- Email: [email protected]
High signal in knee osteophytes is not associated with knee pain
OsteoArthritis and Cartilage (2006) 14, 413e417 ª 2005 OsteoArthritis Research Society International. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.joca.2005.11.012
International Cartilage Repair Society
High signal in knee osteophytes is not associated with knee pain1 M. Sengupta M.D.yz, Y. Q. Zhang D.Sc.yz, J. B. Niu M.D.yz, A. Guermazi M.D.x, M. Grigorian M.D.x, D. Gale M.D.k, D. T. Felson M.D., M.P.Hyz and D. J. Hunter M.B.B.S., Ph.D.yz* y Clinical Epidemiology Research and Training Unit, Boston University School of Medicine, Boston, MA, USA z The Department of Medicine, Boston Medical Center, Boston, MA, USA x Synarc, San Francisco, CA, USA k VA Medical Center, Boston, MA, USA Summary Objective: Our understanding of the local source of pain in osteoarthritis (OA) remains unclear. We undertook this study to determine if the presence of high-signal osteophytes on magnetic resonance imaging (MRI) was associated with pain presence, location or severity. Methods: Subjects were chosen from the Boston Osteoarthritis of the Knee Study, a natural history study of symptomatic knee OA. Assessments included knee MRI, pain assessments and information on weight and height. Osteophyte signal was defined as areas of increased signal intensity in the osteophyte on fat-suppressed T2 weighted images, and graded in the joint margins where osteophyte size is graded. All patients were evaluated with the frequent knee symptoms question for pain presence, the Western Ontario McMasters Osteoarthritis Index (WOMAC) for pain severity, and location of self-reported pain was recorded as present or absent based on locations identified on a standardized diagram. The osteophyte signal measures anywhere within one given knee were summed, creating an osteophyte signal aggregate. Logistic regression was conducted with quartile of osteophyte signal aggregate as the independent predictor and frequent knee symptom question as the dependent outcome. Association between quartile of osteophyte signal aggregate and pain severity on WOMAC was assessed using a linear regression. Logistic regression was used to evaluate the association between compartment-specific high-signal osteophytes aggregates (independent variable) and compartment-specific knee pain (dependent variable). Analyses were adjusted for gender, body mass index (BMI), and age. Results: Two hundred and seventeen subjects were included in this analysis. They were predominantly male and 75% of subjects had radiographic tibio-femoral (TF) OA, and the remainder had patello-femoral (PF) radiographic OA. We did not find any association of high-signal osteophytes with presence of pain, pain severity or self-reported pain location. Conclusion: High-signal osteophytes detected on MRI are not associated with the presence of pain, pain severity or the self-reported location of pain. ª 2005 OsteoArthritis Research Society International. Published by Elsevier Ltd. All rights reserved. Key words: High-signal osteophytes, MRI, Knee osteoarthritis, Pain.
a potential source of pain in knee OA. Despite this knowledge, the local source of pain in OA remains elusive. Osteophytes are considered to be a defining structural feature of knee OA4. They have previously been shown to be correlated with knee pain5. In one prior study, magnetic resonance imaging (MRI) ‘hyperintense osteophytosis’ and ipsilateral ‘tramline’ scintigraphic uptake of osteophytes correlated suggesting increased fat content in ‘active’ osteophytes6. Some degree of osteophytosis was identifiable in all the patients’ scans and several showed areas of osteophytosis which appeared hyperintense with respect to underlying bone marrow6. Locations of osteophytes have also been linked to knee pain. Marginal osteophytes, located at the margin of the joint, have previously been shown to be associated with knee pain5. Osteophytes are formed from a cartilage anlage and then ossified and during this latter process, a neurovascular ingrowth occurs, so that osteophytes have nerve innervation. Also, in those with OA, joint line tenderness may represent tenderness at the most superficial and easily palpable bony prominence, the osteophyte. There are other structures at the joint margin that may be responsible for
Introduction Knee osteoarthritis (OA) is a common cause of pain and disability1. Despite this high prevalence, the local cause of pain in OA remains unclear. It is well recognized that individuals may have radiological evidence of OA without symptoms of pain, while others may have considerable knee pain with moderate or no radiological signs of OA2,3. Loss of hyaline cartilage is the characteristic pathologic finding in OA. However, cartilage does not contain any pain fibers. The periosteum and bone marrow are densely innervated with nociceptive fibers and could represent 1 Supported by the National Institutes of Health (AR47785), and by Arthritis Foundation Clinical Sciences and OA Biomarkers Grants. *Address correspondence and reprint requests to: Dr David John Hunter, M.B.B.S., Ph.D., Room A203, Clinical Epidemiology Research and Training Unit, Boston University School of Medicine, 715 Albany Street, Boston, MA 02118, USA. Tel: 1617-638-5180; Fax: 1-617-638-5239; E-mail: [email protected] Received 28 July 2005; revision accepted 28 November 2005.
