The expert committee evaluated the 2008 OARSI
recommendations on the VAS. As a result, the
minimum and maximum levels for accepting each of
the 25 propositons were found to be between 51% (95%
CI 40-62) and 97% (95% CI 94-100), respectively.
At the end of five Delphi rounds, 19 propositions
(one for general principles, nine for non-pharmacologic
treatments, seven for pharmacologic treatments, and
two for surgical treatments) were suggested as TLAR
evidence-based recommendations for the management
of knee OA (Table 3). The level of evidence (LOE),
consensus level, and strength of recommendation for
these propositions are shown in Table 4.
Click Here to Zoom |
Table 4: TLAR recommendations with level of evidence, consensus rate and strength of recommendations |
General Principles
1. The main goal in the treatment of knee OA should
be directed toward controlling pain, preserving and
improving the function of joints, providing functional
independency, and increasing the quality of life. In
order to reach these goals, management of knee OA
should contain non-pharmacologic, pharmacologic,
and, when necessary, surgical approaches. Treatment
should be tailored for each patient individually.
The daily clinical practice for the management of
knee OA is a combination of pharmacologic and nonpharmcologic
treatment options. The main goal of all
these approaches is to decrease pain and ameliorate the
functional capacity of the patients. This is a generally
accepted proposition although there is no clinical
trial evidence to support it. It was included in all
previously published guidelines eventhough the LOE
is IV.2-5 Similarly, the majority of the experts voted
in favor of accepting this recommendation during
the Delphi tours of our study, and the SOR was 96.8%
(95% CI 93-100).
Non-Pharmacologic Treatment
2. Educational programs about the symptoms of
the disease, composition, and treatment objectives,
designed for individuals and patient groups, may
increase the adherence of patients to the therapy.
These educational programs should contain lifestyle
alterations, joint protection techniques, and diet
and exercise programs which would provide weight
control.
The instructional programs, which include lifestyle
alteration, joint protection principles, and suggestions
for weight reduction, are contained in the main
guidelines for patients with knee OA. These programs
can be given individually or as group education, the
latter being less expensive. Group education followed
up by home exercise programs has been shown to
restore functional capacity and improve pain levels
(LOE Ib).12 Obesity is considered to be a risk factor
for symptomatic knee OA. Each 3 or 4 kg/m2 increase
in body mass index (BMI) doubles the risk of OA.13
Weight reduction was strongly recommended in
previous guidelines, even though controversy existed
about the relationship between weight reduction and
clinical symptoms.2-5 In a recently published report, a
10% reduction of weight in obese individuals showed
significant improvement in the symptoms of knee
OA (LOE III).14 Bliddal et al.15 reported that an 11%
weight reduction in overweight patients provided a small but significant degree of reduction in pain
associated with symptomatic knee OA (LOE Ib). Thus,
an education program that includes weight control
for the patients has great importance. The experts
strongly recommended this proposition (SOR 95.7%,
95% CI 94-98).
3. Patients with knee OA should be informed and
encouraged to use their joints in a manner that allows
the least amount of loading on their joints during
their occupational, sports, and daily living activities.
They should be educated to make this a principle
part of their daily lives. The conditions at their home
and office should be designed according to their
disease. Patients should be advised to avoid climbing
stairs, sitting cross-legged, squatting, kneeling during
“namaz (an Islamic prayer)” or doing any other
activity which would cause loading and enforcement
of knee flexion. Using elevators, sitting without knee
flexion during “namaz”, and using a Western-style
toilet should be advised instead.
Modification of daily living activities is an important
component of non-pharmacologic treatment options
for management of knee OA. Additionally, patients
should learn joint protection principles for when they
do occupational activities or participate in sports.
Activities which put pressure on the knee might
aggrevate symptoms in a dose/response manner.16
Squatting or bending of the knee and similar activities
were reported to possibly contribute to the development
of symptomatic knee OA in both genders (OR was
2.5%, 95% CI 1.4-4.7; and 2.2 % 95% CI 1.2-3.8 for
females and males, respectively). Particularly with
obesity and load lifting activities, the risk was even
higher.13,16 In an epidemiological survey, no association
was found between “namaz” and symptomatic knee
OA (LOE III).1 More comprehensive investigations
concerning these issues are needed. In light of present
scientific data, to avoid knee flexion, using alternative
positions during “namaz” might be more beneficial
(LOE IV). Patients should also be instructed to modify
their sitting positions along with their use of stairs and
the toilet, although there is no adequate evidence to
support this approach.16 This propositon was strongly
recommended (SOR 94%, 95% CI 91-97).
