Association of Severity of Coexisting Patellofemoral Disease With Increased Impairments and Functional Limitations in Patients With Knee Osteoarthritis

Authors


Abstract

Objective

To evaluate the association between severity of coexisting patellofemoral (PF) disease with lower extremity impairments and functional limitations in patients with tibiofemoral (TF) osteoarthritis (OA).

Methods

Radiographic views of the TF and PF compartments, knee extension strength, and knee range of motion were obtained for 167 patients with knee OA. Additionally, knee-specific symptoms and functional limitations were assessed using the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) and the Activities of Daily Living Scale (ADLS).

Results

Moderate/severe PFOA was associated with lower knee extension strength (mean ± SD 1.4 ± 0.5 Nm/body weight [BW]) compared to no PFOA (mean ± SD 1.8 ± 0.5 Nm/BW). Additionally, total knee range of motion was significantly lower for patients with moderate/severe PFOA (mean ± SD 120.8° ± 14.4°) compared to no PFOA (mean ± SD 133.5° ± 10.7°) and mild PFOA (mean ± SD 125.8° ± 13.0°). Moderate/severe PFOA and mild PFOA were also associated with less pain while standing (odds ratio [OR] 0.2, 95% confidence interval [95% CI] 0.1–0.7 and OR 0.2, 95% CI 0.1–0.6, respectively) on the WOMAC, and moderate/severe PFOA was associated with greater difficulty with going downstairs (OR 2.9, 95% CI 1.0–8.1) on the ADLS.

Conclusion

It appears that knees with more severe coexisting PF disease demonstrate features distinct from those observed in TFOA in isolation or in combination with mild PF disease. Treatment strategies targeting the PF joint may be warranted to mitigate the specific lower extremity impairments and functional problems present in this patient population.

INTRODUCTION

Knee osteoarthritis (OA) is a prevalent medical condition in older adults and one of the leading causes of disability in the US (1, 2). For the most part, epidemiologic and clinical studies of knee OA have traditionally focused on the disease status of the tibiofemoral (TF) compartments, and the patellofemoral (PF) joint has seldom been considered. However, findings from several population-based studies of older adults with a history of knee pain suggest that PFOA in isolation or combined with TFOA is commonly observed in this patient population (3, 4). Thus, inclusion of PF joint assessment along with the TF joint has been deemed essential in studies of symptomatic knee OA (4–6). However, the potential clinical impact of coexisting PFOA along with TFOA in individuals with symptomatic disease has not been fully clarified.

An important theme arising from image-based studies of knee OA suggests that structural abnormalities related to PFOA are significantly associated with knee pain (7–11). In a study of 819 older adults with knee pain, greater radiographic disease severity of the PF compartment was independently associated with higher pain and greater functional deficits (5). Additionally, an increase in severity of isolated PFOA has been associated with greater levels of pain, stiffness, and functional limitations (6). However, this conclusion was reached based on the comparison of knees with isolated PF disease and those without radiographic signs of TFOA. Identifying the impact of coexisting PF disease severity along with TFOA may be more important as knees with structural changes in both the TF and PF compartments are commonly observed, are more likely to be painful, and are associated with greater loss of function (3, 4, 9). To this end, Englund and Lohmander (9) reported more symptoms, lower functional status, and worse knee-related quality of life for patients with coexisting PFOA and TFOA many years after meniscal resection. However, Englund and Lohmander did not examine the degree to which PFOA severity may impact these findings. To date, the potential influence of coexisting PFOA severity along with TFOA has not been fully elucidated.

As the odds of symptomatic disease and disability seem to be influenced by the compartmental pattern and severity of knee OA (12–16), clinical recognition of specific features of coexisting PF disease and its severity may have great implications for diagnosis and treatment of patients with knee OA. Therefore, the objective of this study was to evaluate the association between severity of coexisting PFOA with lower extremity impairments and limitations with activities of daily living in patients with TFOA.

Significance & Innovations

  • The potential clinical impact of severity of coexisting patellofemoral (PF) joint osteoarthritis (OA) along with tibiofemoral OA in individuals with symptomatic knee OA has not been fully clarified.

