Table 2 Study characteristics
From: Therapeutic benefits of lower limb prostheses: a systematic review
Author | Participants | Prosthetic comparison | Familiarisation | Task | Outcome | Results |
|---|---|---|---|---|---|---|
Arifin et al. 2014a | 10 TTA • Age: 44.8 ± 13.5 yr • Gender: M = 10, F = 0 • Weight: 77 ± 17.9 kg • TSA: 7.1 ± 6.6 yr • MFCL: K2–K3 • AB-control group: Y/N • Cause of amputation: TR = 4, VA = 5, TU = 1 | Passive vs passive • SACH (P) • Talux (P) • Single axis (P) | 1 week | Standing Standing on BSS platform in 3 conditions (with rigid, compliant & unstable surface) for 20 s per condition | Biomechanical • Overall stability index • Anterior stability index • Posterior stability index • Medial stability index • Lateral stability index | • OSI, APSI, and MLSI indices were not affected by the interaction between prosthetic foot types and surface conditions • OSI ↑ using Talux ↔ SACH on foam surface ↔ firm and unstable support surface (p = 0.04) • Trend of stability indexes: lowest for SACH foot and highest for Talux foot in most of the conditions |
Arifin et al. 2014b | 10 TTA • Age: 44.8 ± 13.5 yr • Gender: M = 10, F = 0 • Weight: 77 ± 17.9 kg • TSA: 7.1 ± 6.6 yr • MFCL: K2–K3 • AB-control group: Y/N • Cause of amputation: TR = 4, VA = 5, TU = 1 | Passive vs passive • SACH (P) • Talux (P) • Single axis (P) | 2 weeks | Standing Standing on BSS platform in two conditions (eyes open & eyes closed) for 20 s per condition | Biomechanical • Overall stability index • Anterior stability index • Posterior stability index • Medial stability index Subjective • ABC-scale | • Control of postural steadiness unaffected by type of prostheses • MLSI > APSI for Talux in both eyes-opened and eyes-closed conditions (p = 0.034 and p = 0.017, respectively) • OSI, APSI and MLSI score > during eyes-closed ↔ eyes-opened condition for all foot types. Differences between the two conditions were only statistically significant in OSI (p = 0.018) and MLSI (p = 0.018) for SACH foot, as well as in OSI (p = 0.043) and APSI (p = 0.027) for Talux foot • ABC-scale: differences occurred between Talux and SACH (p = 0.043) as well as Talux and single axis foot (p = 0.028) |
Childers et al. 2018 | 5 TTA • Age: 44 ± 13.9 yr • Gender: – • Weight: 80.5 ± 13.9 kg • TSA: 11.2 ± 5.3 yr • MFCL: – • AB-control group: Y/N • Cause of amputation: – | Passive vs passive • Proflex (P) • Variflex (P) | 5 min | Treadmill slope walking 1 min of level, incline and decline walking at 1.1 m/s | Biomechanical • Foot angle • Prosthetic ankle and foot power • Whole body COM rate of energy change | • Range of motion ↑ with Pro-Flex foot • Energy return ↑ with Pro-Flex foot • Energy return from Pro-Flex foot ↓ ↔ sound limb ankle–foot system • Energy from Pro-Flex foot affected whole body COM mechanics • ↓ loading on sound limb = unclear |
D'Andrea et al. 2014 | 8 TTA • Age: 47 ± 8 yr • Gender: M = 8, F = 0 • Weight: 98.6 ± 9.7 kg • TSA: 19.4 ± 11.8 yr • MFCL: ≥ K3 • AB-control group: Y/N • Cause of amputation: TR = 8 | Active vs passive • Biom prototype (A) • Participants’ current prosthesis (P) | 1 session with Biom of at least 2 h | Level walking Walking 3 times at 0.75, 1.00, 1.25, 1.50, and 1.75 m/s along 10-m walkway | Biomechanical • Whole-body angular momentum | During the affected leg stance phase • Sagittal whole-body angular momentum ranges > passive prostheses ↔ active prosthesis at 1.25 m/s (ES = 0.25; CI = 0.039–0.047, 0.037–0.045; p = 0.032) and 1.50 m/s (ES = 0.22; CI = 0.034–0.042, 0.031–0.039; p = 0.032) During the unaffected leg stance phase: • Sagittal whole-body angular momentum ranges > passive prosthesis at 0.75 m/s (ES = 0.33; CI = 0.046–0.060, 0.042–0.054; p = 0.031) and 1.75 m/s (ES = 0.33; CI = 0.023–0.031, 0.019–0.027; p = 0.017) ↔ active prosthesis • No differences in frontal whole-body angular momentum ranges between prostheses. no differences in transverse H at any speed, except for 0.75 m/s, transverse H range > passive prosthesis ↔ active prosthesis (ES = 0.11; CI = 0.016–0.026, 0.015–0.025; p = 0.040) |
