Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Arthritis Care Res (Hoboken). Author manuscript; available in PMC 2014 March 1.
Published in final edited form as:
PMCID: PMC3537858

Perception and Presentation of Function in Patients with Unilateral Versus Bilateral Knee Osteoarthritis

Adam R. Marmon, Ph.D., Joesph A. Zeni, Jr., PT, Ph.D., and Lynn Snyder-Mackler, PT, Sc.D.



Lower extremity functional performance and perception of functional abilities influence clinical management in people diagnosed with unilateral or bilateral knee osteoarthritis. The purpose of this study was to determine if there were differences in perception of function and performance during functional tasks between individuals with unilateral and bilateral knee osteoarthritis.


The functional abilities of patients with symptomatic and radiographic diagnosed unilateral (N=84) or bilateral (N=68) knee osteoarthritis were evaluated with self-reports and performance-based tests. Self reports included the Knee Outcome Survey (KOS), Global Rating Scale (GRS), and Physical Component of Short Form-36 (PCS); functional tests included Timed Up-and-Go (TUG), Stair Climbing Test (SCT), and 6-minute Walk (6MW). Separate MANOVAs were performed separately for men and women to determine if perception (self-reports) and performance (functional tests) were dependent on the number of involved knees.


No significant main effects were observed in functional performance between groups for either sex. Similarly, the perception measures did not differ between groups. In general, individuals diagnosed with unilateral and bilateral knee osteoarthritis both performed functional tasks and perceived their functional ability similarly. Conclusion. Regardless of the number of involved knees, individuals with knee osteoarthritis perform and perceive their functional ability similarly, which suggests that clinicians need to consider other factors, such as include how long the disease has been progressing or how functional abilities have changed when treating patients with knee osteoarthritis.


The knee is the most common joint to be diagnosed with osteoarthritis (OA) and progression of the disease in this load bearing joint leads to substantial disability and reduced functional capacity during a large range of daily activities (1; 2). Although the magnitude of functional impairments associated with knee OA is dependent upon the severity of the disease (3), it is unclear if impairment levels are also dependent upon the number of involved knees.

In studies of patients with knee OA, a single limb is often used as the reference by which functional outcomes are measured (4; 5), despite the fact that knee OA frequently affects both lower extremities (6) and this bilaterality may amplify the magnitude of functional impairments (7). Despite previous reports of a relation between bilaterality and greater disability (7,8, 9), only self-assessment surveys of functional ability were used as outcomes to assess functional ability. Self-assessment surveys are commonly used to evaluate functional capacity and are integral in quantifying patient satisfaction and quality of life during treatment for knee OA. These instruments are easy to administer making them the primary outcome measure for many interventional studies or large cross-sectional normative samples (10-14). However, these instruments must be cautiously interpreted as they are generally driven by pain (15-18), with greater pain associated with worse perception of function (11). It is not known whether perception of disability is related to the condition of the worse limb, or whether bilateral involvement would increase the actual or perceived functional deficits that these tools are intended to measure.

Self-report instruments and performance-based tests are used to drive patient care, measure the effectiveness of treatment, and stratify patients in research studies. However, it is unclear if the outcomes from these tests differ for individuals with unilateral compared to bilateral disease, even if the severity of OA is similar. Therefore, the aims of this investigation were to; 1) determine if the presence of bilateral disease amplifies self-reported and performance-based functional deficits, and 2) determine if the number of involved knees affects scores on performance-based and perception-based assessments differently, as these tests measure different domains of impairment. We hypothesized that both performance and perception of performance would be worse in individuals diagnosed with bilateral disease compared to those diagnosed with unilateral disease.



One hundred fifty-two individuals diagnosed by an orthopedic surgeon with unilateral or bilateral knee OA were referred to the University of Delaware Physical Therapy Clinic for evaluation of their functional ability. Data were collected from June 2009 through January 2011. All patients were classified as having knee OA based on radiographic evidence of one or both of their knees with a Kellgren-Lawrence score of ≥3 (19) and pain during activities of daily living. All patients were assessed with the Delaware Osteoarthritis Profile, a comprehensive functional evaluation that includes standard clinical measures, performance-based functional tests, and self-reports of function. These data were a part of a database maintained by the University of Delaware Physical Therapy Clinic and analysis of this de-identified dataset was approved by the Human Subjects Review Board.