4. Age, comorbid diseases, and degree of OA
should be taken into consideration when choosing
an appropriate exercise program for each individual
patient. Patients should be encouraged to do rangeof-
motion, stretching, isometric, isotonic, balance,
proprioception, and aerobic exercises. Aquatic exercises can be planned in concert with the preferences
of the patient and physician. The exercise programs
should be taught to patients in a manner that they
can clearly understand and be able to do on their own.
Initially, they should be under supervision, but when
the patients are able to do them by themselves, home
programs should be started.
Exercise protocols should be tailored according
to the clinical aspects and individual needs of each
patient. The importance of exercise treatments has
been emphasized in previous recommendations and
guidelines.2-6 In an RCT, no difference between
behavioral activity and normal exercise therapy was
found in the long-term primary outcome measures
of OA.17 It was reported that progressive resistive
exercises improved physical acitivity and reduced pain
in patients with knee OA (LOE Ib).18 Tai Chi exercises
might be considered as an optional treatment choice
since they are safe and effective on pain and function,
although there is no significant scientific evidence
to prove its benefits.19 Aquatic exercises were not
deemed to be superior when compared with other
types of exercises, but they might be preferred due
to less pain occurring after exercise.20 According to
the results of an RCT which compared the effects of
loading and non-loading exercises, simple flexionextension
exercises for an eight-week period provided
significant functional improvement for patients using
either type of exercises; however, the beneficial effect in
position sensation was shown only in the group doing
the loading exercises (LOE Ib).21 In another RCT,
simple strengthening exercises designed as a home
program provided significant pain relief and functional
improvement in long-term follow-up. As a result of this
study, weight reduction was found to have no benefit
on pain, but it improved depression (LOE Ib).22 A pilot study was performed in our country concerning
the results of Pilates exercises which were designed
as group and home exercise programs. This study
showed that these exercises led to improvement in
many clinical parameters, especially when performed
as group therapy.23 Lin et al.24 reported significant
improvement in pain and functional Western Ontario
and McMaster Universities Osteoarthritis Index
(WOMAC) scores in an RCT in which strengthening
and proprioceptive exercises without loading were
compared (LOE Ib). Progressive resistive training
programs were found to have beneficial effects on
physical activities, especially in those patients with early
OA.18 Significant improvement in pain and functional
capacity was shown when performing concentriceccentric isokinetic training for the quadriceps muscle
(LOE Ib).25,26 There are many kinds of exercises for
the treatment of knee OA. A detailed evaluation of the
patient along with an appropriate choice of exercise
with respect to the patient’s needs are crucial. The LOE
was good, and the SOR was high for this proposition
(SOR 97%; CI 95-98).
5. Physical medicine and rehabilitaion specialists
may advise the use of an appropriate walking stick,
walker, or similar walking aid after the evaluation of
the patients. These devices may decrease the patients’
pain levels; therefore, they should be educated
regarding the correct use of these devices.
Patients with knee OA may have difficulty in
walking due to a variety of factors. Patients generally
transmit almost all body weight to the medial
compartment of the knee joint. Reduction of the
loading over this site may be an important treatment
goal. Use of walking sticks may be recommended for
this purpose. A study which evaluated the effect of
using walking sticks revealed that for patients who
used them for two months, a small but significant
amount of improvement occurred in their pain and
function (effect sizes 0.18, 95% CI -0.42-0.87 and
0,13, 95% CI -0.11-0.42, respectively) (LOE Ib).27 If
involvement is unilateral, sticks or crutches should be
used contralaterally. In case of bilateral involvement,
simple walkers or walkers with rollers should be
the preference. This proposition has no supportive
evidence at the RCT level, but our experts strongly
recommended it, and the SOR was high (96.4%, 95%
CI 95-98).
6. In knee OA patients with mild to moderate
joint instability, the use of appropriate ortheses may
decrease the risk of falling and may help to restore the
stability. Each patient should be advised to choose
convenient, comfortable, soft-soled shoes. The use of
sole plates may help ambulation by decreasing pain in
patients with knee OA. The application of a laterallywedged
insole in the shoes may have a symptomatic
benefit in patients with medial tibiofemoral OA.
This proposition is also present in the 2008 OARSI
recommendations. Based on a brace study, the authors
concluded that if a brace or sleeve application was
compared with medical therapy alone, there was
limited evidence of additional beneficial effects
on pain and function, with the brace being more
effective than sleeve (LOE Ia).2 In an SR which was
published after this report, non-randomized trials were
also analyzed.28 It was concluded that compressive loading over the medial tibial compartment might be
decreased, proprioception might be improved, and the
isokinetic power of the quadriceps muscle might be
increased by using a valgus brace (LOE III). A shortto-
medium period of using the valgus brace might
provide pain relief and improve quality of life, but the
effect range of this device is variable, and it is difficult
and uncomfortable to use.