Darter et al. 2014 | 6 TTA • Age: 30 ± 4 yr • Gender: M = 5, F = 1 • Weight: 85.4 ± 16.9 kg • TSA: 2.8 ± 1.2 yr • MFCL: ≥ K2 • AB-control group: Y/N • Cause of amputation: – | Quasi-passive vs passive • Proprio (QP) • Participants’ current prosthesis (P) | 3 weeks | Treadmill slope walking Walking at 3 speeds (0.89, 1.11, and 1.34 m/s) at each of three slope conditions (− 5°, 0°, and 5°) | Physiological • VO2 Subjective • RPE | • EE for walking with the current foot was 13.5% > for slope descent ↔ Proprio (on-mode) (p < 0.05) and 10.3% more than with the Proprio (off-mode) (p < 0.05) • No differences were found for EE during level walking and slope ascent • Mean energy cost values ↓ (improved economy) as speed ↑ during slope descent and level grade walking • Prosthetic foot type = significant (p < 0.01) during slope descent → less-economical gait with current prosthesis ↔ Proprio devices [Proprio (on-mode) 14.0%, p < 0.01, Proprio (off-mode) 10.5%, p < 0.05] but no differences between Proprio (on-mode) and Proprio (off-mode) • Perceived difficulty of walking ↑ as walking speed ↑ with significant device effect for slope descent (p < 0.01). RPE values ↓ with the Proprio (on-mode) by an average of 2.2 on the 6–20 scale ↔ current prosthesis (p < 0.01) and 1.8 with the Proprio (off-mode) ↔ current prosthesis (p < 0.01) |
Davot et al. 2021 | 5 TTA • Age: 37.2 ± 15.2 yr • Gender: M = 4, F = 1 • Weight: 76.2 ± 12.2 kg • TSA: 3.4 ± 2.2 yr • MFCL: ≥ K2 • AB-control group: Y/N • Cause of amputation: – | Quasi-passive vs passive • Proprio (QP) • Meridium (QP) • Elan (QP) • Participants’ current prosthesis (P) | 2 weeks | Level walking + slope walking 3 walking conditions at SS speed: on level ground, on a 12% (7°) ramp ascent and on a 12% (7°) ramp descent of 6.2 m long | Biomechanical • ROM • Equilibrium point • Hysteresis (= net energy loss of the system, computed on the entire gait cycle) • Late stance energy released • Quasi-stiffness | • ROM = Elan lowest maximal dorsiflexion in ascent (9°) and maximal plantarflexion in descent (12°). Dorsiflexion differences Meridium ↔ Elan (p = 0.008) and ↔ ESR (p = 0.0027). In every situation, the highest ROM was observed with the Meridium (mean = 19.5° in descent, 20.5° on level ground, 22.6° in ascent) and the lowest ROM with the Elan (mean = 18.9° in descent, 18.9° on level ground and 13.9° in ascent) • Equilibrium point of current prosthesis was similar in the three conditions (no shift of the curve along the X axis). For the Elan, the equilibrium point was not shifted for the first characteristic pattern. For the proprio, a shift could be observed between level ground and ascent; for the Meridium, between level ground and descent + between level ground and ascent • Hysteresis = Proprio and the current prosthesis presented lowest hysteresis in all conditions. The Meridium hysteresis was 2–3 times higher ↔ other 3 feet (p = 0.001) • Elan: the energy released was the lowest in descent and the highest in ascent. On level ground, it was ↑ ↔ descent and↓ ↔ ascent. Meridium had the lowest energy for propulsion • Quasi-stiffness = no differences between devices |
De Asha et al. 2014 | 11 TTA 7 TFA • Age: 45 ± 12.4 yr • Gender: – • Weight:  • TTA: 84.5 ± 17.0 kg  • TFA: 86.3 ± 15.3 kg • TSA: 14.5 ± 14.4 yr • MFCL: ≥ K3 • AB-control group: Y/N • Cause of amputation: TR = 16, TU = 3 | Passive vs passive • Echelon (P) • Participants’ current prosthesis (P) | No familiarisation | Level walking Walking 8 m–walkway at SS speed | Biomechanical • COM • COP • Swing time • Stance time • Inter-limb asymmetry • Step length Performance • Speed | • Walking speed = ↑ with Echelon and ↑ for TTA ↔ TFA • Aggregate negative CoP displacement was ↓ with Echelon. The CoP passed anterior to the shank earlier in stance with the Echelon • Instantaneous COM speed at intact-limb TO was unchanged across foot conditions but instantaneous COM speed minimum during the subsequent prosthetic-limb single support phase was ↑ using the Echelon. As a result, there was less slowing of COM speed (walking speed) during prosthetic-limb single support for both groups when using the Echelon ↔ current prosthesis. Peak COM speed during prosthetic limb stance was unchanged across foot conditions. All instantaneous COM speed values were ↑ for TTA ↔ TFA (p ≤ 0.045) • Swing time was longer for the prosthetic limb ↔ intact-limb and the differences between limbs was ↑ for TFA ↔ TTA • Stance time ↑ intact-limb ↔ prosthetic-limb & differences between limbs ↑ TFA ↔ TTA • Step length ↑ prosthetic limb ↔ intact limb • There were no effects of foot condition (p = 0.84) or group (p = 0.063) on cadence. There were no effects of foot condition on inter-limb asymmetry in swing time, stance time or step length. Swing and stance time inter-limb asymmetry were ↑ TFA ↔ TTA but there was no group effect on step length inter-limb symmetry |
