Clinical Features of Patellofemoral Joint Osteoarthritis
Clinical Features of Patellofemoral Joint Osteoarthritis
This was a cross-sectional diagnostic study in adults aged 50 years and over reporting current or recent knee pain not attributed to inflammatory arthropathy. The reference standard was patellofemoral and/or tibiofemoral joint OA defined using plain radiography. Diagnostic indicators were previously-documented risk factors, and clinical signs and symptoms obtained from a simple, low-cost, non-instrumented assessment.
Participants were recruited from a two-stage cross-sectional postal survey of all adults ages ≥50 years registered with three general practices in North Staffordshire (irrespective of actual consulting patterns). Respondents reporting pain of any duration in or around the knee within the previous 12 months were invited to attend a research clinic at a local National Health Service Hospital Trust. The study protocol was approved by North Staffordshire Local Research Ethics Committee (project number 1430) and details have been published elsewhere. All participants provided written informed consent to undergo clinical and radiographic assessment. In addition, they were asked for consent to medical record review to assist in excluding preexisting inflammatory disease. The inclusion criteria for the current analysis were as follows: age ≥50 years, registered with one of the participating general practices at the time of study, responded to both postal questionnaires, consented to further contact, and attended the research clinic. Participants were excluded if they had not experienced knee pain within the six months prior to clinic attendance, had a pre-existing diagnosis of inflammatory arthropathy in their medical records, or had had a total knee replacement in their most affected knee.
All data were planned and gathered prospectively. Participants underwent a standardized clinical interview and physical examination conducted by one of six research therapists blinded to the findings from radiography, postal questionnaires and medical records. The assessments were abbreviated versions of those developed in an earlier stage of this research through consultation and formal consensus exercises with practising clinicians. Inter- and intra-rater reliability and quality assurance and control procedures have been reported elsewhere.
Participants filled in a brief self-complete questionnaire about their knee symptoms on the day of their clinic attendance. Copies of self-complete questionnaires and detailed protocols for clinical assessment are available on request from the corresponding author.
Plain knee radiographs were obtained on the day of clinic attendance. Three views were taken of each knee: a weight-bearing semi-flexed posteroanterior (PA) view, according to the protocol developed by Buckland-Wright et al., and lateral and skyline views, both in a supine position with the knee flexed to 45°. The tibiofemoral joint was assessed using the PA view and the posterior compartment of the lateral view. The patellofemoral joint was assessed using the skyline and lateral views.
A single reader (RD), blinded to all other information on participants, scored all films. Films were scored for individual radiographic features, including osteophytes, joint space width, sclerosis, subluxation and chondrocalcinosis. The atlas and scoring system developed by Altman et al. were used for the PA and skyline views and the atlas developed by Burnett et al. was used for the lateral view. Additionally, PA and skyline views were assigned a Kellgren and Lawrence (K&L) grade based on these authors' original written descriptions. For PA, K&L score, skyline K&L score and lateral osteophytes, intra- and inter-reader reliability were assessed in a subsample of 50 participants (100 knees) and found to be very good (κ = 0.81 to 0.98 and 0.49 to 0.76, respectively).
Defining Radiographic Patellofemoral Joint OA and Tibiofemoral Joint OA Only one knee per individual was analysed, the "index knee": the single painful knee in participants with unilateral knee pain and the most painful knee in individuals with bilateral knee pain. An individual was allocated to one of four mutually exclusive groups: (1) no radiographic OA, (2) isolated patellofemoral joint OA, (3) isolated tibiofemoral joint OA or (4) combined patellofemoral/tibiofemoral joint OA. We repeated all analyses using two cut-offs for defining the outcome of radiographic OA using the more stringent of the two to attempt to identify 'purer' phenotypes. The operational definitions are provided in Table 1 .
Prior to analysis, a total of 40 potential indicators were identified from information in the two postal questionnaires, the clinical assessment and the brief self-complete questionnaire (Table 2). Potential indicators were chosen, if they were known or suspected risk indicators, for radiographic OA, clinical signs and symptoms with a known or putative link to the occurrence of radiographic OA, or clinical manifestations of alternative diagnoses. All indicators had to be practicable for assessment within a routine primary care consultation. Because of collinearity between items in the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) we selected only the first items from the Pain, Stiffness and Function subscales (pain walking on flat surfaces, stiffness on waking, difficulty descending stairs).
