This article appeared in a journal published by Elsevier. The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
Author's personal copy
Gait analysis after bi-compartmental knee replacement
He Wanga,*, Eric Dugana, Jeff Framea, Lindsey Rolstonb
aBiomechanics Laboratory, School of Physical Education, Sport, and Exercise Science, Ball State University Muncie, IN 47306, USA
bHenry County Center for Orthopedics, New Castle, IN 47362, USA
a r t i c l ei n f o
Received 20 January 2009
Accepted 21 July 2009
a b s t r a c t
Background: It is reported that a majority of the patients with knee osteoarthritis have cartilage degen-
eration in medial and patellofemoral compartments. A bi-compartmental knee replacement system was
designed to treat osteoarthritis at medial and patellofemoral compartments. To date, there is very little
information regarding the knee mechanics during gait after bi-compartmental knee replacement. The
purpose of the study was to evaluate knee strength and mechanics during level walking after knee
Methods: Ten healthy control subjects and eight patients with unilateral bi-compartmental knee replace-
ment participated in the study. Maximal isokinetic concentric knee extension strength was evaluated. 3D
kinematic and kinetic analyses were conducted for level walking. Paired Student t-test was used to deter-
mine difference between surgical and non-involved limbs. One way MANOVA was used to determine dif-
ference between surgical and control groups.
Findings: The surgical knee exhibited less peak torque and initial abduction moment than both the non-
involved and control limbs (P < 0.05). The non-involved limb had less knee extension at stance and
greater knee extensor moment during push-off than both the surgical and control limbs (P < 0.05). No dif-
ferences were found for other typical knee mechanics among the surgical, non-involved, and control
limbs during walking (P > 0.05).
Interpretations: Patients with bi-compartmental knee replacement exhibited good frontal plane knee
mechanics and were able to produce the same level of knee extensor moment as healthy control limbs
during walking. While showing some compensatory patterns during walking, patients with bi-compart-
mental knee replacement largely exhibited normal gait patterns and knee mechanics.
Published by Elsevier Ltd.
Osteoarthritis (OA) is a cartilage degenerative disease and
causes more disability with respect to mobility than any other sin-
gle disease in the elderly (Guccione et al., 1994). Knee OA typically
affects joints in a non-uniform manner; the medial compartment
of the knee is most frequently affected in both men and women
(Windsor and Insall, 1994). Furthermore, the three most common
areas of knee OA distribution are medial compartment, patellofe-
moral compartment, and medial/patellofemoral compartment
overlap (McAlindon et al., 1992). The predominance of medial knee
OA is likely due to the high medial forces generated during weight-
bearing activities (e.g. walking) (Morrison 1970; Schipplein and
Andriacchi, 1991). It is believed that knee adduction torque is
strongly associated with risk of medial knee OA progression (Miya-
zaki et al., 2002).
Knee osteoarthritis (OA) and problems associated with varus
knee alignment are the primary factors leading to total knee
replacements (TKR). However, the damage associated with these
conditions is often limited to cartilage degeneration in the medial
and patellofemoral compartments. One alternative to the TKR,
especially for more active patients, is uni-compartmental knee
(UKR) replacement which leaves the lateral knee compartment
and cruciate ligaments intact while replacing only the affected
medial compartment. UKR leads to faster recovery times, less bone
loss, and better knee kinematics than TKR (Swienckowski and Pen-
nington, 2004; Banks et al., 2005; Fuchs et al., 2005). However, UKR
does not treat the OA at patellofemoral joint which is common in
those patients with compromised medial compartments.
As it was found that patellofemoral joint arthritis is often asso-
ciated with medial compartment arthritis (McAlindon et al., 1992),
a bi-compartmental knee replacement (BKR) was introduced to
treat the medial compartment and patellofemoral joint arthritis
(Journey Deuce systemTM, Smith & Nephew Inc., Memphis, TN,
USA). Similar to a UKR, the BKR preserves the lateral compartment
of the knee and cruciate ligaments while also addressing the
0268-0033/$ - see front matter Published by Elsevier Ltd.
* Corresponding author.
E-mail address: firstname.lastname@example.org (H. Wang).
Clinical Biomechanics 24 (2009) 751–754
Contents lists available at ScienceDirect
journal homepage: www.elsevier.com/locate/clinbiomech
Author's personal copy
patellofemoral articulation. With intact lateral compartment and
cruciate ligaments, the BKR may lead to faster recovery and better
functional outcomes. However, objective data are lacking to sup-
port this hypothesis.
