If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Address correspondence and reprint requests to: D. Schiphof, Erasmus MC, Department of General Practice, Room GK-1044, P.O. Box 2040, 3000 CA Rotterdam, The Netherlands. Tel: 31-10-704-38-15; Fax: 31-10-704-47-66.
Department of Orthopaedics, Erasmus MC, University Medical Center, Rotterdam, The NetherlandsDepartment of Orthopaedics, UMC Utrecht, Utrecht, The NetherlandsDepartment of Rheumatology, UMC Utrecht, Utrecht, The Netherlands
Department of General Practice, Erasmus MC, University Medical Center, Rotterdam, The NetherlandsDepartment of Orthopaedics, Erasmus MC, University Medical Center, Rotterdam, The Netherlands
The patellofemoral joint (PFJ) is important in early detection of knee osteoarthritis (OA). Little is known about the relationship between specific clinical findings and PFJ Magnetic resonance Imaging (MRI) features. The objective was to examine the relationship between (early) clinical findings and PFJ MRI features in females (45–60 years) without knee OA (PFJ or tibiofemoral joint (TFJ) OA) based on a recently suggested MRI definition.
Methods
MRIs of knees of women of a sub-study of the Rotterdam Study were scored with semi-quantitative scoring. Specific patellar tests were performed on physical examination. Current knee pain and history of patellar knee pain were reported. Binomial logistic generalized estimated equations were used to determine the association between clinical findings of OA and PFJ MRI features. All associations were adjusted for age, body mass index (BMI) and TFJ MRI features.
Results
In 888 women (1776 knees, mean age: 55.1 years and mean BMI: 27.0 kg/m2) we found significant associations between crepitus and all PFJ MRI features (Odds ratios (OR) range: 2.61–5.49). A history of patellar pain was significantly associated with almost all PFJ MRI features (ORcartilage: 1.95; ORcysts: 1.86; ORbone marrow lesions: 1.83), except for osteophytes. No significant associations were found between the clinical findings and TFJ MRI features.
Conclusion
Crepitus and history of patellar pain are clinical findings that indicate PFJ lesions seen on MRI. These tests could help to indicate signs of PFJOA. Follow-up data needs to confirm whether these tests have an additional diagnostic value for early knee OA in PFJ or TFJ.
Traditionally, research on knee OA has been primarily focused on the tibiofemoral joint (TFJ), however the awareness of the importance of the patellofemoral joint (PFJ) has increased in recent years
Association between patella alta and the prevalence and worsening of structural features of patellofemoral joint osteoarthritis: the multicenter osteoarthritis study.
. An early recognition of PFJOA might be important to reduce or intervene at the symptoms which occur with knee OA. Different risk factors were found for PFJOA in comparison with TFJOA, which confirms the aberrant role of the PFJ in OA
. The recommendations are based on the whole knee, but there might be clinical findings that are specific for PFJOA or TFJOA. Previous studies have explored the ability of clinical findings to distinguish PFJOA and TFJOA
The association of magnetic resonance imaging (MRI)-detected structural pathology of the knee with crepitus in a population-based cohort with knee pain: the MoDEKO study.
. In a cross-sectional study among OA patients, an association between knee pain and osteophytes scored on Magnetic resonance Imaging (MRI) was found only when an osteophyte was located in the patellofemoral compartment or when more than four osteophytes were present anywhere in the knee
. Several distinctive clinical findings were found for moderate to severe isolated radiographic PFJOA, but there were barely distinctive clinical findings for mild isolated radiographic PFJOA
. Still, little is known about the relationship between specific physical examination findings of the PFJ and MRI features of PFJOA. MRI allows another perspective of structural abnormalities associated with PFJOA. Bone marrow oedema, osteophytes, and cartilage lesions are MRI features that may be associated with specific clinical findings from clinical history and physical examination. These associations may occur even at an early stage and help us to identify PFJOA.
Therefore, the present study aims to examine the relationship between clinical findings of the PFJ and prevalent PFJ MRI features in a population-based middle-aged female cohort.
