin M. Trans-femoral amputee gait: socketpelvis constraints and compensation strategies. Prosthet Orthot Int 2005;29:183–92. [62] Sjodahl C, Jarnlo GB, Soderberg B, Persson BM. Kinematic and kinetic gait analysis in the sagittal plane of trans-femoral amputees before and after special gait re-education. Prosthet Orthot Int 2002;26:101–12. [63] 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. [64] Segal AD, Orendurff MS, Klute GK, McDowell ML, Pecoraro JA, Shofer J, et al. Kinematic and kinetic comparisons of transfemoral amputee gait using C-Leg and Mauch SNS prosthetic knees. J Rehabil Res Dev 2006;43:857–70. [65] Han TR, Chung SG, Shin HI. Gait patterns of transtibial amputee patients walking indoors barefoot. Am J Phys Med Rehabil 2003;82:96–100. [66] Isakov E, Keren O, Benjuya N. Trans-tibial amputee gait: time-distance parameters and EMG activity. Prosthet Orthot Int 2000;24:216–20. [67] Powers CM, Torburn L, Perry J, Ayyappa E. Influence of prosthetic foot design on sound limb loading in adults with unilateral below-knee amputations. Arch Phys Med Rehabil 1994;75:825–9. [68] Nolan L, Lees A. The functional demands on the intact limb during walking for active trans-femoral and trans-tibial amputees. Prosthet Orthot Int 2000; 24:117–25. [69] Rusaw D, Ramstrand N. Sagittal plane position of the functional joint centre of prosthetic foot/ankle mechanisms. Clin Biomech 2010;25:713–20. [70] Seroussi RE, Gitter A, Czerniecki JM, Weaver K. Mechanical work adaptations of above-knee amputee ambulation. Arch Phys Med Rehabil 1996;77:1209–14. [71] Hansen AH, Meier MR, Sessoms PH, Childress DS. The effects of prosthetic foot roll-over shape arc length on the gait of trans-tibial prosthesis users. Prosthet Orthot Int 2006;30:286–99. [72] Sadeghi H, Allard P, Duhaime PM. Muscle power compensatory mechanisms in below-knee amputee gait. Am J Phys Med Rehabil 2001;80:25–32. [73] Gitter A, Czerniecki JM, DeGroot DM. Biomechanical analysis of the influence of prosthetic feet on below-knee amputee walking. Am J Phys Med Rehabil 1991;70:142–8. [74] Silverman AK, Fey NP, Portillo A, Walden JG, Bosker G, Neptune RR. Compensatory mechanisms in below-knee amputee gait in response to increasing steady-state walking speeds. Gait Posture 2008;28:602–9. [75] Arya AP, Lees A, Nirula HC, Klenerman L. A biomechanical comparison of the SACH, Seattle and Jaipur feet using ground reaction forces. Prosthet Orthot Int 1995;19:37–45. [76] Bae TS, Choi K, Hong D, Mun M. Dynamic analysis of above-knee amputee gait. Clin Biomech 2007;22:557–66. [77] Barnett C, Vanicek N, Polman R, Hancock A, Brown B, Smith L, et al. Kinematic gait adaptations in unilateral transtibial amputees during rehabilitation. Prosthet Orthot Int 2009;33:135–47. [78] Buckley JG, Spence WD, Solomonidis SE. Energy cost of walking: comparison of ‘‘intelligent prosthesis’’ with conventional mechanism. Arch Phys Med Rehabil 1997;78:330–3. [79] Chin T, Sawamura S, Shiba R. Effect of physical fitness on prosthetic ambulation in elderly amputees. Am J Phys Med Rehabil 2006;85:992–6. [80] Chin T, Machida K, Sawamura S, Shiba R, Oyabu H, Nagakura Y, et al. Comparison of different microprocessor controlled knee joints on the energy consumption during walking in trans-femoral amputees: intelligent knee prosthesis (IP) versus C-Leg. Prosthet Orthot Int 2006;30:73–80. [81] Chin T, Oyabu H, Maeda Y, Takase I, Machida K. Energy consumption during prosthetic walking and wheelchair locomotion by elderly hip disarticulation amputees. Am J Phys Med Rehabil 2009;88:399–403. [82] Cortes A, Viosca E, Hoyos JV, Prat J, Sanchez-Lacuesta J. Optimisation of the prescription for trans-tibial (TT) amputees. Prosthet Orthot Int 1997;21:168–74. [83] Culham EG, Peat M, Newell E. Below-knee amputation: a comparison of the effect of the SACH foot and single axis foot on electromyographic patterns during locomotion. Prosthet Orthot Int 1986;10:15–22. [84] Doane NE, Holt LE. A comparison of the SACH and single axis foot in the gait of unilateral below-knee amputees. Prosthet Orthot Int 1983;7:33–6.