The primary purpose of this study was to evaluate knee
mechanics during level walking in patients who have undergone
unilateral BKR system. The secondary purpose was to evaluate
knee strength in patients who have undergone unilateral BKR
As the BKR surgery results in less loss of bone and retains the
lateral compartment and cruciate ligaments of the knee, a return
to normal knee extensor strength and normal knee kinematics
was expected after surgery. In addition, the knee extensor moment
arm was expected to be similar to control knees. Therefore, it was
hypothesized that there would be no differences of net knee exten-
sor moments between the BKR, non-involved, and healthy control
limbs during knee strength tests and level walking.
Also, the BKR surgery is expected to result in good longitudinal
alignment between the femur and tibia. The mechanical axis is ex-
pected to pass through the center of the knee joint. Therefore, there
would be no differences of knee abduction/adduction knee joint
moments between the BKR, non-involved, and healthy control
limbs during level walking.
The combination of these potential improvements may also
lead to a reduction or absence of compensatory gait patterns. Spe-
cifically, it was expected there would be no differences of step
length, single support time, and knee flexion angles between the
BKR, non-involved, and healthy control limbs during level walking.
Ten healthy subjects formed a control group. Eight participants
from a pool of patients who had been diagnosed with OA in the
medial and patellofemeral compartments and undergone unilate-
ral BKR surgery (performed by the same surgeon) with the Journey
DeuceTMsystem (Smith & Nephew Inc., Memphis, TN, USA) formed
the BKR group. In addition, these BKR patients were selected based
on the following criteria: (a) no history of major diagnosed health
problems or injuries that would affect performance; (b) medical
clearance from the surgeon; (c) no current symptoms of pain, in-
jury, or muscle soreness; (d) have both the functional score and
knee society score greater than 90; (e) no orthopedic problems,
OA symptoms, or pain in the non-involved knees; (f) no OA symp-
toms in the lateral compartment of the involved limb. All patients
had gone through a standard rehabilitation program after the sur-
gery. The means and SDs of age, body mass, body height of the BKR
and control groups and post-operative time, knee society clinical
score and functional score of the BKR limbs are reported in Table
1. Institutional Research Board (IRB) approval was obtained prior
to commencing the study.
The subjects came to the Ball State University Biomechanics
Laboratory on two separate occasions with one week apart. During
the first session the subject’s maximal knee extensor torque was
measured using a CYBEX NORMTMisokinetic dynamometer (CYBEX
International, Inc., NY, USA) at 60 deg/s. During the second session,
3-dimensional (3D) kinematic and kinetic analyses were con-
ducted while the subjects performed five trials of level walking
at self-selected pace. Ten VICONTMF40 cameras (Vicon, Lake For-
est, CA, USA) at 100 Hz were used to capture the spatial locations
of reflective markers on the participant. The standard full body
plug-in gait model was used for the placement of the markers on
the body. A VICON WorkstationTMsystem (V 4.2) was used to gen-
erate 3D coordinates of the reflective markers. Two AMTITMforce
platforms (Model BP600900-6-1000, Advanced Medical Technol-
ogy, Inc., Watertown, MA, USA) were used to collect the ground
reaction forces at 1000 Hz. The following dependent variables were
analyzed: maximum isokinetic concentric knee extension (KE) tor-
que, knee flexion at foot strike, peak knee flexion, extension and
adduction angles at stance, peak knee abduction moment at foot
strike, peak knee adduction moment during stance, peak knee
extensor moment at push-off, walking speed, single support time,
and step length. In addition, knee extension torques during
strength testing and knee moments during level walking were nor-
malized to body mass.
One-way multiple analysis of variance (MANOVA) was used to
determine kinematic and kinetic difference between BKR and con-
trol groups. Paired Student t-test was used to determine kinematic
and kinetic difference between non-involved limbs and BKR limbs.
Post-hoc power analysis was performed to determine the effect
sizes and achieved power of each dependent variable. Significant
level was set at 0.05.
Effect sizes and achieved power of dependent variables were re-
ported in Table 2.
During strength testing, BKR limbs exhibited less KE torque
than both the non-involved and control limbs (P < 0.05) (Fig. 1).
During level walking, BKR limbs exhibited smaller peak of ini-
tial adduction moment at foot strike than both the non-involved
and control limbs (P < 0.05) (Table 3). Non-involved limbs had less
knee extension at mid-stance and greater normalized knee exten-
sor moment at push-off (Table 3) than the BKR and control limbs
(P < 0.05). No difference was found for walking speed between
the two groups (P > 0.05) (Table 1). No differences were found for
knee flexion at foot strike, peak knee flexion, adduction, and
adduction moment during stance between the control, BKR and
non-BKR limbs (P > 0.05) (Table 2). No differences were found for
Means and SDs of age, body mass, body height, and walking speed of the BKR and
control groups and post-operative time, knee society clinical score and functional
score of the BKR limbs.