Patients and methods
Population
The study population is a subpopulation (RS-III-1) of the Rotterdam Study, a population-based cohort study in which the incidence and risk factors for chronic disabling diseases are investigated. All participants of the RS-III-1 cohort were ≥45 years and live in Rotterdam; the participants were included between 2006 and 2008
. The Medical Ethics committee of the Erasmus Medical Centre approved the study and all participants provided written consent. All participants were interviewed at home for demographic data, and were invited to visit the research centre for a physical examination and radiographs of the knees. Height and weight were measured at the research centre. Of these participants of the Rotterdam Study (RS-III-1) we invited 1116 women, aged 45–60 years, to join a sub-study for investigation of early signs of knee OA. 891 women that participated in this sub-study underwent MRI and physical examination of both knees and filled in a questionnaire. All women were screened for contra-indications for MRI.
Clinical data
Height, weight, body mass index (BMI) and age were determined. Current knee pain was defined if participants experienced pain in the knee at baseline (no/yes, left and/or right). Presence of history of patellar pain was defined as having had pain of the patella in history i.e., sometimes in combination with locking and grinding, especially during walking stairs, headwind cycling, squatting and sitting (no/yes, left and/or right). Two trained researchers both examined approximately half of the participants. The physical exam included assessment of pain at palpation at the medial and lateral patellar edges, the quadriceps tendon and patellar ligament, and the tibial tuberosity. Furthermore, we tested whether the participants reported pain with the patellar compression test
, and if there was crepitus in the knee during active flexion or extension of the knee. Crepitus was defined as a hearable grinding noise and/or palpable vibrations in the knee, detected by the hand of the investigator rested on the patella of the participant.
MRI acquisition
We performed a multi-sequence MRI protocol on a 1.5-T MRI scanner (Signa Excite 2, General Electric Healthcare, Milwaukee, Wisconsin). All participants were scanned with an eight-channel cardiac coil, so that two knees could be scanned at once without repositioning the subject.
The protocol consisted of a sagittal dual echo fast spin echo (FSE) proton density (TR 4900 ms; TE 11 ms) and T2 (TR 4900 ms; TE 90 ms) weighted sequence (slice thickness 3.2 mm, field of view 15 cm2), a sagittal FSE T2 weighted sequence with fat suppression (TR/TE 6800/80, slice thickness 3.2 mm, field of view 15 cm2), a sagittal spoiled gradient echo sequence with fat suppression (TR/TE 20.9/2.3, flip angle 35, slice thickness 3.2 (1.6)mm, field of view 15 cm2) and a fast imaging employing steady state acquisition (FIESTA) sequence (TR/TE 5.7/1.7, flip angle 35, slice thickness 1.6 mm, field of view 15 cm2). This FIESTA sequence was acquired in the sagittal plane, and reformatted in coronal and axial plane. Total scanning time was 27 min for two knees.
MRI interpretation
An extensively trained researcher, who was blinded for any clinical or radiographic data, scored all MRIs of the knees with the semi-quantitative comprehensive scoring system described in detail elsewhere
MRI assessment of knee osteoarthritis: Knee Osteoarthritis Scoring System (KOSS)–inter-observer and intra-observer reproducibility of a compartment-based scoring system.
. The researcher, a human movement scientist with extensive training in anatomy, is trained by a highly experienced musculoskeletal radiologist with 35 years of experience. In addition she had a training at the institute where the used semi-quantitative scoring system, the Knee Osteoarthritis Scoring System (KOSS), is developed. An experienced musculoskeletal radiologist (5 years of experience), also blinded for any clinical and radiographic data, scored a random sample of 30 knees to determine the inter-observer reliability. Cartilage lesions, osteophytes, bone marrow lesions (BML), and subchondral cysts were scored at the following nine locations: crista patellae, medial and lateral patellar facet, medial and lateral trochlear facet (anterior femur), the medial and lateral femoral condyle, and the medial and lateral tibia plateau.