In a study performed in our country, the authors
reported that medial taping of the patella was superior
to other medical therapies in patellofemoral OA (LOE
III).29
Recommending the use of appropriate shoes is not
supported by evidence in controlled trials. Instead, it
was recommended based on expert opinions (LOE IV).
Laterally-wedged insole application for medial
tibiofemoral OA was recommended by many guidelines,
and there are many trials concerning this issue. Two
RCTs were published after the 2010 update of the
OARSI recommendations. In the first trial performed
in our country, the effects of laterally-wedged insole
application on pain and function was evaluated in
patients with knee OA. As a result, it was found
that it was superior to medical therapy consisting
of analgesics and exercise (LOE Ib).30 In the other
RCT, the effects of laterally-wedged insole application
was compared with the use of the valgus brace. Both
treatment modalities led to a significant reduction in
pain, but neither was superior to the other (LOE Ib).31
The strength of recommendation was moderate for this
proposition (SOR 86.4%, 95% CI 80-93).
7. Electrotherapeutic agents such as TENS,
interferential currents, and diadynamic currents may
have beneficial effects on pain, joint function, and
quality of life. Superficial and deep heat (ultrasound,
short wave diathermy) applications may provide
benefits for patients with knee OA who have no active
synovitis. Cold application should be advised in the
case of synovitis.
Physical treatment modalities are widely used and
preferred by patients with knee OA. Superficial and deep
heaters along with analgesic currents may be the only
treatment options, especially for elderly patients who
are potentially intolerant to drugs. Physical modalities
are recommended by all guidelines for the management
of knee OA.2-5 The number of RCTs regarding the
effects of these agents is not sufficient. In a study
performed in our country, a combination of hot pack,
transcutaneous electrical nerve stimulation (TENS), therapeutic ultrasound, quadriceps strengthening
exercises, and diclofenac was compared with the use of
diclofenac alone. After completion of a 10-day session,
painless walking distance and daily living activities
showed greater improvement in the group having a
combination of physical agents (LOE Ib).32 In another
study, the effects of a combination of ultrasound
and TENS combined with exercises was found to be
superior to exercise therapy alone on the restoration of
balance (LOE Ib).33 The 2008 OARSI reccomendations
and many other guidelines have recommended the
use of TENS for knee OA. The efficiency of TENS
by itself on pain and joint function in knee OA was
evaluated in an SR and was found to be insignificant.34
The heterogeneity and inadequate number of patients
within the evaluated trials were the weaknesses of
this SR, and the necessity for well-designed trials was
emphasized. According to a recent study, an RCT
with a combination of exercises, hot compresses, and
TENS created better results on pain and quality of life
scores than the same combination with sham TENS
application (LOE Ib).35 Transcutaneous electrical nerve
stimulation, diadynamic currents or interferential
currents are thought to produce analgesia. These
modalities were investigated in combination with
exercise and/or heat applications in general. Therefore,
the pure efficiency of these currents could not been
estimated individually. However, combination therapy
is used in daily practice.
Superficial and deep heating modalities are widely
used in the management of knee OA. According to
the 2010 OARSI update, application of any thermal
modality in patients with knee OA is also recommended
in the majority of guidelines.6 Mechanical vibrations,
continous passive motion, and thermal applications in
combination with other treatments were reported to
provide significant improvement in pain and functions
(LOE Ib).36
The application of therapeutic ultrasound was
found to be effective for providing symptomatic
and functional improvements in knee OA in two
MA, but the heterogeneity of the methods and
weaknesses in the level of evidence of evaluated trials
created difficulty in predicting the effect size of this
modality (LOE Ia).37,38 In a study performed in our
country, the authors showed that an exercise program
combined either with ultrasound or short-wave
diathermy had significant beneficial effects on pain
and function. However, the low number of subjects
and lack of a control group were limitations of this
study (LOE III).39 The application of a 10-session therapeutic ultrasound was compared with sham
in another RCT, and a significant reduction in pain
scores (48%) together with significant improvements
in the total WOMAC score and 50-meter walking
time were found (LOE Ib).40
Pulsed electromagnetic field treatment for knee
OA was evaluated in a small trial, and a significant
improvement in WOMAC scores was reported
(LOE Ib).41 Based on nine trials with 483 patients,
the authors concluded that pulsed electromagnetic
field therapy improved pain and stiffness scores
in patients with knee OA, but not significantly. In
contrast, the daily living activities and functional
scores improved significantly (LOE Ia).42 It was shown
that pulsed electromagnetic field therapy was not
superior to conventional physical therapy (LOE Ib).43
Pulsed electromagnetic field treatment did not exist
in guidelines published before the 2010 update of the
OARSI recommendations.6 In summary, the beneficial
effect of pulsed electromagnetic field therapy is evident
with respect to function functions, but it is not as clear
considering on pain.
