Arthritis Care & Research Vol. 67, No. 3, March 2015, pp 374 –381 DOI 10.1002/acr.22495 © 2015, American College of Rheumatology
ORIGINAL ARTICLE
Contralateral Knee Effect on Self-Reported Knee-Specific Function and Global Functional Assessment: Data From the Osteoarthritis Initiative SEBASTIAN COTOFANA,1 WOLFGANG WIRTH,1 CLAUDIA PENA ROSSI,2 FELIX ECKSTEIN,1 ¨ NTHER3 AND OLIVER H. GU
Objective. To analyze the effect of contralateral knee pain on sensitivity of patient-reported outcomes and objectively measured functional performance tests in subjects with knee osteoarthritis (OA). Methods. Subjects with discordant knee pain status (i.e., 1 knee being painful [>4 on a numeric pain rating scale (NPRS)], with the contralateral knee being pain free [NPRS 0]) were selected from the Osteoarthritis Initiative and matched to subjects with bilaterally pain-free and painful knees by age, sex, body mass index, and radiographic knee OA. The Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) physical function score, the global Physical Activity Scale for the Elderly (PASE), and objective functional performance tests were cross-sectionally compared in a matched case– control design. Results. A total of 378 subjects with discordant knee pain status were matched to 359 controls with bilaterally pain-free knees and to 323 controls with bilaterally painful knees. WOMAC scores in pain-free knees of discordant knee pain cases significantly differed compared to scores of bilaterally pain-free knees (P ⴝ 0.003). Likewise, scores in painful knees of discordant knee pain cases significantly differed compared to scores of bilaterally painful knees (P < 0.001). PASE levels between these groups were not significantly different (P > 0.68). Functional performance tests differed in subjects with discordant knee pain compared to subjects with bilaterally pain-free knees and when compared to subjects with bilaterally painful knees, with the chair stand test showing the strongest effect size (standardized response mean 0.28 and 0.33, respectively). Conclusion. The WOMAC physical function score, although knee specific, is impacted by the contralateral knee pain status. The repeated chair stand test appears to be the most sensitive assessment in differentiation between groups with different status of knee pain.
INTRODUCTION Knee osteoarthritis (OA) is the most common type of arthritis and is the leading cause of functional disability and impaired health-related quality of life in the elderly in comparison with any other chronic disease (1). Knee pain is known to be associated with both self-reported funcThe Osteoarthritis Initiative is a public-private partnership comprised of 5 contracts (N01-AR-2-2258, N01-AR-22259, N01-AR-2-2260, N01-AR-2-2261, and N01-AR-2-2262) funded by the NIH, a branch of the US Department of Health and Human Services, and conducted by the Osteoarthritis Initiative Investigators. Private funding partners include Merck Research Laboratories, Novartis Pharmaceuticals, GlaxoSmithKline, and Pfizer. Private sector funding for the OAI is managed by the Foundation for the NIH. The specific analysis that supported the current study was funded by Merck KGaA, Darmstadt, Germany. Drs. Cotofana, Wirth, and Eckstein’s work was supported by Merck KGaA.
374
tional limitations (2) and physical performance (3), including difficulties in walking or getting up from a chair (4 – 6). Generally, improvement of knee pain by medical treatment is expected to result in improved physical performance (7), but when symptomatic treatment is adminis1 Sebastian Cotofana, MD, Wolfgang Wirth, PhD, Felix Eckstein, MD, PhD: Institute of Anatomy, Paracelsus Medical University, Salzburg, Austria, and Chondrometrics GmbH, Ainring, Germany; 2Claudia Pena Rossi, MD, PhD: EMD Serono, Rockland, Maryland; 3Oliver H. Gu¨nther, PhD: Merck KGaA, Darmstadt, Germany. Dr. Cotofana has a part-time appointment with Chondrometrics GmbH. Dr. Wirth has a part-time appointment with Chondrometrics GmbH and is co-owner of Chondrometrics GmbH, has received consultant fees, speaking fees, and/or honoraria (less than $10,000) from Merck Serono Switzerland, and owns stock or stock options in Chondrometrics GmbH. Dr. Pena Rossi has patents pending for combination
Knee-Specific Function and Contralateral Knee Effect
Significance & Innovations ●
Regarding the patient-reported Western Ontario and McMaster Universities Osteoarthritis Index physical function score, functional limitations in a painful knee are perceived worse when the contralateral knee is painful, and functional limitations in a pain-free knee are perceived worse when the contralateral knee is painful.
●
Self-assessed levels of the Physical Activity Score in the Elderly do not discriminate between subjects with discordant knee pain status, whereas walking tests, and particularly the chair stand test, displayed high sensitivity in differentiating the pain status of one knee.
●
Objective functional performance tests, particularly the chair stand test, showed more sensitivity in differentiating between the pain status of one knee, even if the pain status in the other (contralateral knee) was similar.
tered by intraarticular injection, only the injected knee can be expected to improve (8). Patient reported outcome (PRO) measures of functional performance in daily activities, as evaluated for each knee by the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) (9), may be biased by functional limitations imposed by the pain status of the contralateral (untreated) knee. Global activity levels and objective functional performance tests critically depend on the interaction and functionality of both knees. Hence, it is unclear to what extent the pain status of the contralateral knee impacts the sensitivity of such instruments or tests, particularly when symptomatic improvement is achieved in one knee by intraarticular injection. The purpose of the current study was to analyze the effect of contralateral knee pain status on the sensitivity of knee-specific (WOMAC physical function score), generic (Physical Activity Scale for the Elderly [PASE]) PRO measures, and global functional performance tests (repeated of BlyS and/or APRIL inhibition and immunosuppressants for treatment of autoimmune disease and for dosing methods for treating autoimmune diseases using a TACI-Ig fusion protein such as atacicept. Dr. Eckstein is Chief Executive Officer and co-owner of Chondrometrics GmbH, has received consultant fees, speaking fees, and/or honoraria (less than $10,000 each) from Merck Serono, Novartis and Abbvie, owns stock or stock options in Chondrometrics GmbH, has received funding support not related to this study from Pfizer, Eli Lilly, Novartis, Merck Serono, GlaxoSmithKline, Wyeth, Centocor, Stryker, AbbVie, Kolon, and Synarc, and has provided educational content for Medtronic. Address correspondence to Sebastian Cotofana, MD, Institute of Anatomy, Paracelsus Medical University, Strubergasse 21, A5020 Salzburg, Austria. E-mail: sebastian. [email protected]. Submitted for publication February 28, 2014; accepted in revised form October 7, 2014.
375 chair stand test, the 20-meter walk test, and the 400-meter walk test). Therefore, we created 2 hypothetical scenarios using data from the Osteoarthritis Initiative (OAI) and analyzed it by using a cross-sectional case– control study design. The first scenario assumes that both knees are painful before treatment and one of them is treated successfully. To estimate the sensitivity of PRO measures and global functional performance tests, we compared subjects with bilaterally painful knees to subjects with a discordant knee pain status (one knee being painful, the other one not). The second scenario assumes that one knee is painful before treatment and this knee is successfully treated. To estimate the sensitivity of PRO measures and global functional performance tests, we compared subjects with discordant pain status to those with bilaterally pain-free knees.
SUBJECTS AND METHODS The OAI. The OAI is a privately and National Institutes of Health–funded longitudinal cohort study of subjects with, or at risk of developing, symptomatic knee OA (10,11). Individuals of multiple ethnicities, ages 45–75 years, were included in this cohort study. Subjects were excluded from the OAI if the following conditions were present: rheumatoid arthritis, bilateral knee arthroplasty or preexisting plans to undergo bilateral (not unilateral) knee arthroplasty in the next 3 years, bilateral Osteoarthritis Research Society International stage 3 (severe) knee OA (12), positive pregnancy test, inability to provide a blood sample, use of ambulatory aids other than a single straight cane for ⬎50% of the time, comorbid conditions that might interfere with 4-year participation, unlikely to reside in the clinic area for at least 3 years, current participation in a double-blind, randomized controlled trial, and being unwilling to sign an informed consent. Men weighing ⬎130 kg and women weighing ⬎114 kg were also excluded, due to specific requirements with the acquisition of magnetic resonance images not relevant to the current study. In total, 27% (n ⫽ 4,796) of all subjects screened (n ⫽ 17,457) were included in the study. The OAI was approved by the institutional review boards and was conducted in accordance with the Declaration of Helsinki. The clinical data are available for public use at http:// oai.epi-ucsf.org/datarelease/About.asp. The study design protocol is available at http://www.oai.ucsf.edu/data release/docs/StudyDesignProtocol.pdf. All variables used in this analysis were sampled from the publicly available OAI database. Sample selection of cases with discordant knee pain status. We used the 2-year followup clinical data from the OAI (version 3.2.1 clinical data) for selecting subjects for the current cross-sectional analysis, since pain status reported at the beginning of a study may be influenced by motivation to participate in that study. Subjects from the entire OAI sample were included based on self-reported knee-specific pain intensity. This rating (i.e., the numeric pain rating scale [NPRS]) was obtained by asking the fol-
376
Cotofana et al
Figure 1. Flow chart showing subject selection for the study. Numeric pain rating scale (NPRS; measures of pain severity according to the self-reported rating of knee pain during the last 7 days, where 0 ⫽ no pain and 10 ⫽ pain as bad as you can imagine); knee pain frequency was assessed according to the self-reported answer of pain occurrence during the past 12 months, ranging from no pain in the past 12 months to infrequent pain (pain in past 12 months but not on most days of a month) and frequent pain (pain on most days of a month in the past 12 months). BMI ⫽ body mass index; KLG ⫽ Kellgren/Lawrence grading.