De Pauw et al. 2018 | 6 TTA 6 TFA • Age:  • TTA: 54 ± 14 yr  • TFA: 53 ± 14 yr • Gender: M = 11, F = 1 • Weight:  • TTA: 80 ± 13 kg  • TFA: 89 ± 16 kg • TSA: – • MFCL: K2–K4 • AB-control group: Y/N • Cause of amputation: – | Quasi-Passive vs passive • AMP-foot 4.0 (QP) • Participants’ current prosthesis (P) | No familiarisation | Treadmill walking 6-min treadmill walking at SS speed, 2-min slow and 2 min fast walking | Physiological • HR • MV • VO2 • VCO2 • RQ • METS Subjective • QUEST • RPE | • At normal speed, no significant differences between groups for MV, VO2, VCO2, RQ, and METS. In TTA, RQ ↑ with AMPfoot ↔ current prosthesis (p = 0.017). At other walking speeds, no differences were found • HR = At fast speed, no differences. At slow speed, HR ↑ in TFA and TTA with AMPFoot ↔ current prosthetic device. In TFA, HR ↑ with current prosthesis and AMP-foot ↔ able-bodied individuals (p = 0.043 and 0.008, respectively). At other speeds, no significant differences were revealed • At normal speed, RPE levels ↑ in TFA and TTA with current prosthesis and AMPFoot ↔ able-bodied individuals at the first (p ≤ 0.016 and p ≤ 0.004) and sixth minute (p ≤ 0.003 and p ≤ 0.004, respectively). No differences were observed between TFA and TTA when wearing the current prosthesis. RPE ↑ with AMPFoot in TFA ↔ TTA (p = 0.027). At slow and fast walking speeds, RPE ↑ for TFA and TTA ↔ able-bodied individuals for current prosthesis and AMPfoot (slow speed: p ≤ 0.004 and p ≤ 0.003, respectively; fast speed: p ≤ 0.005 and p ≤ 0.009, respectively). No differences in RPE were observed between TFA and TTA. In addition, at fast speed RPE ↓ in TTA ↔ TFA with AMPFoot (p = 0.042). In TFA, RPE levels were ↑ with AMPFoot ↔ current prosthesis at 1 and 6 min (p = 0.027 and 0.042, respectively) • QUEST = 10 participants responded positive regarding buying the device if it was available on the market. Only in TFA, significant lower values for satisfaction and weight of AMPFoot ↔ current prosthesis were observed (p = 0.038 and 0.042, respectively) |
De Pauw et al. 2019 | 6 TTA 6 TFA • Age:  • TTA: 54 ± 14 yr  • TFA: 53 ± 14 yr • Gender: M = 11, F = 1 • Weight:  • TTA: 80 ± 13 kg  • TFA: 89 ± 16 kg • TSA: – • MFCL: K2–K4 • AB-control group: Y/N • Cause of amputation: – | Quasi-Passive vs passive • AMP-foot 4.0 (QP) • Participants’ current prosthesis (P) | No familiarisation | Treadmill walking Sustained Attention to Response Task, 6-min walking at SS speed + sustained attention to response task, 2-min walking at SS speed | Physiological • MRCP Performance • Dual-task accuracy | • Dual-task walking: reaction times ↑ for TFA with AMPfoot ↔ AB individuals (p = 0.020). During walking with AMPfoot significant accuracy differences of the no-go stimuli at the middle and end part of the cognitive task were observed • MRCP: no differences for MRCP amplitude and latency measures at electrode Cz between AB individuals and TTA walking with the current or novel prosthetic device. TFA walking with AMPfoot did not exhibit MRCPs, but TFA walking with the current prosthesis showed MRCPs at different electrode locations. No differences in activity of the brain sources of the different MRCP peaks were observed when TTA walked with the current and novel prosthetic device. Additionally, no significant differences were observed when TTA walked with the current prosthetic device ↔ AB individuals. On the other hand, ↔ AB individuals TTA wearing the AMPfoot showed ↑ activity of brain sources at the first positive deflection |
De Pauw et al. 2020 | 6 TTA 6 TFA • Age:  • TTA: 54 ± 14 yr  • TFA: 53 ± 14 yr • Gender: M = 11, F = 1 • Weight:  • TTA: 80 ± 13 kg  • TFA: 89 ± 16 kg • TSA: – • MFCL: K2–K4 • AB-control group: Y/N • Cause of amputation: – | Quasi-Passive vs passive • AMP-foot 4.0 (QP) • Participants’ current prosthesis (P) | No familiarisation | Treadmill walking 2-min walking at SS speed, 2 min at slow (− 25% self-selected) and 2 min at fast (+ 25% self-selected) speeds. 1 min rest in between tasks | Biomechanical • LE joint angles • LE angular velocities • Stride length • Step width • Maximum GRF Performance • Speed | • TFA did not benefit from walking with the novel prosthesis • TTA walking at slow and normal speed with AMPfoot 4.0 → beneficial effects at the level of the ankle and knee • No differences between walking with the current prostheses and AMPfoot 4.0 with respect to force platform data |
