ARTICLE

Neuromuscular Exercise as Treatment of

Degenerative Knee Disease
Eva Ageberg 1 and Ewa M. Roos 2
1
Department of Health Sciences, Lund University, Lund, Sweden; and 2Institute of Sports and Clinical
Downloaded from https://journals.lww.com/acsm-essr by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCywCX1AWnYQp/IlQrHD3fgGGXf0fUkBkas8bzKWjdhSu7k4GMhEZQhWZNCB92sniKa3VIlNEWg== on 06/26/2019

Biomechanics, University of Southern Denmark, Odense, Denmark

AGEBERG, E. and E.M. ROOS. Neuromuscular exercise as treatment of degenerative knee disease. Exerc. Sport Sci. Rev., Vol.
43, No. 1, pp. 14Y22, 2015. Exercise is recommended as first-line treatment of degenerative knee disease. Our hypothesis is that neuro-
muscular exercise is feasible and at least as effective as traditionally used strength or aerobic training but aims to target more closely the
sensorimotor deficiencies and functional instability associated with the degenerative knee disease than traditionally used training methods.
Key Words: exercise therapy, knee joint, osteoarthritis, patient-reported outcomes, performance-based measures

INTRODUCTION disease, from onset of an identifiable risk factor, such as joint

injury, via signs such as pain and a degenerative meniscal tear
The term osteoarthritis (OA) describes a common, age- to radiographically verified stages of OA.
related, heterogeneous group of disorders characterized path- Pain is the chief complaint in patients with degenerative
ologically by focal areas of loss of cartilage in synovial joints, knee disease, but patients also report functional limitations in
associated with varying degrees of osteophytic formation, sub- daily and/or sport and recreation activities, and they are less
chondral bone change, and synovitis (10). The prevalence of physically active than the general population. It is nowadays
OA, however, depends on the definition, with radiographic suggested that both self-reported and objective measures of
case definition representing the highest prevalence estimates function be used to obtain a complete picture of function in
(~53%), whereas self-reported and symptomatic OA defini- patients with OA.
tions have lower but similar estimates (~24%) (25). A combination of patient education, exercise, and weight
OA develops slowly across decades, making it amenable loss is recommended in evidence-based guidelines as first-line
to prevention and early intervention, at least in terms of re- treatment of degenerative knee disease (see, for example, (15)).
ducing symptoms and improving physical function, compa- General exercise, such as aerobic training, and local exercise,
rable to other chronic diseases (36). Radiographically visible such as strength training, show positive effects in terms of re-
structural changes of OA can be viewed as an end result of duced pain and improved physical function (37). Because
a degenerative process starting much earlier. A frequent physical activity also has beneficial effects on general health,
early sign of this process is the degenerative meniscal tear in it’s essential to provide appropriate recommendations regarding
middle-aged individuals (12). In young people, a traumatic exercise for people with degenerative knee disease. Specific
knee injury constitutes the highest risk factor for developing exercises to improve range of motion and flexibility usually are
early-onset OA (22). To acknowledge the slowly developing included in various training programs (15), but there are dif-
disease process, the different definitions used, and the desire ferences between types of exercise with regard to components
to intervene at all stages of the disease, we use the term de- such as goal and principles.
generative knee to encompass both early and late stages of the Available guidelines for aerobic and strengthening exer-
cise, developed by the American Geriatrics Society (2001) for
older people with OA and by the American College of Sports
Address for correspondence: Eva Ageberg, Ph.D., Department of Health Sciences, Lund Medicine for healthy adults (2009), are used currently also
University, P.O. Box 157, SE-221 00 Lund, Sweden (E-mail: eva.ageberg@med.lu.se). for people with different stages of degenerative knee disease.
Accepted for publication: October 2, 2014.
Associate Editor: Walter Herzog, Ph.D. Strength training has formed the cornerstone of specific training
because muscle weakness of the lower extremity is common in
0091-6331/4301/14Y22 people with OA (7), with most research focusing on quadri-
Exercise and Sport Sciences Reviews
Copyright * 2014 by the American College of Sports Medicine ceps strengthening. Findings from recent studies and reviews