lowing question: “Please rate the pain that you’ve had in your right (or left) knee during the past 7 days that best describes the pain at its worst: 0 means no pain and 10 means pain as bad as you can imagine.” This measure of pain intensity has been shown to have a high reliability and strong associations to changes in knee pain status following treatment (13,14). In agreement with previous reports on the patient-acceptable symptom state (15,16), we defined an NPRS of ⱖ4 as painful (i.e., non-acceptable pain), and an NPRS score of 0 as painless (17,18). The selection process further relied on answers given to questions pertaining to knee symptom frequency as follows: “During the past 12 months, have you had pain, aching, or stiffness in or around your right (or left) knee on most days for at least one month? Zero means no pain in the past 12 months, 1 means infrequent pain in the past 12 months, i.e., not on most days of a month, and 2 means frequent pain in past 12 months, i.e., on most days of a month in the past 12 months.” OAI subjects were identified as “discordant knee pain cases” for the current study if 1) one knee exhibited nonacceptable knee pain over the past 7 days (NPRS ⱖ4) and also was frequently (or infrequently) painful over the past 12 months (pain frequency 1 or 2); 2) the other (contralateral) knee exhibited no pain over the past 7 days (NPRS 0) that was not (or infrequently) painful over the past 12 months (pain frequency 0 or 1); and 3) there was complete
information on age, sex, body mass index (BMI), Kellgren/ Lawrence (K/L) grading (19) (from central readings, version 0.6/3.5), and leg dominance status at the 2-year followup time point for matching purposes (see below and Figure 1). Comparison with bilaterally pain-free and painful controls. As controls, we selected subjects who indicated bilaterally pain free or bilaterally painful knee status; bilaterally pain-free subjects had to display an NPRS of 0 and no (or infrequent) pain in both knees. This condition was fulfilled by 898 OAI subjects who also had complete information on age, sex, BMI, K/L grading, and limb dominance status at 2-year followup. Bilaterally painful subjects had to display an NPRS score ⱖ4 and frequent (or infrequent) pain in both knees. This condition was fulfilled by 534 OAI subjects who also had complete information on age, sex, BMI, K/L grading, and limb dominance data at 2-year followup. For comparing discordant knee pain cases with bilaterally painful or painless subjects, we used a matching algorithm that minimizes the squared differences between cases and controls. The pain-free knees in the discordant cases were matched with one of both pain-free knees in the bilaterally pain-free controls, and the painful knee in the discordant cases with one of both painful knees in the bilaterally painful controls by matching for 1) leg domi-
Knee-Specific Function and Contralateral Knee Effect nance status, 2) sex, 3) pain intensity status (when NPRS ⱖ4 preferably same value, but when not available, a maximum of 1–2 units difference), 4) age ⫾ 3 years, 5) BMI ⫾ 3 kg/m2, and 6) K/L grading on same stratum: either K/L grade 0 –1 or K/L grade 2– 4. In the following, the dominant leg was defined as the leg used to kick a ball (data collected and provided by the OAI) and the matched knee was defined as the one used for matching purpose in bilaterally pain free or painful when matched with the respective knee of the discordant knees. Functional measures examined. We included the following PROs and functional performance tests in the analysis (all were part of the OAI study): 1) the knee-specific WOMAC physical function score (scale 0 – 68, where 0 ⫽ no functional limitation and 68 ⫽ worst functional disability) relies on self-assessment (patient-reported) of everyday activities, such as stair use, standing up from a sitting or lying position, standing, bending, walking, getting in and out of a car, shopping, putting on or taking off socks, lying in bed, getting in or out of a bath, sitting, and heavy and light household duties over the past 7 days (9) and has undergone extensive validation (9,20,21), 2) the generic PASE (range 0 – 400 or more, with higher scores indicating greater physical activity) relies on the patient-reported levels of physical activity in subjects age ⱖ65 years and is comprised of items measuring occupational, household, and leisure activities during a 1-week period (22), and 3) the objective performance tests of global lower leg function: the time (in seconds) measured for getting up 5 times from a chair as quickly as possible (chair stand test), walking 20 meters as quickly as possible (20-meter walk test), and walking 400 meters as quickly as possible (400meter walk test). These functional performance tests reflect a variety of physical activities performed by persons with knee OA, are highly reliable (22–25), and cover aspects of endurance, strength, and balance. A more detailed description of the PROs and functional performance tests can be found in the OAI operations manuals (http://www.oai.ucsf.edu/datarelease/Operations Manuals.asp). Statistical analysis. Differences in scores were compared by using unpaired Student’s t-tests. The size of the differences was evaluated using the standardized response mean (SRM; mean difference divided by the SD). An SRM of 0.1 represents a “small” effect size, 0.3 represents a “medium” effect size, and 0.5 represents a “large” effect size (26). No formal hypotheses were tested and, therefore, no correction for multiple testing was performed. All statistical analyses were performed using SPSS, version 21.
RESULTS A “discordant knee pain status,” as defined in Subjects and Methods, was fulfilled by 378 OAI subjects, consisting of 208 women (55%) with a mean ⫾ SD age of 63.9 ⫾ 9.3 years. Of these, 359 knees were appropriately matched to a pain-free knee (of which the contralateral knee also was pain free) and 323 knees were matched to a painful knee
377 (of which the contralateral knee also was painful) (Figure 1). Demographic data of the 3 samples (discordant knee pain status, bilateral pain free knee status, and bilateral painful knee status) are presented in Table 1. Patient-reported, knee-specific WOMAC physical function score. In subjects with discordant knee pain status, the mean ⫾ SD WOMAC physical function score was 11.2 ⫾ 9.6 in the painful knee and 1.2 ⫾ 4.0 in the painfree knee. The mean ⫾ SD pairwise difference was 10.0 ⫾ 9.1 (95% confidence interval [95% CI] 9.1, 10.9, P ⬍ 0.01). In subjects with bilaterally pain-free knee status, the mean ⫾ SD WOMAC physical function score was 0.52 ⫾ 1.7 for the matched knee and 0.53 ⫾ 1.9 for the nonmatched knee. The mean ⫾ SD pairwise difference (matched compared to nonmatched) was ⫺0.004 ⫾ 1.19 (95% CI ⫺0.12, 0.11, P ⫽ 0.95). The mean ⫾ SD WOMAC physical function score was 0.59 ⫾ 1.9 in the right knee and 0.46 ⫾ 1.8 in the left knee (mean ⫾ SD pairwise difference 0.13 ⫾ 1.1 [95% CI 0.01, 0.25], P ⫽ 0.03). The mean ⫾ SD WOMAC physical function score was 0.59 ⫾ 1.9 in the dominant compared to 0.46 ⫾ 1.8 in the nondominant knee (mean ⫾ SD pairwise difference 0.13 ⫾ 1.1 [95% CI 0.01, 0.25], P ⫽ 0.03). In subjects with bilaterally painful knee status, the mean ⫾ SD WOMAC physical function score was 17.0 ⫾ 11.0 in the matched knee and 18.5 ⫾ 11.7 in the nonmatched knee. The mean ⫾ SD pairwise difference (matched compared to nonmatched) was ⫺1.5 ⫾ 8.7 (95% CI ⫺2.4, ⫺0.53, P ⫽ 0.002). The mean ⫾ SD score was 17.8 ⫾ 11.0 in the right and 17.7 ⫾ 11.7 in the left knee (mean ⫾ SD pairwise difference was 0.17 ⫾ 8.9 [95% CI ⫺0.81, 1.1], P ⫽ 0.74). The mean ⫾ SD WOMAC physical function score was 17.8 ⫾ 11.1 in the dominant and 17.7 ⫾ 11.7 in the nondominant knee (mean ⫾ SD pairwise difference 0.1 ⫾ 9.0 [95% CI ⫺0.9, 1.1], P ⫽ 0.85). Compared to the pain-free knee of the discordant cases, the matched pain-free knee of bilaterally pain-free subjects had a significantly lower mean ⫾ SD WOMAC physical function score (i.e., less functional limitations; mean pairwise difference was ⫺0.71 ⫾ 4.5 [95% CI ⫺1.18, ⫺0.24], P ⫽ 0.003). Compared with the painful knee of the discordant cases, the matched painful knee of bilaterally painful subjects had a significantly higher mean ⫾ SD WOMAC physical function score (i.e., more functional limitations; mean ⫾ SD pairwise difference 5.7 ⫾ 12.9 [95% CI 4.2, 7.1], P ⬍ 0.001). We performed a cross-sectional comparison to estimate potential effects of unilateral symptomatic treatment on WOMAC physical function, with the other knee showing similar pain status. The mean ⫾ SD pairwise difference between the painful knee in the discordant cases compared to the one in the bilaterally pain-free subjects was 11.0 ⫾ 9.9 (95% CI 10.0, 12.0, P ⬍ 0.001) with an SRM of 1.12. The mean ⫾ SD pairwise difference of one of the knees in the bilaterally painful subjects compared to the pain-free knee in the discordant cases was 11.5 ⫾ 14.5 (95% CI 9.9, 13.0, P ⬍ 0.001) with an SRM of 0.79. Subjective global activity measure (PASE). The PASE score was similar in subjects with discordant knee pain
378
Cotofana et al
Table 1. Demographic and physical performance data of the selected subjects at 2-year followup of the OAI: discordant knee pain status and bilateral pain-free and bilateral painful knee status*
Female sex, no. (%) Age, years BMI, kg/m2 Dominance right, no. (%) Pain intensity, pain free/painful knee (0–10) K/L grade, no. pain free/no. painful knee 0 1 2 3 4 Pain intensity, matched/nonmatched knee (0–10) K/L grade, no. matched/no. nonmatched knee 0 1 2 3 4 PASE Chair stand time, seconds† 20-meter walk time, seconds 400-meter walk time, seconds
Discordant knee pain status (n ⴝ 378)
Bilateral pain-free knee status (n ⴝ 359)
Bilateral painful knee status (n ⴝ 323)
208 (55) 63.9 ⫾ 9.3 28.7 ⫾ 4.5 342 (90.5) 0/5.5
201 (56) 64.3 ⫾ 8.9 28.5 ⫾ 4.1 333 (92.8)
185 (57) 63.7 ⫾ 8.9 29.0 ⫾ 4.2 292 (90.4)
0/0 100/128 41/86 161/94 52/40 5/6 100/128 145 ⫾ 80.6 9.6 ⫾ 2.3 15.2 ⫾ 2.3 304 ⫾ 44
5.6/6.0 77/86 48/54 93/84 80/71 25/23 77/86 151 ⫾ 81 12.1 ⫾ 3.8 16.3 ⫾ 3.3 323 ⫾ 58
148/91 78/52 97/79 46/86 7/70
149 ⫾ 75.9 10.6 ⫾ 2.8 15.6 ⫾ 2.9 304 ⫾ 66.1
* Values are the mean ⫾ SD unless indicated otherwise. OAI ⫽ Osteoarthritis Initiative; BMI ⫽ body mass index; K/L ⫽ Kellgren/Lawrence; PASE ⫽ Physical Activity Score for the Elderly. † Repeated 5 times.
