1. Research Report
Age-Related Changes in Strength,
Joint Laxity, and Walking Patterns: Are
They Related to Knee Osteoarthritis?
Katherine S Rudolph, Laura C Schmitt, Michael D Lewek
KS Rudolph, PT, PhD, is Assistant
Professor, Department of Physical
Therapy and Program in Biome-
Background and Purpose
chanics and Movement Science, Aging is associated with musculoskeletal changes and altered walking patterns. These
University of Delaware, 301 changes are common in people with knee osteoarthritis (OA) and may precipitate the
McKinly Lab, Newark, DE 19716 development of OA. We examined age-related changes in musculoskeletal structures
(USA). Address all correspondence
to Dr Rudolph at: krudolph@
and walking patterns to better understand the relationship between aging and knee
udel.edu. OA.
LC Schmitt, PT, PhD, is Post-
Doctoral Fellow, Department of
Methods
Pediatrics, University of Cincin- Forty-four individuals without OA (15 younger, 15 middle-aged, 14 older adults) and
nati, College of Medicine; and 15 individuals with medial knee OA participated. Knee laxity, quadriceps femoris
Physical Therapist, Sports Medi- muscle strength (force-generating capacity), and gait were assessed.
cine Biodynamics Center, Cincin-
nati Children’s Hospital Medical
Center, Cincinnati, Ohio. Results
MD Lewek, PT, PhD, is Assistant
Medial laxity was greater in the OA group, but there were no differences between the
Professor, Center for Human middle-aged and older control groups. Quadriceps femoris strength was less in the
Movement Science, Division of older control group and in the OA group. During the stance phase of walking, the OA
Physical Therapy, University of group demonstrated less knee flexion and greater knee adduction, but there were no
North Carolina at Chapel Hill, differences in knee motion among the control groups. During walking, the older
Chapel Hill, NC.
control group exhibited greater quadriceps femoris muscle activity and the OA group
[Rudolph KS, Schmitt LC, Lewek used greater muscle co-contraction.
MD. Age-related changes in
strength, joint laxity, and walking
patterns: are they related to knee
Discussion and Conclusion
osteoarthritis? Phys Ther. 2007: Although weaker, the older control group did not use truncated motion or higher
87:1422–1432.] co-contraction. The maintenance of movement patterns that were similar to the
© 2007 American Physical Therapy subjects in the young control group may have helped to prevent development of
Association knee OA. Further investigation is warranted regarding age-related musculoskeletal
changes and their influence on the development of knee OA.
Post a Rapid Response or
find The Bottom Line:
www.ptjournal.org
1422 f Physical Therapy Volume 87 Number 11 November 2007

2. Age-Related Changes in Strength, Joint Laxity, and Walking Patterns
S
ymptomatic knee osteoarthritis in the presence of quadriceps femo- ing that the material properties of
(OA) is a worldwide problem1– 4 ris weakness, which occurs with ag- ligaments of older adults can lead to
that produces substantial dis- ing,7–11 and in the presence of knee excessive joint laxity. Because
ability in middle-aged and older OA, either the hamstring or the gas- frontal-plane laxity has been related
adults and leads to a tremendous trocnemius muscles may be required to high muscle co-contraction in in-
economic burden on society.5 The to assist with knee control. dividuals with knee OA,24 it is plau-
prevalence of OA among older indi- sible that normal age-related in-
viduals has led some authors6 to re- Activation of muscles surrounding creases in joint laxity also may
gard its development as a normal the knee can occur selectively and contribute to higher muscle co-
part of aging. Loeser and Shakoor,6 with precise timing that allows for contraction patterns and predispose
however, suggested that age-related normal knee motion, or, alterna- individuals to develop knee OA.
changes in musculoskeletal tissue, tively, activation can occur more Whether older adults have greater
such as muscle weakness and liga- generally as a global co-contraction frontal-plane knee laxity coupled
ment laxity, do not directly cause pattern that could limit joint motion. with higher muscle co-contraction is
OA, but may predispose individuals We, therefore, have defined the not known.
to develop the disease. It is possible movement strategy that involves both
that the manner in which people re- increased muscle co-contraction and In this study, we investigated the
spond to these age-related changes reduced knee flexion during walking knee laxity, quadriceps femoris mus-
in musculoskeletal tissues about the as a “stiffening strategy.” Excessive cle strength (force-generating capac-
knee may be related to whether or muscle co-contraction can lead to ity), walking patterns, and muscle ac-
not OA develops in the knees of excessive joint contact forces,18 and tivation patterns in 3 age groups of
older adults. reduced knee motion during weight people without symptomatic or ra-
acceptance can cause higher impact diographic knee OA to examine fac-
The features that are similar between loads in the knee.19,20 Older adults tors that are thought to contribute to
older adults and people with knee are known to walk with less knee the development of knee OA. The
OA include quadriceps femoris mus- flexion,21 but whether they do so results are discussed in relation to
cle weakness and altered knee move- as a result of higher muscle co- characteristics of a group of people
ment during walking. Sarcopenia is contraction is unknown. However, with knee OA. Because the older
well known in older adults and leads not all older adults develop knee OA. adults in our study did not have joint
to quadriceps femoris muscle weak- If older adults who have not devel- degeneration, we hypothesized that
ness,7–11 which has been noted in oped knee OA walk with a knee stiff- older adults who are healthy will
people as early as 40 years of age.9 ening strategy, then the combination have weaker quadriceps femoris
Because both the development of of reduced knee flexion and muscle muscles and increased frontal-plane
knee OA12 and quadriceps femoris co-contraction alone, in the pres- knee laxity but will not exhibit
muscle strength changes9 are initi- ence of quadriceps femoris muscle greater muscle co-contraction pat-
ated during middle age, it is not sur- weakness, is unlikely to contribute terns compared with young or
prising that quadriceps femoris mus- to the development of knee OA. middle-aged people.