14

Copyright © 2014 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
suggest that besides muscle weakness, patients with degenera- structural and symptomatic knee OA, particularly in women
tive knee disease have impairments such as reduced proprio- (7). There still is conflicting evidence for the role of muscle
ceptive acuity and muscle activation deficiency (28). They also strength in OA progression, although higher strength may be
report functional knee instability (16). related to slower progression in women and at the patello-
From this perspective, it can be argued that training pro- femoral joint (7).
grams should address several aspects of sensorimotor function Although research on neural inhibition has focused on
and functional stabilization to improve objective function and the quadriceps, other muscles in the lower extremity, such as
alleviate symptoms. In young people with knee injuries, at high the gastrocnemius, hamstrings, and hip abductors, also are
risk of early-onset knee OA (22), neuromuscular training pro- important for stabilizing the knee joint. For example, weaker
grams have been developed specifically to target the sensori- quadriceps, hamstrings, and hip abductor muscles and poorer
motor deficiencies and functional instability associated with knee joint proprioception are related to a greater func-
knee injury (1). Because people with degenerative knee disease tional decline in OA progression (7). Furthermore, not only
have similar deficiencies in sensorimotor function and also per- strength but also muscle activation patterns and coordinated
ceive functional instability, one could assume that they would timing of muscles during movements influence knee joint
benefit from this form of exercise therapy. load (7).
Our hypothesis is that neuromuscular exercise is feasible The understanding of the impairments present at all levels
and at least as effective in reducing symptoms and improving of the sensorimotor system led to the development of neu-
physical function as traditionally used strength or aerobic train- romuscular training programs in patients with ACL injury
ing but aims to target more closely the sensorimotor deficiencies (1) and meniscal damage (13), with the aim of improving
and functional instability associated with the degenerative knee sensorimotor control and achieving functional stability.
disease than traditionally used training methods. Neuromuscular training programs are found effective in im-
proving function and reducing symptoms in people with
RATIONALE FOR NEUROMUSCULAR EXERCISE knee injury (40) at high risk of early-onset OA. To reduce
the number of severe traumatic knee injuries, there have
Underlying Mechanisms for Sensorimotor Deficiency been an emerging number of studies the past decade on
prevention of such injuries. Today, there is strong evidence
The sensorimotor deficiencies observed in young and
from well-designed, randomized, controlled trials that neu-
middle-aged people at high risk of degenerative knee disease
romuscular training programs are effective in reducing knee
(1,13) and in those with degenerative knee disease (7,28)
injuries (19). Recently, we found exercise according to the
suggest that impairments are present at different levels of the
principles of neuromuscular training to be feasible and to
sensorimotor system V from sensory input through integra-
relieve symptoms in people with degenerative knee disease
tion and processing of information in the central nervous
(2,3,29,38), although there is today not enough knowledge
system to motor output to perform voluntary movements and
to indicate whether neuromuscular training is sufficiently
maintain postural control. This sensorimotor dysfunction also
effective in the prevention and treatment of structural signs
may play a role in the development and progression of de-
of OA (Fig. 1).
generative knee disease.
Rice and McNair (26) summarize the neural mechanism
involved in inhibition of the quadriceps muscle in different Relating Different Types of Exercise to Available
knee joint pathologies, such as anterior cruciate ligament Definitions of Physical Activity
(ACL) injury, meniscal damage, and OA. Neural inhibition We could relate current exercise forms that are used for
caused by factors such as pain, swelling, inflammation, joint people with degenerative knee disease to available definitions
laxity, and damage to sensory receptors in the joint prevents of physical activity, physical fitness, exercise, and exercise
the muscle to be activated fully likely through altered excit- therapy, respectively, as follows: according to the World
ability of spinal and supraspinal pathways (26). This inabi- Health Organization, physical activity is defined as ‘‘any bodily
lity to activate voluntarily a muscle completely (arthrogenic movement produced by skeletal muscles that requires energy
muscle inhibition) is considered a potential contributing factor expenditure’’; physical fitness is the characteristic people have
to the quadriceps weakness associated with ACL injury (24), or can achieve through physical activity or exercise (9); ex-
and weak knee extensor strength may increase the risk of ercise is a Medical Subject Headings (MeSH) term defined as

Figure 1. Neuromuscular training as prevention and treatment of knee injury and degenerative joint disease with the end-stage knee osteoarthritis (OA).

Volume 43 & Number 1 & January 2015 Neuromuscular Exercise and Knee 15

Copyright © 2014 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
 two extra repetitions to the last set,
‘‘physical activity which is usually regular and done with the

Level and Progression of Training

control and quality of movement.

Low to moderate (corresponding to

of, direction, and velocity of the

the load is increased at the next

changing support surface and/or
intention of improving or maintaining physical fitness or

1 RM, ‘‘plus-two-rule’’, i.e., if the

Guided by patient’s sensorimotor

Progression by varying number

using unexpected movements

patient is able to add at least

movements; increasing load;
health’’; and exercise therapy is a MeSH term defined as ‘‘a

12Y14 according to Borg’s

regimen or plan of physical activities designed and pre-
scribed for specific therapeutic goals. Its purpose is to restore
normal musculoskeletal function or to reduce pain caused by

open and/or closed kinetic chains training session.

diseases or injuries.’’ For a person with or at high risk of knee

RPE scale)
injury/OA, aerobic training may be regarded as a form of
physical activity, strength training as a form of exercise, and
neuromuscular training as exercise therapy. More specifically,
in this context, a person with OA walking performs aerobic
training (defined as physical activity when walking at slow

achieve the desired requirement

Walking, cycling, aquatic exercise

one or more muscle groups in
speed and defined as exercise when the intention is to im-

Muscle-strengthening exercises
performed in closed kinetic

weight machines involving

(lying, sitting, standing) to
chains in various positions
and muscle groups, mostly
prove physical fitness), doing exercises in weight machines

Type of Exercises

using free weights and/or

neuromuscular principles, involving multiple joints
performs strength training (exercise), and performing spe-
cific exercises aiming at relearning stair climbing with good

of postural activity
Functional exercises
sensorimotor control and without pain from the joint per-

TABLE 1. Neuromuscular exercise compared with strength training and aerobic exercise
forms neuromuscular training (exercise therapy).
In Table 1, the differences in such as goals, principles, and
structure between neuromuscular exercise, strength training,
and aerobic exercise are given. In the following sections,
focus will be on the sensorimotor deficiency observed in

and functional limitations activation patterns, coordination, patients with knee injury
specifically developed for

according to guidelines

people with or without

Principles of Exercise
people with or at high risk of degenerative knee disease,

to guidelines for older

2Y3 times per week,

per week, according

20Y30 min, 2Y5 times
serving as the basis for using neuromuscular exercise as

65%Y70% of 1 RM,

for healthy adults

Sensorimotor deficiencies, Postural control (postural stability Biomechanical and
treatment.