  • Individuals with more severe coexisting PF joint disease demonstrate features distinct from those observed in tibiofemoral OA in isolation or in combination with mild PFOA.

  • More severe PFOA seems to be associated with lower extremity impairments of reduced knee extension strength and limitations of knee range of motion, as well as increased difficulty with going downstairs.

  • Targeted interventions designed to treat the PF joint should be considered to mitigate functional deficits and impairments in individuals with coexisting PFOA.

PATIENTS AND METHODS

A secondary analysis of baseline data for individuals participating in a randomized clinical trial of exercise therapy for knee OA was performed (17). Individuals were included in the study if they met the 1986 American College of Rheumatology clinical criteria for knee OA (18) and had Kellgren/Lawrence (K/L) (19) grade ≥2 radiographic changes in the TF joint of at least 1 lower extremity. Patients were further asked to identify their most painful knee as the right or the left knee. From a pool of 183 subjects included in the parent clinical trial, 167 subjects with full radiographic and clinical data for the most painful knee were eligible for this secondary analysis. Reasons for ineligibility included absence of TFOA in the most painful knee (n = 5), incomplete radiographic data for the PF joint (n = 3), or missing data for lower extremity impairments, knee symptoms, or functional limitations (n = 8). Ethical approval was obtained for all phases of the study from the University of Pittsburgh Institutional Review Board.

Radiographic views and scoring.

Radiographs of the painful knee in those with unilateral symptoms and the most painful knee in patients with bilateral involvement were analyzed. Three radiographic views of the index knee were obtained for each subject as follows: 1) a semiflexed, weight-bearing, anteroposterior view, 2) a lateral view, and 3) a skyline view. A single trained rheumatologist (CVO) scored all films. The K/L scoring system was used to assign radiographic OA grades for each knee compartment based on the following criteria: grade 0 = no osteophytes; grade 1 = doubtful osteophytes (<1 mm); grade 2 = minimal osteophytes, possibly with joint space narrowing, cysts, and sclerosis; grade 3 = moderate or definite osteophytes and or moderate joint space narrowing; and grade 4 = large osteophytes and/or severe joint space narrowing. TFOA was defined by a K/L score of grade ≥2 on the anteroposterior view. Subjects with TFOA were then stratified into 3 groups of no PFOA (K/L grade ≤1), mild PFOA (K/L grade 2), or moderate/severe PFOA (K/L grade ≥3) based on the severity of their PF disease assessed on the skyline and/or the lateral radiographic views. To determine whether reliable data could be obtained with respect to radiographic scoring of knee OA, K/L scoring was repeated for a subset of 15 knees on 2 different days at least 7 days apart. Unweighted kappa coefficients and exact percentage agreement were calculated. Intrarater reliability scores were excellent for the medial and lateral TF compartments (κ = 0.86, agreement 93.3%) and the PF compartment (κ = 0.80, agreement 86.7%).

Lower extremity impairments.

The maximum voluntary isometric torque output for knee extension was measured using a Biodex System 3 dynamometer. All tests were performed with the subject seated and the tested knee flexed to 60°. A minimum of 3 trials and a maximum of 6 trials were performed. After 3 trials, when a trial had a maximum torque output less than the previous trial, the strength testing was concluded. The highest maximum torque output from all trials was recorded as the knee extension strength score. To allow for between-subject comparisons, all maximum torque measurements were normalized to the subject's body weight (BW). The analysis of 40 subjects' maximum voluntary isometric torque output on 2 different days indicated that this procedure yields reliable measurements. An intraclass correlation coefficient (ICC2,3) of 0.97, with a 95% confidence interval (95% CI) of 0.94–0.98, was obtained for test–retest reliability of our testing procedures. In addition, knee flexion and extension range of motion were measured in degrees using standard goniometric procedures. Excellent ICCs of between 0.84 and 0.93 have been reported for goniometric measurements of knee joint range of motion (20, 21).

Self-reported symptoms and functional status.