BKR (n = 8)Control (n = 10)
Body mass (kg)
Walking speed (m/s)
Post-operative time (mo.)
Knee society clinical score
Achieved power and effect size of dependent variables.
Dependent variablesEffect size (based on
Knee flexion at foot strike
Peak knee flexion during 1st half of
Peak knee extension
Peak knee adduction
Peak knee extensor moment at push-
Peak initial knee abduction moment
at Foot strike
Peak adduction moment
Single support time
Max isokinetic concentric knee
H. Wang et al./Clinical Biomechanics 24 (2009) 751–754
Author's personal copy
step length, single support time between the BKR and non-in-
volved limbs (P > 0.05) (Table 3).
In summary, the purpose of this study was to evaluate knee
mechanics during level walking and knee strength in patients
who had undergone unilateral BKR system. As the BKR surgery re-
sults in less loss of bone and retains the lateral compartment and
cruciate ligaments of the knee, it was expected that knee mechan-
ics and knee strength of the unilateral BKR subjects would not be
different than those of their non-involved limbs or those of healthy
The BKR limbs exhibited less isokinetic knee extensor strength
than both of the non-involved limbs and limbs from the control
group. This result suggests that at approximately one year post-
surgery, BKR knee extensor strength has not yet returned to nor-
mal. As none of the patients involved in the study had participated
in a strength training program, it may be possible to elicit addi-
tional improvements in knee extensor strength for the BKR limbs
if the patients participated in a regular strength training program
after the surgery.
It was also hypothesized that there would be no difference of
knee extensor moment at push-off among the BKR, non-involved,
and control limbs. This hypothesis was partially supported as there
was no difference between the BKR and control limbs. This finding
indicated that the BKR limb was able to produce a similar knee
extensor moment during normal walking compared to that of
healthy control limbs. Although the maximum knee extensor
strength of the BKR limb was still less than both the non-involved
and control limbs, it appears that during daily activities (e.g. nor-
mal walking), the BKR limb had no limitation in its ability to per-
form in a similar fashion to the normal control limb. On the other
hand, the non-involved limb exhibited less knee extension and
greater knee extensor moment at push-off than the BKR and con-
trol limbs. This finding is thought to be a result of a possible reten-
tion of the gait pattern formed prior to the surgery. The
development of knee OA and the associated knee pain may have
led to less usage of the involved limb and increased the depen-
dency on the non-involved limb during gait. This would poten-
tially lead to a reliance on the non-involved limb to produce a
greater knee extensor moment at push-off to compensate for the
‘painful’ knee. While this explanation seems plausible, this study
does not include any pre-surgery analyses, therefore, it is not pos-
sible to determine if this walking pattern was present before the
One of the goals of BKR surgery is to correct the frontal plane
knee deformity. Patients with medial and patello-femoral knee
osteoarthritis typically have significant varus deformity. It was re-
ported that BKR surgery effectively corrects varus deformity and
restores good knee alignment (Rolston and Siewert, 2008). In the
current study, despite the lack of a direct measure on frontal plane
knee alignment, the finding of no differences of peak knee adduc-
tion angle and moment between the BKR, non-involved, and con-
trol limbs, and the results in the previous work on BKR efficacy
in alignment correction (Rolston and Siewert, 2008), suggested
that good frontal plane knee mechanics during walking corre-
sponded to a good knee alignment. As it was reported that both
the increased peak knee adduction moment (Baliunas et al.,
2002; Hunt et al., 2006; Milner and O’Bryan, 2008) and increased
peak adduction angle (Astephen and Deluzio, 2005) were associ-
ated with medial knee OA, the absence of an increased peak knee
adduction angle and moment in the BKR limb indicates that frontal
plane knee mechanics during walking has returned to normal after
the BKR surgery. It is also reported that higher abduction knee mo-
ment exhibited at heel contact is often associated with medial knee
OA and possibly used as a strategy to reduce the high knee adduc-
tion moment occurring at late stance for reducing pain (Munder-
mann et al., 2005). In this study, the reduction of peak abduction
knee moment at heel contact likely indicates a positive change of
gait pattern after BKR surgery as patients with BKR did not exhibit
the typical pain avoidance compensatory pattern typically seen in
individuals with knee OA.
Fig. 1. Maximum isokinetic concentric knee extension torque for control, non-
involved, and the BKR limbs.
Means and SDs of typical variables of knee mechanics and gait kinematics at stance (flexion at foot strike, peak flexion, extension, and adduction, peak extensor moment, initial
peak of adduction moment at foot strike, and adduction moment, step length, and single support time) of the control, non-involved, and BKR limbs.
Variables of knee mechanics and gait kinematics Control Non-involvedBKR
Knee flexion at foot strike (?)