Cartilage lesions were graded as diffuse or focal. The sagittal FSE proton density and the FIESTA scan in all three planes were used to assess the TFJ and PFJ. The surface extent of a lesion was classified by its maximal diameter and was graded as: grade 0, absent; grade 1, minimal (<5 mm); grade 2, moderate (5–10 mm); grade 3, severe (>10 mm). Lesion depth was graded as: grade 0, absent; grade 1, <50% reduction of the cartilage thickness; grade 2, >50% reduction of cartilage thickness; grade 3, full thickness or nearly full thickness cartilage defect.
Osteophytes were assessed in all three planes, using the following scale: grade 0, absent; grade 1, minimal (<3 mm); grade 2, moderate (3–5 mm); grade 3 severe (>5 mm). Subchondral cysts were graded as follows: grade 0, absent; grade 1, minimal (<3 mm); grade 2, moderate (3–5 mm); grade 3, severe (>5 mm). BMLs were scored on the T2 weighted sequence with fat suppression and graded as follow: grade 0, absent; grade 1, minimal (<5 mm); grade 2, moderate (5 mm–2 cm); grade 3; severe (>2 cm). A BML was defined as an ill-defined area of increased intensity.
Degenerative meniscal lesions were scored at the proton density sequence, as follows: grade 0, absent; grade 1, when a small central focus of intermediate signal intensity on the proton density sequence was noticed in the meniscus; grade 2, when the intrameniscal focus of intermediate signal intensity was surrounded by a broad, hypointense peripheral rim; grade 3, when only a thin, hypointense peripheral rim outlined the intermediate signal intensity meniscal center. A horizontal meniscal tear (0-1) was defined as a region of intermediate signal intensity communicating with a superior or inferior margin.
The inter-observer reliability was calculated with the kappa as well as the prevalence adjusted bias adjusted kappa (PABAK). The PABAK allows for the prevalence of a finding and the bias of the observers for that finding and provides therefore a more realistic estimate for agreement than the kappa, if the prevalence of the features is low. The PABAK is calculated as 2po − 1, where po is the observed proportion of agreement
. The inter-observer reliability was moderate to nearly perfect with the PABAK (Table I). For meniscal degeneration the PABAK was 0.47 (k = 0.47) and for meniscal horizontal tear the PABAK was 0.87 (k = 0.0).
. Patellofemoral OA (PFOAMRI) was defined as a definite osteophyte and partial or full thickness cartilage loss in the patella or the trochlea (anterior femur). The MRI definition for tibiofemoral OA (TFOAMRI) was defined as the presence of a definite osteophyte and full thickness cartilage loss, or one of these features and two of the following features: (1) subchondral BML or cyst not associated with meniscal or ligamentous attachments, (2) meniscal subluxation, maceration or degeneration (including a horizontal tear), (3) partial thickness cartilage loss, or (4) bone attrition.
To enable use of these definitions, the scores of the features were dichotomised, although differently than for the separate lesions. Grade 1 and 2 cartilage lesions were classified as partial thickness lesions and grade 3 cartilage lesions were classified as full thickness lesions. Grade 2 or 3 osteophytes were classified as definite osteophyte. BML and cysts were present when scored as grade 1 or higher. To score ‘present’ on the second feature: ‘meniscal subluxation, maceration or degeneration (including a horizontal tear)’ meniscal degeneration had to be assessed with a grade 1 or more or there had to be a horizontal tear. We did not score meniscal subluxation. Bone attrition was disregarded in the definition, because we did not assess bone attrition. A knee still needed two of the three remaining features to be defined as having TFOAMRI.
Knee OA was defined as having TFOAMRI and/or PFOAMRI. Not having knee OA was therefore defined as not having OA in neither PFJ nor TFJ. Knees that did not fulfil the criteria for knee OA (TFOAMRI and/or PFOAMRI) could still have MRI OA features (cartilage lesions, osteophytes, cysts, BMLs) commonly seen in OA. These knees were defined as not having OA, but were considered as having early OA.
Statistical analysis
Descriptive statistics (means, standard deviations (SD)) were applied to describe the participants' characteristics. If a physical exam outcome or feature was found in 1% or less it was not analysed.