Delussu et al. 2016 | 20 TTA • Age: 66.6 ± 6.7 yr • Gender: M = 17, F = 3 • Weight: 78.5 ± 13.2 kg • TSA: – • MFCL: K1–K2 • AB-control group: Y/N • Cause of amputation: TR = 6, VA = 13, TU = 1 | Passive vs passive • 1M10 (P) • SACH (P) | 30 days | Level walking 6MWT along 30-m-long linear course | Physiological • MV • VO2 • RER • HR • REI • Energy cost Performance • SS speed Subjective • RPE • Satisfaction | • No differences for MV, VO2, RER, HR and REI using SACH or 1M10 • Energy cost, SS speed, RPE score and SATPRO improved with the 1M10 compared to the SACH |
Esposito et al. 2014a | 10 TTA • Age: 30.2 ± 5.3 yr • Gender: M = 9, F = 1 • Weight: 95.8 ± 7.3 kg • TSA: – • MFCL: ≥ K3 • AB-control group: Y/N • Cause of amputation: TR = 10 | Active vs passive • BiOM (A) • Participants’ current prosthesis (P) | 3 weeks | Level walking Walked at 3 controlled speeds | Biomechanical • GRF • Knee joint moments • Loading rate Subjective • Rating of ambulation ability Performance • Speed | • The active prosthesis did not ↓ sound limb’s peak adduction moment or its impulse, but did ↓ the external flexor moment, peak vertical force and loading rate as speed ↑ • The active prosthesis ↓ loading rate from AB controls. The sound limb did not display a greater risk for knee osteoarthritis ↔ intact limb or ↔ AB controls in either device • Self-selected walking speeds were not significantly different between prosthesis conditions • Subject rating of ambulation ability using the PEQ was high in both devices |
Esposito et al. 2014b | 6 TTA • Age: 23 ± 5 yr • Gender: M = 5, F = 1 • Weight: 91.4 ± 12.1 kg • TSA: – • MFCL: – • AB-control group: Y/N • Cause of amputation: TR = 6 | Active vs passive • BiOM (A) • Participants’ current prosthesis (P) | 3 weeks | Level walking + slope walking Walking at standardized speed (± 5) over level ground and inclined walkway + 6MWT on treadmill until steady state metabolic rate was achieved for both level and inclined walking | Biomechanical • Step-to-step transition work • LE joint angles • LE joint moments • LE joint power Physiological • Metabolic rate | • Kinetics & kinematics: during level walking, the BiOM ↑ peak ankle plantarflexion angles and powers at push-off ↔ current prosthesis and ↓ peak internal plantar flexor moments. During inclined walking, peak angles and powers ↑ in the BiOM. Peak ankle plantarflexion angles and powers ↓ in current prosthesis ↔ AB controls over level ground, and angles, moments, and powers ↓ on the incline. The BiOM normalized the peak ankle plantarflexion angles on level ground, but moments remained ↓ and powers ↑ ↔ AB controls during level ground and inclined walking • Metabolic rate: during level walking, VO2 was ↓ 16% with BiOM ↔ current prosthesis. ↑ 9% metabolic rates with current prosthesis ↔ able-bodied individuals, but BiOM normalized metabolic rates. On the incline, metabolic rates were not different between BiOM ↔ AB controls or between BiOM ↔ current prosthesis • Step-to-step transition = During level walking, the net step-to-step transition work prosthetic limb ↑ 63% with active ↔ current prosthesis. Active prosthetic trailing limb step-to-step transition work ↑28% ↔ AB controls, while current prosthesis ↓ 22% ↔ AB controls • Net leading limb work during step-to-step transitions inclined walking ↑ 53% with active prosthesis ↔ current. Net trailing limb step-to-step transition work did not differ between AB controls and TTA |
Ferris et al. 2012 | 11 TTA • Age: 29.8 ± 5.3 yr • Gender: M = 10, F = 1 • Weight: 95 ± 7.3 kg • TSA: – • MFCL: – • AB-control group: Y/N • Cause of amputation: TR = 11 | Active vs passive • BiOM (A) • Participants’ current prosthesis (P) | 3 weeks | Level walking + agility and mobility Walking at SS and controlled speed + T-test, Four square step test, Hill and stair Assessment Tests | Biomechanical • GRF • Symmetry • Stance time • Swing time • Cadence • Step length • Stride length • LE joint angles • LE joint moments • LE joint powers Performance • Agility and mobility • Speed Subjective • User satisfaction | • Ankle ROM 30% > active prosthesis ↔ AB current, both < ROM ↔ AB control and intact limbs • Peak ankle power ↓ 40% current prosthesis ↔ active • Peak knee power ↑ 35% active prosthesis ↔ AB control ↑ 125% current → active absorbing 2 × the peak knee power observed in AB control and intact limbs • Peak hip power ↑ 45% active prosthesis ↔ intact limb • Walking speed ↑ active prosthesis ↔ current (not significant) ↔ AB control group • User satisfaction scores → preference for active over current prosthesis |