To explore the comparative clinical features of isolated patellofemoral joint OA, isolated tibiofemoral joint OA and combined patellofemoral/tibiofemoral joint OA, a series of pairwise binary logistic regression models were undertaken based on complete case analyses (missing data on indicators was < 1% except for WOMAC items (< 7%)). The strength of association between each potential indicator and outcome was initially evaluated adjusting for age, gender and measured body mass index. Variables with a P-value < 0.05 for the likelihood ratio test were considered eligible for entry into a multivariable model. Where different items addressed the same underlying clinical construct (for example, patient-perceived swelling) one variable was selected to represent that construct and was entered into the multivariable model. Binary logistic regression was used to fit the multivariable model, with age, gender and body mass index forced into the model, and a backward elimination procedure (P = 0.05) used for variable reduction. The final models were refitted to participants with complete data on the retained predictor variables. Model calibration was checked using the Hosmer-Lemeshow goodness of fit statistic. Model discrimination was summarised by the area under the receiver operator characteristic (ROC) curve (AUC) and was visually displayed using simple histograms of density functions, which show the distribution and overlap in predicted probabilities generated from the logistic regression models.
Finally, to explore the ability of these clinical variables to support a confident diagnosis of patellofemoral and tibiofemoral joint OA we fitted a multinomial regression function, with the isolated patellofemoral group as the reference, using the indicators identified from the above pairwise analyses. For this purpose, categorical indicators were dichotomised. Once again, age, gender and body mass index were forced into the model, and a backward elimination procedure (P = 0.05) was used for variable selection. From the predicted probabilities, we summarised the proportion of participants correctly classified on the 'balance of probabilities' (that is, the category with the highest predicted probability) and the numbers of instances where the predicted probability exceeded the arbitrarily chosen threshold of 80% for a confident diagnosis.
Analyses were undertaken in Stata 11.0 (StataCorp, 2009, College Station, Texas, USA) and PASW 18.0 (PSS Inc., 2010, Chicago, Illinois, USA).
Materials and Methods
Study Design
This was a cross-sectional diagnostic study in adults aged 50 years and over reporting current or recent knee pain not attributed to inflammatory arthropathy. The reference standard was patellofemoral and/or tibiofemoral joint OA defined using plain radiography. Diagnostic indicators were previously-documented risk factors, and clinical signs and symptoms obtained from a simple, low-cost, non-instrumented assessment.
Study Population
Participants were recruited from a two-stage cross-sectional postal survey of all adults ages ≥50 years registered with three general practices in North Staffordshire (irrespective of actual consulting patterns). Respondents reporting pain of any duration in or around the knee within the previous 12 months were invited to attend a research clinic at a local National Health Service Hospital Trust. The study protocol was approved by North Staffordshire Local Research Ethics Committee (project number 1430) and details have been published elsewhere. All participants provided written informed consent to undergo clinical and radiographic assessment. In addition, they were asked for consent to medical record review to assist in excluding preexisting inflammatory disease. The inclusion criteria for the current analysis were as follows: age ≥50 years, registered with one of the participating general practices at the time of study, responded to both postal questionnaires, consented to further contact, and attended the research clinic. Participants were excluded if they had not experienced knee pain within the six months prior to clinic attendance, had a pre-existing diagnosis of inflammatory arthropathy in their medical records, or had had a total knee replacement in their most affected knee.
Data Collection
All data were planned and gathered prospectively. Participants underwent a standardized clinical interview and physical examination conducted by one of six research therapists blinded to the findings from radiography, postal questionnaires and medical records. The assessments were abbreviated versions of those developed in an earlier stage of this research through consultation and formal consensus exercises with practising clinicians. Inter- and intra-rater reliability and quality assurance and control procedures have been reported elsewhere.
Participants filled in a brief self-complete questionnaire about their knee symptoms on the day of their clinic attendance. Copies of self-complete questionnaires and detailed protocols for clinical assessment are available on request from the corresponding author.
Plain knee radiographs were obtained on the day of clinic attendance. Three views were taken of each knee: a weight-bearing semi-flexed posteroanterior (PA) view, according to the protocol developed by Buckland-Wright et al., and lateral and skyline views, both in a supine position with the knee flexed to 45°. The tibiofemoral joint was assessed using the PA view and the posterior compartment of the lateral view. The patellofemoral joint was assessed using the skyline and lateral views.