chronic disease
Sensorimotor Deficiency in People at High Risk
of Degenerative Knee Disease
Sensorimotor Deficiency Target(s)

and orientation), proprioception,

Knee injury is the major risk factor for early-onset OA

in the young to middle-aged people, ‘‘Young patients with old

Not specified, aims to target

muscle weakness, muscle

knees’’ (22). Approximately 50% of these individuals have

functional performance

OA after 10 to 15 yr, that is, when they are approximately

30 to 40 yr old and still have high demands on the joints Muscle weakness

general fitness
for work and recreation requiring physical activity. Particular
focus is on severe traumatic knee injuries because such inju-
ries often lead to several months of absence from physical
activity and requires long-term exercise therapy with or with-
out additional surgical treatment (17). The knowledge on
the sensory role of ligaments reported in the 1990s led to a
functional instability),

shift from strength-based training programs to neuromuscular

Aerobic exercise Improve cardiovascular Low physical activity

training programs for patients with knee injury. It was found

Rationale

symptoms (pain,

Muscle weakness

that the ligaments not only are crucial for passive joint stabi-
lity but also are involved in functional joint stabilization. Thus,
not only the mechanical aspects but also sensorimotor in-
tegration through motor learning were thought to be important
RPE indicates rating of perceived exertion.

in training to effectively use afferent neural input and improve

sensorimotor function.
Improve sensorimotor
control and obtain

Sensorimotor deficiency in terms of reduced strength of

and muscle mass
functional joint

Strength training Increase strength

muscles in the lower extremity, reduced functional perfor-

stabilization
Goal

mance, altered muscle activation patterns and timing, pro-

prioceptive deficiency, and impaired postural control, as well
fitness

as perceived functional instability, were reported in several

studies (1). Sensorimotor deficiencies also were found in the
noninjured leg compared with controls possibly because of
Exercise Type

Neuromuscular

factors such as physical inactivity after the injury, inherently

exercise

poor function, and/or disturbed sensory feedback from the

injured joint with an inhibitory effect of muscle activation
also on the noninjured side. Neuromuscular training programs

16 Exercise and Sport Sciences Reviews www.acsm-essr.org

Copyright © 2014 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
are nowadays integrated and successfully used in the treat- Neuromuscular exercises for the lower extremities typically
ment of knee injuries (40). involve multiple joints and muscle groups performed in func-
tional weight-bearing positions. Emphasis is on the quality and
Sensorimotor Deficiency in People with Degenerative efficiency of movement, as well as alignment of the trunk and
Knee Disease lower limb joints.
Patients with degenerative knee disease have sensorimotor The neuromuscular training method that we describe in
deficiencies in terms of sensory dysfunction (28), lower limb this review is based on biomechanical and neuromuscular
muscle weakness (7), altered muscle activation patterns (7), principles and aims to improve sensorimotor control and
and reduced functional performance (28). These sensorimo- achieve compensatory functional stability (Table 1). Aerobic
tor deficiencies are reported not only for the affected leg and strengthening exercise for people with OA follow guide-
compared with the unaffected leg but also compared with a lines for healthy or older people, while the neuromuscular
reference group from the population (3). Given that these training method was designed specifically to target the sen-
sensorimotor deficiencies are comparable to those observed in sorimotor deficiencies and functional instability associated
young people with knee injury, we applied the principles of with knee injury (1,13). Sensorimotor control (also called
neuromuscular training to middle-aged people with degener- neuromuscular control) is the ability to produce controlled
ative knee disease (13,27,30Y32) and to older people with movement through coordinated muscle activity, and func-
established OA (2,3,6,29,39). tional stability (also called dynamic stability) is the ability of the
The functional instability of the knee that many patients joint to remain stable during physical activity. The neuro-
with OA perceive limits their ability to perform functional muscular training method was evaluated first in younger pa-
tasks (16). In Figure 2, we provide novel data from several tients with ACL injury, summarized in a review (1), and later
of our cohorts, suggesting that patients with a degenerative also in middle-aged patients with meniscectomy (13,27). The
meniscal tear or moderate to severe OA perceive functional biomechanical and neuromuscular principles have been de-
knee instability corresponding to that reported by patients scribed in detail previously (1,13). These principles also apply
with ACL injury. Hallmarks of ACL rupture are increased to other knee injuries/diseases and to other joints in the lower
mechanical sagittal instability and functional instability (give extremities because the training aims at resembling conditions
way) during activity (Fig. 2A). From our preexercise- in daily life and more strenuous activities (1).
postexercise therapy data, it seems like neuromuscular train- In summary, the principles include the following: Active
ing can lead to improvement in knee confidence (Fig. 2B). movements in synergies of all the joints in the injured extremity
are included. The movements start with the uninjured ex-
tremity initiating the normal movement and applying bilateral
PRINCIPLES OF NEUROMUSCULAR EXERCISE transfer effect of motor learning to the injured leg. To improve
sensorimotor control, exercises are performed mainly in closed
There is no uniform definition of neuromuscular exercise, kinetic chains in different positions (e.g., lying, sitting, stand-
and other terms such as functional exercise and propriocep- ing) with the intention to obtain low, evenly distributed
tive, agility, or perturbation training are used in the literature. articular surface pressure by muscular coactivation. The model