The 24-item Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) and the 14-item Activities of Daily Living Scale (ADLS) of the Knee Outcome Survey were used to gather knee-specific information on symptoms and limitations during performance of functional tasks. The WOMAC is a valid and reliable disease-specific measure of pain, stiffness, and physical function for individuals with knee OA (22–25). Each item on the WOMAC has 5 response options (none, mild, moderate, severe, and extreme) and yields total subscale scores for pain (5 items, range 0–20), stiffness (2 items, range 0–8), function (17 items, range 0–68), and a total composite score (24 items, range 0–96). Higher scores on the WOMAC indicate increased severity of symptoms or functional limitations. The ADLS consists of 14 items that measure the full spectrum of symptoms and functional limitations during activities of daily living as a result of a variety of knee pathologies (26). Each item on the ADLS has 6 response options (none, minimal, slight, moderate, severe, and extreme). The ADLS score is transformed to a 0–100-point scale with 100 indicating the absence of symptoms and functional limitations. The ADLS has proven to be a reliable, valid, and responsive instrument for the assessment of functional limitations that result from disorders and impairments of the knee (26, 27).

Statistical analysis.

Descriptive statistics were calculated and the normality of the distributions of all continuous variables was checked. Analysis of variance (ANOVA) and chi-square tests were used to determine group differences in demographics, radiographic knee OA severity, knee-specific impairments, and WOMAC and ADLS scores. Where appropriate, ANOVA tests were adjusted for sex, age (<60 years, 60–70 years, and >70 years), body mass index (BMI; <25, 25–30, and >30 kg/m2), and TFOA disease severity (K/L grade 2, 3, or 4). The association between severity of PFOA and individual items on the WOMAC and ADLS were evaluated using multivariable adjusted logistic regression models. Individual items on the WOMAC, categorized as normal/mild or moderate/severe/extreme for pain, stiffness, or physical function limitations, were the outcome variables for the logistic regression models. The individual items on the ADLS were dichotomized as none/minimal/slight or moderate/severe/extreme for symptoms and difficulty with activities of daily living and were analyzed in a similar manner. The associations between severity of PF disease and the individual items on the WOMAC and the ADLS were expressed as adjusted odds ratios (ORs) accounting for sex, age (<60 years, 60–70 years, and >70 years), BMI (<25, 25–30, and >30 kg/m2), and TFOA disease severity (K/L grade 2, 3, or 4) as covariates in the model. In addition, post hoc analyses using adjusted logistic regression models were performed to evaluate the associations between knee-specific impairments and difficulty with specific items on the WOMAC and the ADLS that were deemed to be problematic due to increases in PFOA severity. All statistical comparisons were 2-tailed, and differences were considered significant at a P value of less than 0.05 using Stata statistical program (version 11.2).

RESULTS

Patients with moderate/severe PFOA had significantly higher mean ± SD BW (89.6 ± 17.4 kg versus 79.4 ± 13.0 kg) and higher mean ± SD BMI (31.9 ± 6.2 kg/m2 versus 28.1 ± 4.2 kg/m2) in comparison to those with no PFOA (Table 1). Patients with mild PFOA also had a higher mean ± SD BMI (30.7 ± 7.2 kg/m2 versus 28.1 ± 4.2 kg/m2) and were significantly older (mean ± SD 65.6 ± 9.1 years and 60.5 ± 8.1 years) compared to the no PFOA group. Additionally, those with moderate/severe PFOA were more likely to be female compared to patients with mild PFOA (77.5% versus 57.1%). The no PFOA group also had a higher proportion of patients with grade 2 TFOA (36.4%) as compared to patients in the mild PFOA (7.9%) and moderate/severe PFOA groups (12.6%). Conversely, the moderate/severe PFOA group had a higher proportion of grade 4 TFOA (43.7%) compared to the no PFOA group (18.2%).

Table 1. Comparisons of patient characteristics*
DemographicsNo PFOA (n = 33)Mild PFOA (n = 63)Moderate/severe PFOA (n = 71)P
  • *

    Values are the number (percentage) unless indicated otherwise. PFOA = patellofemoral osteoarthritis; BMI = body mass index.

  • Significantly different than the no PFOA group.

  • Significantly different than the mild PFOA group.

  • §

    All patients had to have at least a grade 2 tibiofemoral OA to be included in the study.