Iontophoresis and phonophoresis may improve pain
and functional activities. Applications of ibuprofen
iontophoresis and phonophoresis were found to
significantly improve pain and functional levels when
compared with the initial findings, but no difference
was evident between the two modalities.44,45 This
proposition consisted of a variety of physical agents
with different levels of evidence, and our experts
recommended it strongly (SOR 96.5%, 95% CI 94-99).
8. Neuromuscular electrical stimulation is one
of the physical treatment modalities- considered
not only for muscle strenghtening but also as an
alternative modality for alleviating the pain and
functions of patients who are not able to take part in
exercise programs.
In studies in which the effects of neuromuscular
electrical stimulation and isometric exercise of the
quadriceps muscle were compared, it was shown that
both modalities provide improvement in pain and
function related to the initial level. However, the
patients who received electrical stimulation had greater
improvement, especially in walking capacity.46-48
In these studies, other treatment modalities, such
as superficial heating and TENS, were also used.
An enlargement in the mass and an increase in the
perfusion of the quadriceps muscle were demonstrated
by computed tomography (CT) and scintigraphy
(LOE III). Neuromuscular electrical stimulation is widely used in the early period after an arthroplasty
operation. In an MA, it was mentioned that the
evidence level of studies related to electrical stimulation
was low, so the effect should be accepted with caution
(LOE Ia).49 The SOR of this proposition was 88,1%
(95% CI 82-94).
9. If there is no contraindication, balneotherapy
may be recommended for at least two weeks of
treatment because of its thermal and non-thermal
effects. Patients who are advised to have balneotherapy
should be informed about the thermal and mineral
aspects of the water of the center that they plan to
attend. In addition to this treatment, peloidotherapy
may be advised. Balneotherapy may be combined with
other physical treatment modalities and exercises by a
PM&R; specialist.
Balneotherapy and mineral water baths are nonpharmacologic
options that have been used for the
treatment of knee OA for a long time. In an SR
(LOE Ia), the results of nine RCTs with a total of 493
patients who completed 10-24 weeks of treatment
duration were evaluated. It was concluded that the
scores of pain and functional capacity were improved,
and this improvement lasted for 24 weeks. It was
also decided that balneotherapy can be considered
as a safe and effective option when used as part of
a multidisciplinary approach for the management
of knee OA.50 Forestier et al.51 reported that a
three-week period of spa treatment combined with
a pharmacologic and home exercise program was
superior to conventional treatments and exercise alone
at the end of sixth month, and it was better tolerated
(LOE Ib). A trial performed in Gönen/Balıkesir
demonstrated that 20 minutes of treatment duration
applied twice daily for two weeks improved walking
time as well as the general health quality in patients
with knee OA (LOE III).52 Any patient directed
to balneotherapy should be informed concerning
its potential effects, especially the “thermal crisis”,
of this treatment option. The presence of a PM&R;
specialist in balneotherapy centers may decrease such
potential adverse events. Additionally, a combination
of balneotherapy with physical agents and exercises
can also be performed (SOR was 91.2, 95% CI 87-95).
10. Complementary treatment options should not
be used instead of standard pharmacologic and nonpharmacologic
treatments. But if it would be, it
should only be used as an additional treatment. Both
the beneficial and adverse effects of complementary
treatments should be followed carefully.
Acupuncture, which has been accepted as an
alternative or complemantory treatment option, was
considered as a separate proposition in the 2008
OARSI recommendations, but the SOR was low (59%;
95% CI 47-71).2 Acupuncture was not mentioned in
the NICE and AAOS guidelines.4,5 The beneficial
effects of acupuncture were shown in sham-controlled
studies, but the level of evidence and the quality
of evaluated trials were not sufficient (LOE Ia).53
The clinical effectiveness of acupuncture application
might vary depending on the technique used by the
person performing, and some of the neurophysiologic
alterations attributed to acupuncture might be
considered to be a placebo effect.54 The SOR of this
proposition was 93.6%, (95% CI 89-91).
Pharmacologic Treatment
11. Acetaminophen (maximum 3 g/day) as an
initial treatment may have a mild analgesic effect
in patients with knee OA who have a mild/moderate
degree of pain. Alternative treatment options should
be considered in case of inadequate response or the
presence of severe pain and/or inflammation.
Acetaminophen was accepted as the first drug
of choice in the treatment of knee OA by many
of the guidelines.2-5 In the updated 2010 OARSI
recommendations, as a result of a cumulative metaanalytic
evaluation, the effect size of analgesia
provided by acetaminophen was found to be low
but significant. In addition, it was concluded that
acetaminophen had no effect on stiffness and
function of the joint (LOE Ia).6 There is controversy
concerning the recommended acetaminophen dose.
