Background: Individuals with chronic ankle instability (CAI) tend to walk with an overly

inverted foot, which increases the risk of ankle sprains during stance phase. Clinicians could
perform ankle taping using kinesiotape (KT) or athletic tape (AT) to address this issue. Because
KT is elastic while AT is not, the techniques and un-derlying mechanisms for applying these
tapes are different, which may lead to different outcomes.

Research question: To compare the effects of KT and AT interventions on foot motion in the
frontal plane and tibial motion in the transverse plane during stance phase of walking.
Methods: Twenty subjects with CAI were assigned to either KT or AT group, and walked on a
treadmill in no tape and taped conditions. Their foot and tibial motions were captured by 3D
motion analysis system. The main component of KT application was two pieces of tape applied
from the medial aspect of the hindfoot to the lateral to generate a pulling tension towards
eversion. AT was applied to the ankle using the closed basket weave approach. AT was not
stretchable and not able to generate the same pulling tension as KT.

Results: KT increased foot eversion during early stance, but showed no effect during late stance.
AT increased tibial internal rotation during late stance, but showed no effect during early stance.

Significance: Compared to AT, KT better provides a flexible pulling force that facilitates foot
eversion during early stance, while not restricting normal inversion in late stance during walking.
KT may be a useful clinical tool in correcting aberrant motion while not limiting natural
movement in sports.

1. Introduction
Ankle sprains are one of the most common athletic injuries with a pooled cumulative
incidence rate of 11.55 per 1000 exposures [​1​]. Inversion injuries account for 85% of all ankle
sprains [​2​]. Many in-dividuals who had an ankle sprain eventually develop chronic ankle
instability (CAI), which is characterized by recurrent ankle giving way and/or sprains [​3​]. In
high school and Division I athletes, approxi-mately 23% were identified as having CAI [​4​].
Repeated injuries at the ankle due to CAI could cause irreversible degenerative changes, with
68–78% of individuals with CAI developing post-traumatic ankle os-teoarthritis [​5​].
Alteration in gait kinematics may contribute to recurrent ankle sprains in individuals with CAI
[​6–9​]. Specifically, they tend to walk with an overly inverted ankle [​7​,​8​], which could increase
the risk of inversion injuries following initial contact. Landing the foot on the floor with
hyper-inversion moves the center of pressure further towards the lateral border of the heel [​10​].
This increases the moment arm for the ground reaction force to generate greater inversion torque
during loading response [​3​]. Loading response is a gait period when the body weight is fully
transferred onto the stance leg, and naturally the ankle moves into eversion for shock absorption
[​11​]. If the eversion motion does not occur effectively, an inversion sprain can occur.
 Decreasing ankle inversion during loading response may play a role in reducing the risk of
ankle sprains. One approach to achieve this goal is using non-elastic athletic tape (AT) to restrict
ankle inversion [​12​]. Two studies applied AT to the ankle using a traditional closed basket weave
in individuals with CAI, and examined how the intervention affected gait [​13​,​14​]. One study
showed that the intervention restricted ankle inversion during pre-swing (when the foot starts to
push off the ground) but not during loading response [​13​]. However, restricting inversion during
pre-swing may not be ideal, as this motion is required
to lock the midfoot for creating a rigid lever for push off [​15​]. The other study examining
tibia-rear foot coupling implied that the AT inter-vention restricted rear foot eversion in relation
to tibial rotation during loading response [​14​]. However, ankle eversion is critical for shock
absorption during loading response [​15​]. These results together showed that AT has the ability to
“restrict” ankle motion, but has limited ability to “facilitate” a desired ankle motion at the right
time during a gait cycle.
We asked if we can use Kinesiotape (KT) to achieve reduction of ankle inversion during
loading response, while not affecting the same motion during pre-swing. Unlike AT, KT is
elastic and can be stretched to 140% of its original length before being applied to the skin [​16​].
In principle, the elasticity of KT provides two advantages. First, KT can generate pulling force in
the direction that the tape is stretched [​16​,​17​]. During loading response, this pulling force could
be used to provide: (a) a sensory cue to guide active ankle movement towards eversion and (b) a
mechanical assistance that passively moves the ankle towards ever-sion [​18​]. Second, people can
overcome the tension generated by KT, and moves the ankle into inversion as it normally occurs
during pre-swing.
The purpose of this pilot study was to compare effects of AT inter-vention (provide inversion
restriction) and KT intervention (provide eversion facilitation through generating pulling tension)
on foot motion in the frontal plane and tibial motion in the transverse plane during loading
response and pre-swing. In closed chain, foot inversion drives the tibia into external rotation as
the configuration of the subtalar joint is like a mitered hinged joint [​15​]. We hypothesized that
KT has greater ability than AT to reduce foot inversion and increase tibial internal rotation
during loading response, but AT has greater ability than KT to achieve the same during
pre-swing.