status compared to those who showed bilaterally pain-free knee status (mean pairwise difference 2.3 ⫾ 104 [95% CI ⫺8.6, 13.2], P ⫽ 0.68) with an SRM of 0.02. Similarly, there was no significant difference in PASE of discordant compared to bilaterally painful subjects (mean ⫾ SD pairwise difference 2.2 ⫾ 99 [95% CI ⫺8.8, 13.2], P ⫽ 0.693) with an SRM of 0.02. Objective global functional performance tests. In subjects with discordant knee pain status, the chair stand time was significantly longer compared to subjects with bilaterally pain-free knee status (mean ⫾ SD pairwise difference was 0.97 ⫾ 3.5 [95% CI 0.59, 1.36], P ⬍ 0.001) with an SRM of 0.28 (Table 2).
The 20-meter walk test and the 400-meter walk test showed significant differences as well (mean ⫾ SD pairwise difference was 0.38 ⫾ 3.5 [95% CI 0.01, 0.75], P ⫽ 0.042 and 9.0 ⫾ 57.2 [95% CI 2.3, 15.6], P ⫽ 0.008) with an SRM of 0.11 and 0.16, respectively. When comparing subjects with discordant knee pain status to subjects with bilaterally painful knee status, the chair stand time was significantly shorter (mean ⫾ SD pairwise difference was 1.4 ⫾ 4.2 [95% CI 0.88, 1.9], P ⬍ 0.001) with an SRM of 0.33. The 20-meter walk time (mean ⫾ SD pairwise difference was 0.7 ⫾ 3.8 [95% CI 0.26, 1.13], P ⫽ 0.002) and the 400-meter walk time (mean ⫾ SD pairwise difference was 14.6 ⫾ 65.6 [95% CI 6.3, 22.8], P ⫽ 0.001) were significantly shorter (Table 3).
Table 2. Performance measures in matched pairs of individuals in the discordant and bilateral pain-free knee pain status stratum (n ⴝ 359)*
PASE Chair stand time, seconds‡ 20-meter walk time, seconds 400-meter walk time, seconds
Discordant knee pain status
Bilateral pain-free knee pain status
Mean pairwise difference
95% CI
P†
148 ⫾ 73.3 10.6 ⫾ 2.9 15.6 ⫾ 2.9 310 ⫾ 48.2
145 ⫾ 80.8 9.6 ⫾ 2.3 15.2 ⫾ 2.3 301 ⫾ 41.6
2.3 ⫾ 104 0.97 ⫾ 3.5 0.38 ⫾ 3.5 9.0 ⫾ 57.2
⫺8.6, 13.2 0.59, 1.36 0.01, 0.75 2.3, 15.6
0.679 ⬍ 0.001 0.042 0.008
* Values are the mean ⫾ SD unless indicated otherwise. 95% CI ⫽ 95% confidence interval; PASE ⫽ Physical Activity Score for the Elderly (22). † Using paired t-test. ‡ Repeated 5 times.
Knee-Specific Function and Contralateral Knee Effect
379
Table 3. Performance measures in matched pairs of individuals in the bilateral painful and discordant knee pain status stratum (n ⴝ 323)*
PASE Chair stand time, seconds‡ 20-meter walk time, seconds 400-meter walk time, seconds
Discordant knee pain status
Bilateral pain-free knee pain status
Mean pairwise difference
95% CI
P†
151 ⫾ 81 12.0 ⫾ 3.5 16.3 ⫾ 3.3 323 ⫾ 59
149 ⫾ 74 10.6 ⫾ 2.8 15.6 ⫾ 2.6 308 ⫾ 47
2.2 ⫾ 99 1.4 ⫾ 4.2 0.70 ⫾ 3.8 14.6 ⫾ 65.6
⫺8.8, 13.2 0.88, 1.9 0.26, 1.13 6.3, 22.8
0.693 ⬍ 0.001 0.002 0.001
* Values are the mean ⫾ SD unless indicated otherwise. 95% CI ⫽ 95% confidence interval; PASE ⫽ Physical Activity Score for the Elderly (22). † Using paired t-test. ‡ Repeated 5 times.
DISCUSSION The main findings of this study were that of the patientreported WOMAC physical function score; although knee specific, it is impacted by the contralateral knee pain status. That means functional limitations in a painful knee are perceived as worse when the contralateral knee is painful, and functional limitations in a pain-free knee are perceived as worse when the contralateral knee is painful. Nevertheless, because the WOMAC is knee specific, it differentiated best between painful and painless knee status when the contralateral knee pain status was similar. The self-assessed levels of physical activity (PASE) did not show the ability to differentiate between subjects with discordant knee pain status and matched controls, whereas objective functional performance tests, particularly the chair stand test, exhibited more sensitivity in differentiating between the pain status of one knee, even if the pain status in the other (contralateral knee) was similar. A strength of the current analysis is the strict matching procedure of the case– control design, taking into account major confounders in knee OA, i.e., age, sex, BMI, radiographic knee OA, and leg dominance status. Previous studies showed that the WOMAC physical function score is highly correlated to pain in the same knee (23,27) and is unable to differentiate between pain and function in construct validation studies (28,29). Further, it has been shown that the WOMAC physical function score shows a weak to moderate correlation with functional performance tests (28,29). The correlation of the WOMAC physical function score with knee pain intensity is generally higher than the correlation of functional performance tests with knee pain intensity (30). Although our results suggest that, likely because of its specificity to a single knee, the WOMAC physical function score is the most sensitive to a unilateral difference in pain, the evidence in literature indicates that the patient-reported WOMAC physical function score does not provide information that is independent (and hence complementary) from/to pain itself. We have no explanation why the subjective global PASE score displayed less sensitivity in differentiating unilateral knee pain, but our results discourage its use in a clinical study that aims to document the effect of unilateral symptomatic treatment on knee function. However, objective
physical performance tests were capable of detecting unilateral differences in knee pain, with the repeated chair stand test revealing the highest sensitivity compared to short- or long-distance walking tests. Our findings differ from those of Marmon et al (31) who reported that persons with knee OA perform and perceive their functional ability similarly, regardless of the number of knees involved. Our results in discordant cases versus controls with bilaterally painful knees are in line with a recent study by Riddle and Stratford (30) who found a stronger contrast between OAI subjects with severe unilateral compared to bilateral pain for the repeated chair stand than for walking tests. However, Riddle and Stratford (30) included only subjects with K/L grade 3 or 4 in at least 1 of both knees, i.e., a relatively advanced stage of radiographic knee OA at which subjects are not commonly recruited for clinical trials. Further, our study used a different (matched pair) design and also compared subjects with unilateral pain (i.e., discordant pain status) with those who did not have pain in either knee, in order to estimate the effect of symptomatic treatment on function in patients with unilateral pain at baseline. A limitation of the current study is that it relies on cross-sectional analysis, comparing subjects with different combinations of bilateral knee pain. These results must therefore be confirmed in longitudinal treatment trials in which effective symptomatic relief is achieved in one knee, whereas pain in the other knee remains unaltered. Further, because we compared knees with non-acceptable symptom status (NPRS ⱖ4) versus pain-free knees (NPRS 0), and because it is unlikely that treatment can transform a knee from non-acceptable to totally pain free, the current results can only provide an estimate of the “upper margin” of the effects on leg function that can be achieved with unilateral symptomatic treatment. Nevertheless, the relative performance of different measures (subjective, objective, knee-specific, and global) may hold and therefore may provide useful indicators of which measures are preferably used in a clinical trial. Ideally we would have liked to apply more strict criteria (e.g., only subjects with one knee frequently painful and the contralateral knee with no pain at all). However, this would considerably reduce the number of discordant cases and the statistical power of the study. In conclusion, we showed that the patient-reported
380
Cotofana et al
knee-specific WOMAC physical function score is impacted by the contralateral knee pain status. The WOMAC physical function score appears suitable to capture unilateral knee pain; however, given its subjective nature, its weak to moderate correlation with objectively measured functional performance, and its moderate to high correlation with subjective measures of pain, its value to demonstrate benefit in joint function under (unilateral) symptomatic treatment seems to be limited. Objective functional performance tests were suitable to capture differences in unilateral knee pain. The repeated chair stand test was the most appropriate and, based on the results of this study, it would be one of the preferred assessments to be used, for instance, in clinical trials evaluating the effect of unilateral symptomatic treatment on knee function.
