Secondary Logo

Journal Logo


The Effectiveness of Off-Loading Knee Orthoses in the Reduction of Pain in Medial Compartment Knee Osteoarthritis

A Systematic Review

Feehan, Nathanael L. BS; Trexler, Gary S. CO; Barringer, William J. MS, CO

Author Information
JPO Journal of Prosthetics and Orthotics: January 2012 - Volume 24 - Issue 1 - p 39-49
doi: 10.1097/JPO.0b013e318240af8d
  • Free

Osteoarthritis (OA) is a debilitating disease that affects millions of people worldwide. It is the most common type of arthritis and is a major cause of musculoskeletal pain and disability in elderly populations.1 In the United States alone, 6% of the adults aged 30 years and older (roughly 10 million individuals) have symptomatic OA of the knee.2 OA of a single compartment is distinguished from that of both compartments. The single-compartment OA is associated with malalignment and increased risk and progression of knee OA, therefore predicting decline in function. Those with medial compartment OA often have a varus alignment observed during gait evaluation, as the mechanical axis and weightbearing pass through the medial compartment. Patients with OA of the lateral compartment usually have a valgus alignment, and the mechanical axis and weightbearing pass through the lateral compartment.3 With so many affected, multiple treatment modalities have been developed. There are >50 nonpharmacological, pharmacological, and surgical interventions for the treatment of OA. Nonsurgical treatment methodologies are generally tried first. If these relieve pain, increase function, and slow the progression of the disease, then surgical interventions are delayed. Surgical interventions include knee joint replacement, both full and partial; high tibial osteotomy; and fusion as a last resort when joint replacement has failed.4

OA knee bracing is one of these nonsurgical interventions. Introduced in 1989, it has become a popular mode of treatment for the OA knee.5 The orthoses function by the use of a mechanical lever arm that reduces the forces transferred through the affected compartment, “off-loading” those affected compartments with unilateral OA by creating a valgus or varus moment at the knee depending on the affected compartment. This, in turn, reduces pain and increases function.6

Although some research in OA unloading knee orthoses has demonstrated an increase in function with various scales and measurement tools,7,8 probably the most practical and easiest way for the orthotists to demonstrate clinical effectiveness through evidence-based practice is by measuring pain. Evidence-based practice seeks to inform clinical decision making by combining a practitioner's training and experience with evidence established through scientific research.9 If pain reduction is a goal of the orthotic intervention process, using pain as a measurable benchmark can demonstrate that an orthosis is an effective treatment modality. The visual analog scale (VAS) is a common, well-established measuring device to track this.10

With the growth and acceptance of the market for these orthoses, there have been multiple articles and studies published on the topic. Although some articles focus on the biomechanical aspects of the knee and its varus and valgus moments, others consider whether or not function is increased by the use of knee orthoses. Other articles reveal the long-term effects of the orthosis; still others investigate pain reduction of the participant with the orthosis. Some articles include various combinations of these aspects of study. By documenting the ability of the unloading knee orthosis to reduce pain in those suffering from unilateral OA of the knee through research, the authors of this article may prove what clinical practice has already shown to be effective: that OA unloading knee orthoses are effective in the relief of medial compartment OA knee pain.


Using the keywords “knee osteoarthritis” and “orthosis” or “brace,” the PubMed, CINAHL, RECAL, and ISI Web of Knowledge electronic databases were searched for randomized control trials with orthoses or braces for the knee between January 1980 and December 2010. The exact Boolean search phrase used was [(knee osteoarthritis) and (orthosis or brace)]. The RECAL and CINAHL databases were selected for their orthotic-specific search capabilities. The RECAL database was compiled by the National Centre for Prosthetics and Orthotics and the service designed and developed by the Centre for Digital Library Research, both of the University of Strathclyde.11 CINAHL was selected because it included the Journal of Prosthetics and Orthotics.

To standardize the process of literature review and selection, the American Academy of Orthotists and Prosthetists (AAOP) has developed a study design classification scale, which is used to classify the study type when performing a literature review. Research is divided into four categories: structured reviews, (quasi) experimental trials, observational studies, and expert opinion. It was decided that experimental trials would best demonstrate the effectiveness of the OA knee orthosis in relieving pain in the OA knee.12

Only those articles that were experimental or quasi- experimental trials were selected, based on the AAOP State- of-the-Science Evidence Report Guidelines. The studies reviewed needed to be prospective research studies with multiple subjects. They must have some comparison, whether it is between conditions or between a control group and an intervention group. The studies must also have had one or more interventions and the data gathered and documented intervals. The five groups established by the AAOP State-of-the-Science Evidence Report Guidelines are randomized control trial (E1), controlled trial (E2), interrupted time series trial (E3), single-subject experimental trial (E4), or a controlled before-and-after trial (E5). The difference between these is the number of subjects, random assignment of the subject(s) and control(s), and the frequency of data collection. A comparative analysis of the studies will be used to determine the effectiveness of the orthoses in the area of pain.12