Gailey et al. 2012 | 10 TTA • Age:  • Group 1: 60.6 ± 2.3 yr  • Group 2: 51 ± 5.8 yr • Gender: M = 9, F = 1 • Weight:  • Group 1: 105.5 ± 6.4 kg  • Group 2: 92.1 ± 9.7 kg • TSA:  • Group 1: 2.90 ± 1.8 yr  • Group 2: 16.1 ± 17.6 yr • MFCL: K2–K3 • AB-control group: Y/N • Cause of amputation: TR = 5, VA = 5 | Quasi-Passive vs passive • SACH (P) • SAFE foot (P) • Talux (P) • Proprio (QP) • Participants’ current prosthesis (P) | 2 weeks | Level walking Performing LCI-5, 6MWT | Performance • LCI-5 • 6MWT • Steps/day • AMPRO • Hours of daily activity Subjective • PEQ-13 | • PEQ-13, LCI-5, 6MWT, or step activity monitor: no differences between devices • AMPPRO: differences following training with the existing prosthesis in group 1 and between selected feet from baseline testing (p ≤ 0.05). Sign differences were found between group 1 and group 2 (p ≤ 0.05) in the AMPPRO and 6MWT when using the Proprio foot • Self-report measures were unable to detect differences between prosthetic feet |
Gardinier et al. 2017 | 10 TTA • Age: 46.6 ± 15 yr • Gender: M = 10, F = 0 • Weight: 93.2 ± 17.9 kg • TSA: – • MFCL: K3–K4 • AB-control group: Y/N • Cause of amputation: TR = 9, VA = 1 | Active vs passive • BiOM (A) • Participants’ current prosthesis (P) | 8 min | Treadmill walking Walking along 8-m walkway at at SS speed and controlled speed + walking 8-min on treadmill until steady-state energy expenditure is reached | Physiological • Energetic cost • VO2 • Cost of transport Performance • Speed | • No sign differences in VO2 (2.9% difference; P = 0.606, d = 0.26) using the active ankle ↔ current prosthesis • No sign differences in cost of transport (1% difference; P = 0.652, d = 0.23) using the active ankle ↔ current prosthesis • No sign differences in preferred walking speed (1% difference; P = 0.147, d = 0.76) using the active ankle ↔ current prosthesis • Participants classified as having the highest function (MFCL = K4) were sign more likely to exhibit energy cost savings ↔ those classified as having lower function (K3; P = 0.014, d = 2.36) |
Gates et al. 2013 | 11 TTA • Age: 30 ± 5 yr • Gender: M = 10, F = 1 • Weight: 95 ± 7.3 kg • TSA: – • MFCL: – • AB-control group: Y/N • Cause of amputation: TR = 11 | Active vs passive • BiOM (A) • Participants’ current prosthesis (P) | 3 weeks | Walking (rocky surface) Walking across a loose rock surface at three controlled speeds; The rock surface was a 4.2-m long by 1.2-m wide by 10-cm deep pit filled with loose river rocks from a major hardware store | Biomechanical • COM • Minimum margin of stability Performance • Speed | • Walking speed ↑ 10% using active prostheses (1.16 m/s) ↔ current (1.05 m/s; p = 0.031) • Ankle plantarflexion ↑ (p < 0.001), knee flexion ↓ (p = 0.045) on their prosthetic limb using active prostheses ↔ current • Other kinematics of the knee and hip = nearly identical between devices • Medial–lateral motion COM ↓ using active prosthesis ↔ current (p = 0.020), • Medial–lateral margins of stability = no differences between devices (p = 0.662) |
Grabowski et al. 2013 | 7 TTA • Age: 45 ± 6 yr • Gender: – • Weight: 99.5 ± 10.2 kg • TSA: 21.1 ± 11.3 yr • MFCL: ≥ K3 • AB-control group: Y/N • Cause of amputation: – | Active vs passive • Active prototype (A) • Participants’ current prosthesis (P) | 2 h | Level walking Walking at 0.75, 1.00, 1.25, 1.50, and 1.75 m/s along 10 m-walkway | Biomechanical • GRF • Knee joint moments • Loading rates | • Active prosthesis ↓ unaffected leg peak resultant forces by 2–11% at 0.75–1.50 m/s ↔ current • Active prosthesis ↓ first peak knee external adduction moments by 21 and 12% at 1.50 and 1.75 m/s ↔ current • Loading rates = no differences between prostheses |