The physical component score of the Short-Form 36, (SF-36-PCS) was used to assess a patient’s perception of physical limitations and physical health (20; 21). The Knee Outcome Survey-Activities of Daily Living Scale (KOS-ADLS) was used to assess an individual’s perception of performance during activities of daily living (22). The Global Rating Scale (GRS) a single, knee and limb specific, question was used to assess patient’s knee function during usual daily activities. The KOS-ADLS and GRS are joint- and limb-specific questionnaires where subjects provide answers for each limb separately. To compare KOS-ADLS and GRS scores for patients with unilateral and bilateral disease, a single score for the bilateral group was computed, both by averaging the two knees and by taking the worse score of the two knees. Significant correlations were observed between average and worse scores for both the KOS-ADLS (π= 0.832; p<0.001) and GRS (π= 0.926; p<0.001). Data used in subsequent analyses were from the worse limb score.

Functional Tests

For the purposes of this study, three functional tests from the Delaware Osteoarthritis Profile were used to assess functional ability; Timed Up-and-Go (TUG), Stair Climbing Test (SCT), and 6-minute Walk (6MW). The TUG requires subjects rise from the chair, walk three meters, and return to the seated starting position as quickly as possible. The TUG is commonly used to assess outcomes for patients with knee OA and has good reliability (23). The average time needed to complete the test from two trials was used for analysis. The SCT is commonly used to assess functional ability in people with knee osteoarthritis (23; 24). During this test, subjects ascend and then descend a flight of 12 stairs as quickly and safely as possible. Patients were permitted to use the handrail if necessary. The 6MW involved patients walking at a quick, comfortable pace around a 157.3 m square for 6-minutes, with the total distance covered taken as the outcome measure.


Power analyses were performed to estimate the sample size needed to detect a clinically important difference, if a difference were to exist, between patients with unilateral and bilateral knee OA. Consideration was made for data stratified by sex in performance-based and self-reported measures of function. Published minimal detectable change (MDC) scores for the following measures were used from the literature; TUG (25), 6MW (25), SCT (23), and KOS-ADLS (26); sample size estimates were assessed using G-Power (Version 3.1.2) for moderate (0.5) and high (0.8) levels of power. The results of our power analyses indicate that our sample size was sufficiently powered (moderate to large power) to detect clinically meaningful differences between groups, if clinically meaningful differences existed (Table 1).

Table 1
Power analyses for sample size estimation to detect a clinically meaningful difference between individuals with unilateral compared to bilateral knee OA, if one were to exist.

Comparison of the dependent measures (age, body mass index, SCT, TUG, 6MW, SF-36, KOS-ADLS, GRS) between men and women were examined and if the majority of dependent variables differed between the two sexes then separate analyses were performed. Chi-square analyses were performed to determine if the number of co-morbidities was different for patients with unilateral compared to bilateral disease. Multiple analysis of variance (MANOVA) tests were performed to determine if significant differences between subjects with unilateral and bilateral disease existed for performance-based measures and self-report surveys. Separate MANOVAs were conducted for the functional performance tasks (SCT, TUG, 6MW) and self-report surveys (SF-36-PCS, KOS-ADLS, GRS). We were also interested in the ability of each survey to differentiate between individuals with unilateral and bilateral disease, because each survey assessed different aspects of a patient’s perception of function. Specifically, the KOS-ADLS questionnaire pertains to a patient’s perception of how the knee influences activities of daily living, the GRS is a report of a patient’s perception of their global functioning level, and the SF-36-PCS accounts for a patient’s perception of overall physical health. Therefore, independent T-tests were used to determine differences between groups (unilateral versus bilateral disease) for each self-report questionnaire. To protect against type I error, we set a familywise error rate of alphaFW=0.01 for all comparisons (27).


Descriptive statistics of the sample are provided in Table 2. Neither age (p=0.255) nor BMI (p=0.098) were significantly different between unilateral and bilateral groups (Table 2B). Similarly, age (p=0.267) and BMI (p=0.567) were not significantly different between patients with unilateral compared to bilateral disease for men (Table 3A). Additionally, age (p=0.845) and BMI (p=0.049) were not significantly different for women, after adjusting the significance level for multiple comparisons (Table 3B). The quantity of co-morbidities was not different for patients with unilateral compared to bilateral disease (Table 4). Therefore, these variables were not adjusted in the final analyses.