Age, mean ± SD years60.5 ± 8.165.6 ± 9.163.8 ± 8.70.03
Female21 (63.6)36 (57.1)55 (77.5)0.04
Height, mean ± SD cm168.0 ± 8.3168.0 ± 10.2168.0 ± 9.30.99
Weight, mean ± SD kg79.4 ± 13.086.7 ± 21.189.6 ± 17.40.03
BMI, mean ± SD kg/m228.1 ± 4.230.7 ± 7.231.9 ± 6.20.02
Tibiofemoral joint radiographic severity§   0.02
 Grade 212 (36.4)5 (7.9)9 (12.6) 
 Grade 315 (45.5)36 (57.1)31 (43.7) 
 Grade 46 (18.2)22 (34.9)31 (43.7) 
Patellofemoral joint radiographic severity   0.001
 Grade 07 (21.2)00 
 Grade 126 (8.8)00 
 Grade 2063 (100)0 
 Grade 30054 (76.1) 
 Grade 40017 (23.9) 

Comparisons of the composite WOMAC and ADLS scores showed no statistically significant differences between the no PFOA group and the mild PFOA or the moderate/severe PFOA groups (Table 2). Moderate/severe PFOA was associated with lower knee extension strength (mean ± SD 1.4 ± 0.5 Nm/BW) compared to the no PFOA group (mean ± SD 1.8 ± 0.5 Nm/BW). Total knee range of motion, as well as knee flexion range of motion (Table 2), were also significantly lower for patients with moderate/severe PFOA (mean ± SD total motion 120.8° ± 14.4° and mean ± SD flexion motion 126.8° ± 10.9°) compared to those with no PFOA (mean ± SD total motion 133.5° ± 10.7° and mean ± SD flexion motion 136.9° ± 7.8°) and mild PFOA (mean ± SD total motion 125.8° ± 13.0° and mean ± SD flexion motion 130.8° ± 10.9°).

Table 2. Comparisons of the adjusted WOMAC, the ADLS, and the knee-specific impairments*
 No PFOA (n = 33)Mild PFOA (n = 63)Moderate/severe PFOA (n = 71)P
  • *

    Values are the mean ± SD. WOMAC = Western Ontario and McMaster Universities Osteoarthritis Index; ADLS = activities of daily living scale; PFOA = patellofemoral osteoarthritis. All analyses were adjusted for age, sex, body mass index, and radiographic tibiofemoral joint disease severity.

  • Significantly different than the no PFOA group.

WOMAC total score25.8 ± 13.828.0 ± 14.832.5 ± 17.10.64
WOMAC physical function score17.4 ± 10.620.2 ± 11.422.5 ± 12.70.60
WOMAC pain score5.5 ± 2.65.0 ± 3.06.5 ± 3.80.17
WOMAC stiffness score2.9 ± 1.62.8 ± 1.43.4 ± 1.70.48
ADLS score71.5 ± 14.767.1 ± 17.562.7 ± 18.30.50
Knee extension strength, Nm/kg1.8 ± 0.51.7 ± 0.51.4 ± 0.50.03
Flexion range of motion, degrees136.9 ± 7.8130.8 ± 10.9126.8 ± 10.90.01
Extension range of motion, degrees−3.3 ± 4.8−5.0 ± 6.0−6.0 ± 6.10.32
Total knee range of motion, degrees133.5 ± 10.7125.8 ± 13.0120.8 ± 14.40.01

Across the individual WOMAC items (Table 3), logistic regression analyses revealed that more severe PF joint disease was independently associated with lower odds of reporting pain while standing for patients with mild PFOA (OR 0.2, 95% CI 0.1–0.6), as well as those with moderate/severe PFOA (OR 0.2, 95% CI 0.1–0.7), when compared to patients with no PFOA. In addition, mild PFOA was associated with greater difficulty with taking off socks or stockings (OR 4.3, 95% CI 1.2–16.1) in comparison to no PFOA. A post hoc analysis further revealed that difficulty with taking off socks or stockings was also associated with lower knee flexion range of motion (OR 0.95 per 1-degree unit, 95% CI 0.92–0.99, P = 0.01) after adjusting for age, sex, BMI, and TFOA disease severity. However, difficulty with taking off socks or stockings was not associated with normalized quadriceps strength, total knee range of motion, or knee extension range of motion.