Only studies that dealt with medial unilateral compartment knee OA and the use of a knee orthosis were selected. The medial compartment is involved in 91% of unilateral cases of OA, 10 times more often than the lateral compartment.13 This can partly be attributed to the increased load carried by the medial compartment. Approximately 60% to 80% of the load across the knee is transmitted through the medial compartment.14 To keep the demographic population consistent, only those studies that investigated medial compartment OA were evaluated. This eliminated variables associated with varus or valgus moments that affect the biomechanics and function of the orthoses and therefore the effect on pain.15

Methodological quality was assessed using a modified version of the State-of-the-Science Evidence Report Guidelines Quality Assessment Form.12 All items have “yes,” “no,” or “N/A” answer options. Fourteen questions were asked: six on internal validity and eight regarding external validity (Table 1). Because of the nature of many of the studies and the justification for the literature review, it was thought by the reviewers that external validity should be more heavily favored because of the desire for clinical application.

Table 1
Table 1:
Methodological quality assessment.

After the articles were reviewed, they were then scored. Those articles that scored five or below were considered of low quality and were rejected for review. Those articles that scored from 6 to 10 were considered moderate quality, whereas those that scored from 11 to 14 were considered high quality. Both moderate- and high-quality articles were reviewed. A high score indicated that the reviewers evaluated the study to have few to no confounding factors or bias, which would limit the usefulness of the study in its applicability. A study of moderate quality indicated that there are areas of bias introduced into the study, somewhat limiting the value and usefulness of the results. A low score indicated that the reviewers believed that there were many confounding factors and bias introduced in the study and that there were serious problems with the applicability of the study to clinical practice.


The initial Internet database search yielded 306 articles. From CINAHL, 19 articles were located, 64 from the ISI-Web of Knowledge, 100 from the RECAL database, and 123 from the PubMed database. A review of the titles and abstracts of these initial 306 articles resulted in 265 being excluded based on inclusion and exclusion criteria. The remaining 41 were pearled (“pearling” is a search for additional references within references located for review), and 5 additional articles were located. These remaining 46 articles were reviewed, and 31 were excluded from the study for the following reasons. In 16, pain was not considered in the study; 9 were not level E5 to E1; in one study the knee orthosis was used after knee arthroplasty; in one study no knee orthoses were used, and the study was unclear as to the mode of treatment; in one study the methodology was unclear; one was not in English; one was a meeting abstract; and one was a product development paper (Figure 1). Fifteen articles were included for review.

Figure 1
Figure 1:
Literature search results.

Of these 15, there were 3 randomized control trials selected (E1), one interrupted time series trial (E3), six single-subject experimental trials (E4), and five before-and-after trials (E5). A quality analysis resulted in nine studies that were of moderate quality and six studies of high quality. Only one study16 had a perfect score, indicating that the reviewers believed that there was little bias introduced in this study; this study used gait parameters as a primary outcome measure and pain as a side measure. There were no studies of low quality reviewed. Studies tended to score lower on questions of internal validity than external validity. Some of the best organized and well-written studies were funded by manufacturers who introduced additional bias into the studies (Table 2).

Table 2
Table 2:
Methodological quality assessment results.

A brief summary of the applicable data from each reviewed articles is as follows. If numbers were given in the study, then they are reported below, otherwise statistical significance was considered.

Barnes et al. published “Effect of CounterForce Brace on Symptomatic Relief in a Group of Patients With Symptomatic Unicompartmental Osteoarthritis: A Prospective 2-Year Investigation” in the American Journal of Orthopedics in 2002. Thirty patients from the clinical population of Barnes et al. were selected as participants in this study. Patients completed the American Academy of Orthopaedic Surgeons (AAOS) arthritis questionnaire, the short-form SF-36® Health Survey, and VAS on pain levels both on a daily and on a weekly basis. Participants were fitted with the CounterForce ™ (Breg, Vista, CA) orthoses from Breg®. After 8 weeks of treatment, participants again completed the AAOS and SF-36 forms. A follow-up after 2.5 years was also completed. With both the AAOS and SF-36 forms, symptoms improved in all categories during the 8-week period. The patients' responses to questions in the weekly log were also of note. Statistically significant decreases in pain during the duration of that time were documented, as well as a marked decrease in weeks 1 to 3. Pain relief in the orthosis was initially dramatic but seemed to level out after week 3. There was another marked decrease in pain between weeks 7 and 8. There were inconclusive results as to the decreased use of pain medications while using the orthosis. At the long-term follow-up, 21 individuals (73%) indicated that the orthosis was effective in relieving their pain.17