Table 2
Descriptive statistics of the sample with comparisons between women and men (A) and between individuals with unilateral and bilateral disease (B), including; age, body mass index (BMI), Short-Form 36 Physical Component Score (SF-36-PCS), Knee Outcome ...
Table 3
Descriptive statistics of the sample separated for women (A) and men (B) with comparisons between those with unilateral (Uni) and bilateral (Bilat) disease, including; age, body mass index (BMI), Short-Form 36 Physical Component Score (SF-36-PCS), Knee ...
Table 4
Co-morbidities. Comparison of the diagnosed co-morbidities between patients with unilateral and bilateral disease, separated by sex, expressed as a percentage within group. The co-morbidities considered were heart disease, heart attack, high blood pressure, ...

While age and body mass index (BMI) were not significantly different between the men and women, men performed significantly better than women in all three functional measures (TUG, SCT, 6MW; p< 0.001) and perceived their function to be better in all three self-report measures (SF-36-PCS, KOS-ADLS, GRS; p≤ 0.05) (Table 2). Similar proportions of patients diagnosed with unilateral (58%) and bilateral (59%) utilized the handrail during the SCT, however, more women (68%) than men (50.1%) utilized the handrail. There were also significant differences in the distribution of sexes in the unilateral (33 women: 51 men) and bilateral (41 women: 26 men) groups (p=0.01). Therefore, separate analyses (MANOVA and independent T-tests) were performed for each sex.

No main effects were observed for the MANOVAs examining group differences (unilateral vs bilateral disease) in the functional performance measures for either women (p=0.450; Figure 1A) or men (p=0.489; Figure 1B). Similarly, no main effects were observed for the MANOVAs examining group differences (unilateral vs. bilateral disease) in self-reports for either women (p=0.819; Figure 2A) or men (p=0.149; Figure 2A). The findings are consistent when controlling for age and BMI for both the functional performance measures (women, p= 0.606; men, p= 0.482) and for self-reports (women, p= 0.968; men, p= 0.163).

Figure 1
Comparison for functional performance for patients with unilateral and bilateral disease for women (A) and men (B). Error bars are ± standard error.
Figure 2
Comparison for perception of functional performance for patients with unilateral and bilateral disease for women (A) and men (B). Error bars are ± standard error.

The analyses of the individual self-report surveys between subjects with unilateral and bilateral disease revealed a small difference for the SF-36-PCS between the men with unilateral disease (38.8 ± 9.2) compared with bilateral disease (34.4± 7.6; p=0.049), which is smaller than the minimum clinically important difference (MCID= 0.5* standard deviation for the group = 4.46) and not significant after adjusted for multiple comparisons. There were no significant differences in the KOS-ADLS or GRS between groups for men, or in any self-report survey for women diagnosed with unilateral versus bilateral disease (p≥ 0.107).


Functional performance measures and self-reports of function are commonly used to assess individuals diagnosed with knee OA. We hypothesized that patients diagnosed with bilateral disease would perform worse in functional tests and perceive their functional ability to be worse than individuals with unilateral disease. Contrary to our hypotheses, neither the performance-based functional tests nor the self-report questionnaires differed between individuals with unilateral and bilateral disease. Our results challenge the conventional clinical perspective that patients with unilateral disease compensate for impairments using the non-involved limb, which has implications for the management of the disease and for categorization of subjects participating in interventional or clinical research.

One would intuitively presume that patients with unilateral disease rely on their non-involved leg during functional tasks and demonstrate functional performance that surpasses patients with bilateral disease. This presumption was not supported by our findings, although others have reported functional differences between individuals with unilateral and bilateral disease. The incongruous findings may be explained by several critical methodological differences between studies. Two studies indicate greater self-reports of pain and functional impairments in individuals with unilateral compared to bilateral pain (8; 9). However, neither of these studies required any diagnoses of knee osteoarthritis, but surveyed all individuals >50 years registered with a group of general medical practices (8) or by surveying a large, random sampling of individuals >55 years, living within a given community (9). A third study acquired data from a sample of community-dwelling individuals deemed at risk for developing knee OA based on age, female sex, previous knee injury, and high body weight. Additionally, the findings from all three studies were based on self-reports of function (7; 8; 9), compared to the present study that compared both performance-based measures of functional ability between groups along with self-reports. Self-reports of function are driven by pain (17). Therefore, compared to self-report questionnaires, performance-based metrics of functional performance may provide a more accurate indication of functional ability and may more precisely capture advantageous compensatory movement strategies that improve functional performance in patients’ with unilateral knee OA. As our sample consisted only of individuals with severe symptoms, who sought medical treatment, and were diagnosed with knee OA, it is plausible that the relation between unilateral or bilateral involvement with functional limitations is not as strong as has been observed in individuals with mild symptoms.