Four grams per day was suggested by the 2008
OARSI recommendations.2 According to more
recent trials, consumption of acetaminophen at a
dose higher than 3 g/day might be associated with
a high risk of gastrointestinal complications, and
long-term consumption could cause hypertension
and impairment in renal functions.55,56 In our
recommendations, the suggested maximum dose
for acetaminophen is 3 g/day. Using slow-release
acetaminophen might increase the compliance
(LOE III).57 The SOR was 90.6% (95% CI 83-98).
12. Nonselective and selective nonsteroidal antiinflammatory
drugs (NSAIDs) and cyclo-oxygenase
2 (COX-2) inhibitors should be used at their lowest
efficacious doses for conditions in which there is
moderate to severe pain or synovitis and for situations
in which paracetamol is insufficient. Concomitant use
of two NSAIDs should be avoided. Gastroprotective agents should be combined with NSAIDs in patients
with gastrointestinal complaints. Precaution should
be taken with the use of NSAIDs if hypertension exists
or renal or hepatic dysfunction are present.
When symptomatic knee OA has been unresponsive
to paracetamol, NSAIDs have been recommended
by almost all of the guidelines or recommendations.
According to previous investigations, the effect size
for the analgesia provided by NSAIDs was reported to
be between 0.20-0.29 (LOE Ia).6 In an RCT (LOE Ib),
the analgesic effect of diflunisal was evaluated and
was reported to be superior to a placebo.58 After
the 2010 update, three RCTs concerning new
NSAIDs were published (LOE Ib). In the first
RCT, S-adenosylmethionine and nabumetone were
compared in an eight-week study. With both drugs, a
significant reduction in pain scores was found when
compared with the initial findings, with no difference
between the two drugs.59 In the second trial, the
effects of naproxcinod were compared with naproxen
and placebo.60 Naproxcinod and naproxen provided
significant improvement in pain and function over
the placebo. The systemic blood pressure was found
to be higher among patients using naproxen while
there was not any difference between the placebo and
naproxcinod. In a trial (LOE IIb), patients who were
candidates for total knee replacement (TKR) surgery
were treated either with celecoxib or indomethacin
for a four-week period prior to the procedure, and
cartilage and synovial samples were evaluated after
the surgery.61 Proteoglycan synthesis was found to be
significantly increased among celecoxib users but not
in the indomethacin and control groups. Prostaglandin
E 2 (PGE 2) levels were found to be lower in the patients
who used either of the drugs when compared with
the controls. The indomethacin and celecoxib groups
had lower levels of interleukin-1beta (IL-1β) while the
latter group also had lower levels of tumor necrosis factor-alpha (TNF-α). In order to prevent the potential
adverse effects of NSAIDs on the gastrointestinal
system, concomittant use of misoprostol or proton
pump inhibitors were included in almost all of the
guidelines. The use of COX-2 specific inhibitors was
also recommended for patients with a gastrointestinal
risk (LOE Ib).2-5 In our country, however, COX-2
inhibitors are not officially permitted. On the other
hand, potential adverse cardiovascular events related
to COX-2 inhibitors have been demonstrated by several
reports, with the presence of ischemic heart disease or
history of stroke being defined as contraindications
for COX-2 inhibitors.6 These drugs should be used with caution in patients with hypertension, diabetes
mellitus, hyperlipidemia, peripheral arterial disease, or
in those who smoke or have other similar risk factors.
This proposition was strongly recommended (SOR
95.8%, 95% CI 94-98).
13. Topical NSAIDs or capsaicin can be used
in combination with other analgesic and/or antiinflammatory
drugs, or they can be used on their own
if patients are unable to take other drugs.
Topical NSAIDs are widely used by patients
with knee OA, and their use is recommended by
many guidelines, including the 2008 OARSI
recommendations.2-5 The effects of these drugs on
pain and stiffness were found to be moderate and
superior to placebo, but analyzed trials were highly
heterogenous (LOE Ia).2 The adverse event occurrance
was not different than for the placebo. Capsaicin is a
lipophylic alkaloid extracted from chilli peppers. It is
considered to be effective over nociceptive receptors. It
is recommended to be applied topically four times per
day over the painful joint. No significant side effects,
other than itching or a burning sensation on the site
of application, have been reported. The SOR was 95.5%
(95% CI 92-99) for this proposition.
14. Intraarticular glucocorticoid injection, not
more than three times a year, can be applied in cases
of symptomatic knee OA with signs of inflammation
which are unresponsive to other treatment options.
Intraarticular glucocorticoid administration has
been widely used for a long period of time for the
management of knee OA. It has been recommended
by a variety of guidelines, including the 2008 OARSI
recommendations.2-5 The effect size was reported
as 0.58 (95% CI 0.34-0.82) with intraarticular
administrations from the first to the third weeks.