Brouwer et al. published “Brace Treatment for Osteoarthritis of the Knee: A Prospective Randomized Multi-Centre Trial” in Osteoarthritis and Cartilage in 2006. Participants in this study were originally treated with standard conservative treatments. Patients were either originally fit with an OAsys orthoses from Innovation Sports or were in a control group that was not treated with any orthosis but instead continued with standard treatment. One hundred eighteen patients were admitted in the study and randomized into two different groups. Patients were assessed using both the VAS and the Hospital for Special Surgery OA form (HSS). Compared with the controls, pain severities on the VAS scale were less in the orthosis group at each of the three assessment points as well as overall during the 12-month follow-up. At 12 months, the difference was borderline significant; however, in both instances, significant pain reduction versus the control group was not found.3

Dennis et al. published “An In Vivo Analysis of the Effectiveness of the Osteoarthritic Knee Brace During Heel Strike and Midstance of Gait” in the Acta Chirurgiae Orthopaedicae Et Traumatologiae Cechoslovaca, a Czech orthopedic journal, in 1999. In it the researchers analyzed the gait of 40 subjects during heelstrike and midstance, which are both weightbearing conditions, using video fluoroscopy to determine whether unloading knee orthoses actually provide separation of the femoral condyle from the tibial plateau, thereby avoiding excessive loads on the degenerated compartment. Participants were also asked to state if the orthosis reduced their pain, though not rated on a VAS or other measuring device. Thirty-four of the 40 participants indicated that they had pain relief from the orthosis, whereas 6 did not. It was noted by the authors that all of these had a body mass index >20% of their ideal body weight and had no condylar separation. Although the authors documented that lack of subjective pain relief correlated with no condylar separation, no statistical analysis was done to determine whether this was a significant association.18

Dennis et al. published his findings again with a parallel study in 2006 in the Journal of Arthroplasty under the title, “Evaluation of Off-Loading Braces for Treatment of Unicompartmental Knee Arthrosis.” As before, it was documented that 34 of the 40 participants, or 85%, stated they had pain relief from the orthosis, whereas 6 did not. Again, no statistical analysis was done to determine whether this was a significant association between subjective pain relief and condylar separation.19

Draganich et al. published “The Effectiveness of Self- Adjustable Custom and Off-the-Shelf Bracing in the Treatment of Varus Gonarthrosis” in the Journal of Bone and Joint Surgery in 2006. Ten patients were evaluated in two orthoses in a random order. One was custom fabricated for them and the other was an off-the-shelf model. Pain and function were both assessed using the Western Ontario and McMaster Universities Osteoarthritis (WOMAC) index (which measures pain, stiffness, and physical function).20 Measurements were taken before application of the orthoses and after 4 to 5 weeks after application of the orthoses. Pain was significantly reduced from an average baseline of 197 to 120 mm with the off-the-shelf model and to an average of 71 mm with the custom orthosis. There was a significant difference between each of the three testing parameters. Function increased significantly only with the custom orthosis over the baseline. The authors attributed this success of the custom model to the better fit of the custom orthosis. The authors further state that their study suggests that an intimate fit is necessary for improved results and increased pain relief.21

Finger et al. published “Clinical and Biomechanical Evaluation of the Unloading Brace” in the Journal of Knee Surgery in 2002. Twenty-three participants were assessed preapplication of the orthosis and at three months using a 10-point pain scale. At three months, the average resting pain decreased from 4.2 to 1.9, pain with activity decreased from 7.2 to 3.9, and night pain decreased from 3.9 to 2.4. No statistical analysis was done on these numbers to determine whether the changes were significant.22

Gaasbeek et al. published “Valgus Bracing in Patients With Medial Compartment Osteoarthritis of the Knee: A Gait Analysis Study of a New Brace” in Gait and Posture in 2006. Fifteen patients with medial compartment OA were fit with a pneumatic knee orthosis, which provided a medial force that was designed to off-load the medial compartment of the knee. Baseline measurements were taken before the fit of the orthosis and after 6 weeks of continuous wear. The VAS pain scores showed a statistically significant change. The average score was 6.8 ± 2.5 without the orthosis and 4.7 ± 3.0 with the orthosis. The WOMAC score also showed a statistically significant change. The average score was 50.1 ± 17.6 without the orthosis and 63.0 ± 18.4 with the orthosis.23