We found no difference in functional performance measures between patients with unilateral and bilateral disease, suggesting potential movement strategies that reduce the discomfort in the involved limb and take advantage of the integrity of the uninvolved limb do not translate into enhanced functional performance as measured by the TUG, SCT and 6MW.

Sex differences

Significant sex differences existed in all measures of perception and performance of lower extremity functional ability. While it is known that women are more likely to develop OA compared to men (1; 28-30) and a greater percentage of women exhibit symptomatic disease (6), our findings are still surprising in a sample where radiographic disease severity was similar. In general, healthy women are weaker and exhibit accelerated declines in performance of functional tests compared to men (31), but the women in the present study had significantly lower self-reported functional ability as well. These findings support previous work that determined that the functional abilities and perception of disability of women are affected by knee OA to a greater extent than men (32-34). The substantial differences between the rate of OA development, the etiology of the disease, and functional ability between men and women with knee OA combined with the sex differences in self-reports observed here, support the separate analyses for men and women in future studies. The results of this study further support previously identified sex differences, in that regardless of whether it is bilateral or unilateral disease, women were, on average, significantly more symptomatic with greater functional impairments (Table 2). Collectively, these findings suggest that clinicians should consider the treatment for individuals with OA to be similar for patients with unilateral or bilateral disease, while taking into account the patient’s sex when determining treatment options and management of care. Women may require surgical or pharmacological intervention earlier in the course of the disease.


Some limitations exist in our current study. Our cross sectional sample, while initially of a large size (N=151), was substantially reduced because of the necessary separation of sex. Because all data collections occurred at one time, we were not able to determine longitudinal differences between groups. Additionally, we did not control for severity of disease using the radiographic classification (Kellgren and Lawrence score), although all subjects were classified as having KL grades of 3 or greater. Controlling for radiographic severity within this study would have further reduced the sample size as well as our ability to detect differences between groups. Symptomatic classification may be optimal to radiographic classification, as there is only weak correlation between clinical and radiographic severity of OA (35). Examination of the differences in movement strategies (e.g. stepping pattern during the SCT) employed by the two groups were not performed. However, irrespective of the chosen movement strategies employed by the two groups, no group differences in performance outcomes were observed. The data presented here were from individuals with moderate, symptomatic OA, who were diagnosed with Kellgren/Lawerence scores of ≥ 3 and our findings suggest that there are likely other factors that contribute to both the perception and performance in functional tasks. Duration of symptoms, joint effusion, bodily pain and physical activity levels may also contribute to functional performance and self-reported ability, which were not assessed in this paper. Future studies should consider quantifying physical the activity levels of patients with unilateral and bilateral disease, to determine if the similarities in performance and perception of function are associated with general inactivity in both populations.


The similarity in performance of functional measures and responses on self-report instruments for individuals with unilateral and bilateral disease has implications for how clinicians treat patients diagnosed knee osteoarthritis and how subjects are classified in interventional research. Factors that should be considered when evaluating a patient’s functional limitations or perception of their functional ability may include how long the disease has been progressing or how functional abilities have changed that can be attributed to the disease, both will provide the clinician with a sense of the individual’s functional capacity, as well as, providing the clinician with a sense of the individual’s expected functional capabilities. Most importantly, clinicians need to understand that individuals with unilateral and bilateral disease perform and perceive their functional abilities similarly. When separate analyses were performed for men and women, the diagnosis of unilateral versus bilateral disease did not appear to affect performance or self-report differently, even though these tests capture different domains of disability. These results suggest that the worse limb dictates the outcome measures for both functional performance and self-reports.


  • Self-reports of function (perception) do not differ between individuals diagnosed with unilateral compared to those diagnosed with bilateral knee osteoarthritis.
  • Functional ability (performance-based tests) does not differ significantly between individuals diagnosed with unilateral and bilateral knee osteoarthritis.
  • Measures of perception of function and actual functional abilities between individuals diagnosed with unilateral or bilateral knee osteoarthritis do not differ irrespective of sex.
  • Treatment and management of care should focus on differences in patient sex and severity of disease and not the number of involved knees.