The NNT was 5 (95% CI 3-38) (LOE Ia).62 The
beneficial effects of intraarticular glucocorticoids seem
to dissappear by the end of the fourth week. No
significant adverse effects have been reported, Their
use has been recommended especially for patients with
inflammatory findings. Patients should be instructed
not to overuse their joints after the injection in order
to prevent accelerated progression of the disease. The
evidence level was high, and our experts supported this
proposition strongly (SOR 97%, 95% CI 95-99).
15. Hyaluronic acid injections may be beneficial
for patients with mild and moderate OA who are not
overweight, have no instability, and are unresponsive
to non-pharmacologic and pharmacologic treatment
modalities.
Intraarticular administration of hyaluronic acid
(IHA) is used extensively for treatment of knee OA,
despite controversies regarding its efficiency. This
proposition did not exist in the NICE guideline, and
the controversies were emphasized in the AAOS.4,5
Zhang et al.6 evaluated the trials related to IHA
until 2009, and they mentioned that the studies were
done with different formulations, performed weekly
for three to five times weekly, and compared either
with placebo or glucocorticoids. The efficiency was
prompted at the first through the fourth weeks of
treatment (LOE Ia). The effect sizes for pain relief
along with improvement in function and stiffness
were found to be moderate when compared with the
placebo (0.60, 95% CI 0.35-0.87, 0.61, 95% CI 0.37-0.83
and 0.54, 95% CI 0.17-1.26, respectively). The NNT
was 7 (95% CI 3-119). They also emphasized that the
trials were heterogenous. When an elimination was
carried out according to the Jadad quality score, no
significant difference was found in the pain scores.
Pain relief occurred earlier in the glucocorticoid group
(2-4 weeks) while requiring much more time in the
IHA group (5-13 weeks). There is no MA concerning
the effects of IHA published after 2009. According to
an RCT, an application of 6 ml of hylan GF (instead of
weekly injections) provided pain relief at the 26th week
of administration (LOE Ib). No difference was reported between these two applications with respect to adverse
events.63 In another study, the effects of intraarticular
administrations of glucocorticoids and IHA were
compared, and no significant clinical differences were
found.64 The synovial fluid examination revealed a
significant increase in the levels of sodium hyaluronate
(Na-HA) and a significant decrease in the levels of
matrix metalloproteinase-9 in the patients treated
with IHA. It was suggested that IHA not only had
a protective effect on articular cartilage but also an
inhibitor effect on catabolism (LOE III). In a one-year
follow-up study, after five consecutive weeks of IHA
administration in 337 subjects, pain, function, and the
need for paracetamol consumption were evaluated at
three, six, nine and 12 months. The results were found
to be no better than placebo (LOE Ib).65 There have
been a number of trials performed in our country on
this subject. In one of these, five consecutive weeks
of IHA administration in patients with knee OA was
evaluated, and the results were also found to be no
better than placebo (LOE Ib).66 In the second trial,
IHA administration was performed for patients with
synovitis, and the results at the end of first year were
found to be better than placebo.67 In the patients who
had been treated with IHA, short-term improvement in proprioception and isokinetic muscle strength along
with significant improvement in function were also
reported (LOE Ib).68 In brief, the trials concernig the
effects of IHA are heterogenous. Their application may
provide a beneficial effect by selecting appropriate
patients. The SOR was 88.8%, (95% CI 83-95).
16. Glucosamines and/or chondroitin sulfate may
provide symptomatic benefits for patients with knee
OA.
Glucosamine sulphate (GS) and chondroitin
sulphate (CS) are complex sugars within the natural
structure of articular cartilage. Both molecules are
prepared for oral consumption and partially absorbed
in the small intestine. It was reported that they could
be detected in synovial fluid as well as serum after
administration.69 They have been used by patients
with knee OA extensively and have been preferred
by many physicians interested in OA, particularly
in the last 10 years. The recommended dose for GS
is 1500 mg/day for six weeks with a one-week break,
and it should be stopped if there is no response at
the end of six months. There are a number of studies
concerning the use of GS and CS, but their efficacy
is still controversial. It is also not clear whether these
agents modify the structure of articular cartilage or
not. While administration of GS was recommended
in several guidelines concerning the management of
knee OA, CS existed only in a few of them. Neither
of the molecules was recommended in the NICE and
AAOS guidelines.3,4 The use of GS and CS in knee
OA, which was in the 2008 OARSI recommendations,
was dependent on a Cochrane review published in
2005, but variability in the formulations were not
taken into account.70 In the 2010 update of the OARSI
recommendations, the therapeutic effects of GS and
glucosamine hydrochloride (GH) were analyzed
separately.6 The authors suggested that the efficiency
of GH on pain was small and statistically insignificant.