2. Methods
2.1. Subject recruitment and assignment
Twenty subjects with CAI were conveniently recruited from a uni-versity campus. A sample
size of 10–20 was previously suggested for a pilot study [​19​]. Subjects were included if they
aged between 18 and 45, as ankle sprains often happen in the young population [​20​]. Sub-jects
were determined to have CAI if they scored 24 or lower in the Cumberland Ankle Instability
Tool (CAIT), had more than one event of ankle giving way in the past six months, and had an
ankle sprain one year before enrollment [​21​]. Subjects did not have current acute in-juries
affecting their leg joints.
Quasi-randomization was used to assign subjects to either KT or AT group. Specifically, data
collection schedules were set based on the availabilities of a certified KT practitioner
(responsible for all KT ap-plications) and an athletic trainer (responsible for all AT applications).
Each subject then selected an available data collection session based on his/her availability,
without knowing that the session was assigned to the KT practitioner or the athletic trainer. The
KT group consisted of 8 females and 2 males (age: 22.8 ± 1.3 years old) and the AT group
consisted of 7 females and 3 males (age: 23 ± 2.3 years old). Informed consent was obtained
from all subjects, and all procedures were con-ducted in accordance with the Helsinki
Declaration of 1975 and ap-proved by the local Institutional Review Board.
2.2. Procedures
Subjects walked at a self-selected comfortable speed on a treadmill for one minute in both no
tape and taped conditions. The tape was applied to the affected side (for unilateral CAI) or the
more severe side (for bilateral CAI) based on the CAIT score. Subjects’ leg kinematics were
captured using Qualisys motion capture system (Göteborg, Sweden) with a sampling frequency
of 100 Hz. Reflective markers were placed on the major bony landmarks of the shank (over the
skin) and the foot (over the shoe) to create a shank model and a single-segment foot model. The
marker placement was consistent with our previous protocols [​9​,​22​]. In the taped condition, we
removed the shoe before applying the tape, and all markers on the shoe stayed in the same
lo-cation without moving. After the tape was applied, the subject put the shoe back onto the foot.
Subjects wore their own comfortable sneakers during data collection to avoid potential changes
in typical ankle ki-nematic patterns due to a new shoe configuration
2.3. Taping techniques
The main component of KT application was two pieces of tape ap-plied from the medial to the
lateral aspect of the lower leg to generate eversion tension (​Fig. 1​A). Tension was applied to
these two pieces of tape at 75% (i.e. the tape was stretched to 75%). This tension was suggested
to provide sensory stimulation and mechanical assistance to facilitate motion [​18​]. The two tapes
were anchored by multiple figure-of-eights without tension (​Fig. 1​B).

The traditional closed basket weave technique was used to apply 1.5-inch AT to the ankle
[​12​]. This technique consisted of medial-lat-eral stirrups to control the hind foot motion in the
frontal plane (​Fig. 2​A), along with continuous heel locks with a figure-of-eight pat-tern (​Fig.
2​B). Unlike KT, AT was not stretchable and was applied in its original length for every
component. The ankle was placed in the neutral position during both AT and KT applications.
2.4. Data analysis
Visual3D (C-Motion, MD) was used to calculate foot/shank position based on the marker
data. Zeni et al.’s coordinate-based treadmillalgorithm was used to determine initial contact and
toe off [​23​]. All position trajectories were segmented into cycles. We selected the first 20 stance
phases for analysis. Each stance phase was interpolated to 200 points and evenly divided into 10
zones. Each outcome variable was averaged within each zone. Zone 1 approximated loading
response while zone 10 approximated pre-swing.