The authors would like thank Dr. Michael J. Axe, M.D. and First State Orthopedics for their assistance in patient recruitment and the University of Delaware Physical Therapy Clinic for treating our subjects. We would also like to thank our funding source the National Institutes of Health (grant 5P20RR016458 and 5P20RR016458-S1).

Funding: National Institutes of Health (grant 5P20RR016458 and 5P20RR016458-S1)


1. Oliveria SA, Felson DT, Reed JI, Cirillo PA, Walker AM. Incidence of symptomatic hand, hip, and knee osteoarthritis among patients in a health maintenance organization. Arthritis & Rheumatism. 1995;38:1134–1141. [PubMed]
2. Guccione AA, Felson DT, Anderson JJ, Anthony JM, Zhang J, Wilson PWF, Kelly-Hayes M, Wolf PA, Kreger BE, Kannel WB. The Effects of Specific Medical Conditions on the Functional Limitations of Elders in the Framingham Study. American Journal of Public Health. 1994;84:351–358. [PubMed]
3. Sadosky AB, Bushmakin AG, Cappelleri JC, Lionberger DR. Relationship between patient-reported disease severity in osteoarthritis and self-reported pain, function and work productivity. Arthritis Research & Therapy. 2010;12:R162. [PMC free article] [PubMed]
4. Dieppe PA, Cushnaghan J, Shepstone L. The Bristol “OA500” Study: progression of osteoarthritis (OA) over 3 years and the relationship between clinical and radiographic changes at the knee joint. Osteoarthritis and Cartilage. 1997:587–97. [PubMed]
5. Sharma L, Cahue S, Song J, Hayes K, Pai YC, Dunlop D. Physical functioning over three years in knee osteoarthritis: role of psychosocial, local mechanical, and neuromuscular factors. Arthritis and Rheumatism. 2003;48:3359–70. [PubMed]
6. Felson DT. The incidence and natural history of knee osteoarthritis in the elderly; The Framingham Study. Arthritis and Rheumatism. 1995;38:1500–1505. [PubMed]
7. White DK, Zhang Y, Felson DT, Niu J, Keysor JJ, Nevitt MC, et al. The Independent Effect of Pain in One Versus Two Knees on the Presence of Low Physical Function in a Multicenter Knee Osteoarthritis Study. Arthritis Care & Research. 2010;62:938–943. [PMC free article] [PubMed]
8. Jinks C, Jordan K, Croft P. Measuring the population impact of knee pain and disability with the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Pain. 2002;100:55–64. [PubMed]
9. Keenan AM, Tennant A, Fear J, Emery P, Conaghan PG. Impact of multiple joint problems on daily living tasks in people in the community over age fifty-five. Arthritis and Rheumatism. 2006;55:757–64. [PubMed]
10. Saleh KJ, Mulhall KJ, Bershadsky B, Ghomrawi HM, White LE, Buyea CM, et al. Development and validation of a lower-extremity activity scale. Use for patients treated with revision total knee arthroplasty. The Journal of Bone and Joint Surgery. American volume. 2005;87:1985–94. [PubMed]
11. Mizner RL, Petterson SC, Clements KE, Zeni J a, Irrgang JJ, Snyder-Mackler L. Measuring Functional Improvement After Total Knee Arthroplasty Requires Both Performance-Based and Patient-Report Assessments A Longitudinal Analysis of Outcomes. The Journal of Arthroplasty. 2010;26(5):728–37. [PMC free article] [PubMed]
12. Salaffi F. Reliability and validity of the Western Ontario and McMaster Universities (WOMAC) Osteoarthritis Index in Italian patients with osteoarthritis of the knee. Osteoarthritis and Cartilage. 2003;11:551–560. [PubMed]
13. Rosenberg IH. Symposium : Sarcopenia : Diagnosis and Mechanisms Sarcopenia : Origins and Clinical Relevance 1. Nutrition Research. 1997:990–991.
14. Grindrod KA, Marra CA, Colley L, Cibere J, Tsuyuki RT, Esdaile JM, et al. After patients are diagnosed with knee osteoarthritis, what do they do? Arthritis Care & Research. 2010;62:510–5. [PubMed]
15. Stratford PW, Kennedy DM. Does parallel item content on WOMAC’s pain and function subscales limit its ability to detect change in functional status? BMC Musculoskeletal Disorders. 2004;9:517. [PMC free article] [PubMed]