For the association between the various PFJ MRI features and clinical findings (including current knee pain and history of patellar pain) we used a binomial logistic Generalized Estimating Equations (GEE) regression model, which takes into account the correlations between the right and left knee within a person. We adjusted all associations for age, BMI, and the presence of TFJ MRI features (cartilage lesions, osteophytes, cysts and BMLs). The odds ratio (OR) represents the odds that the clinical finding will be present in the knees with the PFJ MRI feature compared to the odds that the clinical finding will be present in the knees without the PFJ MRI feature, regardless of age, BMI and presence of TFJ MRI features. We performed the same analyses for associations between TFJ MRI features and the clinical findings (adjusted for age, BMI, and PFJ MRI-lesions) to show the difference between PFJ and TFJ. The ORs were reported with corresponding 95% confidence intervals (CI). Analyses were performed in knees without knee OA (defined as TFOAMRI and/or PFOAMRI) to find out if the clinical findings are indicative for OA at an early stage. The current population is a population at risk for OA, women aged 45–60 years. In supplementary data online analyses of all participants (with and without knee OA) were shown. There were participants with OA in one knee and no OA in the other knee. The analyses were performed on the knee-level; therefore it was possible that there was only one knee of a participant in the analyses without knee OA. Also in supplementary data were shown the associations of the independent MRI features with clinical findings in participants without knee OA additionally adjusted for the other PFJ MRI features.
Furthermore the pre- and post-test probabilities of the clinical findings that showed a significant association with an MRI feature were calculated to show the probability of having any MRI lesion when testing positive on a clinical test. Since we performed multiple testing, we consider the results with a P-value less than .001 statistically significant. All analyses were performed with PASW Statistics 20 (SPSS Inc, Chicago, USA).
Results
Of 1116 invited women, 891 women were included (1782 knees) and underwent MRI of the knees. Most important reasons for non-participation were: no time/no interest (51.1%) and fear of MRI (24.4%). Other reasons (24.5%) were sickness, moving out of the region or country, language problems and unattainable. Fig. 1 shows the flowchart of the inclusion of the study population. Of the 891 included, three women were excluded for the present study because of missing data on age and/or BMI. Characteristics of the remaining 888 women are shown in Table II, as well as the physical exam outcomes and prevalence of MRI-OA-features of their knees. Mean age was 55.1 year (SD: 3.7), mean BMI was 27.0 kg/m2 (SD: 4.8).
Fig. 1Flowchart of the inclusion of the study population.
Of 888 women 180 women (20.3% of all women) had TFOAMRI and/or PFOAMRI in one or both knees. Fifty-five women had isolated PFOAMRI (6.2% of all women), 73 women had isolated TFOAMRI (8.2% of all women) and 52 women had PFOAMRI and TFOAMRI in one or both knees (5.9% of all women). The numbers are different when analysed per knee: 242 knees (13.6%) were defined as having knee OA of the 1776 knees. Eighty-one knees (4.6%) were defined as having isolated PFOAMRI, 105 knees (5.9%) were defined as having isolated TFOAMRI, and 56 knees (3.2%) were defined as having both PFOAMRI and TFOAMRI [Fig. 1].
MRI features
Of those who did not fulfil the criteria for PFOAMRI or TFOAMRI (824 women and 1518 knees) 15% (235 knees) still had cartilage defects, 25% (384 knees) had osteophytes in the PFJ; and 10% (162 knees) had cartilage defects and 24% (347 knee) osteophytes in the TFJ (Table II). Pain at the tibial tuberosity was found in only 1% of the knees and we therefore removed this feature from further analyses. Table III, Table IV show the ORs (95%CI) of the associations between the physical exam outcomes and the MRI features of the PFJ and TFJ, respectively, in the knees without knee OAMRI.
Table IIIAssociations (OR (95%CI)) between clinical findings and PFJ MRI features in participants without any knee OA based on MRI
Clinical findings and MRI features in the PFJ (Table III)
The presence of crepitus was significantly associated with all PFJ MRI features (ORcartilage lesions = 5.49 (3.79–7.94); ORosteophytes = 2.61 (2.00–3.40); ORcysts = 2.82 (2.00–3.98); ORBML = 3.70 (2.71–5.04)).