Table 3. Absolute frequencies, percentages, and the associations between severity of PFOA and moderate/severe/extreme levels of symptoms or difficulty performing the tasks on the WOMAC*
 No PFOA, frequency (%)Mild PFOAModerate/severe PFOA
Frequency (%)OR (95% CI)Frequency (%)OR (95% CI)
  • *

    PFOA = patellofemoral osteoarthritis; WOMAC = Western Ontario and McMaster Universities Osteoarthritis Index; OR = odds ratio; 95% CI = 95% confidence interval.

  • Reference group.

  • All analyses were performed in reference to the no PFOA group and were adjusted for age, sex, body mass index, and radiographic tibiofemoral joint disease severity.

  • §

    Significant at P < 0.05.

Pain subscale     
 Pain walking on a flat surface?8 (24.2)20 (31.7)1.5 (0.5–4.2)20 (28.2)0.9 (0.3–2.6)
 Pain going up or down stairs?16 (48.5)34 (54.0)1.1 (0.4–2.9)52 (73.2)1.7 (0.7–4.4)
 Pain at night while in bed?6 (18.2)12 (19.0)0.9 (0.3–2.9)22 (31.0)1.3 (0.4–4.0)
 Pain while sitting or lying?6 (18.2)7 (11.1)0.8 (0.2–3.0)15 (21.1)1.0 (0.3–3.2)
 Pain while standing?16 (48.5)13 (20.6)0.2 (0.1–0.6)§27 (38.0)0.2 (0.1–0.7)§
Stiffness subscale     
 Stiffness after waking up in the morning?17 (51.5)27 (42.9)0.8 (0.3–1.9)41 (57.7)0.7 (0.3–2.0)
 Stiffness after sitting/lying or resting during the day?13 (39.4)27 (42.9)1.3 (0.5–3.7)39 (54.9)1.3 (0.5–3.5)
Function subscale     
 Difficulty going down stairs?15 (45.5)32 (50.8)0.9 (0.4–2.5)49 (69.0)1.9 (0.7–4.9)
 Difficulty going up stairs?11 (33.3)32 (50.8)2.0 (0.8–5.4)41 (57.7)1.6 (0.6–4.2)
 Difficulty getting up from a sitting position?10 (30.3)29 (46.0)1.6 (0.6–4.7)40 (56.3)1.7 (0.6–4.9)
 Difficulty while standing?9 (27.3)17 (27.0)0.8 (0.3–2.2)22 (31.0)0.7 (0.2–2.0)
 Difficulty while bending to the floor?13 (39.4)36 (57.1)2.3 (0.9–6.3)39 (54.9)1.8 (0.7–4.7)
 Difficulty with walking on a flat surface?6 (18.2)15 (23.8)1.2 (0.4–3.8)20 (28.2)1.2 (0.4–3.7)
 Difficulty getting in or out of the car?12 (36.4)26 (41.3)1.1 (0.4–3.0)31 (43.7)0.6 (0.2–1.8)
 Difficulty going shopping?9 (27.3)25 (39.7)1.2 (0.4–3.5)21 (29.6)0.5 (0.2–1.5)
 Difficulty putting on socks or stockings?5 (15.2)21 (33.3)2.2 (0.7–7.2)19 (26.8)1.3 (0.4–4.5)
 Difficulty getting out of bed?13 (39.4)17 (27.0)0.6 (0.2–1.5)27 (38.0)0.7 (0.3–1.9)
 Difficulty taking off socks or stockings?4 (12.1)24 (38.1)4.3 (1.2–16.1)§20 (28.2)2.0 (0.6–7.3)
 Difficulty while lying in bed?3 (9.1)11 (17.5)2.5 (0.6–11.3)14 (19.7)2.1 (0.5–8.9)
 Difficulty getting in or out of the bathtub?6 (18.2)25 (39.7)2.4 (0.8–7.5)33 (46.5)2.4 (0.8–7.5)
 Difficulty while sitting?5 (15.2)11 (17.5)1.2 (0.3–4.2)18 (25.4)1.3 (0.4–4.6)
 Difficulty with getting on or off the toilet?10 (30.3)19 (30.2)0.7 (0.2–2.1)28 (39.4)0.9 (0.3–2.5)
 Difficulty with heavy household chores?18 (54.5)37 (58.7)0.9 (0.3–2.6)50 (70.4)1.0 (0.4–2.7)
 Difficulty with light household chores?6 (18.2)16 (25.4)1.1 (0.3–3.9)17 (23.9)0.7 (0.2–2.4)