cle weakness has been implicated in
the development of knee OA.13–15 Another possible precursor to knee Method
OA is excessive frontal-plane laxity, Subjects
Quadriceps femoris muscle weak- which is common in people with Fifty-nine people were recruited
ness also is associated with adapta- existing knee OA.22–24 Specifically, from the community or were re-
tions in walking patterns that are the- greater frontal-plane knee laxity is ferred by a local orthopedic surgeon
orized to put articular cartilage at observed in both the involved and to participate in the study. All sub-
risk. For instance, subjects with knee uninvolved knees of people with OA jects signed an informed consent
OA who have weaker quadriceps compared with control subjects, sug- statement approved by the Human
femoris muscles exhibit less stance- gesting that laxity may precede the Subjects Review Board of the Univer-
phase knee motion during walking.16 development of knee OA.22 Addi- sity of Delaware. Forty-four partici-
At self-selected walking speeds, it is tionally, a significant correlation be- pants who reported no history of
the role of the quadriceps femoris tween frontal-plane laxity and age knee OA (confirmed by radiograph)
muscles to control knee flexion dur- has been observed in individuals or previous lower-extremity injury
ing weight acceptance while the without evidence of knee OA.22 This comprised 3 control groups (15
hamstring and gastrocnemius mus- finding is consistent with the find- younger individuals [ages 18 –25
cles are typically silent.17 However, ings of other studies,22,25,26 indicat- years], 15 middle-aged individuals
November 2007 Volume 87 Number 11 Physical Therapy f 1423

3. Age-Related Changes in Strength, Joint Laxity, and Walking Patterns
Table 1.
Group Characteristics
Young Middle-aged Older Osteoarthritis
Control Group Control Group Control Group Group
(n 15) (n 15) (n 14) (n 15)
Age (y), X (range) 20.6 (18–25) 49.2 (40–57) 68.8 (60–80) 49.2 (39–57)
Sex (female/male) 8/7 7/8 10/4 7/8
2 a,b a c
Body mass index (kg/m ), X (SD) 24.3 (2.8) 28.7 (5.5) 24.7 (2.5) 30.7 (4.8)b.c
Alignment (°), X (SD) Not tested 0.1 (1.58)d valgus 1.0 (2.09)d varus 6.33 (2.39)d varus
a
P .030.
b
P .001.
c
P .002.
d
P .001.
[ages 40 –59 years], and 14 older in- anterior (approximately 30° of knee was assessed by repeated testing on
dividuals [ages 60 – 80 years]) flexion) radiographs of the knees of a subset of 8 subjects and showed
(Tab. 1). The middle-aged individuals the middle-aged and older control high reliability for medial (intraclass
were matched by age and sex to 15 groups were obtained as an added correlation coefficient [ICC] .96)
people with symptomatic, medial precaution to rule out the presence and lateral laxity (ICC .98).
knee OA (Tab. 1). The subjects with of knee OA. Radiographs were not
medial knee OA were part of a larger taken of the knees of the young sub-
study of people who were going to jects because they had no knee
undergo a high tibial osteotomy. symptoms and no history of knee
They had no history of knee liga- injury and were unlikely to have un-
ment injury; however, those individ- diagnosed knee OA based on the
uals with a history of meniscectomy above definition.
were included. Data on some of
the people with knee OA and data Varus and valgus stress radiographs
on the middle-aged control group were taken of the tested lower ex-
have been reported previously.24 tremity in the middle-aged and older
Radiographic information, isometric control groups as well as the OA
quadriceps femoris strength, and ki- group. Subjects were positioned su-
nematic, kinetic, and electromyo- pine on a radiograph table with the
graphic (EMG) data during walking knee flexed to 20 degrees and the
were collected from the more- patella facing anteriorly. The x-ray
involved limb of the subjects with tube was centered approximately
OA and a randomly chosen limb of 100 cm above the knee joint. A
the control subjects. The test limb of TELOS* stress device was used to ap-
the control subjects was chosen ran- ply a 150-N force in the varus or
domly to avoid any possible influ- valgus direction (Fig. 1). Medial and
ence of limb dominance. lateral joint spaces were measured
at the narrowest location in both
Procedure compartments using calipers. X-ray
Radiographs. The diagnosis of OA beams were adjusted for magnifica-
Figure 1.
is based on the presence of knee tion using a known distance from the Setup for stress radiographs. The top im-
pain in conjunction with age over 50 TELOS device that was visible in ev- ages show the limb alignment in the
years and either radiographic evi- ery image. Medial and lateral joint TELOS device (top left) and the resulting
dence of OA (eg, osteophytes) or laxities were calculated as described radiograph (top right), and the method of
other symptoms such as stiffness or in Figure 1.28 Interrater reliability calculating medial laxity is shown in the
lower images. Lateral laxity was calculated
crepitus.27 Although none of our similarly but with subtraction of the lateral
control subjects complained of knee * Austin & Associates, 1109 Sturbridge Rd, joint space in valgus from lateral joint
pain or stiffness, standing posterior- Fallston, MD 21047. space in varus.