Figure 2. A. Knee confidence (item Q3 from the knee-related quality of life subscale in the Knee injury and Osteoarthritis Outcome Score Questionnaire)
reported by patients with anterior cruciate ligament (ACL) injury (17), severe primary knee osteoarthritis (OA) (3), degenerative meniscus tear (30), medial
knee OA (6), and meniscectomy (13,27) at baseline before exercise therapy. The patients with severe primary OA were all on the waiting list for total knee
replacement (3), patients with degenerative meniscus tear were all under consideration for arthroscopic surgery (30), patients with medial knee OA were
recruited among community volunteers via advertisements (6), and patients with meniscectomy were recruited through the surgical code system (13,27).
Thus, the different inclusion criteria in the studies may affect their knee symptoms and knee confidence. B. Self-reported knee confidence reported by
patients at baseline and follow-up (Fup) after 12 wk (3,6,30) or 16 wk (13,27) of neuromuscular exercise.

Volume 43 & Number 1 & January 2015 Neuromuscular Exercise and Knee 17

Copyright © 2014 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
 patients are told that pain is allowed up to 5 on a 0-to-10 scale
during and after the training session. They also are told that,
the day after training, pain should subside to ‘‘pain as usual.’’
If pain does not subside, the level of training is reduced. This
pain-monitoring system is part of the NEMEX-TJR concept
as described (see Additional file 1 in Ageberg et al. (2)).
We have named the neuromuscular training method
NEuroMuscular EXercise (NEMEX). A suffix is added to in-
dicate the group of patients to which that program applies;
for example, NEMEX-TJR, where TJR stands for total joint
replacement (2).

NEUROMUSCULAR TRAINING PROGRAMS

The principles of NEMEX have been applied to middle-

Figure 3. A. ‘‘Knee over toe position,’’ that is, joints in lower extremity aged people with degenerative knee disease (13,18,31) and for
well aligned with appropriate position of knee over foot. B. ‘‘Knee medial older people with end-stage OA (3,6,39). The complete
to foot position,’’ that is, joints in lower extremity not aligned; knee is NEMEX-TJR training program is given in the Additional file
placed inappropriately medial to foot. in Ageberg et al. (2), the NEXA training program in the
Additional file in Bennell et al. (5), the neuromuscular and
strength training program in the Appendix in Stensrud et al.
(31), and the ALIGN program in Table 1 in Hall et al. (18).
emphasizes the enhancement of antigravity postural functions Examples of exercises with increasing level of difficulty are
of weight-bearing muscles and the provocation of postural described and illustrated in Table 2.
reactions (feed-forward and feedback control) in the injured
leg by using voluntary movements in the other lower ex-
tremity, trunk, and arms. The quality of the performance in each FEASIBILITY AND EFFECTS OF NEUROMUSCULAR
exercise with an appropriate position of the joints in relation EXERCISE
to each other (postural orientation), that is, with the hip,
knee, and foot well aligned, is emphasized (Fig. 3). Feasibility
The goal is to obtain equilibrium of loaded segments in Two studies have reported on the feasibility of NEMEX
static and dynamic situations and acquire postural control (2,29). Seventy-six patients (mean, 69 yr; SD, 4.3) with se-
in situations resembling conditions of daily life and more vere primary knee (n = 38) or hip (n = 38) OA underwent
demanding activities. Emphasis is put on efficiency and supervised neuromuscular training (NEMEX-TJR) in groups
quality of movements of each exercise. Several aspects of sen- with individualized level and progression of training for a
sorimotor function, such as strength, coordination, balance, median of 11 wk before they had a total joint replacement.
and proprioception, are included in the exercises, but focus The training program was found feasible in terms of reduced
can be, for example, balance in one exercise and strength in or unchanged pain during the training period, few joint-
another. To achieve the desired requirement of postural activ- specific adverse events, and achieved progression of training
ity, patients perform exercises in various positions, that is, lying, level during the training period (2). Reductions in pain also
sitting, and standing. were observed in a pilot study including 27 patients (mean
The training is individualized because symptoms and age, 59 yr; range, 56Y65 yr) with clinical knee OA who had
functional limitations are heterogeneous in people with an two sessions of patient education and 6 wk of NEMEX-TJR
injury or disease. The level of training and progression is (29). In succession of the second pilot study and the effect
guided by the patient’s sensorimotor function, taking into studies described below, physical therapists in Denmark cur-
account various factors related to the individual (e.g., symp- rently are educated in delivering NEMEX for patients with
toms, age, gender, previous and target activity level) and the knee and hip OA. In the first 15 months of the program (Good
injury/disease (e.g., affected joint structures and type and se- Life with Arthritis in Denmark, GLA:D), more than 200
verity of injury/disease). Progression is provided by varying physical therapists have learned the concept and trained more
the number of, direction, and velocity of the movements; than 1700 patients in clinical practice (www.glaid.dk). The
increasing the load; changing the support surface; and/or results of these studies indicate that neuromuscular training is
using unexpected movements. feasible in patients at different stages of degenerative knee
Training may take place in groups to use the positive effects disease, and that it can be taught and implemented nationwide
of group training in terms of more effective learning compared for patients with degenerative knee and hip disease.
with individual practice sessions. To ensure an appropriate
level and progression, training takes place under the supervi- Effects on Sensorimotor Function
sion of experienced physical therapists specializing in training In a first controlled before-and-after study, 87 patients
of musculoskeletal disorders. with severe knee (n = 49) or hip OA (n = 38) had neuro-
Because pain is a major symptom for patients with knee muscular training (NEMEX-TJR) for a mean of 12 wk be-
OA, we use a scale for monitoring pain during training. The fore they underwent TJR (3). The improvements obtained in