Finally, moderate/severe PFOA was independently associated with greater limitations with going downstairs (ORadj 2.9, 95% CI 1.0–8.1) on the ADLS (Table 4). Further post hoc analyses revealed that greater difficulty with going down stairs on the ADLS was significantly associated with lower knee extension strength (OR 0.41 per 1 Nm/kg unit, 95% CI 0.16–0.99, P < 0.05) after adjusting for age, sex, BMI, and TFOA disease severity. However, knee flexion or extension range of motion or the total available knee range of motion was not associated with reports of difficulty with going down stairs on the ADLS.

Table 4. Absolute frequencies, percentages, and associations between severity of PFOA and moderate/severe/extreme levels of symptoms or difficulty performing ADLS tasks*
 No PFOA, frequency (%)Mild PFOAModerate/severe PFOA
Frequency (%)OR (95% CI)Frequency (%)OR (95% CI)
  • *

    PFOA = patellofemoral osteoarthritis; ADLS = activities of daily living scale; OR = odds ratio; 95% CI = 95% confidence interval.

  • Reference group.

  • All analyses were adjusted for age, sex, body mass index, and radiographic tibiofemoral join disease severity.

  • §

    Significant at P < 0.05.

Pain affecting daily activity?26 (78.8)47 (74.6)1.0 (0.3–3.2)51 (71.8)0.4 (0.1–1.2)
Stiffness affecting daily activity?20 (60.6)43 (68.3)1.8 (0.6–5.0)50 (70.4)0.8 (0.3–2.4)
Swelling affecting daily activity?12 (36.4)17 (27.0)0.7 (0.3–2.1)27 (38.0)0.9 (0.3–2.3)
Giving way/buckling/shifting of knee affecting daily activity?11 (33.3)23 (36.5)1.1 (0.4–3.1)27 (38.0)1.3 (0.5–3.4)
Weakness affecting daily activity?14 (42.4)30 (47.6)1.0 (0.4–2.6)35 (49.3)1.0 (0.4–2.7)
Limping affecting daily activity?14 (42.4)27 (42.9)0.9 (0.3–2.3)29 (40.8)0.8 (0.3–2.1)
Limitations with walking?10 (30.3)24 (38.1)1.1 (0.4–2.9)28 (39.4)0.9 (0.3–2.5)
Limitations with going up stairs?12 (36.4)32 (50.8)1.3 (0.5–3.6)43 (60.6)1.3 (0.4–3.7)
Limitations with going down stairs?14 (42.4)32 (50.8)1.1 (0.4–3.1)53 (74.6)2.9 (1.0–8.1)§
Limitations with standing?7 (21.2)16 (25.4)0.8 (0.3–2.7)23 (32.4)1.1 (0.4–3.3)
Limitations with kneeling on the front of the knee?23 (69.7)41 (65.1)1.2 (0.4–3.6)60 (84.5)2.4 (0.7–8.0)
Limitations with squatting?23 (69.7)45 (71.4)0.8 (0.3–2.6)58 (81.7)1.2 (0.4–4.0)
Limitations with sitting with knee bent?6 (18.2)18 (28.6)1.6 (0.5–5.1)26 (36.6)2.0 (0.6–6.1)
Limitations with rising from a chair?11 (33.3)29 (46.0)1.3 (0.5–3.9)38 (53.5)1.4 (0.5–4.2)

DISCUSSION

Coexisting PF and TF disease is a common radiographic pattern of knee OA observed in older adults with knee pain (3, 4). In addition, symptomatic disease and reduced function are more likely to be found if radiographic OA changes are present in both the TF and PF compartments (9). The findings from the current study suggest that the severity of coexisting PF disease may also be important. It appears that knees with coexisting moderate/severe PF disease demonstrate features that are distinctly different from those observed in TFOA without presence of PFOA. More specifically, moderate/severe PFOA seems to be associated with lower extremity impairments of lower knee extension strength and limitations of knee range of motion.