1424 f Physical Therapy Volume 87 Number 11 November 2007

4. Age-Related Changes in Strength, Joint Laxity, and Walking Patterns
Skeletal alignment of the tested limb tion was measured as the highest sampled at 1,920 Hz. Ground reac-
was measured with a standing long volitional force (N) during the con- tion force data were used to calcu-
cassette radiograph for the middle- traction and was normalized to body late moments about the knee and for
aged and older control groups and mass index (BMI) (N/BMI). In tests determination of heel-strike and
the OA group. Subjects stood, with- on 10 subjects who were healthy, toe-off.
out footwear, with the tibial tuber- repeated testing of the MVIC re-
cles facing forward and the x-ray vealed an intraclass correlation coef- Electromyographic data were col-
beam centered at the knee from a ficient (2,1) of .98.33 lected with a 16-channel electromyo-
distance of 2.4 m. Alignment was graphy system (model MA-300-16#)
measured as the angle formed by the Gait and electromyographic sampled at 1,920 Hz. After skin prep-
intersection of the mechanical axes (EMG) data. To determine knee aration, surface electrodes with par-
of the femur and tibia.29 –31 A knee motion during walking, the motions allel, circular detection surfaces
was in varus alignment when the in- of the lower-extremity segments (1.14 cm in diameter, 2.06 cm apart),
tersection of the lines was 0 de- were collected by a 6-camera, pas- a common mode rejection ratio (100
grees in the varus direction and was sive, 3-dimensional motion analysis dB at 65 Hz), and a signal detection
in valgus alignment when the inter- system (Vicon 512)§ at 120 Hz. Cam- range of less than 2 V for the
section of the lines was 0 degrees eras were calibrated to detect mark- built-in preamplifier were placed
in the valgus direction.30 ers within a volume that was 1.5 over the mid-muscle bellies of the
2.4 1.5 m. Calibration residuals lateral quadriceps femoris (LQ), me-
Quadriceps femoris muscles func- were kept below 0.6 mm. The cam- dial quadriceps femoris (MQ), lateral
tion. Quadriceps femoris muscle eras detected retroreflective markers hamstring (LH), medial hamstring
force output was measured with an (2.5 cm in diameter) placed on the (MH), lateral gastrocnemius (LG),
isokinetic dynamometer (Kin-Com tested lower extremity. Markers and medial gastrocnemius (MG) mus-
500H).† Each subject sat with the were placed bilaterally over the cles. Electromyographic data were
knee and hip flexed to 90 degrees, greater trochanters, the lateral femo- recorded for 2 seconds at rest and
the knee joint axis aligned with the ral condyles, and lateral malleoli for during an MVIC for each muscle
dynamometer axis, and the trunk identification of appropriate joint group for normalization.
fully supported. Thigh and hip straps centers. Thermoplastic shells with 4
secured each subject in the seat, rigidly attached markers were used Motion, force, and EMG data were
while an ankle strap secured the to track segment motion. The shells collected simultaneously as subjects
shank to the dynamometer. Subjects were secured on the posterior-lateral walked at a self-selected speed along
performed a maximal volitional iso- aspects of the thigh and shank. Pre- a 9-m walkway for 10 trials. Walking
metric contraction (MVIC) on which vious work in our laboratory (unpub- speed was recorded from 2 photo-
a supramaximal burst of electrical lished data collected April 2002) has electric beams to ensure that speed
current (Grass S48 stimulator‡) (100 revealed good reliability for kine- did not vary more than 5% from their
pulses/second, 600-microsecond matic variables with ICCs ranging self-selected speed during the trials.
pulse duration, 10-pulse tetanic from .6343 to .9969. The ICCs for Trials were only accepted if the sub-
train, 130 V) was applied. The burst the kinematic variables used in the ject walked at a consistent speed and
superimposition was used for the present study ranged from .9721 to walked across the force platform
measurement if the subjects were .9969. Errors in estimating bone without adjusting their stride in any
providing maximum activation of movement from skin mounted mark- way to contact the force platform.
the quadriceps femoris muscles.32 A ers for sagittal and frontal plane mo-
central activation ratio (CAR) is a ra- tions are approximately 2 to 3 de- Data management. Marker tra-
tio between the highest volitional grees during the stance phase of jectories and ground reaction forces
force (measured as the peak force walking.34,35 Vertical, medial-lateral, were collected over the stance phase
before the electrical burst was ap- and anterior-posterior ground reac- of one limb (heel-strike to toe-off
plied) and the force achieved during tion forces were collected from a on the force platform) and were fil-
the electrically elicited burst. Maxi- 6-component force platform (Bertec tered with a second-order, phase-
mum volitional isometric contrac- force platform, model 60905 ) and corrected Butterworth filter with a
cutoff frequency of 6 Hz for the
† Isokinetic International, 6426 Morning Glory video data and 40 Hz for the force-
Dr, Harrison, TN 37341. §Oxford Metrics, 14 Minns Business Park,
‡ Grass Instrument Division, Astro-Med Inc, West Way, Oxford OX2 0JB, United Kingdom.
600 East Greenwich Ave, West Warwick, RI Bertec Corp, 6171 Huntley Rd, Ste J, Colum- # Motion Lab Systems, 15045 Old Hammond
02893. bus, OH 43229 Hwy, Baton Rouge, LA 70816.
November 2007 Volume 87 Number 11 Physical Therapy f 1425

5. Age-Related Changes in Strength, Joint Laxity, and Walking Patterns
plate data. Sagittal- and frontal-plane contraction was operationally de- group (P .030), and subjects in the
knee angles and external knee mo- fined as the simultaneous activation OA group had higher BMI values
ments were calculated with Euler of a pair of opposing muscles and than the subjects in the young and
angles and inverse dynamics, respec- was calculated using an equation de- older control groups (P .002)
tively (Visual 3D**). Data were ana- veloped in our laboratory37: (Tab. 1). The knees of the subjects in
lyzed using custom-written com- the middle-aged and older control
puter programs based on strict Average co-contraction value groups were in less varus than the
criteria (eg, thresholds for initial con- n
knees of the subjects in the OA
tact, time of peak adduction mo- lowerEMGi
lowerEMGi higherEMGi
group (P .001) (Tab. 1).
higherEMGi
ment) to eliminate tester bias. Data i 1
were analyzed during the loading in- n Knee Laxity
terval, which we defined as from 100 Subjects in the OA group had signif-
milliseconds prior to initial contact where i is the sample number and n icantly greater medial laxity than the
(to account for electromechanical is the total number of samples in the subjects in the middle-aged and older
delay)36 through the first peak knee interval. Co-contraction values were control groups (P .001) (Fig. 2).