18 Exercise and Sport Sciences Reviews www.acsm-essr.org

Copyright © 2014 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
 TABLE 2. Examples of exercises adapted from neuromuscular training of knee bends in 30 s), and in muscle power measures (knee
programs (4,35,40), here shown with an increasing level extension, hip extension, hip abduction, multijoint leg
of difficulty (from A to C). In all exercises, patients are
extension), but only the improvement for hip abduction
encouraged to maintain a knee-over-foot position and
to perform each exercise with good quality.
was large enough to show differences compared with the
control group (data provided in supplementary Table S1 in
Villadsen et al. (39)).
In a small RCT on middle-aged patients with meniscectomy
(13,27), improvements were observed for hop performance
(single-leg hop for distance), quadriceps endurance, and ham-
string muscle strength after 16 wk of neuromuscular training
compared with a control group without any intervention (13).
However, no between-group differences were noted for ham-
string muscle endurance or quadriceps strength. In a recent
report on 20 middle-aged patients with degenerative meniscus
tears included in an ongoing RCT (31), improvements were
noted in both functional performance (single-leg hop for
distance, 6-m timed hop, maximum number of knee bends in
30 s) and quadriceps and hamstring muscle strength by 12 wk
of neuromuscular exercise (31). The same method for
assessing knee muscle strength (isokinetic peak torque five
repetitions at 60 degrees per second) was used in these two
studies (13,31). Although both studies included supervised
neuromuscular training with an individualized level and pro-
gression (13,31), the training program in the study by Stensrud
et al. (31) also added single-leg concentric and eccentric ex-
ercises in both weight-bearing and nonYweight-bearing posi-
tions in weight training machines. The addition of strength
exercises to the neuromuscular training program may explain
that Stensrud et al. (31) found the improvements in both
quadriceps and hamstring muscle strength.
There currently is not enough data to indicate whether one
exercise type is more beneficial than the other for people with
degenerative knee disease (37). In a recent RCT comparing
neuromuscular exercise (NEXA) with quadriceps strength
(QS) (6), 12 wk of exercise provided similar improvements
in both groups in knee and hip muscle strength and in
performance-based measures (timed stair climb, sit-to-stand
In all exercises, patients are encouraged to maintain a knee-over-foot po- test, balance test, step test, and four square step test). How-
sition and to perform each exercise with good quality. ever, it could be that different exercises target different
symptoms and functional limitations. A secondary analysis
from this trial showed that NEXA resulted in greater pain
reduction than QS in patients with varus thrust, visually ob-
patient-reported outcomes (Knee injury and Osteoarthritis served during walking, whereas effects were opposite for
Outcome Score/Hip disability and Osteoarthritis Outcome nonthrusters. This suggests that neuromuscular exercise may
Score (KOOS/HOOS)), performance-based measures (chair be the best type of exercise for pain relief in those with varus
stands, 20-m walk test), and knee extensor strength (3) in- thrust (4).
dicated that effects from neuromuscular training could be seen
even in patients with severe OA about to have TJR. The EVALUATION TOOLS AND OUTCOMES IN
feasibility of 8 wk of NEMEX-TJR was confirmed recently in a NEUROMUSCULAR EXERCISE TRIALS
randomized controlled trial (RCT) in 164 patients before
they had TJR (38). Immediately after the intervention, the 81 From the patient’s perspective, it is most important to
patients with knee disease who underwent neuromuscular evaluate the effect from exercise on perceived pain and
exercise therapy improved in all self-reported outcomes, in functional limitations. Patients with degenerative joint dis-
most performance-based measures (chair stands, 20-m walk ease who seek medical care most often do so because of pain
self-chosen speed, number of knee bends every 30 s), and in and other symptoms. Commonly, they also report difficulty
muscle power, measuring both force and velocity (e.g., single- with physical function, and objective testing reveals impair-
joint hip abduction) compared with the control group re- ments related to muscle strength and sensorimotor control,
ceiving an educational package (care-as-usual) (38). At resulting in activity limitations measured as decreased walking
3 months after surgery, improvements were maintained in speed and worse performance during functional tests. Meta-
most functional tasks (chair stands, 20-m walk test, number analyses including data from patients with knee OA included

