the AB group (
p ≤ 0.001), but showed low deviation preva-
lence in the TTA group.
Hip
The greatest prevalence of deviations at the hip of the
intact limb was with abduction (56.3%). All sagittal plane
hip kinematic measures of the prosthetic limb had low
deviation prevalence (12.5
–18.8%), but unlike the knee of
the prosthetic limb, the measures are skewed in a single
direction relative to the NRR. Moderate deviation prevalence
was observed for intact limb hip extension during preswing
and
flexion during terminal swing, however, only the latter
TABLE III.
Prevalence (%) of Individuals within the TTA Group with Kinematic, Kinetic, or Temporal Spatial Deviations of the
Prosthetic or Intact Limbs as Identi
fied Using an AB NRR for Each Measure. “High” (>50%), Moderate (25–49%), and Low (<25%)
Prevalence is Indicated for Each Measure. Arrows Represent a Signi
ficant Increase (↑) or Decrease (↓) in the TTA Mean Group Measure
Relative to the AB Group and Correspond to Differences Presented in Table II. For Each Peak Kinematic and Kinetic Measure,
the Timing of the Peak in the Gait Cycle is identi
fied
Deviation Prevalence (%)
Prosthetic
Intact
Deviation Prevalence (%)
Prosthetic
Intact
Ankle Angle
Ankle Moment
Plantar
flexion: LR
18.8%
6.3%
Dorsiflexion: LR
↑
12.5%
↑
12.5%
Dorsi
flexion: TSt
6.3%
0.0%
Plantarflexion: TSt
6.3%
0.0%
Plantar
flexion: ISw
↓
100.0%
12.5%
Ankle Powers
Sagittal ROM
↓
81.3%
6.3%
Absorption: LR
6.3%
↑
43.8%
Absorption: TSt
12.5%
12.5%
Generation: TSt
↓
50.0%
12.5%
Knee Angle
Knee Moment
Flexion: IC
18.8%
6.3%
Flexion: LR
↓
6.3%
↑
6.3%
Flexion: LR
25.0%
6.3%
Extension: MSt
↓
0.0%
0.0%
Extension: TSt
18.8%
0.0%
Flexion: TSt
12.5%
0.0%
Flexion: MSw
31.3%
↓
12.5%
Extension: TSt
6.3%
0.0%
Sagittal ROM
↓
68.8%
↓
18.8%
Varus: LR
↓
18.8%
6.3%
Valgus: LR
12.5%
12.5%
Knee Powers
Generation: MSt
↓
87.5%
12.5%
Absorption: LR
↓
0.0%
0.0%
Generation: TSt
↓
0.0%
0.0%
Absorption: TSt
0.0%
6.3%
Hip Angle
Hip Moment
Flexion: LR
12.5%
18.8%
Extension: LR
↓
0.0%
↑
6.3%
Extension: PSw
18.8%
31.3%
Flexion: TSt
6.3%
↓
6.3%
Flexion: TSw
12.5%
↓
31.3%
Extension: Sw
31.3%
↑
18.8%
Sagittal ROM
18.8%
0.0%
Abductor
6.3%
31.3%
Adduction
12.5%
0.0%
Adductor
↓
12.5%
0.0%
Abduction
31.3%
↓
56.3%
Hip Powers
Generation: MSt
↑
0.0%
6.3%
Absorption: TSt
12.5%
12.5%
Generation: TSt
6.3%
6.3%
Pelvic Angle
Trunk-Pelvic Angle
Anterior Tilt
↓
31.3%
↓
31.3%
Sagittal ROM
37.5%
N/A
Posterior Tilt
31.3%
31.3%
Frontal ROM
↓
6.3%
N/A
Sagittal ROM
↑
25.0%
↑
25.0%
Transverse ROM
6.3%
N/A
Contralateral Drop
↓
37.5%
0.0%
Trunk-Lab Angle
Contralateral Elevation
0.0%
↓
12.5%
Sagittal ROM
0.0%
N/A
Frontal ROM
↓
0.0%
↓
0.0%
Frontal ROM
25.0%
N/A
Hip Forward
6.3%
0.0%
Transverse ROM
12.5%
N/A
Hip Back
0.0%
12.5%
Transverse ROM
6.3%
0.0%
Temporal
Spatial
Stance Time
18.8%
6.3%
Step Length
31.3%
43.8%
Swing Time
6.3%
↓
25.0%
Step Width
0.0%
0.0%
Step Time
31.3%
37.5%
Stride Length
18.8%
12.5%
Signi
ficant between group differences of kinematic and kinetic measures are after Bonferroni–Holm correction with the smallest p-value cutoff of 0.0009.
Signi
ficant between group differences of temporal spatial measures are after Bonferroni–Holm correction with the smallest p-value cutoff of 0.0083.
MILITARY MEDICINE, Vol. 181, November/December Supplement 2016
34
The Prevalence of Gait Deviations in Individuals With Transtibial Amputation
was found to be signi
ficantly different from the AB group
(
p ≤ 0.001).