The primary outcome variables were foot inversion/eversion and tibial internal/external
rotation in zones 1 and 10 for hypothesis testing. The secondary variables were the same position
data in zones 2–9. Paired t-tests were used to compare position data between no tape and taped
conditions within each group. We also calculated the change score between no tape and taped
conditions for each group (change score = taped value − no tape value), and compared the
change scores between the groups using independent t-tests. Prior to the t-tests, we conducted
Shapiro-Wilks tests to check the normality assumption. When the normality assumption was
violated, we alternatively per-formed non-parametric tests for comparison (Wilcoxon
signed-rank for within-group or Mann-Whitney U for between-group). In the results section, p
values obtained from non-parametric tests were noted. Cohen’s d was used to quantify the effect
size for each t-test. The effect size for each non-parametric comparison was calculated as: ​r = Z​ ​N ,
where Z is the Z score of the comparison and N is the number of total observations [​24​].
3. Results
3.1. Foot position
Subjects demonstrated a less inverted foot when walking with KT compared to walking
without KT during the early stance phase (​Fig. 3​A). Such change only reached significance in
zone 1 (without KT: 6.97 ± 3.1°; with KT: 5.28 ± 3.1°; p = 0.03; d = 0.82). The foot po-sition
change observed from zone 2 to zone 10 did not reach statistical significance.
Subjects showed minimal changes in foot position during each stance phase zone when
walking with AT compared to walking without AT (​Fig. 3​B), and none of these changes reached
statistical significance.

Fig. 3​C ​shows the change score of the foot position between taped ​and no tape conditions
during stance phase. Comparing the change score between groups, we found a significant
difference in zone 1 (KT: −1.69 ± 2.1°; AT: 0.43 ± 2.4°, p = 0.049, d = 0.9). No significant
differences in the change score were detected in any other zones.
3.2. Tibial position
Subjects demonstrated minimal changes in tibial rotation in each stance phase zone when
walking with KT compared to walking without KT (​Fig. 4​A), and none of these changes reached
statistical significance.

Subjects demonstrated an increase in tibial internal rotation during late stance phase when
walking with AT compared to walking without AT (​Fig. 4​B). Significant differences were found
in zone 7 (p = 0.03; d = 0.81), zone 8 (p = 0.01; d = 1), zone 9 (p < 0.01; d = 1.1), and zone 10 (p
= 0.03; d = 0.79). The differences observed from zone 1 to zone 6 did not reach statistical
significance.

Fig. 4​C ​shows the change score of the tibial position in the trans-verse plane during stance
phase. Comparing the change score between groups, we found significant differences in zone 7
(KT: −0.03 ± 1.9°; AT: 1.39 ± 3°; p = 0.03; d = 1.1), zone 8 (KT: 0.05 ± 1.8°; AT:

1.06 ± 2.8°; p = 0.03, Mann-Whitney U test; r = 0.5), and zone 9 (KT: −0.01 ± 1.7°; AT: 0.13 ±
2.4°; p = 0.04; d = 0.97). No significant differences in the change score were detected from zone
1 to zone 6 and in zone 10.
4. Discussion
This study had two major findings. First, KT reduced foot inversion (or increased foot
eversion) right after initial contact during walking in individuals with CAI, but the same effect
was not observed in AT. Second, AT increased tibial internal rotation (or reduced tibial external
rotation) during late stance phase in walking in individuals with CAI, but the same effect was not
shown in KT. Based on effect size estimates, all significant results had a medium to large effect
[​25​].

The KT and AT interventions showed different effects on foot po-sition during loading response.
In zone 1, subjects in the KT group showed an average of 6.97° of foot inversion when walking
normally in the no tape condition, and this angle significantly decreased to 5.28° in the taped
condition. This suggests that the KT intervention may have the ability to increase foot eversion
during loading response. In con-trast, subjects in the AT group showed an average of 5° of foot
inversion when walking normally in the no tape condition in zone 1. This angle slightly
increased to 5.43° in the taped condition, although this increase did not reach statistical
significance. The results suggest that the AT intervention may have limited effeect on altering
foot position during loading response.
Two mechanisms may contribute to the increase in foot eversion due to KT. First, the pulling
force generated by KT could provide a sensory cue to guide active foot movement towards
eversion. Second, the pulling force could provide a mechanical assistance that passively moves
the foot towards eversion. These mechanisms were proposed by the developers of KT [​18​],
although the supporting evidence is con-troversial. For example, a previous study showed that
KT application reduced rather than increased the activity of tibialis anterior and peroneous
longus [​26​]. In addition, we are not aware of empirical evidence to support the required tension
for KT to move a body seg-ment, which weakens the idea of mechanical assistance.
AT did not affect foot position in the frontal plane during loading response. While AT can
provide mechanical restraint to the joint/seg-ment, the restraint may only happen in the extreme
range and may have no effect on influencing the kinematics of the joints or segments when they
are in the functional range of motion [​27​]. In addition, the closed basket weave approach
consisted of multiple layers of medial-lateral stirrups and figure-of-eights [​12​]. Together, these
components mainly generate compression sensation around the ankle, rather than creating a
directional cue to guide eversion. Also, AT has limited elasticity and cannot generate a dynamic
pulling tension to assist motion like KT.