History of patellar pain was significantly associated with cartilage lesions, cysts and BMLs of the PFJ (ORcartilage lesions = 1.95 (1.39–2.72); ORcysts = 1.86 (1.32–2.61); ORBMLs = 1.83 (1.33–2.50)). Current knee pain was only associated with BMLs (OR = 2.09 (1.38–3.15)). No significant associations were found between PFJ MRI features and pain at the patellar edge, pain at the quadriceps tendon, pain at the patellar ligament, and with a positive patellar compression test.
Clinical findings and MRI features in the TFJ (Table IV)
No significant associations were found between any clinical finding and TFJ MRI features.
There were no knees that had both cysts and pain at the patellar edge. Therefore no OR could be calculated for this association.
Pre- and post-test probabilities of crepitus, history of patellar pain and the combination on MRI-OA-features
Table V shows the pre- and post-test probabilities of having any lesion on the PFJ or the TFJ in the participants without knee OAMRI. We tested crepitus and history of patellar pain and the combination of these two (having both presence of crepitus and a history of patellar pain). In general, all post-test probabilities were higher compared to the pre-test values, although all post-test probabilities for any defect in PFJ are higher than the post-test probabilities for any defect in TFJ.
Table VPre- and post-test probability values of crepitus, history of patellar pain and the combination on any defect in PFJ or TFJ in participants without knee OA
The MRI definition for knee OA (TFOAMRI and/or PFOAMRI) is described in the Method section.
Any defect in PFJ
Any defect in TFJ
Pre-test-PV (%) (n/N)
Post-test-PV (%) (n/N)
Pre-test-PV (%) (n/N)
Post-test-PV (%) (n/N)
Presence of crepitus
43.8 (651/1486)
64.3 (388/603)
43.1 (642/1488)
51.9 (314/605)
History of patellar pain
43.8 (661/1508)
54.7 (202/369)
43.2 (653/1510)
50.4 (186/369)
Combination
43.9 (661/1506)
71.8 (135/188)
43.2 (652/1508)
63.0 (119/189)
Pre-test-PV, pre-test predictive value; Post-test-PV, post-test predictive value; n, number of knees with presence of clinical findings; N, number of total knees in analysis.
∗ The MRI definition for knee OA (TFOAMRI and/or PFOAMRI) is described in the Method section.
Our study shows that the presence of crepitus in the knee and history of patellar pain are significantly associated with all prevalent MRI features of the PFJ in women aged 45–60 years without knee OA. No significant associations were found between any clinical finding and prevalent TFJ MRI features. The post-test probabilities of crepitus and history of patellar pain for any lesion in PFJ were higher than the pre-test values and higher than the post-test probabilities for any lesion in the TFJ. Testing positive on both of these tests gives almost 72% certainty of having an osteoarthritic lesion in the PFJ.
To illustrate that these clinical findings are associated with knee OA the analyses were also performed in all knees (with and without OA). These results are shown in Supplementary data (S1) and show similar or higher associations between the clinical findings and PFJ MRI features or TFJ MRI features. In addition, the association between history of patellar pain and osteophytes in PFJ and in TFJ is significant. Furthermore, the subsequent analyses of the independent association of each MRI feature with the clinical findings show that crepitus is still associated with cartilage lesions in the PFJ when additionally adjusted for the other PFJ MRI features (data shown in Supplementary file [Table S1(d)]), but with a slightly lower estimate, probably due to the correlations between the separate features. This strengthens our conclusion that crepitus is an indicator of osteoarthritis (OA) lesions in the PFJ. Similar analysis for the history of patellar pain showed the same trend.
To our knowledge this is the first study that investigated the association between clinical findings of the PFJ alone and PFJ MRI features. Several previous studies examined the relationship between clinical findings (such as crepitus) and radiological OA in all compartments of the knee
. Duncan et al. investigated how radiographic severity and compartmental involvement influenced symptoms in the knee, showing that radiographic PFJOA is associated with symptoms
. Crema et al. studied compartment specific crepitus and showed that crepitus in the PFJ compartment associated with osteophytes in PFJ, but not with cartilage damage
The association of magnetic resonance imaging (MRI)-detected structural pathology of the knee with crepitus in a population-based cohort with knee pain: the MoDEKO study.