Pelvis/Trunk
Deviation prevalence was low to moderate for all pelvic and
trunk measures of the TTA group. Five individuals (31.3%)
with TTA accounted for all deviations of anterior and poste-
rior tilt. Overall, the TTA group ambulated with less anterior
pelvic tilt and had a signi
ficantly larger sagittal ROM than
the AB group (
p ≤ 0.001). In the AB group, the frontal
plane motion was symmetrical, with 3.1 (SD 1.5) degrees
of contralateral drop and 3.7(SD 2.1) degrees of contralat-
eral elevation relative to the stance limb. However, the
TTA group demonstrated asymmetric pelvic motion with
signi
ficantly reduced contralateral pelvic drop on the pros-
thetic limb and intact limb elevation (
p ≤ 0.001). The
frontal pelvic ROM of the TTA group was signi
ficantly
decreased (
p ≤ 0.001), although there were no values out-
side the NRR.
DISCUSSION
The current study is the
first, to our knowledge, to identify
and characterize deviations in individuals with a TTA using
both group mean comparisons and NRRs. The use of NRR
and prevalence is presented as an additional method for
identifying and quantifying deviations in individuals with a
TTA likely to be encountered in a treatment environment.
This approach allows direct comparisons of each individual
to the NRR. This is in contrast to the group mean approach
which, as evident in (Fig. 1), requires a high proportion of
values be consistently greater or less than the AB group
mean, or a few individuals with a TTA with large devia-
tions in the same direction. When using the NRR approach,
a relatively large sample of AB individuals is needed to
provide an accurate estimate of NRRs. In this study, data
from 40 AB males were the largest sample available at the
time of analysis and presented in Table II for use by others.
Prosthetic Limb
Similar to previous reports, individuals with a TTA in this
study exhibited signi
ficant kinematic
16
and kinetic
10,19
devi-
ations at the prosthetic ankle. Kinematic deviations such
as decreased peak plantar
flexion during initial swing and
decreased sagittal ROM were highly prevalent. The decrease
in peak plantar
flexion was due to an inability of the pros-
thetic devices to actively plantar
flex. When unloaded, these
passive devices return to their aligned position of approxi-
mately 5 degrees of dorsi
flexion. The use of passive pros-
thetic devices and associated decrease in plantar
flexion
contributed to the signi
ficant decrease in prosthetic ankle
power generation at terminal stance. Similar to earlier
works,
10,19
individuals with a TTA on average exhibited
an approximate 50% reduction in prosthetic ankle power
generation at terminal stance compared to the intact ankle
and AB group. However, this deviation was only prevalent
in 50% of our TTA group. This suggests that, although pros-
thetic ankle ROM was limited, 50% of the TTA group were
capable of achieving push-off powers within the NRR. This
demonstrates that performance within normative ranges is
possible, and interventions which allow individuals with a
TTA to more effectively load and store energy in the foot
are warranted.
Individuals with a TTA in the present study walked
with a signi
ficant decrease in prosthetic limb sagittal knee
ROM, which was highly prevalent and may be related to
an extended knee posture observed in the prosthetic limb
throughout gait. In agreement with previous reports, the
effects of the extended knee posture were greatest at initial
contact when individuals with a TTA displayed a signi
fi-
cantly decreased knee
flexor moment
15,16
and a signi
ficant
and highly prevalent reduction in knee power genera-
tion.
15
Signi
ficant decreases in sagittal knee moments and
powers measured later in stance were consistent with ear-
lier works,
11,15,16,19,27
however, demonstrated 0% devia-
tion prevalence meaning all values were within the NRR.
These kinetic differences may be related to a compensatory
extended knee posture, which prevents the knee from col-
lapsing
16
during stance.
Similar to previous literature,
12,16
individuals with a TTA
demonstrated a trend toward greater hip extension during
prosthetic limb stance. This was associated with a signi
fi-
cant increase in hip power generation during stance
10,12,19
and used to control knee
flexion in the prosthetic limb and
aid forward progression.
19
Despite signi
ficant kinetic group
mean differences, the TTA group exhibited little to no prev-
alence of hip deviations. Longitudinal analysis of our partici-
pants would provide insight into when and if compensations
at the hip develop.
Intact Limb
Individuals with a TTA in the present study exhibited no
signi
ficant differences in intact ankle kinematics compared
to the AB group. However, a signi
ficant increase in intact
ankle power absorption at initial contact was observed with
43.8% prevalence. We speculate this was in response to an
abrupt transition off the prosthetic ankle at terminal stance
and onto the intact limb at initial contact.
28
Individuals with
a TTA must transition off the prosthetic foot more quickly
due to a shortened roll-over arc of the prosthetic foot
28
and
lack of active plantar
flexion. This is consistent with the sig-
ni
ficant reduction in intact limb swing times observed in the
TTA group. The increase in intact ankle power absorption
during loading response may be associated with reports of
increased intact limb loading
1,3,5,29
thought to contribute to
secondary musculoskeletal disorders.
The results from the intact limb knee are in contrast to
previous works, which reported increases in swing knee
flexion,
11
maximum extension moments,
11,16
and stance knee
MILITARY MEDICINE, Vol. 181, November/December Supplement 2016
35
The Prevalence of Gait Deviations in Individuals With Transtibial Amputation