Volume 43 & Number 1 & January 2015 Neuromuscular Exercise and Knee 19

Copyright © 2014 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
in exercise trials reveal a moderate effect on pain and func- that is inversely related to the glucosaminoglycan content of
tional limitations, an effect size that was established already the cartilage. The improvement in cartilage matrix quality was
in 2004 and practically has been unchanged by additional paralleled by improved functional performance, but no im-
studies since (37). The effect from neuromuscular exercise provements in quadriceps strength or aerobic capacity were
on pain and function (3,6,39) is comparable to the effects detected (13). These findings have two important implica-
seen from other forms of exercise. tions, the first being that neuromuscular exercise seems to im-
Functional tests are recommended to assess more complex prove function through mechanisms other than those
activities of daily living and sport in those with or at risk of commonly addressed with strength training and aerobic exer-
degenerative knee disease (11,21). Functional tasks measured cise, and the second being that neuromuscular exercise seems
in distance, height, or frequency usually are used. However, to improve cartilage quality. In an 11-yr follow-up of the same
the quality of movement also may be important such as ob- subjects, it was reported that a worse dGEMRIC value at
serving and scoring the position of the knee relative to the baseline was associated with greater radiographic joint space
foot during the performance of functional tasks. Results in- narrowing at follow-up (23), indicating dGEMRIC being an
dicate that tests of movement quality can be used to dis- important evaluation method in those at risk of future OA. It
criminate healthy versus those with a knee injury (34) and also was found that greater quadriceps strength at baseline
that such assessments seem to measure aspects of sensorimotor (before exercise) protected from development of radiographic
function that are not captured in the more commonly used OA (14).
measures of function and strength (35). However, it is still A noninvasive evaluation method is three-dimensional
unknown to what extent movement quality performance is movement analysis, including the use of reflective markers
responsive to change from neuromuscular exercise. and an 8 to 16 infrared camera system coupled with one or
On a population level, the correlation between structural several force platforms. Inverse dynamics is used to calculate
changes of degenerative knee disease (meniscal tears on moments of all three movement planes related to the knee.
magnetic resonance imaging (MRI) and radiographic changes The most commonly used outcome is the knee adduction
of OA) and perceived symptoms is poor (12,25), indicating moment (KAM), a measure of mediolateral force distribution
structural changes being of no or little importance to patients’ across the knee that has been shown to predict radiographic
perceptions. Despite the correlation to symptoms improving OA development. In a first uncontrolled pilot study, 8 wk
with increasing degree of radiographic disease, interestingly of neuromuscular exercise was found to decrease KAM by
and importantly, the degree of radiographic changes of OA 13% in middle-aged patients with mild radiographic OA (33).
was found not to impact on the effect seen from exercise, A later RCT comparing the effect from 8 wk of neuromus-
indicating patients benefitting to a similar degree from exer- cular exercise and quadriceps strengthening in patients with
cise, despite the severity of radiographic changes (20). Taken severe OA and malalignment found similar improvements in
together, from the patients’ perspective, it is most relevant to self-reported pain and function but failed to show any differ-
assess the impact from exercise on pain and function. ence in KAM between groups or any improvements in KAM
From a disease modification perspective, it is however of within groups (6). Recent studies suggest extensor and flexor
interest to evaluate the effect of exercise on joint structure. moments being of, at least, similar importance as the KAM
This approach is associated with a number of methodological when used for evaluation of treatment effects from biome-
difficulties, and no study has this far been able to combine chanical interventions in knee OA (8).
successfully a rigorous study design with a large-enough
number of patients, sufficient adherence to exercise, a long-
enough follow-up time, and a sensitive outcome method to SUMMARY AND FUTURE DIRECTIONS
assess structural changes of the joint. The possible effect of
exercise on joint structure is, therefore, still not well under- In this review, we argue that neuromuscular exercise is
stood. The few studies that have attempted to research the feasible and at least as effective as traditionally used strength
question have showed either no effect or positive effect using or aerobic exercise for people with degenerative knee disease
a variety of outcomes. A few studies have addressed the and should, therefore, be part of the physical rehabilitation for
question in relation to neuromuscular exercise. The hypoth- this population. Data also suggest that neuromuscular training
esized underlying mechanism would be that neuromuscular targets the functional knee instability that many patients
exercise improves functional (dynamic) stability and, thus, perceive. The neuromuscular training method that we de-
contributes to a reduced or more evenly distributed joint load scribe is based on biomechanical and neuromuscular princi-
across the weight-bearing surfaces. ples and aims to improve sensorimotor control and achieve
A first RCT in middle-aged patients having had a partial compensatory functional stability. The training method was
medial meniscectomy, and, therefore, at high risk of future OA, developed specifically to target the sensorimotor deficiencies
found that the cartilage matrix quality was improved signifi- and functional instability associated with knee injury in
cantly after four months of supervised neuromuscular exer- young people. Because people with degenerative knee disease
cise compared with a no-intervention control group (27). In have similar deficiencies in sensorimotor function and per-
this study, delayed gadolinium-enhanced MRI of the ceive functional instability, neuromuscular training also may
cartilage (dGEMRIC) was used to assess indirectly the be a relevant exercise therapy for them. We have applied the
glucosaminoglycan content of the femoral weight-bearing car- principles of neuromuscular training to middle-aged people
tilage. The patient is injected with gadolinium 2 h before MRI, with degenerative knee disease and to older people with es-
and the dGEMRIC value is a function of the relaxation time tablished OA, with promising results with regard to reduced