Graham et al. 2007 | 6 TFA • Age: 40.3 ± 6.3 yr • Gender: – • Weight: 88.5 ± 9.4 kg • TSA: – • MFCL: – • AB-control group: Y/N • Cause of amputation: – | Passive vs passive • VariFlex (P) • Multiflex (P) | 3–6 weeks | Level walking Timed walking test along 207.3-m oval circuit including outdoor and indoor walking with the resultant variations of camber and surface | Biomechanical • Step-length ratio • Stance time • Vertical GRF • Ankle dorsiflexion • Knee flexion • Hip flexion/extension • Transverse pelvic rotation • Ankle power • Hip power Performance • Speed Subjective • Prosthetic socket fit comfort score | • VariFlex speed ↑ + ↑ equal step lengths at fast speed ↔ multiflex • VariFlex ↑ peak ankle dorsiflexion at push-off on the prosthetic side (18.3° + − 4.73°, P < 0.001) + ↑ 3 × power from the prosthetic ankle at push-off (1.13 + − 0.22 W/kg, P < 0.001) ↔ multiflex • No sign differences in temporal symmetry or loading of the prosthetic limb, in the timed walking test with each foot, or in the comfort score |
Graham et al. 2008 | 6 TFA • Age: 40.3 ± 6.3 yr • Gender: – • Weight: 88.5 ± 9.4 kg • TSA: – • MFCL: – • AB-control group: Y/N • Cause of amputation: – | Passive vs passive • VariFlex (P) • Mutliflex (P) | 3–10 weeks | Treadmill walking 2-min walking tests; Speeds increases every 2 min starting at 0.83 m/s then 0.94 m/s, 1.1 m/s, 1.25 m/s, 1.39 m/s, 1.53 m/s, 1.67 m/s and 1.81 m/s until subjects find the treadmill speed too fast | Physiological • Mean VO2 Performance • Speed | • VariFlex ↓ mean VO2 ↔ Multiflex at all speeds, although the differences were only sign at speeds of 0.83 and 1.1 m/s. The estimated differences across all speeds was 3.54 mL/kg min |
Heitzmann et al. 2018 | 11 TTA • Age: 37.9 ± 12.3 yr • Gender: M = 9, F = 2 • Weight: 81.1 ± 17.4 kg • TSA: 11.9 ± 10.6 yr • MFCL: K3–K4 • AB-control group: Y/N • Cause of amputation: TR = 4, VA = 2, TU = 4, O = 1 | Passive vs passive • Proflex pivot (P) • Participants’ current prosthesis (P) | 30–45 min | Level walking Walking along 10 m-walkway at SS speed | Biomechanical • Ankle ROM • Peak ankle moment, peak ankle power • Peak external knee varus moment • Peak vertical GRF Performance • Speed | • Proflex ↓ walking speed (1.33 ± 0.16 m/s) ↔ current prosthesis (1.39 ± 0.17 m/s). AB controls did not walk sign faster ↔ TTA • Proflex ↑ prosthetic ankle ROM by 12.5° ↔ current prosthesis • Angle ROM and peak dorsiflexion of 18.8° ↔ current prosthesis + no sign differences ↔ AB controls • Peak external ankle dorsi-flexion moment < AB controls (proflex: 28%, current prosthesis: 36% + no sign differences in peak external ankle dorsi-flexion moment between prosthetic feet • Peak positive ankle power < current prosthesis (by 66%) and Proflex (by 33%) ↔ AB controls + Proflex ↑ peak ankle power ↔ current prosthesis • External knee varus moment and the peak vertical GRF for Proflex ↓ ↔ current prosthesis & AB controls |
Houdijk et al. 2018 | 15 TTA • Age: 58.8 ± 11.1 yr • Gender: – • Weight: 86 ± 12.6 kg • TSA: – • MFCL: K3 • AB-control group: Y/N • Cause of amputation: TR = 12 | Passive vs passive • SACH (P) • Variflex (P) | 1 day | Level walking Walking along 10-m walkway at a fixed speed | Biomechanical • Work • Vertical COM • Step length intact • Step length symm • Backward Margin of stability | • Push-off work ↑ Variflex ↔ SACH • COM speed at toe-off ↑ Variflex ↔ SACH • Intact step length and step length symmetry ↑ without ↓ the backward margin of stability Variflex ↔ SACH |
Hsu et al. 2006 | 8 TTA • Age: 36 ± 15 yr • Gender: M = 8, F = 0 • Weight: 81.7 ± 9.6 kg • TSA: – • MFCL: – • AB-control group: Y/N • Cause of amputation: – | Passive vs passive • C-walk (P) • Flex foot (P) • SACH (P) | 4 weeks | Treadmill walking 2 min walking at SS speed | Physiological • Gait efficiency • VO2 • %APMHR Performance • Steps/day Subjective • RPE | • C-Walk had a trend of ↑ physiologic responses ↔ SACH • Flex foot: no sign differences in EE and gait efficiency, but ↓ %APMHR & RPE ↔ C-Walk and SACH |
Johnson et al. 2014 | 21 TTA • Age: 48.2 ± 12.8 yr • Gender: M = 18, F = 3 • Weight: 87.4 ± 13.2 kg • TSA: 8.8 ± 14 yr • MFCL: – • AB-control group: Y/N • Cause of amputation: – | Passive vs passive • Echelon (P) • Participants’ current prosthesis (P) | 45 min | Level walking Walking along 8 m-walkway | Biomechanical • MTC • LE joint angles • Prosthetic limb hip-hiking Performance • Speed | • Mean MTC ↑ on both limbs with Echelon ↔ current prosthesis (p = 0.03) • Walking speed ↑ Echelon ↔ current prosthesis (p = 0.001) + ≈ ↑ swing-limb hip flexion on the prosthetic side Echelon ↔ current prosthesis (p = 0.04) • Variability in MTC ↑ on the prosthetic side with Echelon (p = 0.03), but this did not ↑ risk of tripping |