Peak knee flexion during 1st half of stance (?)
Peak knee extension (?)
Peak knee adduction (?)
Peak knee extensor moment at push-off (N m/kg)
Initial peak of knee adduction moment at foot strike (N m/kg)
Peak adduction moment (N m/kg)
Step length (m)
Single support time (s)
*Indicates significant difference (P < 0.05) between the BKR and non-involved limbs.
**Indicates significant difference (P < 0.05) between the BKR and the control limbs.
***Indicates significant difference (P < 0.05) between the non-involved and control limbs.
H. Wang et al./Clinical Biomechanics 24 (2009) 751–754
Author's personal copy Download full-text
Finally, level walking was well performed by the BKR group.
This was supported by the similar walking speed between the
BKR and control groups and similar knee joint kinematics and gait
kinematics between the BKR and non-involved limbs.
In conclusion, the BKR patients exhibited good frontal plane
knee mechanics and were able to produce the same level of knee
extensor moment as healthy control limbs during walking. While
showing some compensatory patterns, the BKR patients had lar-
gely returned to normal gait patterns and knee mechanics.
Some limitations must be addressed here. Firstly, the BKR pa-
tients’ preoperative gait data were lacking. It is not known how
much of an improvement achieved in leg strength and gait after
the BKR surgery. Secondly, we did not include a TKR group in the
study. Thus, we could not make comparisons between the BKR
and TKR limbs during gait. Future studies should focus on the
above two issues.
Conflict of interest
There is no conflict of interest in this study.
BSU S.E.E.T Fund.
Astephen, J.L., Deluzio, K.J., 2005. Changes in frontal plane dynamics and the loading
response phase of the gait cycle are characteristic of severe knee osteoarthritis
application of a multidimensional analysis technique. Clin. Biomech. 20 (2),
Baliunas, A.J., Hurwitz, D.E., et al., 2002. Increased knee joint loads during walking
are present in subjects with knee osteoarthritis. Osteoarthr. Cartilage 10 (7),
Banks, S.A., Fregly, B.J., et al., 2005. Comparing in vivo kinematics of unicondylar and
bi-unicondylar knee replacements. Knee Surg Sports Traumatol. Arthrosc. 13
Fuchs, S., Tibesku, C.O., et al., 2005. Clinical and functional comparison of uni- and
bicondylar sledge prostheses. Knee Surg Sports Traumatol. Arthrosc. 13 (3),
Guccione, A.A., Felson, D.T., et al., 1994. The effects of specific medical conditions on
the functional limitations of elders in the Framingham Study. Am. J. Public
Health 84 (3), 351–358.
Hunt, M.A., Birmingham, T.B., et al., 2006. Associations among knee adduction
moment, frontal plane ground reaction force, and lever arm during walking in
patients with knee osteoarthritis. J. Biomech. 39 (12), 2213–2220.
McAlindon, T.E., Cooper, C., et al., 1992. Knee pain and disability in the community.
Br. J. Rheumatol. 31 (3), 189–192.
Milner, C.E., O’Bryan, M.E., 2008. Bilateral frontal plane mechanics after unilateral
total knee arthroplasty. Arch. Phys. Med. Rehabil. 89 (10), 1965–1969.
Miyazaki, T., Wada, M., et al., 2002. Dynamic load at baseline can predict
radiographic disease progression in medial compartment knee osteoarthritis.
Ann. Rheum. Dis. 61 (7), 617–622.
Morrison, J.B., 1970. The mechanics of the knee joint in relation to normal walking.
J. Biomech. 3 (1), 51–61.
Mundermann, A., Dyrby, C.O., et al., 2005. Secondary gait changes in patients with
medial compartment knee osteoarthritis: increased load at the ankle, knee, and
hip during walking. Arthritis Rheum. 52 (9), 2835–2844.
Rolston, L., Siewert, K., 2008. Assessment of knee alignment after bicompartmental
knee arthroplasty. J. Arthroplasty (Epub ahead of print).
Schipplein, O., Andriacchi, T., 1991. Interaction between active and passive knee
stabilizers during level walking. J. Orthop. Res. 9 (1), 113–119.
Swienckowski, J.J., Pennington, D.W., 2004. Unicompartmental knee arthroplasty in
patients sixty years of age or younger. J. Bone Joint Surg. Am. 86-A (Suppl. 1(Pt
Windsor, R., Insall, J., 1994. Surgery of the knee. In: Sledge, C., Ruddy, S., Harris, E.,
Kelley, W. (Eds.), Arthritis Surgery. W.B. Saunders, Philadelphia, pp. 794–817.
H. Wang et al./Clinical Biomechanics 24 (2009) 751–754