Overall, although a trend towards clinically important differences in WOMAC (28) and ADLS (29) scores were noted in our cohort, these group differences did not reach statistical significance. This finding is most likely due to large within-group variability and lack of statistical power of our study after adjusting for multiple confounders. However, as overlap in item content associated with the composite WOMAC and ADLS scores may have masked specific symptoms and functional limitation (30), an individual item analysis was also performed. This analysis revealed that the severity of coexisting PF disease is associated with greater odds of having difficulty with going downstairs on the ADLS scale but not on the WOMAC. The lack of concordance between the 2 instruments may have been related to variability in patient responses due to differences in phraseology used in describing similar items and/or the difference in item response structure (i.e., 5 options for WOMAC versus 6 options for the ADLS). Alternatively, given that the proportion of patients reporting difficulty with going downstairs on both instruments was very similar (Tables 3 and 4), it could also be argued that our study might have been underpowered to detect differences for this item on the WOMAC.

The specific item analysis also revealed that greater severity of PF compartment disease is associated with lower odds of having pain while standing on the WOMAC pain subscale after adjusting for age, sex, BMI, and TF disease severity. This finding was unexpected and difficult to explain. However, it could be reasoned that the fully extended knee position during standing disengages the patella from the femur to unload sensitive PF compartment structures. Biomechanical studies support this notion, as a reduction in PF joint contact pressures have been reported when the knee is fully extended compared to when the knee is in greater degrees of flexion during weight-bearing tasks (31–33).

Another unexpected finding from our specific item analysis was greater odds of having difficulty with taking off socks or stockings for patients with mild PFOA compared to those with no PFOA. Because high degrees of knee flexion range of motion are needed to perform this task, we hypothesized that limited knee joint mobility may explain the greater reports of difficulty. To explore this hypothesis, a post hoc analysis of the association between knee flexion range of motion and reports of difficulty with taking off socks or stockings on the WOMAC was performed. The results of this analysis revealed that difficulty with taking off socks or stockings on the WOMAC is significantly associated with less knee flexion range of motion. Interestingly, patients with moderate/severe PFOA did not demonstrate increased odds of having difficulty with this task. It is plausible that patients with more severe PFOA may have developed compensatory strategies to perform this task without requiring high degrees of knee flexion. For example, to avoid pain and difficulties, patients with more severe PFOA and limited knee flexion range of motion could potentially take off their socks or stockings by flexing their hips and/or trunk instead of flexing their knees.

Physical disability and functional impairments have frequently been reported for patients with knee OA (34, 35). As such, a number of factors including impairment of the lower extremity have been proposed as possible explanations for such functional limitations (36–38). For instance, lower extremity muscle weakness, particularly affecting the quadriceps, has been suggested as a risk factor, which is associated with both pain severity and disability in patients with knee OA (39–42). In addition, quadriceps weakness may be related to disease progression over the lateral PF compartment and worsening of symptoms and functional limitations over time (43). Limited knee flexion range of motion has also been associated with increased disability in this patient population (38). The findings from the current study suggest that compared to patients with TFOA who have no PFOA, those with coexisting moderate/severe PF disease demonstrate significantly lower knee extension strength and diminished total knee range of motion. Given that our analyses were adjusted for severity of radiographic TFOA, the group differences in knee extension strength and range of motion could be attributed to the severity of PFOA.

The limitations in function reported in individuals with more severe PFOA in our study are consistent with those reported previously (5, 6, 9). Englund and Lomander (9) reported more symptoms, lower functional status, and worse knee-related quality of life for patients with combined PFOA and TFOA many years after undergoing meniscal resection. However, Englund and Lomander did not examine the potential impact of PF disease severity on these findings. The current study adds to previous work by demonstrating that the severity of coexisting PF disease in individuals with TFOA may be associated with additional increases in symptoms and limitations with specific activities of daily living beyond those caused by severity of TFOA. Our observations also corroborate the findings by Duncan et al (5), who reported that higher radiographic OA severity in the PF compartment is independently associated with lower functional scores.