413
414
M. Sengupta et al.: High-signal osteophytes and knee pain
joint line tenderness such as the capsule and synovium. Thus, both through statistical and clinical associations, marginal osteophytes may be associated with pain. Not only marginal osteophytes are associated with pain, but bone lesions with high signal on fat-suppressed T2 weighted MRIs (so-called bone marrow lesions (BMLs)) are linked to knee pain7. Given the proven relation of osteophytes to pain and the association of BMLs with pain, we hypothesized that osteophytes which show ‘edema’ changes on MRI would be especially strongly related to knee pain. Given these findings we took advantage of an MRI natural history study of persons with symptomatic OA to determine if the presence of high-signal osteophytes on MRI was associated with pain presence, location or severity.
Design and methods STUDY POPULATION
The subjects in this study were a subsample chosen from the Boston Osteoarthritis of the Knee Study, a natural history study of symptomatic knee OA. The subsample chosen here was randomly selected from the whole sample (365 persons). The recruitment methods for the whole sample are described in detail elsewhere7. Briefly, men were recruited from Veterans Administration (VA) hospital records and outpatient clinics. The women were recruited from VA hospital records and outpatient clinics, Boston Medical Center outpatient clinics and newspaper advertisements. The study included a baseline examination and follow-up examinations at 15 and 30 months. At baseline, patients who did not have contraindications to MRIs had an MRI of the more symptomatic knee. MRIs of the same knee were also performed at 15 and 30 months follow-up visits.
Patients were also weighed with shoes off on a balancebeam scale, and height was assessed. The institutional review boards of Boston University Medical Center and the Veterans Administration Boston Health Care System approved the examinations. Symptomatic knee OA was defined as knee pain, aching or stiffness on most days plus the presence of a definite osteophyte in the symptomatic knee on knee X-ray. Persons with inflammatory arthritis were excluded. Subjects were men aged 45 years and older and women aged 50 years and older. MRI MEASUREMENTS
All studies were performed with a Signa 1.5 T MRI system (General Electric Corp., Milwaukee, Wisconsin) using a phased-array knee coil. A positioning device was used to ensure uniformity among patients. Coronal, sagittal, and axial images were obtained. Fat suppressed spinecho (FSE) proton density and T2-weighted images (repetition time, 2200 ms; echo time, 20/80 ms) with a slice thickness of 3 mm, a 1-mm interslice gap, one excitation, a field of view of 11e12 cm, and a matrix of 256 128 pixels were obtained. The T2 weighted sagittal sequence on which the majority of the scoring was performed was 7 min and 20 s in duration. Osteophyte size, BMLs, synovosis and attrition were assessed using a semiquantitative, multi-feature scoring method, whole-organ magnetic resonance imaging score (WORMS), for whole-organ evaluation of the knee that is applicable to conventional MRI techniques8. Osteophytes along the anterior (a), central weight bearing (c) and posterior (p) margins of the femoral condyles and tibial plateaus, and along the medial (M) and lateral (L) margins of the patella were graded from 0 to 7 using
Fig. 1. Osteophyte signal within a medial tibial marginal osteophyte.
415
Osteoarthritis and Cartilage Vol. 14, No. 5 the following scale: 0 ¼ none; 1 ¼ equivocal; 2 ¼ small; 3 ¼ small-moderate; 4 ¼ moderate; 5 ¼ moderate-large; 6 ¼ large; and 7 ¼ very large. Osteophytes were also assessed for increased signal intensity on fat suppressed images in the axial, sagittal and coronal planes in locations analogous to where they are scored for size. MRIs of all patients were reviewed for the presence of osteophytes and their size and location. Then the slice of MRI ( one slice) which had the largest osteophyte was scored for signal intensity (see Fig. 1). A scale of 0e2 was used in terms of signal intensity (0 ¼ normal intensity, 1 ¼ possible high-intensity, and 2 ¼ probable highintensity). The osteophyte signal measures anywhere within one given knee were summed, creating an osteophyte signal aggregate. The intra-observer correlation coefficient (ICC) for osteophyte signal aggregate was 0.74. For this analysis the predictor (osteophyte signal) was read at baseline.