Horlick et al. published “Valgus Knee Bracing for Medial Gonarthrosis” in the Clinical Journal of Sports Medicine in 1993. Thirty-nine participants were fit with the generation II unloader orthosis, some with the hinge on the lateral side and some with the hinge on the medial side. Participants were then randomly assigned to one of four treatment sequences: (a) brace in neutral, brace in valgus, and no brace; (b) brace in neutral, no brace, and brace in valgus; (c) brace in valgus, no brace, and brace in neutral; and (d) brace in valgus, brace in neutral, and no brace. Although an option, Horlick decided not to start participants out in the no-brace stage because it did not differ from the pretreatment status. During the study, participants filled out a daily pain and function log. Twenty participants had laterally hinged orthoses, and 19 participants had medially hinged orthoses. The VAS showed a statistically significant decrease in the pain level from baseline to valgus and a marginally significant decrease from baseline to neutral, using the lateral hinge. Somewhat stronger evidence of differences was seen with the medial hinge. Both baseline to valgus and baseline to neutral were statistically significant. Something that was noted by the authors was a carryover effect of the orthosis into the no-brace, washout phase. Continuing relief of pain was experienced in the no-brace phase of the study by both those with the valgus and neutral orthoses.24

Kirkley et al. compared the neoprene knee sleeve, a custom-made unloader orthosis, and no treatment, in their study, “The Effect of Bracing on Varus Gonarthrosis,” published in the Journal of Bone and Joint Surgery in 1999. One hundred ten patients participated in the study, with 33 in the control group, 36 in the neoprene knee sleeve group, and 41 in the unloader group. The researchers found a significant difference between treatment groups as far as pain. The six-month assessment showed a significant difference among the treatment groups with regard to the mean aggregate change score for the WOMAC. There was an increase in pain equal to 13.1 mm for the control group, a decrease in pain equal to 13.1 mm with the neoprene sleeve group, and 43.2 mm with the unloader orthosis group. It was remarkable that most of the pain relief came in the first 6 weeks with continued relief tapering off in the following weeks.25

In 1997, Lindenfeld et al. published “Joint Loading With Valgus Bracing in Patients With Varus Gonarthrosis” in Clinical Orthopaedics and Related Research. The researchers investigated the generation II knee orthosis by applying it to 11 participants. Baseline measurements were taken before applications and after 4 to 6 weeks of orthosis wear using the 10-point VAS. Pain symptoms decreased significantly with brace wear (48%) from 6.3 to 3.6. The authors also noted significant increases in function and activity levels, which they attributed to the decreased pain levels.26

Matsuno et al. published “Generation II Knee Bracing for Severe Medial Compartment Osteoarthritis of the Knee” in Archives of Physical Medical and Rehabilitation in 1997. Using a modified knee scoring system developed by the Japanese Orthopaedic Association, 20 participants were evaluated before orthosis application and each month thereafter for 1 year. The modified knee scoring scale used in the study considered pain while engaging in walking and stair climbing. Nineteen of the 20 patients (95%) reported decreased pain during activities. All scores significantly improved with the application of the orthosis and continued to improve throughout the observation period. Unfortunately, the results were not clearly stated.27

Pagani et al. published “Short-Term Effects of a Dedicated Knee Orthosis on Knee Adduction Moment, Pain, and Function in Patients With Osteoarthritis” in the Archives of Physical Medical and Rehabilitation in 2010. Eleven participants were incorporated in this study. Participants were tested in three different sessions by using a crossover design: before orthosis wear, after 2 weeks of wearing the orthosis with a neutral setting, and before and after the orthosis set with 4° of valgus. The order of the two different orthosis conditions was randomly assigned, and patients were blinded to the different adjustments. Before the participant was fitted and wore the orthosis, the researchers used the WOMAC scale to measure pain and function. There were statistically significant improvements in pain from the control group to the orthosis in neutral and from the orthosis in neutral to the 4° of valgus conditions. Each was higher than the previous condition. The same was true for the function measure of the WOMAC scale.28

Ramsey et al. published “A Mechanical Theory of the Effectiveness of Bracing for Medial Compartment Osteoarthritis of the Knee” in the Journal of Bone and Joint Surgery in 2007. Sixteen participants with medial compartment OA were referred from a local orthopedic practice for participation in the study. Baseline measurements were taken. Each participant was fit with a generation II unloading orthosis with an initial setting of 0° of correction. This was worn for 2 weeks, and measurements were taken again. The orthosis was removed for 2 weeks and then reapplied with 4° of valgus added. There were no significant results between the washout and the valgus condition or between the two bracing conditions with respect to pain and the activities of daily living (ADL). However, the symptoms were worse with the 4° of valgus condition than the initial setting of 0°. The orthoses control condition changes were significantly different than the baseline measurements. The authors attributed this to changes in the muscle action at the knee, which were also measured in this study. The authors concluded that the use of an orthosis was a cost-effective way to relieve pain, while not exclusively recommending the use of an OA unloading orthosis.29

Richards et al. published “A Comparison of Knee Braces During Walking for the Treatment of Osteoarthritis of the Medial Compartment of the Knee” in the Journal of Bone and Joint Surgery (British Volume) in 2005. Twelve patients were recruited and randomly fit with either a hinged knee orthosis or the generation II knee orthosis. After 6 months, each received the second type of orthosis for another 6 months. Measurements were taken before application and at the end of each 6-month period, using the VAS for resting, standing, walking, and stair climbing. It was discovered that only significant differences were found for the valgus orthosis condition, and no significant difference was discovered between the simple hinged orthosis and the baseline measurements.30