Prakash et al. 2020 | 15 TTA • Age: 33.3 ± 5.5 yr • Gender: – • Weight: – • TSA: – • MFCL: – • AB-control group: Y/N • Cause of amputation: – | Passive vs passive • SACH (P) • Passive prototype ESR (P) | 15 min | Level walking 10‑m walk test + 5 min of strolling at SS speed | Biomechanical • Stride length • Cadence Physiological • PCI Performance • Speed | • Stride length, cadence, speed, and PCI ↓ SACH ↔ current prosthesis |
Paradisi et al. 2015 | 20 TTA • Age: 66.7 ± 6.7 yr • Gender: M = 17, F = 3 • Weight: 78.7 ± 13.2 kg • TSA: 9.8 ± 13.5 yr • MFCL: – • AB-control group: Y/N • Cause of amputation: TR = 6, VA = 13, O = 1 | Passive vs passive • 1M10 (P) • SACH (P) | 4 weeks | Level walking, slope walking and stair climbing Performing 6MWT, LCI-5, HAI, SAI, BBS | Performance • Score on BBS, LCI-5, HAI, SAI • Time • Speed • Upright Gait Stability Subjective • PEQ | • Walking speed ↑ 1M10 ↔ SACH (p < 0.05) maintaining the same upright gait stability • BBS, LCI-5, and SAI times and 4 of 9 subscales of the PEQ ↑ 1M10 ↔ SACH |
Rábago et al. 2016 | 10 TTA • Age: 30.2 ± 5.3 yr • Gender: M = 9, F = 1 • Weight: 96.1 ± 6.8 kg • TSA: – • MFCL: – • AB-control group: Y/N • Cause of amputation: – | Active vs passive • BiOM (A) • Participants’ current prosthesis (P) | 3 weeks | Slope walking walking along 5 m long, 5˚ sloped ramp at controlled speed | Biomechanical • GRF • Stance time • Step length • Stride length • Swing time • LE joint angles • LE joint moments and powers • Second vertical peak • Braking • Propulsion Performance • Speed | • During slope ascent, the BiOM ↑ prosthetic ankle plantarflexion and push-off power generation ↔ current prosthesis + matched AB controls more closely • Similar deviations and compensations between both feet • Transitioning off the prosthetic limb → ↑ ankle plantarflexion and push-off power with BiOM → ↓ intact limb knee extensor power production → ↓ demand on the intact limb knee ↔ current prosthesis |
Riveras et al. 2020 | 13 TTA • Age: 38.2 ± 13.2 yr • Gender: M = 10, F = 3 • Weight: 75.1 ± 15.4 kg • TSA: 10.8 ± 13.1 yr • MFCL: – • AB-control group: Y/N • Cause of amputation: TR = 10, VA = 2, O = 1 | Passive vs quasi-passive • Esprit (P) • Echelon (QP) • Elan (QP) | 1 h | Slope walking walking along 6 m, 5° inclination ramp at SS speed | Biomechanical • Tripping probability • Coefficient of variation • Minimum toe clearance | • MTC median values for ascending (P ≤ 0.001, W = 0.58) and descending the ramp (P = 0.003, W = 0.47) in the prosthetic limb ↑ Elan ↔ Esprit and Echelon • CV ↓ on the prosthetic limb for descending the ramp (P = 0.014, W = 0.45) using the Echelon and Elan ↔ Esprit • Elan = Lowest TP for the prosthetic leg in three conditions evaluated • On the sound limb results showed the median MTC was ↑ (P = 0.009, W = 0.43) and CV ↓ (P = 0.005, W = 0.41) during ascent using Echelon and Elan ↔ Esprit |
Schmalz et al. 2019 | 4 TTA • Age: 56 ± 12 yr • Gender: M = 4, F = 0 • Weight: 79 ± 8.0 kg • TSA: – • MFCL: K3 – K4 • AB-control group: Y/N • Cause of amputation: TR = 3, VA = 1 | Passive vs quasi-passive • Meridium (QP) • Participants’ current prosthesis (P) | 2 weeks | Slope walking Walking along circuit of 3 m downhill walkway (10° inclination) followed by specific uphill and downhill elements with opposite inclination angles of 10 | Biomechanical • GRF • LE joint moments • LE joint angles | • Meridium ↑ ankle adaptation to the abruptly changing inclination, reflected by a ↑ stance phase dorsiflexion ≈ to AB controls ↔ current prosthesis • Peak value of the knee extension moment on the prosthetic side was ↑ with current prosthesis, whereas it was almost normal with Meridium (current prosthesis: 0.71 ± 0.13 Nm/kg, Meridium: 0.42 ± 0.12 Nm/kg, NA: 0.36 ± 0.07 Nm/kg, p < 0.05 and p < 0.01) • External knee adduction moment was ↓ for TTA and did not show differences between prostheses |