During late stance, the ankle naturally moves into inversion to allow the foot to become a rigid
lever for push off [​15​]. Our results showed that this natural motion was not affected by KT but
was affected by AT, particularly in the tibia. The elasticity of KT allowed subjects to over-come
the pulling force in eversion and move the foot into inversion. In contrast, a previous study
showed that AT reduced ankle inversion during late stance [​13​]. Our results provided additional
information, showing that AT restricted tibial external rotation more than foot in-version. The
ground reaction force directly applies to the foot during stance phase, and may play a role in
counteracting the restricting force generated by AT. The ground reaction force may not achieve
the same effect on the tibia as it does not directly apply to this segment.
Our results suggest that KT does not restrict normal ankle kine-matics during walking. This is
beneficial because many essential tasks in walking (e.g., shock absorption and propulsion) are
achieved by al-tering the flexibility of the foot through changing the ankle position [​15​]. Similar
functionality is required in many sports like running to achieve successful performance [​28​].
Non-elastic taping prevents ankle sprains by restricting ankle mobility, but some evidence
suggests that it may negatively affect agility [​29​] and running/jumping performance [​30​] during
a sport. With its flexibility, KT may be a potential solution for this issue. Future study is
warranted to examine if our results on KT can be replicated in sporting tasks.
This study was motivated by previous findings that individuals with CAI tend to walk with an
overly inverted ankle [​7​,​8​], which could in-crease the risk of inversion injuries during loading
response. We tested if KT could be used to address this issue. While we found that KT could
statistically significantly reduce foot inversion during loading response, the mean degree of
reduction was only ∼1.7°. It is unclear if the amount of reduction is clinically significant, because
to our best knowledge, the minimum clinically important change in this case has not been
established. In [​7​,​8​], the difference in ankle inversion during loading response between healthy
and CAI subjects was approximately 3–6°, which was higher than the effect of KT shown in our
study. Future study is warranted to examine the clinical significance of our current findings.
A limitation of this study was that we did not control for the var-iation in subject’s shoes.
While all subjects wore typical sneakers, subtle differences in shoe structures could potentially
affect the results. Future study can examine if our findings can be replicated using standardized
shoes or in a barefoot condition. We used a quasi-randomized method for group assignment due
to limited time availability of some research team members and subjects, but this method
increased the risk of se-lection bias. We will use true randomization in our future large scale
study to address this issue. We used a between-group rather than a within-group design.
Applying tapes to the ankle essentially generates force perturbation to affect the joint kinematics.
Previous study showed that changes in ankle kinematics induced by force perturbation could
retain even after the perturbation was removed [​22​]. A between-group design was therefore used
to avoid potential after effects transferring from one condition to another. However,
between-group differences in subject characteristics could potentially confound the results. This
was a pilot study and we focused on examining the feasibility of using KT to affect ankle
kinematics. In future study, we will collect additional variables such as electromyography and
subjects’ subjective feedback to determine the underlying mechanism of KT.
5. Conclusion
Compared to AT, KT better provides a flexible pulling force that facilitates foot eversion
during early stance, while not restricting nat-ural inversion during late stance in walking. The
foot and ankle must move in different directions to achieve different functions in walking. The
elasticity allows KT to facilitate motion in one direction, while not limiting motion in the
opposite direction. This cannot be achieved by AT, which has a non-elastic property. The
clinical significance of our results was inconclusive. While KT increased ankle eversion during
loading response, the magnitude was small, which may or may not be able to reduce the risk of
inversion injuries. We used walking as a model to test KT’s ability to facilitate foot eversion
during early stance while not restricting natural inversion during late stance. It is clinically
sig-nificant to examine if the current results are replicable in running in which the foot/ankle also
need to move in different directions to achieve different functions. KT may be a useful tool in
correcting aberrant motion while not limiting natural movement in sports.
Conflict of interest statement
The authors affirm that there is no conflicts of interest that may have influenced the
preparation of this manuscript.
Acknowledgements
We thank all students who assisted in subject preparation and data collection. We thank Dr.
Stephen Clark for his assistance in athletic taping.