. This is in contrast to the results of the present study, but there are differences between the two studies that may explain the discrepancies in outcomes. The most important difference is the symptomatic study population (inclusion criteria: knee pain) of Crema et al.
The association of magnetic resonance imaging (MRI)-detected structural pathology of the knee with crepitus in a population-based cohort with knee pain: the MoDEKO study.
in contrast our study population, with only 14% knee pain. Furthermore, in the present study crepitus was tested in active flexion and extension of the knee, whereas Crema et al.
The association of magnetic resonance imaging (MRI)-detected structural pathology of the knee with crepitus in a population-based cohort with knee pain: the MoDEKO study.
The association of magnetic resonance imaging (MRI)-detected structural pathology of the knee with crepitus in a population-based cohort with knee pain: the MoDEKO study.
Fat-suppressed three-dimensional spoiled gradient-echo MR imaging of hyaline cartilage defects in the knee: comparison with standard MR imaging and arthroscopy.
Fat-suppressed three-dimensional spoiled gradient-echo MR imaging of hyaline cartilage defects in the knee: comparison with standard MR imaging and arthroscopy.
The association of magnetic resonance imaging (MRI)-detected structural pathology of the knee with crepitus in a population-based cohort with knee pain: the MoDEKO study.
used the grade≥2 as cut-off), whereas grade 1 in our grading is described as a reduction of cartilage thickness (a contour defect). Because of this difference in grading systems, we repeated the analysis with grade 2 as cut-off for cartilage lesion and osteophytes and found no differences for the association with crepitus; similarly strong associations between crepitus and cartilage lesions, and crepitus and osteophytes in the PFJ and no association for the TFJ (data shown in Supplementary data S2). For history of patellar pain the associations became stronger in the PFJ as well as in the TFJ. Furthermore, we also analysed grade 1 vs no cartilage lesion (or osteophyte) and even in this analysis we found significant associations between crepitus and cartilage lesions, and crepitus and osteophytes. This all strengthens our conclusion. Crema et al.
The association of magnetic resonance imaging (MRI)-detected structural pathology of the knee with crepitus in a population-based cohort with knee pain: the MoDEKO study.
found an association between medial meniscal damage and knee crepitus; we did an additional analysis adjusting our results for meniscal tears and/or meniscal degeneration. None of these features changed the results (data not shown).
Crepitus of the knee is often described as a grinding noise with a clearly palpable vibration, which could indicate cartilage damage in the PFJ. It is one of the signs for diagnosis of both TFJ and PFJOA especially useful in primary care, as described in the EULAR recommendation for diagnosis of knee OA
found a similar frequency of crepitus. In these symptomatic cohorts a significant association between crepitus and radiological OA (K&L ≥ 2) was found. Cibere and colleagues
showed, however, that crepitus was not associated with pre-radiological OA, which was based on contour defect of cartilage thickness seen on MRI in combination with K&L ≤ 2 for all knee compartments combined. In our study we analysed the PFJ and TFJ separately. This enabled us to demonstrate the association between crepitus and cartilage lesions for the PFJ alone. In knees with OA (n = 242 knees) we found that crepitus still was associated with PFJ MRI features and not with TFJ MRI features (data not shown).
To our knowledge history of patellar pain has never been investigated in relation to MRI features of OA. The association between history of patellar pain and all MRI features could indicate that the changes in the knee seen on MRI (damage of the cartilage, osteophyte, cysts, BMLs, or joint effusion) started at an earlier time-point. Information about the cause and circumstances of the patellar pain were unknown. Adjusting for knee injury in history (“did you ever had a knee injury with a swollen knee or contacted a medical doctor for a knee injury?” yes/no) did not change the results in significance of MRI features (data not shown). History of patellar pain could therefore also be an indicator of early PFOAMRI, as already suggested in the review of Thomas et al.