ACKNOWLEDGMENTS The authors would like thank the OAI participants, investigators, technicians, and the coordinating center for providing public access to the OAI data. AUTHOR CONTRIBUTIONS All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be submitted for publication. Dr. Cotofana had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study conception and design. Cotofana, Rossi, Eckstein, Gu¨nther. Analysis and interpretation of data. Cotofana, Wirth, Rossi, Eckstein, Gu¨nther.
ROLE OF THE STUDY SPONSOR The Osteoarthritis Initiative study sponsors (Merck Research Laboratories, Novartis Pharmaceuticals Corporation, GlaxoSmithKline, and Pfizer) had no role in the study design, data analysis, data interpretation, writing of this manuscript, or the decision to submit the manuscript for publication. Publication of this article was not contingent upon approval by these sponsors. Merck KGaA was involved in the study design, data collection, data analysis and interpretation, the writing of the manuscript, and in the decision to submit the manuscript for publication. Publication of this article was contingent upon approval by Merck KGaA.
ADDITIONAL DISCLOSURE Author Cotofana is a part-time employee with Chondrometrics GmbH. Author Wirth is a part-time employee and co-owner of Chondrometrics GmbH. Author Pena Rossi was an employee at EMD Serono at the time of the study. Author Eckstein is Chief Executive Officer and co-owner of Chondrometrics GmbH. Author Gu¨nther is an employee of Merck KGaA.
REFERENCES 1. Guccione AA, Felson DT, Anderson JJ, Anthony JM, Zhang Y, Wilson PW, et al. The effects of specific medical conditions on the functional limitations of elders in the Framingham Study. Am J Public Health 1994;84:351– 8. 2. Perrot S, Poiraudeau S, Kabir-Ahmadi M, Rannou F. Correlates of pain intensity in men and women with hip and knee osteoarthritis: results of a national survey. The French ARTHRIX study. Clin J Pain 2009;25:767–72.
3. Stratford PW, Kennedy D, Pagura SM, Gollish JD. The relationship between self-report and performance-related measures: questioning the content validity of timed tests. Arthritis Rheum 2003;49:535– 40. 4. McAlindon TE, Cooper C, Kirwan JR, Dieppe PA. Determinants of disability in osteoarthritis of the knee. Ann Rheum Dis 1993;52:258 – 62. 5. Jordan J, Luta G, Renner J, Dragomir A, Hochberg M, Fryer J. Knee pain and knee osteoarthritis severity in self-reported task specific disability: the Johnston County Osteoarthritis Project. J Rheumatol 1997;24:1344 –9. 6. Jinks C, Jordan K, Croft P. Osteoarthritis as a public health problem: the impact of developing knee pain on physical function in adults living in the community: (KNEST 3). Rheumatology (Oxford) 2007;46:877– 81. 7. Zhang W, Moskowitz RW, Nuki G, Abramson S, Altman RD, Arden N, 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– 62. 8. Bannuru RR, Vaysbrot EE, Sullivan MC, McAlindon TE. Relative efficacy of hyaluronic acid in comparison with NSAIDs for knee osteoarthritis: a systematic review and meta-analysis. Semin Arthritis Rheum 2014;43:593–9. 9. Bellamy N, Buchanan WW. A preliminary evaluation of the dimensionality and clinical importance of pain and disability in osteoarthritis of the hip and knee. Clin Rheumatol 1986;5: 231– 41. 10. Eckstein F, Kwoh CK. Imaging in rheumatology in 2013: from images to data to theory [review]. Nat Rev Rheumatol 2014; 10:69 –70. 11. Lester G. The Osteoarthritis Initiative: an NIH public-private partnership. HSS J 2012;8:62–3. 12. Altman RD, Hochberg M, Murphy WA Jr, Wolfe F, Lequesne M. Atlas of individual radiographic features in osteoarthritis. Osteoarthritis Cartilage 1995;3 Suppl A:3–70. 13. Pautex S, Herrmann F, Le Lous P, Fabjan M, Michel JP, Gold G. Feasibility and reliability of four pain self-assessment scales and correlation with an observational rating scale in hospitalized elderly demented patients. J Gerontol A Biol Sci Med Sci 2005;60:524 –9. 14. Jensen MP. The validity and reliability of pain measures in adults with cancer. J Pain 2003;4:2–21. 15. Tubach F, Dougados M, Falissard B, Baron G, Logeart I, Ravaud P. Feeling good rather than feeling better matters more to patients. Arthritis Rheum 2006;55:526 –30. 16. Tubach F, Ravaud P, Baron G, Falissard B, Logeart I, Bellamy N, et al. Evaluation of clinically relevant states in patient reported outcomes in knee and hip osteoarthritis: the patient acceptable symptom state. Ann Rheum Dis 2005;64:34 –7. 17. Moisio K, Eckstein F, Chmiel JS, Guermazi A, Prasad P, Almagor O, et al. Denuded subchondral bone and knee pain in persons with knee osteoarthritis. Arthritis Rheum 2009;60: 3703–10. 18. Ornetti P, Dougados M, Paternotte S, Logeart I, Gossec L. Validation of a numerical rating scale to assess functional impairment in hip and knee osteoarthritis: comparison with the WOMAC function scale. Ann Rheum Dis 2011;70:740 – 6. 19. Kellgren JH, Lawrence JS. Radiological assessment of rheumatoid arthritis. Ann Rheum Dis 1957;16:485–93. 20. Bellamy N, Buchanan WW, Goldsmith CH, Campbell J, Stitt LW. 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– 40. 21. Bellamy N, Kirwan J, Boers M, Brooks P, Strand V, Tugwell P, et al. Recommendations for a core set of outcome measures for future phase III clinical trials in knee, hip, and hand osteoarthritis: consensus development at OMERACT III. J Rheumatol 1997;24:799 – 802. 22. Washburn RA, Smith KW, Jette AM, Janney CA. The Physical Activity Scale for the Elderly (PASE): development and evaluation. J Clin Epidemiol 1993;46:153– 62.
Knee-Specific Function and Contralateral Knee Effect 23. Stratford PW, Kennedy DM, Woodhouse LJ. Performance measures provide assessments of pain and function in people with advanced osteoarthritis of the hip or knee. Phys Ther 2006;86:1489 –96. 24. Curb JD, Ceria-Ulep CD, Rodriguez BL, Grove J, Guralnik J, Willcox BJ, et al. Performance-based measures of physical function for high-function populations. J Am Geriatr Soc 2006;54:737– 42. 25. Gill S, McBurney H. Reliability of performance-based measures in people awaiting joint replacement surgery of the hip or knee. Physiother Res Int 2008;13:141–52. 26. Cohen J. A power primer. Psychol Bull 1992;112:155–9. 27. Parent E, Moffet H. Comparative responsiveness of locomotor tests and questionnaires used to follow early recovery after total knee arthroplasty. Arch Phys Med Rehabil 2002;83:70 – 80.
381 28. Thumboo J, Chew LH, Soh CH. Validation of the Western Ontario and McMaster Universities Osteoarthritis Index in Asians with osteoarthritis in Singapore. Osteoarthritis Cartilage 2001;9:440 – 6. 29. Faucher M, Poiraudeau S, Lefevre-Colau MM, Rannou F, Fermanian J, Revel M. Algo-functional assessment of knee osteoarthritis: comparison of the test-retest reliability and construct validity of the WOMAC and Lequesne indexes. Osteoarthritis Cartilage 2002;10:602–10. 30. Riddle DL, Stratford PW. Unilateral vs bilateral symptomatic knee osteoarthritis: associations between pain intensity and function. Rheumatology (Oxford) 2013;52:2229 –37. 31. Marmon AR, Zeni JA Jr, Snyder-Mackler L. Perception and presentation of function in patients with unilateral versus bilateral knee osteoarthritis. Arthritis Care Res (Hoboken) 2013;65:406 –13.