Consistent with our findings, Duncan et al (6) also reported a strong association for functional difficulty with going downstairs for patients with isolated PFOA compared to those with normal knee radiographs. However, these investigators also reported greater difficulty with getting in/out of the bath and getting in/out of the car for patients with more severe isolated PFOA, which was not evident in our cohort. This discrepancy may be the result of using different reference comparison groups. Whereas we used a comparison cohort with TFOA, Duncan and colleagues used a control group with no radiographic disease (6). As such, a good proportion of our patients in the no PFOA group had difficulty with getting in/out of a car (36.4%), getting in/out of the bath (18.2%), and going downstairs (45.5%) due to their TFOA. Therefore, using a reference comparison group with greater impairments and functional deficits may have prevented us from showing a relationship between severity of PFOA with less demanding bilateral activities such as getting in/out of a bath or a car, which require only 50% weight-bearing by the symptomatic extremity. On the other hand, the observed association in our study between more severe PF disease and difficulty with going downstairs may be due to the higher weight-bearing requirements of this single extremity activity, which requires 100% weight-bearing by the symptomatic extremity.

From a biomechanical perspective, stair climbing places extremely high demands on the quadriceps and requires greater range of knee motion when compared to level walking. For instance, it has been reported that the demands placed on the quadriceps while climbing stairs could be 3 times higher than those encountered during walking on a flat surface (44). In addition, peak weight-bearing knee flexion angles of 70–92° are required for climbing stairs, compared to only 15–20° needed during walking on a flat surface (33, 45). Although the available knee range of motion reported for our cohort is adequate for meeting the demands of stair climbing, the reduced knee extension strength associated with more severe PF disease may make climbing stairs more difficult. To explore this hypothesis, a post hoc analysis of the association between knee extension strength and reports of difficulty with going downstairs was performed. The results of this analysis revealed that, on the ADLS, difficulty with going downstairs is significantly associated with lower knee extension strength.

Greater knee flexion angles and higher quadriceps demands during weight-bearing tasks have been previously associated with greater PF joint compression forces (31) and pain (46) in individuals with PF dysfunction. Therefore, it could alternatively be argued that the individuals with PF disease in our study may have reduced the degree to which they flexed their knees during stair climbing in an attempt to lessen the compressive forces and pain. Such compensatory reduction in knee flexion during stair descent will most likely contribute to the difficulties reported with going downstairs in this patient population. However, this assertion cannot be verified in the current study given the lack of quantitative information about how the stair descent task was actually performed by the patients. Future biomechanical analyses are needed to quantify the degree of knee flexion during stair descent in patients with coexisting TF and more severe PFOA.

Although progression of disease severity could lead to loss of functional independence and poor quality of life, the cross-sectional nature of this study limits our ability to predict whether progression of PF joint disease is associated with change in knee-specific impairments and functional limitations. A longitudinal study would be needed to tease out the potential influence of PF joint disease progression on lower extremity impairments and functional limitations. We also had a small number of subjects in our no PFOA group (n = 33), which may have limited our findings. Another limitation of our study is that information about symptoms and functional limitation was taken from self-report measures that represent the patient's perception of their ability to perform a given task. Future work should consider performance-based measures of function to better quantify potential group differences in functional deficits.

In conclusion, our results indicate that presence of more severe coexisting PFOA is associated with significantly lower knee extension strength and range of motion, as well as difficulty with going down stairs. Therefore, it appears that knees with more severe PF disease demonstrate features distinct from those observed in TFOA in isolation or in combination with mild PF disease. As structural abnormalities and symptoms related to the PFOA are frequently observed in individuals with symptomatic knee OA (3, 4), epidemiologic and clinical studies of knee OA should also consider evaluation of the PF compartment. In addition, targeted interventions designed to treat the PF joint should be considered to mitigate functional deficits present in this patient population.

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. Farrokhi 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. Piva, Gil, Oddis, Fitzgerald.

Acquisition of data. Piva, Gil, Oddis, Fitzgerald.

Analysis and interpretation of data. Farrokhi, Piva, Gil, Brooks, Fitzgerald.

Ancillary