Segal et al. 2015 | 7 TTA • Age: 52.3 ± 12 yr • Gender: – • Weight: 80.9 ± 9.9 kg • TSA: – • MFCL: – • AB-control group: Y/N • Cause of amputation: TR = 7 | Passive vs quasi-passive • Participants’ current prosthesis (P) • Lightfoot2 (P) • Prototype: Controlled Energy Storage and Return prosthetic foot (QP) | 5 min | Level walking Walking on a treadmill at the target speed of 1.14 m/s for 10 min, until they reached steady state. + walking along a 10 m-walkway at same speed | Biomechanical • GRF • COM • LE joint powers • Work during gait Physiological • VO2 | • ↑ energy storage during early stance, ↑ prosthetic foot peak push-off power and work, ↑ prosthetic limb COM push-off work and ↓ intact limb COM collision work with Controlled Energy Storage and Return prosthetic foot ↔ Lightfoot2 and current prosthesis • Biological contribution of the positive COM work for Controlled Energy Storage and Return prosthetic foot was ↓ ↔ Lightfoot2 and current prosthesis • Net metabolic cost for Controlled Energy Storage and Return prosthetic foot did not change comp ↔ Lightfoot2 and ↑ ↔ current prosthesis |
Struckov et al. 2016 | 9 TTA • Age: 41.2 ± 12.9 yr • Gender: M = 9, F = 0 • Weight: 74.1 ± 15.7 kg • TSA: – • MFCL: K3 • AB-control group: Y/N • Cause of amputation: – | Passive vs quasi-passive • Elan (QP) • Epirus (P) | 20 min | Slope walking Ramp descent at slow and customary speed | Biomechanical • Residual-knee loading, response flexion • Single-support minimum flexion • Time to foot flat • CoP • Prosthetic-limb shank mean angular velocity during single-support • Single-support residual-knee moment impulse • Single-support negative mechanical work at the residual hip and knee joints • Unified deformable segment | • Foot-flat was attained fastest with the Epirus and second fastest with the Elan (P < 0.001) • Prosthetic shank single-support mean rotation speed ↓ (p = 0.006), flexion (P < 0.001) ↓, negative work done at the residual knee (P = 0.08) ↓, and negative work done by the ankle–foot ↑ (P < 0.001) with Elan ↔ Epirus and Elan in off-mode |
Underwood et al. 2012 | 11 TTA • Age: 42.5 ± 13.5 • Gender: M = 8, F = 3 • Weight: 80.3 ± 14.3 kg • TSA: 11.1 ± 13.3 yr • MFCL: – • AB-control group: Y/N • Cause of amputation: – | Passive vs passive • FlexWalk (P) • SAFE FOOT 2 (P) | 30 min | Level walking Walking along 10 m-walkway at SS speed | Biomechanical • LE peak joint moments and power • Perceived stability and mobility | • The majority of the kinetic differences that occurred due to the changing of prosthetic foot type were limited to ankle joint variables in the sagittal plane with ↑ peak moments and power during propulsion for the Flex foot ↔ SAFE foot • Effects were also found at joints proximal to the prosthesis (e.g., knee) and differences were also found in the kinetics of the sound limb |
Wezenberg et al. 2014 | 15 TTA • Age: 55.8 ± 11.1 yr • Gender: M = 15, F = 0 • Weight: 86 ± 12.6 kg • TSA: – • MFCL: – • AB-control group: Y/N • Cause of amputation: TR = 15 | Passive vs passive • SACH (1D10) (P) • Variflex (P) | 1 day | Level walking Walking along 10 m-walkway at SS speed | Biomechanical • GRF • COM mechanical work • Work at push-off • COP • Step length • Symmetry Performance • Speed | • Positive mechanical work COM performed by the trailing prosthetic limb was ↑ (33%, p = 0.01) and the negative work performed by the leading intact limb ↓ (13%, p = 0.04) with Variflex ↔ SACH foot • ↓ step-to-step transition cost & ↑ mechanical push-off power and extended forward progression of the COP with Variflex ↔ SACH |
Yang et al. 2017 | 10 TTA • Age: 63.8 ± 2.5 yr • Gender: M = 10, F = 0 • Weight: – • TSA: 3.1 ± 0.8 yr • MFCL: K2–K3 • AB-control group: Y/N • Cause of amputation: – | Passive vs passive • 1C30 Trias (P) • 1C60 Triton (P) | 1 week | Level walking Walking along 10 m-walkway at SS speed | Biomechanical • Cadence • Step width • Step length • Stance and swing phase ratio • LE joint angles • Ankle plantarflexion moment at end of stance Performance • Speed | • Cadence asymmetry with Trias was observed. Ankle plantarflexion at the end of stance and ankle supination at the onset of pre-swing ↓ with both prosthetic feet ↔ intact side. Other spatiotemporal, kinematic, and kinetic data showed no sign differences in a side-to-side comparison • In a comparison between the two prosthetics, stance and swing ratio and ankle dorsiflexion through mid-stance was closer to normal with Triton ↔ Trias. Other spatiotemporal, kinematic, and kinetic data showed no statistically sign differences between prosthetics |