Patellofemoral Compartment
Medial Tibiofemoral Compartment
PAIN MEASUREMENTS
Pain was assessed in three ways.
Lateral Tibiofemoral Compartment
(1) Presence/absence: Participants were evaluated using the frequent knee symptoms question, ‘‘During the past 30 days, have you had pain, aching, or stiffness in or around your right knee on most days?’’ A similar question was used to assess frequent knee symptoms in the left knee. (2) Severity: All patients were evaluated for the Western Ontario McMasters Osteoarthritis Index (WOMAC), an outcome measure validated for the evaluation of pain in patients with knee OA9, recommended by the Osteoarthritis Research Society in clinical trials evaluating symptoms related to knee OA10. (3) Location: Location of self-reported pain in both knees was recorded dichotomously as present or absent based on locations identified on a standardized diagram (Fig. 2).
Statistical analysis We examined the relation of high-signal osteophytes to the pain status (presence or absence), severity of knee pain, and location of knee pain, respectively. The details were described below.
1. Pain status: The osteophyte signal measures anywhere within one given knee were summed, creating an osteophyte signal aggregate. The osteophyte signal aggregate was used to create quartiles of osteophyte signal aggregate. Knee pain presence was assessed using the frequent knee symptom question (on most days, do you have pain, aching or stiffness in your x joint?), as a dichotomous outcome; pain present/absent. Logistic regression was conducted with quartile of osteophyte signal aggregate as the independent predictor and frequent knee symptom question as the dependent outcome adjusted for gender, body mass index (BMI), and age. 2. Pain severity: Knee pain severity was assessed using the WOMAC pain subscale. Association between quartile of osteophyte signal aggregate and pain severity was assessed using a linear regression with analyses adjusting for covariates including attrition, osteophyte size, synovitis, BMI, and age.
Fig. 2. Diagram used for location of self-reported pain.
3. Pain regional location: The third analysis assessed if region-specific self-reported knee pain location was associated with osteophyte signal on MRI. Compartment-specific knee pain (i.e., patello-femoral (PF), medial tibio-femoral (TF), and lateral TF), was assessed based on a self-reported pain location identified on a knee diagram. The scores for all high-signal osteophytes in each compartment of the knee (i.e., PF compartment [medial/lateral anterior femur/tibia and patella], medial TF compartment [central/posterior medial femur and central/posterior medial tibia], and lateral TF compartment [central/posterior lateral femur and central/posterior lateral tibia]) were summed. Logistic regression was used to evaluate the association between compartment-specific high-signal osteophytes aggregates (independent variable) and compartment-specific knee pain (dependent variable). Analyses were adjusted for covariates including attrition, osteophyte size, synovitis, gender, BMI, and age. Table I Characteristics of study population (n ¼ 217) Characteristics Age (years): mean SD % Male % K&L grade > 2 Proportion of subjects with osteophytes (Y/N) BMI: mean SD Percentage with knee pain WOMAC pain: mean SD Osteophyte signal aggregate: mean SD
67.3 9.1 70.0 74.5 94.4 31.0 5.1 89.8 7.3 3.9 5.5 5.2
SD: Standard Deviation; K&L: Kellgren and Lawrence.
416
M. Sengupta et al.: High-signal osteophytes and knee pain Table II Association between osteophyte signal quartile and knee pain presence Osteophyte signal aggregate quartile
Osteophyte aggregate Number of subjects Percentage with pain PR (95% CI)
0
1
2
3
0e1 58 94.83 1.0
2e4 56 83.93 0.89 (0.78e1.00)
5e8 47 91.49 0.97 (0.86e1.10)
9e25 55 89.09 0.94 (0.83e1.06)
PR: Prevalence ratio; CI: Confidence Interval; Adjusted for age, sex, and BMI.