20 Exercise and Sport Sciences Reviews www.acsm-essr.org

Copyright © 2014 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
pain and improvements in self-reported outcomes, sensori- 16. Fitzgerald GK, Piva SR, Irrgang JJ. Reports of joint instability in knee
motor function, and functional stability. However, so far, nei- osteoarthritis: its prevalence and relationship to physical function. Ar-
thritis Rheum. 2004; 51(6):941Y6.
ther neuromuscular nor strengthening exercises seem sensitive 17. Frobell RB, Roos EM, Roos HP, Ranstam J, Lohmander LS. A random-
to changes in lateral-to-medial knee joint load. A challenge ized trial of treatment for acute anterior cruciate ligament tears. N. Engl.
for future studies is to explore the best type of exercise ther- J. Med. 2010; 363(4):331Y42.
apy for improving sensorimotor function, alleviating symptoms, 18. Hall M, Hinman RS, Wrigley TV, et al. The effects of neuromuscular
and possibly slowing the disease process in different subgroups exercise on medial knee joint load post-arthroscopic partial medial
meniscectomy: SCOPEX, a randomised control trial protocol. BMC
of patients with degenerative knee disease. Musculoskelet. Disord. 2012; 13233.
19. Hubscher M, Zech A, Pfeifer K, Hansel F, Vogt L, Banzer W. Neuro-
Acknowledgments muscular training for sports injury prevention: a systematic review. Med.
Sci. Sports Exerc. 2010; 42(3):413Y21.
E. Ageberg is supported by grants from the Swedish Research Council 20. Juhl C, Christensen R, Roos EM, Zhang W, Lund H. Impact of exercise
(2009-1447), the Crafoord Foundation, the Faculty of Medicine, Lund Uni- type and dose on pain and disability in knee osteoarthritis: a systematic
versity, Skåne Regional Council, and the Swedish Rheumatism Association. review and meta-regression analysis of randomized controlled trials. Ar-
Conflicts of interest: None declared. thritis Rheum. 2014; 66(3):622Y36.
21. Kroman SL, Roos EM, Bennell KL, Hinman RS, Dobson F. Measurement
properties of performance-based outcome measures to assess physical
function in young and middle-aged people known to be at high risk of hip
References and/or knee osteoarthritis: a systematic review. Osteoarthritis Cartilage.
2014; 22(1):26Y39.
1. Ageberg E. Consequences of a ligament injury on neuromuscular function 22. Lohmander LS, Englund PM, Dahl LL, Roos EM. The long-term conse-
and relevance to rehabilitation-using the anterior cruciate ligament- quence of anterior cruciate ligament and meniscus injuries: osteoarthritis.
injured knee as model. J. Electromyogr. Kinesiol. 2002; 12(3):205Y12. Am. J. Sports Med. 2007; 35(10):1756Y69.
2. Ageberg E, Link A, Roos EM. Feasibility of neuromuscular training 23. Owman H, Ericsson YB, Englund M, et al. Association between delayed
in patients with severe hip or knee OA: the individualized goal- gadolinium-enhanced MRI of cartilage (dGEMRIC) and joint space
based NEMEX-TJR training program. BMC Musculoskelet. Disord. narrowing and osteophytes: a cohort study in patients with partial
2010; 11126. meniscectomy with 11 years of follow-up. Osteoarthritis Cartilage. 2014;
3. Ageberg E, Nilsdotter A, Kosek E, Roos EM. Effects of neuromuscular 22(10):1537Y41.
training (NEMEX-TJR) on patient-reported outcomes and physical 24. Palmieri-Smith RM, Thomas AC. A neuromuscular mechanism of
function in severe primary hip or knee osteoarthritis: a controlled before- posttraumatic osteoarthritis associated with ACL injury. Exerc. Sport Sci.
and-after study. BMC Musculoskelet. Disord. 2013; 14(1):232. Rev. 2009; 37(3):147Y53.
4. Bennell K, Egerton T, Wrigley T, et al. Type of exercise and presence 25. Pereira D, Peleteiro B, Araujo J, Branco J, Santos RA, Ramos E. The
of varus thrust influences pain outcomes in people with medial knee effect of osteoarthritis definition on prevalence and incidence estimates: a
osteoarthritis. J. Sci. Med. Sport. 2013; 16(Suppl. 1):e89Ye90. systematic review. Osteoarthritis Cartilage. 2011; 19(11):1270Y85.
5. Bennell KL, Egerton T, Wrigley TV, et al. Comparison of neuromuscu- 26. Rice DA, McNair PJ. Quadriceps arthrogenic muscle inhibition: neural
lar and quadriceps strengthening exercise in the treatment of varus mechanisms and treatment perspectives. Semin. Arthritis Rheum. 2010;
malaligned knees with medial knee osteoarthritis: a randomised con- 40(3):250Y66.
trolled trial protocol. BMC Musculoskelet. Disord. 2011; 12276. 27. Roos EM, Dahlberg L. Positive effects of moderate exercise on glycos-
6. Bennell KL, Kyriakides M, Metcalf B, et al. Neuromuscular versus aminoglycan content in knee cartilage: a four-month, randomized, con-
quadriceps strengthening exercise in patients with medial knee osteoar- trolled trial in patients at risk of osteoarthritis. Arthritis Rheum. 2005;
thritis and varus malalignment: a randomized controlled trial. Arthr. 52(11):3507Y14.
Rheumatol. 2014; 66(4):950Y9. 28. Roos EM, Herzog W, Block JA, Bennell KL. Muscle weakness, afferent
7. Bennell KL, Wrigley TV, Hunt MA, Lim BW, Hinman RS. Update on sensory dysfunction and exercise in knee osteoarthritis. Nat. Rev.
the role of muscle in the genesis and management of knee osteoarthritis. Rheumatol. 2011; 7(1):57Y63.
Rheum. Dis. Clin. North Am. 2013; 39(1):145Y76. 29. Skou ST, Odgaard A, Rasmussen JO, Roos EM. Group education and
8. Boyer KA, Angst MS, Asay J, Giori NJ, Andriacchi TP. Sensitivity of exercise is feasible in knee and hip osteoarthritis. Danish Med. J. 2012;
gait parameters to the effects of anti-inflammatory and opioid treatments 59(12):A4554.
in knee osteoarthritis patients. J. Orthop. Res. 2012; 30(7):1118Y24. 30. Stensrud S, Risberg MA, Roos EM. Knee function and knee muscle
9. Caspersen CJ, Powell KE, Christenson GM. Physical activity, exercise, strength in middle-aged patients with degenerative meniscal tears eli-
and physical fitness: definitions and distinctions for health-related re- gible for arthroscopic partial meniscectomy. Br. J. Sports Med. 2014;
search. Public Health Rep. 1985; 100(2):126Y31. 48(9):784Y8.
10. Dieppe PA, Lohmander LS. Pathogenesis and management of pain in 31. Stensrud S, Roos EM, Risberg MA. A 12-week exercise therapy program
osteoarthritis. Lancet. 2005; 365(9463):965Y73. in middle-aged patients with degenerative meniscus tears: a case series
11. Dobson F, Hinman RS, Hall M, Terwee CB, Roos EM, Bennell KL. with 1-year follow-up. J. Orthop. Sports Phys. Ther. 2012; 42(11):919Y31.
Measurement properties of performance-based measures to assess physical 32. Thorstensson CA, Henriksson M, von Porat A, Sjödahl C, Roos EM.
function in hip and knee osteoarthritis: a systematic review. Osteoarthritis The effect of eight weeks of exercise on knee adduction moment in early
Cartilage. 2012; 20(12):1548Y62. knee osteoarthritis V a pilot study. Osteoarthritis Cartilage. 2007; 15(10):
12. Englund M, Roemer FW, Hayashi D, Crema MD, Guermazi A. Meniscus 1163Y70.
pathology, osteoarthritis and the treatment controversy. Nat. Rev. 33. Thorstensson CA, Roos EM, Petersson IF, Ekdahl C. Six-week high-
Rheumatol. 2012; 8(7):412Y9. intensity exercise program for middle-aged patients with knee osteoar-
13. Ericsson YB, Dahlberg LE, Roos EM. Effects of functional exercise thritis: a randomized controlled trial [ISRCTN20244858]. BMC
training on performance and muscle strength after meniscectomy: a ran- Musculoskelet. Disord. 2005; 627.
domized trial. Scand. J. Med. Sci. Sports. 2009; 19(2):156Y65. 34. Trulsson A, Garwicz M, Ageberg E. Postural orientation in subjects with
14. Ericsson YB, Owman H, Roos E, Englund M, Dahlberg LE. Thigh muscle anterior cruciate ligament injury: development and first evaluation of a
weakness 4 years after partial meniscectomy is associated with radio- new observational test battery. Knee Surg. Sports Traumatol. Arthrosc.
graphic features of osteoarthritis 11 years later. Osteoarthritis Cartilage 2010; 18(6):814Y23.
2014; 22(Suppl.):S329. 35. Trulsson A, Roos EM, Ageberg E, Garwicz M. Relationships between
15. Fernandes L, Hagen KB, Bijlsma JW, et al. EULAR recommendations postural orientation and self reported function, hop performance and
for the non-pharmacological core management of hip and knee osteoar- muscle power in subjects with anterior cruciate ligament injury. BMC
thritis. Ann. Rheum. Dis. 2013; 72(7):1125Y35. Musculoskelet. Disord. 2010; 11(1):143.