ORIGINAL ARTICLE
Contralateral Knee Effect on Self-Reported Knee-Specific Function and Global Functional Assessment: Data From the Osteoarthritis Initiative SEBASTIAN COTOFANA,1 WOLFGANG WIRTH,1 CLAUDIA PENA ROSSI,2 FELIX ECKSTEIN,1 ¨ NTHER3 AND OLIVER H. GU
Objective. To analyze the effect of contralateral knee pain on sensitivity of patient-reported outcomes and objectively measured functional performance tests in subjects with knee osteoarthritis (OA). Methods. Subjects with discordant knee pain status (i.e., 1 knee being painful [>4 on a numeric pain rating scale (NPRS)], with the contralateral knee being pain free [NPRS 0]) were selected from the Osteoarthritis Initiative and matched to subjects with bilaterally pain-free and painful knees by age, sex, body mass index, and radiographic knee OA. The Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) physical function score, the global Physical Activity Scale for the Elderly (PASE), and objective functional performance tests were cross-sectionally compared in a matched case– control design. Results. A total of 378 subjects with discordant knee pain status were matched to 359 controls with bilaterally pain-free knees and to 323 controls with bilaterally painful knees. WOMAC scores in pain-free knees of discordant knee pain cases significantly differed compared to scores of bilaterally pain-free knees (P ⴝ 0.003). Likewise, scores in painful knees of discordant knee pain cases significantly differed compared to scores of bilaterally painful knees (P < 0.001). PASE levels between these groups were not significantly different (P > 0.68). Functional performance tests differed in subjects with discordant knee pain compared to subjects with bilaterally pain-free knees and when compared to subjects with bilaterally painful knees, with the chair stand test showing the strongest effect size (standardized response mean 0.28 and 0.33, respectively). Conclusion. The WOMAC physical function score, although knee specific, is impacted by the contralateral knee pain status. The repeated chair stand test appears to be the most sensitive assessment in differentiation between groups with different status of knee pain.
INTRODUCTION Knee osteoarthritis (OA) is the most common type of arthritis and is the leading cause of functional disability and impaired health-related quality of life in the elderly in comparison with any other chronic disease (1). Knee pain is known to be associated with both self-reported funcThe Osteoarthritis Initiative is a public-private partnership comprised of 5 contracts (N01-AR-2-2258, N01-AR-22259, N01-AR-2-2260, N01-AR-2-2261, and N01-AR-2-2262) funded by the NIH, a branch of the US Department of Health and Human Services, and conducted by the Osteoarthritis Initiative Investigators. Private funding partners include Merck Research Laboratories, Novartis Pharmaceuticals, GlaxoSmithKline, and Pfizer. Private sector funding for the OAI is managed by the Foundation for the NIH. The specific analysis that supported the current study was funded by Merck KGaA, Darmstadt, Germany. Drs. Cotofana, Wirth, and Eckstein’s work was supported by Merck KGaA.
374
tional limitations (2) and physical performance (3), including difficulties in walking or getting up from a chair (4 – 6). Generally, improvement of knee pain by medical treatment is expected to result in improved physical performance (7), but when symptomatic treatment is adminis1 Sebastian Cotofana, MD, Wolfgang Wirth, PhD, Felix Eckstein, MD, PhD: Institute of Anatomy, Paracelsus Medical University, Salzburg, Austria, and Chondrometrics GmbH, Ainring, Germany; 2Claudia Pena Rossi, MD, PhD: EMD Serono, Rockland, Maryland; 3Oliver H. Gu¨nther, PhD: Merck KGaA, Darmstadt, Germany. Dr. Cotofana has a part-time appointment with Chondrometrics GmbH. Dr. Wirth has a part-time appointment with Chondrometrics GmbH and is co-owner of Chondrometrics GmbH, has received consultant fees, speaking fees, and/or honoraria (less than $10,000) from Merck Serono Switzerland, and owns stock or stock options in Chondrometrics GmbH. Dr. Pena Rossi has patents pending for combination
Knee-Specific Function and Contralateral Knee Effect
Significance & Innovations ●
Regarding the patient-reported Western Ontario and McMaster Universities Osteoarthritis Index physical function score, functional limitations in a painful knee are perceived worse when the contralateral knee is painful, and functional limitations in a pain-free knee are perceived worse when the contralateral knee is painful.
●
Self-assessed levels of the Physical Activity Score in the Elderly do not discriminate between subjects with discordant knee pain status, whereas walking tests, and particularly the chair stand test, displayed high sensitivity in differentiating the pain status of one knee.
●
Objective functional performance tests, particularly the chair stand test, showed more sensitivity in differentiating between the pain status of one knee, even if the pain status in the other (contralateral knee) was similar.
tered by intraarticular injection, only the injected knee can be expected to improve (8). Patient reported outcome (PRO) measures of functional performance in daily activities, as evaluated for each knee by the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) (9), may be biased by functional limitations imposed by the pain status of the contralateral (untreated) knee. Global activity levels and objective functional performance tests critically depend on the interaction and functionality of both knees. Hence, it is unclear to what extent the pain status of the contralateral knee impacts the sensitivity of such instruments or tests, particularly when symptomatic improvement is achieved in one knee by intraarticular injection. The purpose of the current study was to analyze the effect of contralateral knee pain status on the sensitivity of knee-specific (WOMAC physical function score), generic (Physical Activity Scale for the Elderly [PASE]) PRO measures, and global functional performance tests (repeated of BlyS and/or APRIL inhibition and immunosuppressants for treatment of autoimmune disease and for dosing methods for treating autoimmune diseases using a TACI-Ig fusion protein such as atacicept. Dr. Eckstein is Chief Executive Officer and co-owner of Chondrometrics GmbH, has received consultant fees, speaking fees, and/or honoraria (less than $10,000 each) from Merck Serono, Novartis and Abbvie, owns stock or stock options in Chondrometrics GmbH, has received funding support not related to this study from Pfizer, Eli Lilly, Novartis, Merck Serono, GlaxoSmithKline, Wyeth, Centocor, Stryker, AbbVie, Kolon, and Synarc, and has provided educational content for Medtronic. Address correspondence to Sebastian Cotofana, MD, Institute of Anatomy, Paracelsus Medical University, Strubergasse 21, A5020 Salzburg, Austria. E-mail: sebastian. [email protected]. Submitted for publication February 28, 2014; accepted in revised form October 7, 2014.
375 chair stand test, the 20-meter walk test, and the 400-meter walk test). Therefore, we created 2 hypothetical scenarios using data from the Osteoarthritis Initiative (OAI) and analyzed it by using a cross-sectional case– control study design. The first scenario assumes that both knees are painful before treatment and one of them is treated successfully. To estimate the sensitivity of PRO measures and global functional performance tests, we compared subjects with bilaterally painful knees to subjects with a discordant knee pain status (one knee being painful, the other one not). The second scenario assumes that one knee is painful before treatment and this knee is successfully treated. To estimate the sensitivity of PRO measures and global functional performance tests, we compared subjects with discordant pain status to those with bilaterally pain-free knees.
SUBJECTS AND METHODS The OAI. The OAI is a privately and National Institutes of Health–funded longitudinal cohort study of subjects with, or at risk of developing, symptomatic knee OA (10,11). Individuals of multiple ethnicities, ages 45–75 years, were included in this cohort study. Subjects were excluded from the OAI if the following conditions were present: rheumatoid arthritis, bilateral knee arthroplasty or preexisting plans to undergo bilateral (not unilateral) knee arthroplasty in the next 3 years, bilateral Osteoarthritis Research Society International stage 3 (severe) knee OA (12), positive pregnancy test, inability to provide a blood sample, use of ambulatory aids other than a single straight cane for ⬎50% of the time, comorbid conditions that might interfere with 4-year participation, unlikely to reside in the clinic area for at least 3 years, current participation in a double-blind, randomized controlled trial, and being unwilling to sign an informed consent. Men weighing ⬎130 kg and women weighing ⬎114 kg were also excluded, due to specific requirements with the acquisition of magnetic resonance images not relevant to the current study. In total, 27% (n ⫽ 4,796) of all subjects screened (n ⫽ 17,457) were included in the study. The OAI was approved by the institutional review boards and was conducted in accordance with the Declaration of Helsinki. The clinical data are available for public use at http:// oai.epi-ucsf.org/datarelease/About.asp. The study design protocol is available at http://www.oai.ucsf.edu/data release/docs/StudyDesignProtocol.pdf. All variables used in this analysis were sampled from the publicly available OAI database. Sample selection of cases with discordant knee pain status. We used the 2-year followup clinical data from the OAI (version 3.2.1 clinical data) for selecting subjects for the current cross-sectional analysis, since pain status reported at the beginning of a study may be influenced by motivation to participate in that study. Subjects from the entire OAI sample were included based on self-reported knee-specific pain intensity. This rating (i.e., the numeric pain rating scale [NPRS]) was obtained by asking the fol-
376
Cotofana et al
Figure 1. Flow chart showing subject selection for the study. Numeric pain rating scale (NPRS; measures of pain severity according to the self-reported rating of knee pain during the last 7 days, where 0 ⫽ no pain and 10 ⫽ pain as bad as you can imagine); knee pain frequency was assessed according to the self-reported answer of pain occurrence during the past 12 months, ranging from no pain in the past 12 months to infrequent pain (pain in past 12 months but not on most days of a month) and frequent pain (pain on most days of a month in the past 12 months). BMI ⫽ body mass index; KLG ⫽ Kellgren/Lawrence grading.