adduction moment. Data during the averaged across the trials, and the There were no differences in lateral
loading interval were time normal- average was used for analysis. This laxity between the subjects in the
ized to 100 data points and averaged method does not identify which OA group and the subjects in the
across each subject’s trials. Knee mo- muscle is more active; rather, it rep- middle-aged and older control
ments were normalized to body resents a relative activation of 2 mus- groups (P .272) (Fig. 2).
mass height and are expressed as cles while accounting for the magni-
external moments. In addition to dis- tudes of both muscles. Co- Quadriceps Femoris
crete variables, we calculated knee contraction was calculated between Muscle Strength
flexion excursion (from initial con- the LQ and LH (LQH), MQ and MH The subjects in the young control
tact to peak knee flexion) and knee (MQH), LQ and LG (LQG), and MQ group produced greater volitional
adduction excursion during loading. and MG (MQG) muscles. quadriceps femoris muscle force
than the subjects in the older control
All EMG data were band-pass filtered Data Analysis group (P .001) or the subjects in
from 20 to 350 Hz. A linear envelope Group means and standard devia- the OA group (P .001) (Fig. 3). Sub-
was created with full-wave rectifica- tions were calculated for all data. jects in the middle-aged control
tion and filtering with a 20-Hz low- One-way analysis of variance group generated more force than the
pass filter (eighth-order, phase- (ANOVA) was used to detect group subjects in the older control group
corrected Butterworth filter). The differences in BMI, quadriceps fem- (P .002) or the subjects in the OA
linear envelope was normalized to oris muscle strength and CAR, and group (P .003) (Fig. 3). There were
the maximum EMG signal obtained radiograph variables. Because walk- no differences between the sub-
during a MVIC for each muscle. ing speed can influence lower- jects in the older control group and
extremity kinematic and kinetic the subjects in the OA group
Custom-designed software (Labview, data,38 – 40 analysis of covariance (AN- (P 1.0) or between the subjects in
version 8.0††), using the same kine- COVA), with walking velocity as a the young and middle-aged control
matic and kinetic events as stated covariate, was used to detect group groups (P .974) (Fig. 3). No dif-
above, was used to analyze all EMG differences in kinematic and kinetic ferences in CAR were observed
data. Magnitude of muscle activity variables. For strength, radiograph, among the control groups (young
and co-contraction between oppos- and kinematic and kinetic data, sig- 0.93 .038 [X SD], middle-aged
ing muscle groups were analyzed nificance was established when 0.93 .027, older 0.94 .052; P .84).
over the loading interval after it was P .05. To detect group differences
time normalized to 100 points. Mag- in magnitudes of muscle activity and Gait Characteristics
nitude of muscle activity was ex- in co-contraction variables, 95% con- The subjects in the middle-aged con-
pressed as the average rectified value fidence intervals were used to evalu- trol group walked faster than the
across the loading interval. Co- ate differences in the mean values. subjects in the OA group (P .023)
and there were no other statistical
Results differences among the groups
** C-Motion Inc, 15821-A Crabbs Branch Way, Subjects in the middle-aged control (young control group 1.39 0.08
Rockville, MD 20855.
†† National Instruments, 11500 N Mopac group had greater BMI values than [X SD] m/s, middle-aged control
Expwy, Austin, TX 78759-3504. the subjects in the young control group 1.51 0.15 m/s, older con-
1426 f Physical Therapy Volume 87 Number 11 November 2007

6. Age-Related Changes in Strength, Joint Laxity, and Walking Patterns
groups, but was reduced in subjects
in the OA group compared with the
subjects in the young (P .006) and
older (P .039) control groups.
There were no differences in frontal-
plane knee motions or moments
among the young, middle-aged, and
older control groups. The subjects in
the OA group exhibited greater ad-
duction compared with the subjects
in the young, middle-aged, and older
control groups at initial contact
(P .004) and at peak adduction dur-
ing loading (P .001). The OA group
showed greater adduction excursion
compared with the young control
group (P .049) and the older con-
trol group (P .055); however, the
Figure 2. latter was not statistically significant
Medial and lateral joint laxity. MA middle-aged control group, O older control group, at the P .05 level. The OA group
OA group with osteoarthritis. * P .001. Error bars represent standard deviation.
showed greater peak knee adduction
moments compared with the young,
middle-aged, and older control
groups (P .002).
Muscle Activity
There was a large degree of variabil-
ity in the muscle activation and co-
contraction as is evident in the large
range of the 95% confidence inter-
vals shown in Figures 4 and 5. Dur-
ing loading response, there was a
tendency for the subjects in the
older control group to use higher
lateral quadriceps femoris activity
than the subjects in the young and
middle-aged control groups and a
tendency for higher medial gastroc-
nemius muscle activity in the sub-
jects in the OA group and the older
control group than in the subjects in
Figure 3. the young and middle-aged control
Quadriceps femoris muscle force production. Y young control group, MA middle- groups. However, the overlap of the
aged control group, O older control group, OA group with osteoarthritis, 95% confidence intervals indicate
N/BMI highest volitional force during contraction normalized to body mass index. that a larger sample size is needed to
* P .000, † P .000, ‡ P .002, § P .003. Error bars represent standard deviation. determine with more certainty
whether the population means are
trol group 1.45 0.10 m/s, OA older control groups, but the OA different. In terms of muscle co-
group 1.38 0.12 m/s). group showed less knee flexion ex- contraction, there were no differ-
cursion compared with all 3 control ences among the control groups, al-
Results of kinematic and kinetic vari- groups (P .036). The peak knee though the OA group showed higher
ables are shown in Table 2. Knee flexion moment was no different co-contraction than the subjects in
flexion excursion was not different among the subjects in the young, the young control group in the LQG
among the young, middle-aged, and middle-aged, and older control and MQG muscle pairs (Fig. 5).