Schmalz et al. published “Analysis of Biomechanical Effectiveness of Valgus-Inducing Knee Brace or Osteoarthritis of Knee” in the Journal of Rehabilitation Research and Development in 2010. The research included 16 patients who wore a prefabricated knee orthosis for 4 weeks. Although gait parameters were the main focus of this study, pain while walking was assessed using the VAS. The initial pain while walking was 6.4 ± 1.7, which was reduced to 3.3 ± 1.9. This was a significant change. Significant changes in walking speed (meter per second) and cadence (step per minute) were documented; although the step length did not change, both walking speed and cadence increased. This points to a significant increase in function. The authors attributed these gait pattern changes to the lack of a need for protection against pain, therefore increasing function.16 van Raaij et al. published “Medial Knee Osteoarthritis Treated by Insoles or Braces: A Randomized Trial” in Clinical Orthopedics and Related Research in 2010. Ninety-one participants with medial knee OA took part in this study. Of these, 45 were treated with laterally wedged insoles and 46 with the MOS Genu knee orthosis. Although this knee orthosis was not exclusively designed for OA, it has the ability to off-load the medial compartment. In this study, the pain severity in the braced group decreased from a mean 5.6 on the VAS to 4.6, and the WOMAC increased from 46.8 to 50.8. These measurements were taken 6 months after application of the orthosis. Both of these results were statistically significant. It was noted that skin irritation was the main complaint, with 10 participants reporting skin irritation and seven patients commented on poor fit. Each of these could have contributed to less than ideal results.31


The importance of pain reduction cannot be underestimated. According to the American Pain Society, “Pain is the leading public health problem in this country and the most common symptom that leads to medical care, resulting in more than 50 million lost workdays each year. The cost of pain, including medical bills and lost workdays, is estimated at $100 billion per year in the United States. Back pain alone produces chronic disability in 1 percent of the U.S. population and is the leading cause of disability in Americans under 45 years old.”32 Because of this, the reduction of pain and its debilitating effects was the goal of this review. The pain caused by knee OA is especially limiting because it affects mobility and walking.

The 15 studies that met the inclusion criteria indicate that there is strong evidence that pain is reduced with the use of unloading osteoarthritic knee orthoses. Only one study, Brouwer et al.,3 had less than significant results, and several lacked statistical analysis of the data, but data were reported in percentages. It is also interesting to note that the study by Brouwer et al. was also the largest single study with 117 participants. Most studies presented the results of their pain measurements with a mean. Matsuo noted one patient who did not receive any pain relief and Dennis noted six. Only one study reported that all participants noted pain relief.30 Therefore, of the 567 combined participants of these studies, at least seven of them had no pain relief whatsoever. This was marginally >1% of participants. On the basis of the studies reviewed, 98.6% of patients experienced pain relief when fitted with an unloading knee orthosis.

Because these studies had different measures of pain, VAS scale, WOMAC index or a modified version of that scale, HSS score, and using different orthoses and different methods in the setup of the studies, the authors were not able to perform any meta-analysis of the combined results. Percentages indicate that 76% of those participants in studies who did statistical analysis had significant reductions in pain. Excluding the 88 participants reporting reductions in pain in studies that did not do statistical analysis to determine significance leaves 479 who were involved in studies that did provide statistically significant data. All these patients reported some pain relief. This percentage would undoubtedly be higher if an appropriate statistical analysis was done in each case or a meta-analysis was completed of these compiled studies.

Immediate pain relief would be expected from the application of the unloading knee orthosis. This was demonstrated by three studies in an interesting way. In the study by Barnes et al., pain measurements were taken each week. There was significant improvement until week 3 in each case and a leveling out of relief after that point.17 Kirkley et al.25 measured their treatment for a total of 24 weeks and found that pain relief leveled out at 12 weeks with the greatest relief happening in the first 6 weeks. Finally, Brouwer et al. noted that at the 3-month, 6-month, and 12-month follow-up, the difference between the two control groups did not change markedly but rather remained within a 20-point range. Although these were not significant at the studies p < 0.01 significance level, most were below p < 0.053. These three studies indicate that most of the pain relief applied by the orthosis is immediate. The theory behind the tapering off pain is the reduction in inflammation due to decreased irritation.33

Of interest was the pain relief during the no-brace phase of Horlick's study. The researchers noted that although the pain relief was more profound in the valgus bracing phase than the brace in neutral phase of the study, there was relief during the no-brace washout period or the no-brace period after wearing the orthoses.24 This “carryover” effect has been documented in other orthotic literature, with the effects of the orthosis continuing after discontinuation of the orthosis. With scoliosis orthoses, there is often a carryover effect that is considered during routine radiographs. If the physician chooses to take radiographs of the patient out of his or her orthosis, the physician looks for what the body is doing when the orthosis is not being worn. There is always a time period between the radiograph and the last wearing time to mitigate the residual or carryover effects of the orthosis.34 A review of ankle-foot orthoses also noted a carryover effect in several studies.35 This residual effect may be one answer for the carryover effect noted by Horlick in this study.