lowing question: “Please rate the pain that you’ve had in your right (or left) knee during the past 7 days that best describes the pain at its worst: 0 means no pain and 10 means pain as bad as you can imagine.” This measure of pain intensity has been shown to have a high reliability and strong associations to changes in knee pain status following treatment (13,14). In agreement with previous reports on the patient-acceptable symptom state (15,16), we defined an NPRS of ⱖ4 as painful (i.e., non-acceptable pain), and an NPRS score of 0 as painless (17,18). The selection process further relied on answers given to questions pertaining to knee symptom frequency as follows: “During the past 12 months, have you had pain, aching, or stiffness in or around your right (or left) knee on most days for at least one month? Zero means no pain in the past 12 months, 1 means infrequent pain in the past 12 months, i.e., not on most days of a month, and 2 means frequent pain in past 12 months, i.e., on most days of a month in the past 12 months.” OAI subjects were identified as “discordant knee pain cases” for the current study if 1) one knee exhibited nonacceptable knee pain over the past 7 days (NPRS ⱖ4) and also was frequently (or infrequently) painful over the past 12 months (pain frequency 1 or 2); 2) the other (contralateral) knee exhibited no pain over the past 7 days (NPRS 0) that was not (or infrequently) painful over the past 12 months (pain frequency 0 or 1); and 3) there was complete
information on age, sex, body mass index (BMI), Kellgren/ Lawrence (K/L) grading (19) (from central readings, version 0.6/3.5), and leg dominance status at the 2-year followup time point for matching purposes (see below and Figure 1). Comparison with bilaterally pain-free and painful controls. As controls, we selected subjects who indicated bilaterally pain free or bilaterally painful knee status; bilaterally pain-free subjects had to display an NPRS of 0 and no (or infrequent) pain in both knees. This condition was fulfilled by 898 OAI subjects who also had complete information on age, sex, BMI, K/L grading, and limb dominance status at 2-year followup. Bilaterally painful subjects had to display an NPRS score ⱖ4 and frequent (or infrequent) pain in both knees. This condition was fulfilled by 534 OAI subjects who also had complete information on age, sex, BMI, K/L grading, and limb dominance data at 2-year followup. For comparing discordant knee pain cases with bilaterally painful or painless subjects, we used a matching algorithm that minimizes the squared differences between cases and controls. The pain-free knees in the discordant cases were matched with one of both pain-free knees in the bilaterally pain-free controls, and the painful knee in the discordant cases with one of both painful knees in the bilaterally painful controls by matching for 1) leg domi-
Knee-Specific Function and Contralateral Knee Effect nance status, 2) sex, 3) pain intensity status (when NPRS ⱖ4 preferably same value, but when not available, a maximum of 1–2 units difference), 4) age ⫾ 3 years, 5) BMI ⫾ 3 kg/m2, and 6) K/L grading on same stratum: either K/L grade 0 –1 or K/L grade 2– 4. In the following, the dominant leg was defined as the leg used to kick a ball (data collected and provided by the OAI) and the matched knee was defined as the one used for matching purpose in bilaterally pain free or painful when matched with the respective knee of the discordant knees. Functional measures examined. We included the following PROs and functional performance tests in the analysis (all were part of the OAI study): 1) the knee-specific WOMAC physical function score (scale 0 – 68, where 0 ⫽ no functional limitation and 68 ⫽ worst functional disability) relies on self-assessment (patient-reported) of everyday activities, such as stair use, standing up from a sitting or lying position, standing, bending, walking, getting in and out of a car, shopping, putting on or taking off socks, lying in bed, getting in or out of a bath, sitting, and heavy and light household duties over the past 7 days (9) and has undergone extensive validation (9,20,21), 2) the generic PASE (range 0 – 400 or more, with higher scores indicating greater physical activity) relies on the patient-reported levels of physical activity in subjects age ⱖ65 years and is comprised of items measuring occupational, household, and leisure activities during a 1-week period (22), and 3) the objective performance tests of global lower leg function: the time (in seconds) measured for getting up 5 times from a chair as quickly as possible (chair stand test), walking 20 meters as quickly as possible (20-meter walk test), and walking 400 meters as quickly as possible (400meter walk test). These functional performance tests reflect a variety of physical activities performed by persons with knee OA, are highly reliable (22–25), and cover aspects of endurance, strength, and balance. A more detailed description of the PROs and functional performance tests can be found in the OAI operations manuals (http://www.oai.ucsf.edu/datarelease/Operations Manuals.asp). Statistical analysis. Differences in scores were compared by using unpaired Student’s t-tests. The size of the differences was evaluated using the standardized response mean (SRM; mean difference divided by the SD). An SRM of 0.1 represents a “small” effect size, 0.3 represents a “medium” effect size, and 0.5 represents a “large” effect size (26). No formal hypotheses were tested and, therefore, no correction for multiple testing was performed. All statistical analyses were performed using SPSS, version 21.
RESULTS A “discordant knee pain status,” as defined in Subjects and Methods, was fulfilled by 378 OAI subjects, consisting of 208 women (55%) with a mean ⫾ SD age of 63.9 ⫾ 9.3 years. Of these, 359 knees were appropriately matched to a pain-free knee (of which the contralateral knee also was pain free) and 323 knees were matched to a painful knee
377 (of which the contralateral knee also was painful) (Figure 1). Demographic data of the 3 samples (discordant knee pain status, bilateral pain free knee status, and bilateral painful knee status) are presented in Table 1. Patient-reported, knee-specific WOMAC physical function score. In subjects with discordant knee pain status, the mean ⫾ SD WOMAC physical function score was 11.2 ⫾ 9.6 in the painful knee and 1.2 ⫾ 4.0 in the painfree knee. The mean ⫾ SD pairwise difference was 10.0 ⫾ 9.1 (95% confidence interval [95% CI] 9.1, 10.9, P ⬍ 0.01). In subjects with bilaterally pain-free knee status, the mean ⫾ SD WOMAC physical function score was 0.52 ⫾ 1.7 for the matched knee and 0.53 ⫾ 1.9 for the nonmatched knee. The mean ⫾ SD pairwise difference (matched compared to nonmatched) was ⫺0.004 ⫾ 1.19 (95% CI ⫺0.12, 0.11, P ⫽ 0.95). The mean ⫾ SD WOMAC physical function score was 0.59 ⫾ 1.9 in the right knee and 0.46 ⫾ 1.8 in the left knee (mean ⫾ SD pairwise difference 0.13 ⫾ 1.1 [95% CI 0.01, 0.25], P ⫽ 0.03). The mean ⫾ SD WOMAC physical function score was 0.59 ⫾ 1.9 in the dominant compared to 0.46 ⫾ 1.8 in the nondominant knee (mean ⫾ SD pairwise difference 0.13 ⫾ 1.1 [95% CI 0.01, 0.25], P ⫽ 0.03). In subjects with bilaterally painful knee status, the mean ⫾ SD WOMAC physical function score was 17.0 ⫾ 11.0 in the matched knee and 18.5 ⫾ 11.7 in the nonmatched knee. The mean ⫾ SD pairwise difference (matched compared to nonmatched) was ⫺1.5 ⫾ 8.7 (95% CI ⫺2.4, ⫺0.53, P ⫽ 0.002). The mean ⫾ SD score was 17.8 ⫾ 11.0 in the right and 17.7 ⫾ 11.7 in the left knee (mean ⫾ SD pairwise difference was 0.17 ⫾ 8.9 [95% CI ⫺0.81, 1.1], P ⫽ 0.74). The mean ⫾ SD WOMAC physical function score was 17.8 ⫾ 11.1 in the dominant and 17.7 ⫾ 11.7 in the nondominant knee (mean ⫾ SD pairwise difference 0.1 ⫾ 9.0 [95% CI ⫺0.9, 1.1], P ⫽ 0.85). Compared to the pain-free knee of the discordant cases, the matched pain-free knee of bilaterally pain-free subjects had a significantly lower mean ⫾ SD WOMAC physical function score (i.e., less functional limitations; mean pairwise difference was ⫺0.71 ⫾ 4.5 [95% CI ⫺1.18, ⫺0.24], P ⫽ 0.003). Compared with the painful knee of the discordant cases, the matched painful knee of bilaterally painful subjects had a significantly higher mean ⫾ SD WOMAC physical function score (i.e., more functional limitations; mean ⫾ SD pairwise difference 5.7 ⫾ 12.9 [95% CI 4.2, 7.1], P ⬍ 0.001). We performed a cross-sectional comparison to estimate potential effects of unilateral symptomatic treatment on WOMAC physical function, with the other knee showing similar pain status. The mean ⫾ SD pairwise difference between the painful knee in the discordant cases compared to the one in the bilaterally pain-free subjects was 11.0 ⫾ 9.9 (95% CI 10.0, 12.0, P ⬍ 0.001) with an SRM of 1.12. The mean ⫾ SD pairwise difference of one of the knees in the bilaterally painful subjects compared to the pain-free knee in the discordant cases was 11.5 ⫾ 14.5 (95% CI 9.9, 13.0, P ⬍ 0.001) with an SRM of 0.79. Subjective global activity measure (PASE). The PASE score was similar in subjects with discordant knee pain
378
Cotofana et al
Table 1. Demographic and physical performance data of the selected subjects at 2-year followup of the OAI: discordant knee pain status and bilateral pain-free and bilateral painful knee status*
Female sex, no. (%) Age, years BMI, kg/m2 Dominance right, no. (%) Pain intensity, pain free/painful knee (0–10) K/L grade, no. pain free/no. painful knee 0 1 2 3 4 Pain intensity, matched/nonmatched knee (0–10) K/L grade, no. matched/no. nonmatched knee 0 1 2 3 4 PASE Chair stand time, seconds† 20-meter walk time, seconds 400-meter walk time, seconds
Discordant knee pain status (n ⴝ 378)
Bilateral pain-free knee status (n ⴝ 359)
Bilateral painful knee status (n ⴝ 323)
208 (55) 63.9 ⫾ 9.3 28.7 ⫾ 4.5 342 (90.5) 0/5.5
201 (56) 64.3 ⫾ 8.9 28.5 ⫾ 4.1 333 (92.8)
185 (57) 63.7 ⫾ 8.9 29.0 ⫾ 4.2 292 (90.4)
0/0 100/128 41/86 161/94 52/40 5/6 100/128 145 ⫾ 80.6 9.6 ⫾ 2.3 15.2 ⫾ 2.3 304 ⫾ 44
5.6/6.0 77/86 48/54 93/84 80/71 25/23 77/86 151 ⫾ 81 12.1 ⫾ 3.8 16.3 ⫾ 3.3 323 ⫾ 58
148/91 78/52 97/79 46/86 7/70
149 ⫾ 75.9 10.6 ⫾ 2.8 15.6 ⫾ 2.9 304 ⫾ 66.1
* Values are the mean ⫾ SD unless indicated otherwise. OAI ⫽ Osteoarthritis Initiative; BMI ⫽ body mass index; K/L ⫽ Kellgren/Lawrence; PASE ⫽ Physical Activity Score for the Elderly. † Repeated 5 times.