November 2007 Volume 87 Number 11 Physical Therapy f 1427

7. Age-Related Changes in Strength, Joint Laxity, and Walking Patterns
Table 2.
Mean Values (Adjusted for Walking Speed) and 95% Confidence Interval (in Parentheses) for Sagittal- and Frontal-Plane
Kinematics and Kinetics
Young Middle-aged Older Osteoarthritis Group P
Control Group Control Group Control Group (n 15)
(n 15) (n 15) (n 14)
Kinematics (°)
Sagittal-plane knee angle 4.97 ( 7.76, 2.18) 3.59 ( 6.48, 0.70) 5.56 ( 8.41, 2.70) 4.68 ( 7.50, 1.85) .800
at initial contact
(negative is flexion)
Knee flexion excursion 16.75 (14.42, 19.09) 16.97 (14.55, 19.39) 17.94 (15.55, 20.33) 11.98 (9.62, 14.35) .036a
during loading
Frontal-plane knee angle 0.434 ( 2.5, 1.64) 2.30 ( 4.45, 0.15) 0.833 ( 2.96, 1.29) 4.83 (2.73, 6.93) .004a
at initial contact
(positive is adduction)
Peak frontal-plane knee 2.62 (0.40, 4.84) 2.35 (0.05, 4.65) 2.18 ( 0.10, 4.45) 9.93 (7.68, 12.18) .001a
angle during loading
(positive is adduction)
Knee adduction 3.06b (1.99, 4.12) 4.65 (3.54, 5.76) 3.00c (1.92, 4.10) 5.10b,c (4.02, 6.18) .049b
excursion during .055
loading
Kinetics (N m/kg m)
Peak knee flexion 0.36b ( 0.29, 0.44) 0.27 ( 0.19, 0.35) 0.33c ( 0.25, 0.41) 0.17b,c ( 0.09, 0.25) .006b
moment during .039c
loading
Peak knee adduction 0.28a (0.23, 0.32) 0.33a (0.28, 0.37) 0.26a (0.21, 0.31) 0.45a (0.41, 0.50) .002a
moment during
loading
a
Osteoarthritis group different from all control groups.
b
Osteoarthritis group different than young control group.
c
Osteoarthritis group different than older control group.
Discussion and Conclusions tablish an environment in which OA sample size, these findings suggest
Most studies of age-related differ- could develop. The results set the that the older adults included in this
ences in movement and muscle acti- stage for future research into how study demonstrate movement strate-
vation patterns include samples of age-related musculoskeletal changes gies similar to those of younger indi-
young subjects in their 20s and older might influence the development of viduals, which may have helped
adults over 60 years of age; yet, age- knee OA. to prevent the development of
related changes in characteristics knee OA as they aged; these find-
such as muscle strength or neuro- The results of this study indicate that ings, however, warrant further
muscular responses can occur in healthy aging was associated with a investigation.
middle age7–11 and may coincide considerable loss of quadriceps fem-
with the development of knee OA. oris muscle strength in the older As age-related muscle weakness de-
As a result, we intended to investi- adults, although we did not observe velops, individuals must adapt their
gate characteristics in individuals increased frontal-plane laxity in movements and muscle activity pat-
who are healthy that are purported those subjects. Despite quadriceps terns to accommodate the dimin-
to be associated with the develop- femoris muscle weakness, the older ished force-generating capacity of
ment of knee OA across a range of adults participating in this study did their aging muscles to maintain a cer-
ages, including middle age. The not adopt a knee stiffening strategy tain level of function. As such, we
novel nature of this approach and (ie, reduced knee motion and high propose that adaptations allowing
the findings of this study provide muscle co-contraction) that we spec- for the continuation of normalized
some insights into how changes in ulate may contribute to damage of joint mechanics and muscle activa-
musculoskeletal function might es- articular cartilage. Despite the small tion patterns are less likely to predis-
1428 f Physical Therapy Volume 87 Number 11 November 2007

8. Age-Related Changes in Strength, Joint Laxity, and Walking Patterns
pose the joint to articular cartilage
damage. A failure to adapt to
strength declines might contribute
to the development of movement
patterns similar to individuals with
quadriceps femoris muscle weak-
ness due to knee joint patholo-
gy.41– 43 Because the older adults in
this study exhibited similar move-
ment and muscle activity patterns to
those in the younger age groups, it
appears that they have discovered a
successful approach to maintaining
normal knee function despite their
quadriceps femoris muscle strength
decline.
In particular, the older control sub-
jects exhibited significantly weaker
quadriceps femoris muscles com-
pared to the younger cohorts, yet
they showed no differences in knee Figure 4.
motion during weight acceptance Mean electromyographic (EMG) muscle activation during loading and 95% confidence
compared with the young control interval (indicated by bars). MVIC maximal voluntary isometric contraction,
LQ lateral quadriceps femoris muscle, MQ medial quadriceps femoris muscle,
subjects. Quadriceps femoris muscle LH lateral hamstring muscle, MH medial hamstring muscle, LG lateral gastrocne-
weakness has previously been asso- mius muscle, MG medial gastrocnemius muscle.
ciated with reduced knee motion
during walking in the presence of
joint pathology.33,44 Electromyo-
graphic data suggest that the older
adults in this study have compen-
sated for the quadriceps femoris
muscle weakness by selectively in-
creasing quadriceps activity during
loading. Although adequate muscle
activity is necessary to ensure joint
stability, too much activation can re-
sult in limited knee flexion and in-
creased impact load on the knee.19
Whether increased activation would
be a positive or negative adaptation
during walking, therefore, would de-
pend on the end result of the muscle
activity. The older adults in this
study were able to maintain normal-
ized knee motion, comparable to
younger subjects, with increased
quadriceps femoris activity. The abil-
ity to maintain normalized knee joint
mechanics may have contributed to
the lack of knee OA in this older Figure 5.