A limitation of the present study is the lack of reproducibility of the physical exam. In the literature the reported reproducibility of physical examination of the knee is generally low or undetermined
. Furthermore, we did not assess meniscal subluxation and bone attrition. Little is known about meniscal subluxation in an early stage of knee OA, and conflicting evidence is found for the association between meniscal subluxation and pain
Prevalence of abnormalities in knees detected by MRI in adults without knee osteoarthritis: population based observational study (Framingham Osteoarthritis Study).
. In the present study these features are probably underestimated which results in an underestimation of knee OA detected by the MRI definition.
We used the PABAK for the inter-observer reliability of the semi-quantitative scoring of the features of the MRI. The normal kappa is affected by the distribution of data across the categories. In our population the prevalence of the features is low, and distribution of the data across the categories is uneven. Most knees were scored 0 and therefore distributed in the zero category of both definitions, and there were features which had zero findings in certain categories. This is why the difference between the kappa and PABAK sometimes is big, for example for definite osteophyte in the patella (k = −0.1 en PABAK = 0.60).
In summary, the presence of crepitus is an important clinical finding which is an indicator for osteoarthritic lesions in the PFJ seen on MRI in women without knee OA, but not for the TFJ. History of patellar pain is also an indicator of early osteoarthritic lesions of PFJ seen on MRI in these women. Follow-up data need to confirm whether these tests have an additional diagnostic value on early stage knee OA.
Author contribution
DS was responsible for the study design, the data collection, monitoring of the data collection, scoring of the MRIs, cleaning and analysing of the data, interpretation of data, draft and final version of the paper. MvM was responsible for the interpretation of the data, the draft, critical revision and final approval of the paper. BMdK was responsible for the data collection, monitoring of the data collection, critical revision and final approval of the paper. EHGO was responsible for scoring of the MRIs, interpretation of the data and critical revision and final approval of the paper. AH was responsible for the study conception, obtaining of the funding, collection of the data and final approval of the paper. BWK was responsible for the critical revision and final approval of the paper. HHW was responsible for the design of the study, obtaining of the funding, interpretation of the data, the critical revision and final approval of the paper. SMAB-Z was responsible for the design of the study, obtaining of the funding, monitoring the collection of the data, the interpretation, the draft, critical revision and final approval of the paper.
Role of the funding source
This study is funded by the Netherlands Organisation for Scientific Research (NWO) (project no. 917.66.350). The Rotterdam Study is financially supported by the Netherlands Organisation of Scientific Research NWO Investments (no. 175.010.2005.011, 911-03-012), the Research Institute for Diseases in the Elderly (014-93-015; RIDE2), the Netherlands Genomics Initiative (NGI)/Netherlands Organisation for Scientific Research (NWO) (project no. 050-060-810), Netherlands Consortium of Healthy Aging, and the Erasmus Medical Center and Erasmus University Rotterdam. This study is partly funded by a program grant of the Dutch Arthritis Foundation for their centre of excellence “Osteoarthritis in primary care”.
The study sponsors had no further involvement than funding the study.
Conflict of interest
None of the authors has a conflict of interest.
Acknowledgements
The authors would like to thank all the participants and staff of the Rotterdam Study.
Association between patella alta and the prevalence and worsening of structural features of patellofemoral joint osteoarthritis: the multicenter osteoarthritis study.
The association of magnetic resonance imaging (MRI)-detected structural pathology of the knee with crepitus in a population-based cohort with knee pain: the MoDEKO study.
MRI assessment of knee osteoarthritis: Knee Osteoarthritis Scoring System (KOSS)–inter-observer and intra-observer reproducibility of a compartment-based scoring system.
Fat-suppressed three-dimensional spoiled gradient-echo MR imaging of hyaline cartilage defects in the knee: comparison with standard MR imaging and arthroscopy.
Prevalence of abnormalities in knees detected by MRI in adults without knee osteoarthritis: population based observational study (Framingham Osteoarthritis Study).
00048-X&id=gr1.jpg){kind=link}