status compared to those who showed bilaterally pain-free knee status (mean pairwise difference 2.3 ⫾ 104 [95% CI ⫺8.6, 13.2], P ⫽ 0.68) with an SRM of 0.02. Similarly, there was no significant difference in PASE of discordant compared to bilaterally painful subjects (mean ⫾ SD pairwise difference 2.2 ⫾ 99 [95% CI ⫺8.8, 13.2], P ⫽ 0.693) with an SRM of 0.02. Objective global functional performance tests. In subjects with discordant knee pain status, the chair stand time was significantly longer compared to subjects with bilaterally pain-free knee status (mean ⫾ SD pairwise difference was 0.97 ⫾ 3.5 [95% CI 0.59, 1.36], P ⬍ 0.001) with an SRM of 0.28 (Table 2).
The 20-meter walk test and the 400-meter walk test showed significant differences as well (mean ⫾ SD pairwise difference was 0.38 ⫾ 3.5 [95% CI 0.01, 0.75], P ⫽ 0.042 and 9.0 ⫾ 57.2 [95% CI 2.3, 15.6], P ⫽ 0.008) with an SRM of 0.11 and 0.16, respectively. When comparing subjects with discordant knee pain status to subjects with bilaterally painful knee status, the chair stand time was significantly shorter (mean ⫾ SD pairwise difference was 1.4 ⫾ 4.2 [95% CI 0.88, 1.9], P ⬍ 0.001) with an SRM of 0.33. The 20-meter walk time (mean ⫾ SD pairwise difference was 0.7 ⫾ 3.8 [95% CI 0.26, 1.13], P ⫽ 0.002) and the 400-meter walk time (mean ⫾ SD pairwise difference was 14.6 ⫾ 65.6 [95% CI 6.3, 22.8], P ⫽ 0.001) were significantly shorter (Table 3).
Table 2. Performance measures in matched pairs of individuals in the discordant and bilateral pain-free knee pain status stratum (n ⴝ 359)*
PASE Chair stand time, seconds‡ 20-meter walk time, seconds 400-meter walk time, seconds
Discordant knee pain status
Bilateral pain-free knee pain status
Mean pairwise difference
95% CI
P†
148 ⫾ 73.3 10.6 ⫾ 2.9 15.6 ⫾ 2.9 310 ⫾ 48.2
145 ⫾ 80.8 9.6 ⫾ 2.3 15.2 ⫾ 2.3 301 ⫾ 41.6
2.3 ⫾ 104 0.97 ⫾ 3.5 0.38 ⫾ 3.5 9.0 ⫾ 57.2
⫺8.6, 13.2 0.59, 1.36 0.01, 0.75 2.3, 15.6
0.679 ⬍ 0.001 0.042 0.008
* Values are the mean ⫾ SD unless indicated otherwise. 95% CI ⫽ 95% confidence interval; PASE ⫽ Physical Activity Score for the Elderly (22). † Using paired t-test. ‡ Repeated 5 times.
Knee-Specific Function and Contralateral Knee Effect
379
Table 3. Performance measures in matched pairs of individuals in the bilateral painful and discordant knee pain status stratum (n ⴝ 323)*
PASE Chair stand time, seconds‡ 20-meter walk time, seconds 400-meter walk time, seconds
Discordant knee pain status
Bilateral pain-free knee pain status
Mean pairwise difference
95% CI
P†
151 ⫾ 81 12.0 ⫾ 3.5 16.3 ⫾ 3.3 323 ⫾ 59
149 ⫾ 74 10.6 ⫾ 2.8 15.6 ⫾ 2.6 308 ⫾ 47
2.2 ⫾ 99 1.4 ⫾ 4.2 0.70 ⫾ 3.8 14.6 ⫾ 65.6
⫺8.8, 13.2 0.88, 1.9 0.26, 1.13 6.3, 22.8
0.693 ⬍ 0.001 0.002 0.001
* Values are the mean ⫾ SD unless indicated otherwise. 95% CI ⫽ 95% confidence interval; PASE ⫽ Physical Activity Score for the Elderly (22). † Using paired t-test. ‡ Repeated 5 times.
DISCUSSION The main findings of this study were that of the patientreported WOMAC physical function score; although knee specific, it is impacted by the contralateral knee pain status. That means functional limitations in a painful knee are perceived as worse when the contralateral knee is painful, and functional limitations in a pain-free knee are perceived as worse when the contralateral knee is painful. Nevertheless, because the WOMAC is knee specific, it differentiated best between painful and painless knee status when the contralateral knee pain status was similar. The self-assessed levels of physical activity (PASE) did not show the ability to differentiate between subjects with discordant knee pain status and matched controls, whereas objective functional performance tests, particularly the chair stand test, exhibited more sensitivity in differentiating between the pain status of one knee, even if the pain status in the other (contralateral knee) was similar. A strength of the current analysis is the strict matching procedure of the case– control design, taking into account major confounders in knee OA, i.e., age, sex, BMI, radiographic knee OA, and leg dominance status. Previous studies showed that the WOMAC physical function score is highly correlated to pain in the same knee (23,27) and is unable to differentiate between pain and function in construct validation studies (28,29). Further, it has been shown that the WOMAC physical function score shows a weak to moderate correlation with functional performance tests (28,29). The correlation of the WOMAC physical function score with knee pain intensity is generally higher than the correlation of functional performance tests with knee pain intensity (30). Although our results suggest that, likely because of its specificity to a single knee, the WOMAC physical function score is the most sensitive to a unilateral difference in pain, the evidence in literature indicates that the patient-reported WOMAC physical function score does not provide information that is independent (and hence complementary) from/to pain itself. We have no explanation why the subjective global PASE score displayed less sensitivity in differentiating unilateral knee pain, but our results discourage its use in a clinical study that aims to document the effect of unilateral symptomatic treatment on knee function. However, objective
physical performance tests were capable of detecting unilateral differences in knee pain, with the repeated chair stand test revealing the highest sensitivity compared to short- or long-distance walking tests. Our findings differ from those of Marmon et al (31) who reported that persons with knee OA perform and perceive their functional ability similarly, regardless of the number of knees involved. Our results in discordant cases versus controls with bilaterally painful knees are in line with a recent study by Riddle and Stratford (30) who found a stronger contrast between OAI subjects with severe unilateral compared to bilateral pain for the repeated chair stand than for walking tests. However, Riddle and Stratford (30) included only subjects with K/L grade 3 or 4 in at least 1 of both knees, i.e., a relatively advanced stage of radiographic knee OA at which subjects are not commonly recruited for clinical trials. Further, our study used a different (matched pair) design and also compared subjects with unilateral pain (i.e., discordant pain status) with those who did not have pain in either knee, in order to estimate the effect of symptomatic treatment on function in patients with unilateral pain at baseline. A limitation of the current study is that it relies on cross-sectional analysis, comparing subjects with different combinations of bilateral knee pain. These results must therefore be confirmed in longitudinal treatment trials in which effective symptomatic relief is achieved in one knee, whereas pain in the other knee remains unaltered. Further, because we compared knees with non-acceptable symptom status (NPRS ⱖ4) versus pain-free knees (NPRS 0), and because it is unlikely that treatment can transform a knee from non-acceptable to totally pain free, the current results can only provide an estimate of the “upper margin” of the effects on leg function that can be achieved with unilateral symptomatic treatment. Nevertheless, the relative performance of different measures (subjective, objective, knee-specific, and global) may hold and therefore may provide useful indicators of which measures are preferably used in a clinical trial. Ideally we would have liked to apply more strict criteria (e.g., only subjects with one knee frequently painful and the contralateral knee with no pain at all). However, this would considerably reduce the number of discordant cases and the statistical power of the study. In conclusion, we showed that the patient-reported
380
Cotofana et al
knee-specific WOMAC physical function score is impacted by the contralateral knee pain status. The WOMAC physical function score appears suitable to capture unilateral knee pain; however, given its subjective nature, its weak to moderate correlation with objectively measured functional performance, and its moderate to high correlation with subjective measures of pain, its value to demonstrate benefit in joint function under (unilateral) symptomatic treatment seems to be limited. Objective functional performance tests were suitable to capture differences in unilateral knee pain. The repeated chair stand test was the most appropriate and, based on the results of this study, it would be one of the preferred assessments to be used, for instance, in clinical trials evaluating the effect of unilateral symptomatic treatment on knee function.