The studies related to GS were heterogenous, and
when they were evaluated according to quality, the
effect size for pain was found to also be small but
significant (0.29, 95% CI 0.0003-0.57) (LOE Ia). The heterogeneity of the studies was a major problem.
The studies concerning the effects of CS were also
heterogenous. The results of these studies, if analyzed
in accordance with quality assesment, indicated that
the effect on pain was small and insignificant. No
significant adverse event was reported (RR/OR= 0.97
and 0.99) (LOE Ia).6 In two MAs, the long-term effects
of GS and CS were evaluated based on the modification
of the structure and progression of the joints.71,72 In one MA based on six RCTs, Lee et al.71 analyzed a
total of 1502 patients and concluded that no significant
differences were found in joint space narrowing by the
end of the first year in patients who used GS, and a
small to moderate difference was observed in the third
year of treatment (effect size 0.43, 95% CI 0.24-0.63)
(LOE Ia). Similarly, patients who used CS had a small
but significant difference in joint space by the end of
the second year (effect size 0.26, 95% CI 0.13-0.39).
As a result of this MA, the authors suggested that
administration of GS for three years and CS for two
years might delay the radiologic progression of the
disease.71 In the other MA, the results of two RCTs
were analyzed, and the authors concluded that a small
but significant difference in joint space narrowing was
found in the patients taking CS 800 mg/day for two
years when compared with placebo (effect size 0.23,
95% CI 0.11-0.35).72 In one of the studies from this MA,
Kahan et al.73 suggested that pain relief was significant
and occurred faster in patients using CS compared
with the placebo, and this was observed as early as the
sixth month of therapy. The NNT was found to be 8
(95% CI 5-17), and no adverse event was reported. No
significant effect on joint space narrowing was shown
in a small sample-sized study with the use of GH.74 In
another trial with LOE IIb, GS administration delayed
the development of OA when compared with placebo in
rats and decreased nociception. The authors reported
that modification of chondrocyte metabolism might
possibly have occurred by increasing the inhibition
of the p38 kinase and the c-Jun N terminal kinase (JNK) or by increasing the release of an extracellular
signal-related kinase (ERK).75 In another RCT
(LOE Ib), cartilage turnover was evaluated, and patients
on a quadriceps strengthening exercise program for 12
weeks were given either GS 1500 mg/ day, ibuprofen
1200 mg/day, or placebo. A significant decrease in the
serum level of cartilage oligometric protein (COMP)
was determined in the patients taking GS.76 According
to the results of a two-year follow-up study in which the
effects of GH at a dose of 1500 mg/day, CS 1200 mg/day,
celecoxib 200 mg/day and placebo were compared, no
superiority on the WOMAC pain scores was found
in GS and CS users compared with celecoxib and the
placebo (LOE Ib).77 In a five-year observational study,
total knee replacement (TKR) was found to be less
frequent among patients who had taken GS for at least
12 months compared with placebo.78 It should be kept
in mind, however, that there are various factors which
would affect the indication for TKR. As one can see,
there is still much controversy regarding the efficacy of
these molecules.
Although there is no evidence concerning the
provocation of diabetes mellitus or asthma with the
consumption of glucosamines, precaution should be
taken if the drug is used in high risk patients. An
MA evaluating this issue was published in 2011
based on 11 trials (six RCTs and five prospective
trials).79 A significant affect of glucosamines on
glucose metabolism was shown in two RCTs. In
several trials performed in obese patients, an affect
was found on glucose metabolism while no affect was
found in patients with diabetes mellitus. However, the
formulations of the glucosamines were not mentioned,
and administration was heterogenous (single or
divided daily doses). The types of laboratory tests
used for glucose metabolism were also not identical.
Thus, further investigation is required. It should not
be ignored that these molecules might provoke latent
diabetes.
Glucosamines are also present in combinations that
include other nutraceuticals in the market. According
to a study performed in Italy, the amount and quality
of the CS in these nutraceuticals varied. The authors
emphasized that the health authority did not have
strict regulations regarding manufacturers and
controllers.80 This is, of course, not only a problem of
Italy, but also of our country beside many others. This
controversial proposition consisting of heterogenous
trials was moderately supported by our experts (SOR
87.3%; 95% CI 80-95).
17. Administration of weak opioids or narcotic
analgesics can be considered for patients with knee
OA who are resistant to or have contraindications
for treatment with other pharmacologic agents.
Treatment with non-pharmacologic modalities should
be continued in these patients, and appropriate
surgical options should be considered.