[85] Gailey RS, Wenger MA, Raya M, Kirk N, Erbs K, Spyropoulos P, et al. Energy expenditure of trans-tibial amputees during ambulation at self-selected pace. Prosthet Orthot Int 1994;18:84–91. [86] Genin JJ, Bastien GJ, Franck B, Detrembleur C, Willems PA. Effect of speed on the energy cost of walking in unilateral traumatic lower limb amputees. Eur J Appl Physiol 2008;103:655–63. [87] Goh JC, Lee PV, Chong SY. Comparative study between patellar-tendon-bearing and pressure cast prosthetic sockets. J Rehabil Res Dev 2004;41:491–501. [88] Hsu MJ, Nielsen DH, Yack HJ, Shurr DG. Physiological measurements of walking and running in people with transtibial amputations with 3 different prostheses. J Orthop Sports Phys Ther 1999;29:526–33. [89] Hurley GRB, McKenney R, Robinson M, Zadravec M, Pierrynowski MR. The role of the contralateral limb in below-knee amputee gait. Prosthet Orthot Int 1990;14:33–42. [90] Isakov E, Susak Z, Becker E. Energy expenditure and cardiac response in above-knee amputees while using prostheses with open and locked knee mechanisms. Scand J Rehabil Med 1985;17:108–11. [91] Isakov E, Burger H, Krajnik J, Gregoric M, Marincek C. Knee muscle activity during ambulation of trans-tibial amputees. J Rehabil Med 2001;33: 196–9. [92] Jaegers SM, Arendzen JH, de Jongh HJ. An electromyographic study of the hip muscles of transfemoral amputees in walking. Clin Orthop Relat Res 1996;328:119–28. [93] Lacroix B, Janin F, Fouquet B, Grandmontagne M, Pastene J, Jehl JL, et al. Energy consumption of walking in young traumatic leg amputated patients upper the knee according to different types of knee protheses. Ann Phys Rehabil Med 1992;35:27–31. [94] Lee S, Hong J. The effect of prosthetic ankle mobility in the sagittal plane on the gait of transfemoral amputees wearing a stance phase controlled knee prosthesis. In: Proceedings of the Institution of Mechanical Engineers Part H. J Eng Med 2009; 223: 263–271. [95] Lemaire ED, Fisher FR, Robertson DG. Gait patterns of elderly men with transtibial amputations. Prosthet Orthot Int 1993;17:27–37. [96] Michaud SB, Gard SA, Childress DS. A preliminary investigation of pelvic obliquity patterns during gait in persons with transtibial and transfemoral amputation. J Rehabil Res Dev 2000;37:1–10. [97] Murray MP, Sepic SB, Gardner GM, Mollinger LA. Gait patterns of above-knee amputees using constant-friction knee components. Bull Prosthet Res 1980; 10:35–45. [98] Murray MP, Mollinger LA, Sepic SB, Gardner GM, Linder MT. Gait patterns in above-knee amputee patients: hydraulic swing control vs constant-friction knee components. Arch Phys Med Rehabil 1983;64:339–45. [99] Nyska M, Shabat S, Arya A, McCabe C, Linge K, Klenerman L. A comparative study of different below-knee prostheses by dynamic foot pressure analysis. Int J Rehabil Res 2002;25:341–4. [100] Pagliarulo MA, Waters R, Hislop HJ. Energy cost of walking of below-knee amputees having no vascular disease. Phys Ther 1979;59:538–43. [101] Pinzur MS, Perona P, Patwardhan A, Havey R. Loading of the contralateral foot in peripheral vascular insufficiency below-knee amputees. Foot Ankle 1991;11:368–71. [102] Pinzur MS, Gold J, Schwartz D, Gross N. Energy demands for walking in dysvascular amputees as related to the level of amputation. Orthopedics 1992;15:1033–6. [103] Royer TD, Wasilewski CA. Hip and knee frontal plane moments in persons with unilateral, trans-tibial amputation. Gait Posture 2006;23:303–6. [104] Segal AD, Orendurff MS, Czerniecki JM, Shofer JB, Klute GK. Transtibial amputee joint rotation moments during straight-line walking and a common turning task with and without a torsion adapter. J Rehabil Res Dev 2009;46:375–83. [105] Snyder RD, Powers CM, Fontaine C, Perry J. The effect of five prosthetic feet on the gait and loading of the sound limb in dysvascular below-knee amputees. J Rehabil Res Dev 1995;32:309–15. [106] Torburn L, Powers CM, Guiterrez R, Perry J. Energy expenditure during ambulation in dysvascular and traumatic below-knee amputees: a comparison of five prosthetic feet. J Rehabil Res Dev 1995;32:111–9. [107] Traballesi M, Porcacchia P, Averna T, Brunelli S. Energy cost of walking measurements in subjects with lower limb amputations: a comparison study between floor and treadmill test. Gait Posture 2008;27:70–5. [108] Van Jaarsveld HWL, Grootenboer HJ, De Vries J. Accelerations due to impact at heel strike using below-knee prosthesis. Prosthet Orthot Int 1990;14:63–6. [109] Zhang M, Lee WC. Quantifying the regional load-bearing ability of trans-tibial stumps. Prosthet Orthot Int 2006;30:25–34.