16. Stratford PW, Kennedy DM, Woodhouse LJ. Performance measures provide assessments of pain and function in people with advanced osteoarthritis of the hip or knee. Physical Therapy. 2006;86:1489–96. [PubMed]
17. Stratford PW, Kennedy DM. Performance measures were necessary to obtain a complete picture of osteoarthritic patients. Journal of Clinical Epidemiology. 2006;59:160–7. [PubMed]
18. Maly MR, Costigan PA, Olney SJ. Determinants of self-report outcome measures in people with knee osteoarthritis. Archives of Physical Medicine and Rehabilitation. 2006;87:96–104. [PubMed]
19. Kellgren JH, Lawrence JS. Radiological assessment of osteo-arthrosis. Annals of the Rheumatic Diseases. 1956:494–502. [PMC free article] [PubMed]
20. Ware JE, Jr, Kosinski M, Bayliss MS, McHorney CA, Rogers WH, Raczek A. Comparison of methods for the scoring and statistical analysis of SF-36 health profile and summary measures: summary of results from the Medical Outcomes Study. Med Care. 1995 Apr;33(4 Suppl):AS264–79. [PubMed]
21. Ware J, Sherbourne C. The MOS 36-Item Short Form Health Survey (SF-36): I Conceptual framework and item selection. Med Care. 1992;30:473–483. [PubMed]
22. Irrgang JJ, Snyder-Mackler L, Wainner RS, Fu FH, Harner CD. Development of a Patient-Reported Measure of Function of the Knee. The Journal of Bone and Joint Surgery. 1998;80:1132–1145. [PubMed]
23. Kennedy DM, Stratford PW, Wessel J, Gollish JD, Penney D. Assessing stability and change of four performance measures: a longitudinal study evaluating outcome following total hip and knee arthroplasty. BMC Musculoskeletal Disorders. 2005;6:3. [PMC free article] [PubMed]
24. Mizner RL, Snyder-Mackler L. Altered loading during walking and sit-to-stand is affected by quadriceps weakness after total knee arthroplasty. Electromyography. 2005;23:1083–1090. [PubMed]
25. Mangione KK, Craik RL, McCormick AA, Blevins HL, White MB, Sullivan-Marx EM, et al. Detectable changes in Physical Performance Measures in Elderly African Americans. 2010;90(6):921–7. [PubMed]
26. Collins NJ, Misra D, Felson DT, Crossley KM, Roos EM. Measures of knee function. Arthritis Care & Research. 2011;63(S11):S208–28. [PubMed]
27. Keppel G. Design and analysis: A researcher s handbook. Prentice-Hall; Englewood Cliffs, N.J.: 1982.
28. Blagojevic M, Jinks C, Jeffery A, Jordan KP. Risk factors for onset of osteoarthritis of the knee in older adults: a systematic review and meta-analysis. Osteoarthritis and Cartilage / OARS, Osteoarthritis Research Society. 2010;18:24–33. [PubMed]
29. Hochberg M. Bone mineral density and osteoarthritis: Data from the Baltimore Longitudinal Study of Aging1. Osteoarthritis and Cartilage. 2004:1245–48. [PubMed]
30. Felson DT, Zhang Y. Osteoarthritis with a view to prevention. Arthritis & Rheumatism. 1998;41:1343–1355. [PubMed]
31. Samson MM, Meeuwsen IB, Crowe A, Dessens JA, Duursma SA, Verhaar HJ. Relationships between physical performance measures, age, height and body weight in healthy adults. Age and Ageing. 2000;29:235–42. [PubMed]
32. Thomas SG, Pagura SMC, Kennedy D. Physical activity and its relationship to physical performance in patients with end stage knee osteoarthritis. The Journal of Orthopaedic and Sports Physical Therapy. 2003;33:745–54. [PubMed]
33. Petterson SC, Raisis L, Bodenstab A, Snyder-Mackler L. Disease-specific gender differences among total knee arthroplasty candidates. The Journal of Bone and Joint Surgery. American volume. 2007;89:2327–33. [PubMed]
34. Ritter MA, Wing JT, Berend ME, Davis KE, Meding JB. The clinical effect of gender on outcome of total knee arthroplasty. The Journal of Arthroplasty. 2008;23:331–6. [PubMed]
35. Bedson J, Croft PR. The discordance between clinical and radiographic knee osteoarthritis: a systematic search and summary of the literature. BMC Musculoskeletal Disorders. 2008;9:116. [PMC free article] [PubMed]