Mean muscle co-contraction index during loading and 95% confidence interval (indi-
adult cohort. Our conclusions are cated by bars). LQH lateral quadriceps femoris-lateral hamstring, MQH medial quad-
limited by the cross-sectional design riceps femoris-medial hamstring, LQG lateral quadriceps femoris-lateral gastrocne-
of this study. A longitudinal study mius, and MQG medial quadriceps femoris-medial gastrocnemius muscle pairs.
November 2007 Volume 87 Number 11 Physical Therapy f 1429

9. Age-Related Changes in Strength, Joint Laxity, and Walking Patterns
would be required to further investi- means to maintain normal move- muscle co-contraction strategies.41,43
gate the effect of age-related muscu- ment strategies. Additional research is required to de-
loskeletal changes on movement lineate whether consistent differ-
strategies in terms of the develop- The similarity in the knee motion ences in muscle activation patterns
ment of knee OA. among the control groups might be exist in people with knee OA or
unexpected because other research- whether there are several strategies
Despite prior evidence of reduced ers21 have shown that older adults that people use to help control the
stiffness and ligament strength with walk with less knee motion during knee in the face of a pathologic con-
advancing age,25 we were unable to loading when walking at the same dition. Whether one strategy is more
detect increases in frontal plane lax- speed as younger subjects. It is pos- detrimental than another remains to
ity with aging in the control subjects. sible that our finding of similar be seen and should be further
The OA group, however, exhibited sagittal-plane knee kinematics investigated.
increased frontal-plane laxity. Al- among the young, middle-aged, and
though subjects were carefully older control groups is due to a small There are several limitations to the
screened for a history of ligament number of subjects in our sample or study design that the readers should
injury, we included individuals with related to our method of subject re- take into consideration. First, testers
a history of meniscal damage in the cruitment. Some of the older adults were not blinded to group assign-
OA group. The subjects with OA had in our study were recruited from lo- ment, which may have created bias
no history of an incident ligamentous cal fitness and senior centers and in recording data. However, the use
injury, and studies45,46 suggest that may represent a more active older of the TELOS device to apply uni-
meniscal injury in the absence of a adult compared with a typical older form stress during the stress radio-
traumatic event is a part of the de- adult, and this may have enabled the graphs reduced the influence of
generative process of knee OA. In older subjects in our study to better tester bias for this measure. Testers
individuals with knee OA, the pres- control more knee motion as the attempted to provide equivalent ver-
ence of increased frontal plane laxity limb accepted weight. Future studies bal encouragement to all subjects
is known to degrade the relationship may consider measuring daily activ- equally when testing quadriceps
between strength and physical ity levels to account for potential in- femoris muscle strength. In addition,
function.47 fluences on walking speeds and the discomfort of the superimposed
movement patterns. burst was motivation for all subjects
Because the older adults who were to perform to their best ability to
healthy did not have to cope with It is interesting to note that differ- avoid repeat testing. During the
strength loss in a lax joint, they may ences in muscle co-contraction val- movement analysis testing, similar
have had the ability to adopt move- ues were found only between the instructions were provided to all
ment strategies that remain normal- subjects with OA and young control subjects to walk at a comfortable
ized and may be “joint sparing.” We subjects. In this study, the subjects speed. Custom-written computer al-
can speculate that the subjects with with OA used greater lateral gas- gorithms were used to determine
OA did not have such an option, be- trocnemius muscle activity during data points used in the analysis of
cause they had to contend with loading and greater quadriceps kinematics, kinetics, and EMG data
strength loss in a lax joint, making femoris-gastrocnemius muscle co- to reduce tester bias. Second, the
joint stabilization a primary determi- contraction on the medial and lateral distribution of male and female sub-
nant in their adopted control strat- sides compared with the young con- jects in the groups was not the same
egy. These findings suggest that trol subjects, but no differences and we did not account for the level
quadriceps femoris muscle weak- were observed between the subjects of physical activity in the subjects in
ness is associated with reduced with OA and the middle-aged control each group, both of which could
stance-phase knee motion in the subjects. This is in contrast to other have influenced the results. Finally,
presence of other factors, such as work in our lab in which subjects the subjects all walked faster than
increased knee laxity or pain, as was with OA were found to use higher has been reported elsewhere,48 and
evident in the OA group. Such a co-contraction between the quadri- walking speed—although used as a
speculation would suggest that age- ceps femoris-gastrocnemius muscles covariate—may have influenced the
related changes to musculoskeletal on the medial side only compared results.
tissues alone are insufficient to lead with age-matched control subjects.24
to the development of knee OA, pro- It is possible that people with patho- The finding that the older adults in
vided the aging individual has the logic conditions in the knee may our study used what can be consid-
limit knee motion through different ered a favorable movement pattern
1430 f Physical Therapy Volume 87 Number 11 November 2007

10. Age-Related Changes in Strength, Joint Laxity, and Walking Patterns
may suggest why they did not de- the Science of Joint Preservation Course, Uni- 13 Slemenda C, Brandt KD, Heilman DK, et al.
versity of Louisville, Louisville, Ky, April 28 – Quadriceps weakness and osteoarthritis of
velop knee OA. We speculate that the knee. Ann Intern Med. 1997;127:
29, 2006.
the manner in which middle-aged in- 97–104.
dividuals compensate for age-related The contents of this article are solely the 14 Brandt KD, Heilman DK, Slemenda C, et al.
responsibility of the authors and do not nec- Quadriceps strength in women with radio-
neuromuscular changes might influ- graphically progressive osteoarthritis of
essarily represent the official views of the
ence the future integrity of the the knee and those with stable radio-
NCRR or NIH. graphic changes. J Rheumatol. 1999;26:
knee’s articular cartilage. An alterna- 2431–2437.
tive interpretation of our results is This article was submitted May 12, 2006, and
15 Hurley MV. The role of muscle weakness
was accepted July 11, 2007.
that the process of knee OA may in the pathogenesis of osteoarthritis.