Results Two hundred and seventeen subjects were included in this analysis. They are predominantly males reflecting the VA community from which they were drawn (see Table I). Thirty-five percentage of knees had isolated radiographic TF ROA, 8% knees had isolated radiographic PF ROA, 65% knees had radiographic PF ROA (isolated or with radiographic TF OA), 75% of knees had radiographic TF OA (isolated or with radiographic PF OA), and 57% knees had both radiographic TF and radiographic PF ROAs. We did not find any association of high-signal osteophytes (subjects were divided into quartiles of osteophyte signal aggregate) with the presence of pain (Table II). Despite relatively small numbers of subjects without pain on most days there was no apparent trend observed. Similarly there was no association between high-signal osteophytes (subjects were divided into quartiles of osteophyte signal aggregate) and pain severity (Table III). It was expected that patients with increasing pain severity (measured by the WOMAC scale) would have higher numbers of high-signal osteophytes. Although there seemed to be a trend towards increasing WOMAC scores in the highest quartile of osteophyte signal, this was not statistically significant. There was no association found between high-signal osteophytes (subjects were divided into tertiles of compartment-specific osteophyte signal aggregate) and pain location (Table IV). There was no statistically significant association found between the self-reported location of pain and the presence of high-signal osteophytes. In the PF joint there was a significant association found in the reverse direction to what we anticipated. The analyses were also conducted without adjusting for BMLs with similar results.
Discussion We have examined whether the presence of high-signal osteophytes on MRI was associated with pain presence,
location or severity. Our research demonstrates that highsignal osteophytes are not associated with pain presence, location or severity. The structural determinants of pain in knee OA are not well understood, but probably involve a multitude of interactive pathways. The current practice of monitoring only a few of these features (usually radiographic joint-space narrowing and osteophytes) provides only a keyhole view of the disease process and heretofore has only been weakly related to symptoms in persons with knee OA. Because OA is a disease of all the tissues in the joint11, measurements of structure need to be seen broadly and capture a broad number of important anatomic features, such as osteophytes, effusions, meniscal tears, subchondral bone architectural changes or ligamentous instability, in addition to cartilage loss. Most of these structures cannot be seen on plain radiography, whereas they are clearly visualized on MRI. This is a critical area for further research, as it is likely that future treatments in OA will target structural abnormalities known to have pain/function consequences. Recent MRI data have been insightful in providing information on the etiology of pain in knee OA7,12e15. Whilst osteophytes are considered to be a defining structural feature of knee OA4, and have previously been shown to be correlated with knee pain on plain radiographs5, high signal within osteophytes does not appear to be associated with pain. The fact that BMLs are associated with pain and high-signal osteophytes are not may be the result of the osteophyte signal being much smaller than the BMLs associated with pain7. The comparative histology of BMLs and high signal within osteophytes is unknown and whilst they are both high-signal lesions on MRI they may be very different from a histopathologic perspective. The findings in the PF joint are interesting and difficult to explain. If it were true it would suggest that the presence of osteophyte signal is inversely associated with pain and contrary to our hypothesis when we undertook this investigation. Moreover they are borderline statistically significant. There are a number of limitations of the current investigation that warrant mentioning. This is a cross-sectional study
Table III Association between osteophyte signal quartile and knee pain severity Osteophyte signal aggregate quartiles
Osteophyte signal aggregate WOMAC pain subscale: mean (SD) B-coefficient *(P value)
0
1
2
3
0e2 7.39 (3.84) Ref.
2e5 6.67 (3.82) 0.777 (0.73)
5e10 7.02 (4.02) 0.140 (0.82)
10e25 8.16 (3.79) 0.556 (0.88)
Adjusted for sex, gender, BMI, BML, attrition, synovosis, and osteophyte size. *B-coefficient represents the difference in WOMAC pain score (0e20 scale) between the quartile and the referent. Numbers less than 0 represent lower WOMAC scores than the lowest quartile.
417
Osteoarthritis and Cartilage Vol. 14, No. 5 Table IV Association between osteophyte signal quartile and knee pain location Location of self-reported pain
Tertiles 0
1
2
PF
Osteophyte signal aggregate No. of subjects Percentage with pain PR (95% CI)
0 76 52 Ref.
1e2 71 46 0.83 (0.59, 1.18)
3e12 86 34 0.62 (0.43, 0.89)
Medial TF
Osteophyte signal aggregate No. of subjects Percentage with pain PR (95% CI)
0 89 35 Ref.
1e2 54 33 0.91 (0.57, 1.45)
3e8 79 36 0.70 (0.43, 1.13)
Lateral TF
Osteophyte signal aggregate No. of subjects Percentage with pain PR (95% CI)
0 140 25 Ref.