Scoring of Plain Radiographs
A single reader (RD), blinded to all other information on participants, scored all films. Films were scored for individual radiographic features, including osteophytes, joint space width, sclerosis, subluxation and chondrocalcinosis. The atlas and scoring system developed by Altman et al. were used for the PA and skyline views and the atlas developed by Burnett et al. was used for the lateral view. Additionally, PA and skyline views were assigned a Kellgren and Lawrence (K&L) grade based on these authors' original written descriptions. For PA, K&L score, skyline K&L score and lateral osteophytes, intra- and inter-reader reliability were assessed in a subsample of 50 participants (100 knees) and found to be very good (κ = 0.81 to 0.98 and 0.49 to 0.76, respectively).
Statistical Analysis
Defining Radiographic Patellofemoral Joint OA and Tibiofemoral Joint OA Only one knee per individual was analysed, the "index knee": the single painful knee in participants with unilateral knee pain and the most painful knee in individuals with bilateral knee pain. An individual was allocated to one of four mutually exclusive groups: (1) no radiographic OA, (2) isolated patellofemoral joint OA, (3) isolated tibiofemoral joint OA or (4) combined patellofemoral/tibiofemoral joint OA. We repeated all analyses using two cut-offs for defining the outcome of radiographic OA using the more stringent of the two to attempt to identify 'purer' phenotypes. The operational definitions are provided in Table 1 .
Potential Indicators of Patellofemoral Joint OA and Tibiofemoral Joint OA
Prior to analysis, a total of 40 potential indicators were identified from information in the two postal questionnaires, the clinical assessment and the brief self-complete questionnaire (Table 2). Potential indicators were chosen, if they were known or suspected risk indicators, for radiographic OA, clinical signs and symptoms with a known or putative link to the occurrence of radiographic OA, or clinical manifestations of alternative diagnoses. All indicators had to be practicable for assessment within a routine primary care consultation. Because of collinearity between items in the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) we selected only the first items from the Pain, Stiffness and Function subscales (pain walking on flat surfaces, stiffness on waking, difficulty descending stairs).
To explore the comparative clinical features of isolated patellofemoral joint OA, isolated tibiofemoral joint OA and combined patellofemoral/tibiofemoral joint OA, a series of pairwise binary logistic regression models were undertaken based on complete case analyses (missing data on indicators was < 1% except for WOMAC items (< 7%)). The strength of association between each potential indicator and outcome was initially evaluated adjusting for age, gender and measured body mass index. Variables with a P-value < 0.05 for the likelihood ratio test were considered eligible for entry into a multivariable model. Where different items addressed the same underlying clinical construct (for example, patient-perceived swelling) one variable was selected to represent that construct and was entered into the multivariable model. Binary logistic regression was used to fit the multivariable model, with age, gender and body mass index forced into the model, and a backward elimination procedure (P = 0.05) used for variable reduction. The final models were refitted to participants with complete data on the retained predictor variables. Model calibration was checked using the Hosmer-Lemeshow goodness of fit statistic. Model discrimination was summarised by the area under the receiver operator characteristic (ROC) curve (AUC) and was visually displayed using simple histograms of density functions, which show the distribution and overlap in predicted probabilities generated from the logistic regression models.
Finally, to explore the ability of these clinical variables to support a confident diagnosis of patellofemoral and tibiofemoral joint OA we fitted a multinomial regression function, with the isolated patellofemoral group as the reference, using the indicators identified from the above pairwise analyses. For this purpose, categorical indicators were dichotomised. Once again, age, gender and body mass index were forced into the model, and a backward elimination procedure (P = 0.05) was used for variable selection. From the predicted probabilities, we summarised the proportion of participants correctly classified on the 'balance of probabilities' (that is, the category with the highest predicted probability) and the numbers of instances where the predicted probability exceeded the arbitrarily chosen threshold of 80% for a confident diagnosis.
Analyses were undertaken in Stata 11.0 (StataCorp, 2009, College Station, Texas, USA) and PASW 18.0 (PSS Inc., 2010, Chicago, Illinois, USA).