Rheum Dis Clin North Am. 1999;25:283–
cause changes in movement and DOI: 10.2522/ptj.20060137 298, vi.
muscle activation patterns. The ab- 16 Bennell KL, Hinman RS, Metcalf BR. Asso-
sence of reduced knee motion and References ciation of sensorimotor function with
knee joint kinematics during locomotion
higher co-contraction in the middle- 1 Carmona L, Ballina J, Gabriel R, Laffon A; in knee osteoarthritis. Am J Phys Med Re-
EPISER Study Group. The burden of
aged and older control subjects may musculoskeletal diseases in the general
habil. 2004;83:455– 463; quiz 464 – 456,
491.
have been a consequence of not hav- population of Spain: results from a na-
tional survey. Ann Rheum Dis. 2001;60: 17 Perry J. Gait Analysis: Normal and Patho-
ing developed OA rather than the 1040 –1045. logical Function. Thorofare, NJ: Slack;
reason they did not develop OA. Res- 2 Felson DT, Zhang Y, Hannan MT, et al.
1992.
olution of such a question would en- Risk factors for incident radiographic knee 18 Hodge WA, Fijan RS, Carlson KL, et al.
osteoarthritis in the elderly: the Framing- Contact pressures in the human hip joint
tail a large-scale longitudinal study to ham study. Arthritis Rheum. 1997;40: measured in vivo. Proc Natl Acad Sci
track changes in movement and mus- 728 –733. U S A. 1986;83:2879 –2883.
cle activation patterns along with ar- 3 Kannus P, Jarvinen M, Kontiala H, et al. 19 Lafortune MA, Hennig EM, Lake MJ. Dom-
Occurrence of symptomatic knee osteo- inant role of interface over knee angle for
thritic changes in large numbers of arthrosis in rural Finland: a prospective cushioning impact loading and regulating
subjects. follow up study. Ann Rheum Dis. initial leg stiffness. J Biomech. 1996;29:
1987;46:804 – 808. 1523–1529.
4 McAlindon TE, Wilson PW, Aliabadi P, et 20 Cook TM, Farrell KP, Carey IA, et al. Ef-
Dr Rudolph and Dr Schmitt provided writing al. Level of physical activity and the risk of fects of restricted knee flexion and walk-
and project management. Dr Schmitt and Dr radiographic and symptomatic knee osteo- ing speed on the vertical ground reaction
arthritis in the elderly: the Framingham force during gait. J Orthop Sports Phys
Lewek provided data collection. All authors study. Am J Med. 1999;106:151–157. Ther. 1997;25:236 –244.
provided data analysis. Dr Rudolph provided
5 Peat G, McCarney R, Croft P. Knee pain 21 DeVita P, Hortobagyi T. Age causes a re-
fund procurement, facilities/equipment, and and osteoarthritis in older adults: a review distribution of joint torques and powers
institutional liaisons. Dr Lewek provided con- of community burden and current use of during gait. J Appl Physiol. 2000;88:
sultation (including review of manuscript be- primary health care. Ann Rheum Dis. 1804 –1811.
fore submission). The authors acknowledge 2001;60:91–97. 22 Sharma L, Lou C, Felson DT, et al. Laxity in
William Newcomb, MD, for providing sub- 6 Loeser RF, Shakoor N. Aging or osteoar- healthy and osteoarthritic knees. Arthritis
ject recruitment and Laurie Andrews, RTR, thritis: which is the problem? Rheum Dis Rheum. 1999;42:861– 870.
Clin North Am. 2003;29:653– 673. 23 Wada M, Imura S, Baba H, Shimada S. Knee
for assistance with radiographs.
7 Akima H, Kano Y, Enomoto Y, et al. Mus- laxity in patients with osteoarthritis and
This study was approved by the Human cle function in 164 men and women aged rheumatoid arthritis. Br J Rheumatol.
Subjects Review Board of the University 20 – 84 yr. Med Sci Sports Exerc. 2001;33: 1996;35:560 –563.
220 –226. 24 Lewek MD, Rudolph KS, Snyder-Mackler
of Delaware.
8 Frontera WR, Hughes VA, Lutz KJ, Evans L. Control of frontal plane knee laxity dur-
This study was funded, in WJ. A cross-sectional study of muscle ing gait in patients with medial compart-
strength and mass in 45- to 78-yr-old men ment knee osteoarthritis. Osteoarthritis
part, by a grant (P20-
and women. J Appl Physiol. 1991;71: Cartilage. 2004;12:745–751.
RR16458) from the Na- 644 – 650. 25 Noyes FR, Grood ES. The strength of the
tional Center for Research
9 Lindle RS, Metter EJ, Lynch NA, et al. Age anterior cruciate ligament in humans and
Resources (KSR), a component of the Na- and gender comparisons of muscle Rhesus monkeys. J Bone Joint Surg Am.
tional Institutes of Health (NIH); a grant strength in 654 women and men aged 1976;58:1074 –1082.
(T32HD007490) from the NIH (LCS, MDL); a 20 –93 yr. J Appl Physiol. 1997;83: 26 Woo SL, Hollis JM, Adams DJ, et al. Tensile
PODS II grant from the Foundation for Phys- 1581–1587. properties of the human femur-anterior
ical Therapy (LCS, MDL); and a Doctoral Re- 10 Roos MR, Rice CL, Connelly DM, Vander- cruciate ligament-tibia complex: the ef-
voort AA. Quadriceps muscle strength, fects of specimen age and orientation.
search Grant from the American College of
contractile properties, and motor unit fir- Am J Sports Med. 1991;19:217–225.