1e2 45 26 1.07 (0.60, 1.93)
3e6 39 28 0.93 (0.41, 2.07)
Adjusted for BMI, osteophyte size, BML, synovosis, attrition, and age. PR: Prevalence ratio.
and we have not evaluated the relation between change in signal and change in pain. There were relatively few asymptomatic subjects recruited into this study making the exploration of the association with pain presence somewhat tenuous. This study began in 1997, and the MRI pulse sequences used at that time do not have the same resolution and signal to noise ratio as pulse sequences used in current investigation that may have limited our ability to measure osteophyte signal. In summary high-signal osteophytes detected on MRI are not associated with the presence of pain, pain severity or the self-reported location of pain.
Acknowledgments We would like to thank the participants and staff of the BOKS study, without whom this research would not have been possible. WORMS MRI reading performed by OARG with assistance from John Lynch. Arthritis Foundation Clinical Sciences Grant. National Institutes of Health (AR47785).
References 1. McAlindon TE, Cooper C, Kirwan JR, Dieppe PA. Knee pain and disability in the community. Br J Rheumatol 1992;31(3):189e92. 2. Creamer P, Hunt M, Dieppe P. Pain mechanisms in osteoarthritis of the knee: effect of intraarticular anesthetic. J Rheumatol 1996;23(6):1031e6. 3. Creamer P, Lethbridge-Cejku M, Hochberg MC. Where does it hurt? Pain localization in osteoarthritis of the knee. Osteoarthritis Cartilage 1998;6(5):318e23. 4. Spector TD, Hart DJ, Byrne J, Harris PA, Dacre JE, Doyle DV. Definition of osteoarthritis of the knee for epidemiological studies. Ann Rheum Dis 1993;52(11): 790e4. 5. Cicuttini FM, Baker J, Hart DJ, Spector TD. Association of pain with radiological changes in different compartments and views of the knee joint. Osteoarthritis Cartilage 1996;4(2):143e7.
6. McAlindon TE, Watt I, McCrae F, Goddard P, Dieppe PA. Magnetic resonance imaging in osteoarthritis of the knee: correlation with radiographic and scintigraphic findings. Ann Rheum Dis 1991;50(1):14e9. 7. Felson DT, Chaisson CE, Hill CL, Totterman SM, Gale ME, Skinner KM, et al. The association of bone marrow lesions with pain in knee osteoarthritis [see comments]. Ann Intern Med 2001;134(7):541e9. 8. Peterfy CG, Guermazi A, Zaim S, Tirman PF, Miaux Y, White D, et al. Whole-Organ Magnetic Resonance Imaging Score (WORMS) of the knee in osteoarthritis. Osteoarthritis Cartilage 2004;12(3):177e90. 9. 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(12):1833e40. 10. Pham T, van der Heijde D, Altman R, Anderson J, Bellamy N, Hochberg M, et al. OMERACT-OARSI Initiative: Osteoarthritis Research Society International set of responder criteria for osteoarthritis clinical trials revisited. Osteoarthritis Cartilage 2004;12(5):389e99. 11. Martin JA, Buckwalter JA. Roles of articular cartilage aging and chondrocyte senescence in the pathogenesis of osteoarthritis [Review] [34 Refs]. Iowa Orthop J 2001;21:1e7. 12. Bhattacharyya T, Gale D, Dewire P, Totterman S, Gale ME, McLaughlin S, et al. The clinical importance of meniscal tears demonstrated by magnetic resonance imaging in osteoarthritis of the knee [comment]. J Bone Joint Surg Am 2003;85-A(1):4e9. 13. Felson DT, McLaughlin S, Goggins J, LaValley MP, Gale ME, Totterman S, et al. Bone marrow edema and its relation to progression of knee osteoarthritis. Ann Intern Med 2003;139(5 Pt 1):330e6. 14. Gale DR, Chaisson CE, Totterman SM, Schwartz RK, Gale ME, Felson D. Meniscal subluxation: association with osteoarthritis and joint space narrowing. Osteoarthritis Cartilage 1999;7(6):526e32. 15. Hill CL, Gale DG, Chaisson CE, Skinner K, Kazis L, Gale ME, et al. Knee effusions, popliteal cysts, and synovial thickening: association with knee pain in osteoarthritis. J Rheumatol 2001;28(6):1330e7.