Volume 43 & Number 1 & January 2015 Neuromuscular Exercise and Knee 21

Copyright © 2014 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
36. United States Bone and Joint Initiative NF P. Report of the May 2012 osteoarthritis of the hip or knee: a secondary analysis from a randomized
Chronic Osteoarthritis Management Initiative (COAMI) Work group controlled trial. J. Rheumatol. 2014; 10.3899/jrheum.130642
meeting and call to action. 2012 [cited 2014 May 30]. Available from: 39. Villadsen A, Overgaard S, Holsgaard-Larsen A, Christensen R, Roos EM.
http://www.usbji.org/programs/coami. Postoperative effects of neuromuscular exercise prior to hip or knee
37. Uthman OA, van der Windt DA, Jordan JL, et al. Exercise for lower limb arthroplasty: a randomised controlled trial. Ann. Rheum. Dis. 2014;
osteoarthritis: systematic review incorporating trial sequential analysis 73(6):1130Y7.
and network meta-analysis. BMJ. 2013; 347:f5555. 40. Zech A, Hubscher M, Vogt L, Banzer W, Hansel F, Pfeifer K. Neuromus-
38. Villadsen A, Overgaard S, Holsgaard-Larsen A, Christensen R, Roos EM. cular training for rehabilitation of sports injuries: a systematic review.
Immediate efficacy of neuromuscular exercise in patients with severe Med. Sci. Sports Exerc. 2009; 41(10):1831Y41.

22 Exercise and Sport Sciences Reviews www.acsm-essr.org

Copyright © 2014 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.