Sports Medicine (MDL). ing rates in young and old men. Muscle 27 Altman R, Asch E, Bloch D, et al. Develop-
Nerve. 1999;22:1094 –1103. ment of criteria for the classification and
This work was presented as a platform pre-
11 Hurley MV, Rees J, Newham DJ. Quadri- reporting of osteoarthritis: classification of
sentation at the Combined Sections Meeting
ceps function, proprioceptive acuity and osteoarthritis of the knee. Diagnostic and
of the American Physical Therapy Associa- functional performance in healthy young, Therapeutic Criteria Committee of the
tion; February 14, 2003; Tampa, Fla; as a middle-aged and elderly subjects. Age Age- American Rheumatism Association. Ar-
platform presentation at XVth Biannual In- ing. 1998;27:55– 62. thritis Rheum. 1986;29:1039 –1049.
ternational Society of Electrophysiology and 12 Felson DT, Zhang Y. An update on the
Kinesiology Conference; June 20, 2004; Bos- epidemiology of knee and hip osteoarthri-
ton, Mass; and at the Functional Fitness and tis with a view to prevention. Arthritis
Rheum. 1998;41:1343–1355.
November 2007 Volume 87 Number 11 Physical Therapy f 1431

11. Age-Related Changes in Strength, Joint Laxity, and Walking Patterns
28 Moore TM, Meyers MH, Harvey JP Jr. Col- 35 Benoit DL, Ramsey DK, Lamontagne M, et 43 Rudolph KS, Axe MJ, Buchanan TS, et al.
lateral ligament laxity of the knee: long- al. Effect of skin movement artifact on Dynamic stability in the anterior cruciate
term comparison between plateau frac- knee kinematics during gait and cutting ligament deficient knee. Knee Surg Sports
tures and normal. J Bone Joint Surg Am. motions measured in vivo. Gait Posture. Traumatol Arthrosc. 2001;9:62–71.
1976;58:594 –598. 2006;24:152–164. 44 Lewek M, Rudolph K, Axe M, Snyder-
29 Chao EY, Neluheni EV, Hsu RW, Paley D. 36 Vos EJ, Harlaar J, van Ingen Schenau GJ. Mackler L. The effect of insufficient quad-
Biomechanics of malalignment. Orthop Electromechanical delay during knee ex- riceps strength on gait after anterior cru-
Clin North Am. 1994;25:379 –386. tensor contractions. Med Sci Sports Exerc. ciate ligament reconstruction. Clin
1991;23:1187–1193. Biomech (Bristol, Avon). 2002;17:56 – 63.
30 Hsu RW, Himeno S, Coventry MB, Chao
EY. Normal axial alignment of the lower 37 Rudolph KS, Axe MJ, Snyder-Mackler L. 45 Englund M. Meniscal tear: a feature of os-
extremity and load-bearing distribution at Dynamic stability after ACL injury: who teoarthritis. Acta Orthop Scand Suppl.
the knee. Clin Orthop Relat Res. 1990; can hop? Knee Surg Sports Traumatol Ar- 2004;75:1– 45, backcover.
225:215–227. throsc. 2000;8:262–269. 46 Englund M, Roos EM, Lohmander LS. Im-
31 Tetsworth K, Paley D. Malalignment and 38 Andriacchi TP, Ogle JA, Galante JO. Walk- pact of type of meniscal tear on radio-
degenerative arthropathy. Orthop Clin ing speed as a basis for normal and abnor- graphic and symptomatic knee osteo-
North Am. 1994;25:367–377. mal gait measurements. J Biomech. 1977; arthritis: a sixteen-year follow-up of menis-
10:261–268. cectomy with matched controls. Arthritis
32 Kent-Braun JA, Le Blanc R. Quantitation of Rheum. 2003;48:2178 –2187.
central activation failure during maximal 39 Crowinshield RD, Brand RA, Johnston RC.
voluntary contractions in humans. Muscle The effects of walking velocity and age on 47 Sharma L, Hayes KW, Felson DT, et al.
Nerve. 1996;19:861– 869. hip kinematics and kinetics. Clin Orthop Does laxity alter the relationship between
Relat Res. 1978;132:140 –144. strength and physical function in knee
33 Snyder-Mackler L, Binder-Macleod SA, Wil- osteoarthritis? Arthritis Rheum. 1999;42:
liams PR. Fatigability of human quadriceps 40 Kirtley C, Whittle MW, Jefferson RJ. Influ- 25–32.
femoris muscle following anterior cruciate ence of walking speed on gait parameters.
ligament reconstruction. Med Sci Sports J Biomed Eng. 1985;7:282–288. 48 Lopopolo RB, Greco M, Sullivan D, et al.
Exerc. 1993;25:783–789. Effect of therapeutic exercise on gait
41 Chmielewski TL, Rudolph KS, Fitzgerald speed in community-dwelling elderly peo-
34 Reinschmidt C. Three-Dimensional Tibio- GK, et al. Biomechanical evidence sup- ple: a meta-analysis. Phys Ther. 2006;86:
calcaneal and Tibiofemoral Kinematics porting a differential response to acute 520 –540.
During Human Locomotion Measured ACL injury. Clin Biomech (Bristol, Avon).
With External and Bone Markers [doc- 2001;16:586 –591.
toral dissertation]. Calgary, Alberta, Can- 42 Mundermann A, Dyrby CO, Andriacchi TP.
ada: University of Calgary; 1996. Secondary gait changes in patients with
medial compartment knee osteoarthritis:
increased load at the ankle, knee, and hip
during walking. Arthritis Rheum. 2005;
52:2835–2844.
1432 f Physical Therapy Volume 87 Number 11 November 2007