All but two of the studies were underpowered, meaning that the number of participants was less than ideal and thereby raising the possibility of a type 2 error. This means that the chance of not reporting a difference in the intervention when there was a difference is more likely. Most of the studies were of moderate design, indicating that there were areas that may have introduced bias and therefore affected the outcomes of the studies. Important elements fundamental to randomized control trials, such as blinding and randomization, were not possible or were not provided for in the studies reviewed. Because of this, further bias may have been introduced.

Although not a primary focus in this project, decreased function is a known complication of OA. Some of the studies reviewed investigated increases in function. Eight of the studies in some way investigated function. All of them noted significant increases in the function of participants.3,16,17,25,26,28,29,31 There were two studies that are of note: studies by Lindenfeld et al. and Schmalz et al. Gait pattern changes seem to be a protective mechanism by the patient with knee OA to reduce knee pain.36 The study by Schmalz et al resulted in more symmetrical gait patterns in the participants, noting that their gait lab measurements were improved with the unloading knee orthosis. Lindenfeld et al. indicated that the participants in their study went from moderate symptoms with ADLs to walking several hours per day without appreciable symptoms. According to the authors of the studies reviewed, the decrease in pain scores was attributed to the use of the knee orthosis. The knee orthosis in each case caused an increase in the knee adduction moment at heelstrike and loading response, thereby relieving the pressure on the medial compartment of the knee.

Although the ability of this review to analyze which orthoses were more effective than others is extremely limited, some interesting observations can be made based on the differences in study morphology, pain measurement scales, and interval of measurements. The most common type of orthosis used was the generation II unloader orthosis. This orthosis was among the first commercially available OA knee orthoses, which was undoubtedly the reason it was the most commonly studied orthosis. Horlick24 and Matsuno et al.27 were among the first to investigate the use of this unloading orthosis. As the clinical effectiveness of these types of orthoses became known, flurries of articles were published around the turn of the century on a variety of knee orthoses. Two worth mentioning are Dennis and Draganich. Dennis compared the effectiveness of four commercially available off-the-shelf orthoses. The Bledsoe Thruster had the greatest amount of relief of symptoms from medial knee OA.19 Draganich et al. examined custom versus off-the-shelf models of the unloading knee orthosis and found a significant difference between the two orthoses. Custom knee orthoses provided significantly better relief of symptoms. Draganich et al.21 attributed this to a better fitting orthosis.

Another interesting observation, and warranting further study, was the importance of patient perception and placebo effect. Demonstrated in both studies done by Brouwer et al.3 and Kirkley et al.,25 some participants in the control group of both studies withdrew because they wanted to be in the knee orthosis group rather than in the control group. In the study by Kirkley et al., seven (18%) withdrew from the control group and two (5%) from the neoprene sleeve group. In the study by Brouwer et al., “One patient withdrew immediately because of dissatisfaction with the randomization outcome.” Furthermore, no patients withdrew from the unloader orthosis control groups in either study. Patient perception and the placebo effect produce a valid effect on pain perception and play a real part in clinical care. Although it may be part of the results, even if the placebo worked and decreased pain, it should be considered a viable option for the treatment of knee OA.37

As insurance companies tighten reimbursement, justifying the use of orthoses over other treatments is important. One thing that was not evaluated in these studies was how many of the participants later underwent corrective surgery, either femoral osteotomy or knee replacement. One way of evaluating the effectiveness of the orthotic intervention is whether or not patients later have to undergo corrective surgery, which always entails risks. In the research by Barnes et al., it was noted how many patients went on to some type of surgery and how many were able to continue using the orthosis to lessen their pain. At a 2-year follow-up, only 24% of the participants had undergone arthroplasty, whereas 41% were still using their orthosis.17 van Raaij et al. noted that of those participating in their study, three in the no-orthosis group underwent surgery at 6 months, whereas one had undergone surgery in the orthosis group.31 In the randomized control trial, Brouwer et al.3 followed up patients for a year and found that in the orthosis group, 11 (18%) patients underwent surgery that year and that in the control group 13 of the remaining 53 (25%) participants underwent some sort of surgery. Each of these studies had similar results of decreased incidence of surgery in the orthosis group versus a control group. Although these numbers are very small, they may signal an interesting trend and warrant further research.