Treatment with opioids was recommended in
almost all guidelines, including the 2008 OARSI
recommendations.2-5 In an MA, the effect size of opioid
treatment on pain and functional scores were reported
as 0.78 (95% CI 0.59-0.98) and 0.31 (95% CI 0.24-
0.39), respectively (LOE Ia).81 However, the trials were
heterogenous with respect to methods and formulations
in general. Nausea, constipation, sleep disturbance,
dizziness, and vomiting are widely encountered
adverse effects associated with opioid therapy, and
these are the major causes for the discontinuation of
the drug. There is no data concerning the long-term
use of these drugs, especially related to dependence.
For patients with moderate to severe pain, a significant reduction in pain and improvement in function might
be achieved by the administration of tramadol once
a day (LOE Ia).82 In a Cochrane review published
in 2009, the advantages and disadvantages of opioid
therapy (oral or transdermal) were compared versus
placebo in patients with knee or hip OA (LOE Ia).
Various kinds of opioid formulations were superior
when compared with the placebo, but there were no
differences between the two types of opioid therapy.83
The authors concluded that non-tramadol opioids had
significant low to moderate benefits but that significant
adverse effects also existed. They suggested that these
formulations should not be used, even in cases suffering
from severe pain. Weak opioids, such as tramadol,
tramadol/paracetamol, codeine, and propoxyphen,
should be preferred initially in resistant cases in which
other types of treatments could not be applied. The
stronger opioids, such as oxymorphone, oxycodone,
phentanyl, and morphine sulphate, should be reserved
for extraordinary conditions. The SOR was moderate
(86.8%, 95% CI 80-93).
Surgical Treatment
18. Osteotomy can be applied in middle-aged,
active, unicompartmental knee OA patients with
malalignment for the aim of biomechanical correction.
Based on a Cochrane SR, the authors mentioned
some beneficial effects of high tibial osteotomy (HTO)
on pain and function, despite there being no study
which has compared the efficacy and safety of an
osteotomy with a placebo or conservative treatment
in unicompartmental knee OA (LOE IIa).84 It is
difficult to compare and evaluate the results of studies
concerning this procedure since there are various
techniques used for a high tibial osteotomy. In another
SR, unicompartmental knee arthroplasty (UKA) was
compared with HTO and TKR.85 The results were
found to be similar witih respect to function, but
complications, such as deep venous thrombosis, were
reported less frequently after UKA (not statistically
significant), and revision rates were lower when
compared with HTO. In brief, HTO and UKA can
be accepted as surgical options for relatively young
patients with involvement of the medial compartment,
but the results are still controversial. There were no
comparative trials with conservative treatments. The
different surgical techniques used in the trials made
a comparison difficult. Selecting the right patient and
taking him/her expectations into account are thought to
be important for positive outcome. It was reported that
only a small number of patients with knee OA could be described as right candidates for this type of surgical
approach (being 60-65 years old, moderately active,
non-obese, having 5-10 degrees of varus malalignment,
with no instability and no limitation in range of motion
and with moderate unicompartmental involvement).86
The subject is still a matter of discussion, and more
comprehensive studies are required. This proposition
was supported moderately by the committee (SOR
86.9%, 95% CI 80-94).
19. Total knee replacement should be considered
for patients with advanced knee OA who have resistant
pain to pharmacologic and non-pharmacologic
treatments and impaired quality of life. Not only
the radiologic images but also the degree of pain and
functional limitation of patients should be taken
into consideration during the course of decision for
surgery.
Total knee replacement was recommended in almost
all guidelines for the management of the patients with
advanced knee OA. Varying degrees of difference
in pain relief and improvement in the function and
quality of life have been reported. The cumulative
rate of revision surgery was estimated to be 10% in
several trials.87,88 Limitations in function, low mental
scores, and comorbidities were reported as negative
factors which determined the outcome for patients who
underwent TKR.89 Total knee replacment was found to
be cost-effective with respect to life-long expenses and
quality of life and was more expensive and less efficient
when performed in small centers, according to a study
from the United States.90 The SOR was found to be
89.3% (95% CI 84-95).
In summary, an exact cure for knee OA is not
possible yet. Preserving and/or improving the structure
and function of joints along with providing symptom
relief are the main targets in the management of knee
OA with various treatment modalities. In this study,
scientific evidence was reviewed and with contribution
of experts in the field, evidence-based recommendations
for the management of knee OA were developed for the
first time in our country. The recommendations should
be updated regularly according to new evidences and
insights. We hope that physicians who are interested
in knee OA will benefit from this report in their daily
clinical practice.
Acknowledgement
We are grateful to Prof. Weiya Zhang, PhD, for his
scientific and moral support along with his expertise
regarding methodology.
Declaration of conflicting interests
The authors declared no conflicts of interest with
respect to the authorship and/or publication of this
article.
Funding
The authors received no financial support for the
research and/or authorship of this article.
){kind=link}
){kind=link}
){kind=link}
){kind=link}