ACKNOWLEDGMENTS The authors would like thank the OAI participants, investigators, technicians, and the coordinating center for providing public access to the OAI data. AUTHOR CONTRIBUTIONS All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be submitted for publication. Dr. Cotofana had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study conception and design. Cotofana, Rossi, Eckstein, Gu¨nther. Analysis and interpretation of data. Cotofana, Wirth, Rossi, Eckstein, Gu¨nther.
ROLE OF THE STUDY SPONSOR The Osteoarthritis Initiative study sponsors (Merck Research Laboratories, Novartis Pharmaceuticals Corporation, GlaxoSmithKline, and Pfizer) had no role in the study design, data analysis, data interpretation, writing of this manuscript, or the decision to submit the manuscript for publication. Publication of this article was not contingent upon approval by these sponsors. Merck KGaA was involved in the study design, data collection, data analysis and interpretation, the writing of the manuscript, and in the decision to submit the manuscript for publication. Publication of this article was contingent upon approval by Merck KGaA.
ADDITIONAL DISCLOSURE Author Cotofana is a part-time employee with Chondrometrics GmbH. Author Wirth is a part-time employee and co-owner of Chondrometrics GmbH. Author Pena Rossi was an employee at EMD Serono at the time of the study. Author Eckstein is Chief Executive Officer and co-owner of Chondrometrics GmbH. Author Gu¨nther is an employee of Merck KGaA.
REFERENCES 1. Guccione AA, Felson DT, Anderson JJ, Anthony JM, Zhang Y, Wilson PW, et al. The effects of specific medical conditions on the functional limitations of elders in the Framingham Study. Am J Public Health 1994;84:351– 8. 2. Perrot S, Poiraudeau S, Kabir-Ahmadi M, Rannou F. Correlates of pain intensity in men and women with hip and knee osteoarthritis: results of a national survey. The French ARTHRIX study. Clin J Pain 2009;25:767–72.
3. Stratford PW, Kennedy D, Pagura SM, Gollish JD. The relationship between self-report and performance-related measures: questioning the content validity of timed tests. Arthritis Rheum 2003;49:535– 40. 4. McAlindon TE, Cooper C, Kirwan JR, Dieppe PA. Determinants of disability in osteoarthritis of the knee. Ann Rheum Dis 1993;52:258 – 62. 5. Jordan J, Luta G, Renner J, Dragomir A, Hochberg M, Fryer J. Knee pain and knee osteoarthritis severity in self-reported task specific disability: the Johnston County Osteoarthritis Project. J Rheumatol 1997;24:1344 –9. 6. Jinks C, Jordan K, Croft P. Osteoarthritis as a public health problem: the impact of developing knee pain on physical function in adults living in the community: (KNEST 3). Rheumatology (Oxford) 2007;46:877– 81. 7. Zhang W, Moskowitz RW, Nuki G, Abramson S, Altman RD, Arden N, 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– 62. 8. Bannuru RR, Vaysbrot EE, Sullivan MC, McAlindon TE. Relative efficacy of hyaluronic acid in comparison with NSAIDs for knee osteoarthritis: a systematic review and meta-analysis. Semin Arthritis Rheum 2014;43:593–9. 9. Bellamy N, Buchanan WW. A preliminary evaluation of the dimensionality and clinical importance of pain and disability in osteoarthritis of the hip and knee. Clin Rheumatol 1986;5: 231– 41. 10. Eckstein F, Kwoh CK. Imaging in rheumatology in 2013: from images to data to theory [review]. Nat Rev Rheumatol 2014; 10:69 –70. 11. Lester G. The Osteoarthritis Initiative: an NIH public-private partnership. HSS J 2012;8:62–3. 12. Altman RD, Hochberg M, Murphy WA Jr, Wolfe F, Lequesne M. Atlas of individual radiographic features in osteoarthritis. Osteoarthritis Cartilage 1995;3 Suppl A:3–70. 13. Pautex S, Herrmann F, Le Lous P, Fabjan M, Michel JP, Gold G. Feasibility and reliability of four pain self-assessment scales and correlation with an observational rating scale in hospitalized elderly demented patients. J Gerontol A Biol Sci Med Sci 2005;60:524 –9. 14. Jensen MP. The validity and reliability of pain measures in adults with cancer. J Pain 2003;4:2–21. 15. Tubach F, Dougados M, Falissard B, Baron G, Logeart I, Ravaud P. Feeling good rather than feeling better matters more to patients. Arthritis Rheum 2006;55:526 –30. 16. Tubach F, Ravaud P, Baron G, Falissard B, Logeart I, Bellamy N, et al. Evaluation of clinically relevant states in patient reported outcomes in knee and hip osteoarthritis: the patient acceptable symptom state. Ann Rheum Dis 2005;64:34 –7. 17. Moisio K, Eckstein F, Chmiel JS, Guermazi A, Prasad P, Almagor O, et al. Denuded subchondral bone and knee pain in persons with knee osteoarthritis. Arthritis Rheum 2009;60: 3703–10. 18. Ornetti P, Dougados M, Paternotte S, Logeart I, Gossec L. Validation of a numerical rating scale to assess functional impairment in hip and knee osteoarthritis: comparison with the WOMAC function scale. Ann Rheum Dis 2011;70:740 – 6. 19. Kellgren JH, Lawrence JS. Radiological assessment of rheumatoid arthritis. Ann Rheum Dis 1957;16:485–93. 20. Bellamy N, Buchanan WW, Goldsmith CH, Campbell J, Stitt LW. 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– 40. 21. Bellamy N, Kirwan J, Boers M, Brooks P, Strand V, Tugwell P, et al. Recommendations for a core set of outcome measures for future phase III clinical trials in knee, hip, and hand osteoarthritis: consensus development at OMERACT III. J Rheumatol 1997;24:799 – 802. 22. Washburn RA, Smith KW, Jette AM, Janney CA. The Physical Activity Scale for the Elderly (PASE): development and evaluation. J Clin Epidemiol 1993;46:153– 62.
Knee-Specific Function and Contralateral Knee Effect 23. Stratford PW, Kennedy DM, Woodhouse LJ. Performance measures provide assessments of pain and function in people with advanced osteoarthritis of the hip or knee. Phys Ther 2006;86:1489 –96. 24. Curb JD, Ceria-Ulep CD, Rodriguez BL, Grove J, Guralnik J, Willcox BJ, et al. Performance-based measures of physical function for high-function populations. J Am Geriatr Soc 2006;54:737– 42. 25. Gill S, McBurney H. Reliability of performance-based measures in people awaiting joint replacement surgery of the hip or knee. Physiother Res Int 2008;13:141–52. 26. Cohen J. A power primer. Psychol Bull 1992;112:155–9. 27. Parent E, Moffet H. Comparative responsiveness of locomotor tests and questionnaires used to follow early recovery after total knee arthroplasty. Arch Phys Med Rehabil 2002;83:70 – 80.
381 28. Thumboo J, Chew LH, Soh CH. Validation of the Western Ontario and McMaster Universities Osteoarthritis Index in Asians with osteoarthritis in Singapore. Osteoarthritis Cartilage 2001;9:440 – 6. 29. Faucher M, Poiraudeau S, Lefevre-Colau MM, Rannou F, Fermanian J, Revel M. Algo-functional assessment of knee osteoarthritis: comparison of the test-retest reliability and construct validity of the WOMAC and Lequesne indexes. Osteoarthritis Cartilage 2002;10:602–10. 30. Riddle DL, Stratford PW. Unilateral vs bilateral symptomatic knee osteoarthritis: associations between pain intensity and function. Rheumatology (Oxford) 2013;52:2229 –37. 31. Marmon AR, Zeni JA Jr, Snyder-Mackler L. Perception and presentation of function in patients with unilateral versus bilateral knee osteoarthritis. Arthritis Care Res (Hoboken) 2013;65:406 –13.