Only a few of the studies reported how the orthoses were fit and if a qualified individual was responsible for fit and follow-up. Most of the studies did not state who fit or how the orthosis was fit. Appropriate fit is an often-heard complaint among wearers of the orthoses.24,31 A trained orthotist can properly fit an orthosis, increase compliance, and further increase the effectiveness of the orthosis.

The Osteoarthritis Research Society International guidelines for knee OA indicate that “emphasis should be placed on encouraging adherence to the regimen of nonpharmacological therapy” in the treatment of knee OA. It was recommended for patients with mild to moderate knee OA for the reduction of symptoms including pain.38 These guidelines are consistent with the findings of the literature review in that they would recommend the OA knee orthosis for treatment of knee OA.

Areas of further research are apparent based on the results of this review. Further research on the effectiveness of one orthosis over another by an unbiased source would help practitioners provide the best possible care. The possible carryover effect of the knee unloading orthosis would have implications for the duration of wear time of the orthosis. If there is indeed a carryover effect, then possibly it is not necessary to wear the orthosis full time but only during activities that cause pain. Can the need for patients to receive knee replacements or other invasive surgeries to correct knee OA be lessened by the wearing the unloading knee orthosis? If this is true, then wearing the unloading knee orthosis may indeed reduce the incidence of surgery. This would be a justification for the orthosis, proving that the orthosis would reduce the overall healthcare cost of the treatment of medial compartment OA.


OA can be a disabling disease. On the basis of the articles reviewed, an OA or unloading knee orthosis is an effective way to relieve pain in the osteoarthritic knee. Pain relief was documented to help in 98.6% of patients fitted with unloading orthoses for medial compartment OA of the knee. With decreased pain comes increased function and quality of life. When compared with surgery, these orthoses are a cost-effective means of treating OA.


The authors thank Jonathan Day, CPO, Samuel Feehan, CPO, and Sonya Feehan, BS, for reviewing the draft manuscript.


1. Peat G, McCarney R, Croft P. Knee pain and osteoarthritis in older adults: a review of community burden and current use of primary health care. Ann Rheum Dis 2001;60:91–97.
2. Felson DT, Lawrence RC, Dieppe PA, et al.. Osteoarthritis: new insights. Part 1: the disease and its risk factors. Ann Intern Med 2000;133:635–646.
3. Brouwer RW, van Raaij TM, Verhaar JAN, et al.. Brace treatment for osteoarthritis of the knee: a prospective randomized multi-centre trial. Osteoarthritis Cartilage 2006;14:777–783.
4. Zhang W, Moskowitz RW, Nuki G, et al.. OARSI recommendations for the management of hip and knee osteoarthritis, part I: critical appraisal of existing treatment guidelines and systematic review of current research evidence. Osteoarthritis Cartilage 2007;15:981–1000.
5. Pollo FE. Bracing and heel wedging for unicompartmental osteoarthritis of the knee. Am J Knee Surg 1998;11:47–50.
6. Pollo FE, Jackson RW. Knee bracing for unicompartmental osteoarthritis. J Am Acad Orthop Surg 2006;14:5–11.
7. Divine JG, Hewett TE. Valgus bracing for degenerative knee osteoarthritis—relieving pain, improving gait, and increasing activity. Phys Sportsmed 2005;33:40–46.
8. Chew KTL, Lew HL, Date E, Fredericson M. Current evidence and clinical applications of therapeutic knee braces. Am J Phys Med Rehabil 2007;86:678–686.
9. Geil MD. Assessing the state of clinically applicable research for evidence-based practice in prosthetics and orthotics. J Rehabil Res Dev 2009;46:305–313.
10. Price DD, McGrath PA, Rafii A, Buckingham B. The validation of visual analog scales as ratio scale measures for chronic and experimental pain. Pain 1983;17:45–56.
11. The Recal Legacy [database on the Internet]. Available at: Accessed March 31, 2011.
12. Hafner BP. American Academy of Orthotists & Prosthetists (AAOP), State-of-the-Science Evidence Report Guidelines. American Academy of Orthotists and Prosthetists; 2008. Available at: Accessed March 31, 2011.
13. Ahlback S. Osteoarthrosis of the knee. A radiographic investigation. Acta Radiol Diagn (Stockh) 1968:Suppl 277:7–72.
14. Prodromos CC, Andriacchi TP, Galante JO. A relationship between gait and clinical changes following high tibial osteotomy. J Bone Joint Surg Am 1985;67:1188–1194.
15. Self BP, Greenwald RM, Pflaster DS. A biomechanical analysis of a medial unloading brace for osteoarthritis in the knee. Arthritis Care Res 2000;13:191–197.
16. Schmalz T, Knopf E, Drewitz H, Blumentritt S. Analysis of biomechanical effectiveness of valgus-inducing knee brace for osteoarthritis of knee. J Rehabil Res Dev 2010;47:419–429.
17. Barnes CL, Cawley PW, Hederman B, et al.. Effect of CounterForce brace on symptomatic relief in a group of patients with symptomatic unicompartmental osteoarthritis: a prospective 2-year investigation. Am J Orthop (Belle Mead NJ) 2002;31:396–401.
18. Dennis AD, Komistek RD. An in vivo analysis of the effectiveness of the osteoarthritic knee brace during heel strike and midstance of gait. Acta Chir Orthop Traumatol Cech 1999;66:323–327. Analyza ucinnosti in vivo ortez pri lecbe gonartrozy behem jednotlivych fazi chuze (“heel-strike” a “mid-stance”).
19. Dennis DA, Komistek RD, Nadaud MC, Mahfouz M. Evaluation of off-loading braces for treatment of unicompartmental knee arthrosis. J Arthroplasty 2006;21:2–8.
20. Bellamy N, Buchanan WW, Goldsmith CH, et al.. Validation study of WOMAC: a health status instrument for measuring clinically important patient relevant outcomes to antirheumatic drug therapy in patients with osteoarthritis of the hip or knee. J Rheumatol 1988;15:1833–1840.
21. Draganich L, Reider B, Rimington T, et al.. The effectiveness of self-adjustable custom and off-the-shelf bracing in the treatment of varus gonarthrosis. J Bone Joint Surg Am 2006;88:2645–2652.
22. Finger S, Paulos LE. Clinical and biomechanical evaluation of the unloading brace. J Knee Surg 2002;15:155–158; discussion 159.
23. Gaasbeek RDA, Groen BE, Hampsink B, et al.. Valgus bracing in patients with medial compartment osteoarthritis of the knee—a gait analysis study of a new brace. Gait Posture 2007;26:3–10.
24. Horlick SG. Valgus knee bracing for medial gonarthosis. Clin J Sports Med 1993;3:251–255.
25. Kirkley A, Webster-Bogaert S, Litchfield R, et al.. The effect of bracing on varus gonarthrosis. J Bone Joint Surg Am 1999;81:539–548.
26. Lindenfeld TN, Hewett TE, Andriacchi TP. Joint loading with valgus bracing in patients with varus gonarthrosis. Clin Orthop Relat Res 1997;344:290–297.
27. Matsuno H, Kadowaki KM, Tsuji H, et al.. Generation II knee bracing for severe medial compartment osteoarthritis of the knee. Arch Phys Med Rehabil 1997;78:745–749.
28. Pagani CH, Bohle C, Potthast W, Bruggemann GP. Short-term effects of a dedicated knee orthosis on knee adduction moment, pain, and function in patients with osteoarthritis. Arch Phys Med Rehabil 2010;91:1936–1941.
29. Ramsey DK, Briem K, Axe MJ, Snyder-Mackler L. A mechanical theory for the effectiveness of bracing for medial compartment osteoarthritis of the knee. J Bone Joint Surg Am 2007;89:2398–2407.
30. Richards JD, Sanchez-Ballester J, Jones RK, et al.. A comparison of knee braces during walking for the treatment of osteoarthritis of the medial compartment of the knee. J Bone Joint Surg Br 2005;87:937–939.
31. van Raaij TM, Reijman M, Brouwer RW, et al.. Medial knee osteoarthritis treated by insoles or braces: a randomized trial. Clin Orthop Relat Res 2010;468:1926–1932.
32. Society AP. Media backgrounder. Internet: American Pain Society; 2011. Available at: Accessed May 1, 2011.
33. Marieb ENR, Hoehn KM. Anatomy & Physiology. 7th ed. San Francisco, CA: Pearson Benjamin Cummings; 2007:1159.
34. Carol J, Hentges C. The team approach to the orthotic treatment of idiopathic scoliosis and Scheuermann's kyphosis. J Orthot Prosthet 2003;15:49–52.
35. Kobayashi TP, Leung AKLP, Hutchins SWP. Design and effect of ankle-foot orthoses proposed to influence muscle tone: a review. J Orthot Prosthet 2011;23:52–58.
36. White A, Foster NE, Cummings M, Barlas P. Acupuncture treatment for chronic knee pain: a systematic review. Rheumatology (Oxford) 2007;46:384–390.
37. Pagani CHF, Potthast W, Bruggemann GP, et al.. The effect of valgus bracing on the knee adduction moment during gait and running in male subjects with varus alignment. Clin Biomech 2010;25:70–76.
38. Zhang W, Moskowitz RW, Nuki G, et al.. OARSI recommendations for the management of hip and knee osteoarthritis, part II: OARSI evidence-based, expert consensus guidelines. Osteoarthritis Cartilage 2008;16:137–162.

osteoarthritis; knee orthoses; brace; knee; pain; function

© 2012 American Academy of Orthotists & Prosthetists