International Cartilage Repair Society
High signal in knee osteophytes is not associated with knee pain1 M. Sengupta M.D.yz, Y. Q. Zhang D.Sc.yz, J. B. Niu M.D.yz, A. Guermazi M.D.x, M. Grigorian M.D.x, D. Gale M.D.k, D. T. Felson M.D., M.P.Hyz and D. J. Hunter M.B.B.S., Ph.D.yz* y Clinical Epidemiology Research and Training Unit, Boston University School of Medicine, Boston, MA, USA z The Department of Medicine, Boston Medical Center, Boston, MA, USA x Synarc, San Francisco, CA, USA k VA Medical Center, Boston, MA, USA Summary Objective: Our understanding of the local source of pain in osteoarthritis (OA) remains unclear. We undertook this study to determine if the presence of high-signal osteophytes on magnetic resonance imaging (MRI) was associated with pain presence, location or severity. Methods: Subjects were chosen from the Boston Osteoarthritis of the Knee Study, a natural history study of symptomatic knee OA. Assessments included knee MRI, pain assessments and information on weight and height. Osteophyte signal was defined as areas of increased signal intensity in the osteophyte on fat-suppressed T2 weighted images, and graded in the joint margins where osteophyte size is graded. All patients were evaluated with the frequent knee symptoms question for pain presence, the Western Ontario McMasters Osteoarthritis Index (WOMAC) for pain severity, and location of self-reported pain was recorded as present or absent based on locations identified on a standardized diagram. The osteophyte signal measures anywhere within one given knee were summed, creating an osteophyte signal aggregate. Logistic regression was conducted with quartile of osteophyte signal aggregate as the independent predictor and frequent knee symptom question as the dependent outcome. Association between quartile of osteophyte signal aggregate and pain severity on WOMAC was assessed using a linear regression. Logistic regression was used to evaluate the association between compartment-specific high-signal osteophytes aggregates (independent variable) and compartment-specific knee pain (dependent variable). Analyses were adjusted for gender, body mass index (BMI), and age. Results: Two hundred and seventeen subjects were included in this analysis. They were predominantly male and 75% of subjects had radiographic tibio-femoral (TF) OA, and the remainder had patello-femoral (PF) radiographic OA. We did not find any association of high-signal osteophytes with presence of pain, pain severity or self-reported pain location. Conclusion: High-signal osteophytes detected on MRI are not associated with the presence of pain, pain severity or the self-reported location of pain. ª 2005 OsteoArthritis Research Society International. Published by Elsevier Ltd. All rights reserved. Key words: High-signal osteophytes, MRI, Knee osteoarthritis, Pain.
a potential source of pain in knee OA. Despite this knowledge, the local source of pain in OA remains elusive. Osteophytes are considered to be a defining structural feature of knee OA4. They have previously been shown to be correlated with knee pain5. In one prior study, magnetic resonance imaging (MRI) ‘hyperintense osteophytosis’ and ipsilateral ‘tramline’ scintigraphic uptake of osteophytes correlated suggesting increased fat content in ‘active’ osteophytes6. Some degree of osteophytosis was identifiable in all the patients’ scans and several showed areas of osteophytosis which appeared hyperintense with respect to underlying bone marrow6. Locations of osteophytes have also been linked to knee pain. Marginal osteophytes, located at the margin of the joint, have previously been shown to be associated with knee pain5. Osteophytes are formed from a cartilage anlage and then ossified and during this latter process, a neurovascular ingrowth occurs, so that osteophytes have nerve innervation. Also, in those with OA, joint line tenderness may represent tenderness at the most superficial and easily palpable bony prominence, the osteophyte. There are other structures at the joint margin that may be responsible for
Introduction Knee osteoarthritis (OA) is a common cause of pain and disability1. Despite this high prevalence, the local cause of pain in OA remains unclear. It is well recognized that individuals may have radiological evidence of OA without symptoms of pain, while others may have considerable knee pain with moderate or no radiological signs of OA2,3. Loss of hyaline cartilage is the characteristic pathologic finding in OA. However, cartilage does not contain any pain fibers. The periosteum and bone marrow are densely innervated with nociceptive fibers and could represent 1 Supported by the National Institutes of Health (AR47785), and by Arthritis Foundation Clinical Sciences and OA Biomarkers Grants. *Address correspondence and reprint requests to: Dr David John Hunter, M.B.B.S., Ph.D., Room A203, Clinical Epidemiology Research and Training Unit, Boston University School of Medicine, 715 Albany Street, Boston, MA 02118, USA. Tel: 1617-638-5180; Fax: 1-617-638-5239; E-mail: [email protected] Received 28 July 2005; revision accepted 28 November 2005.
413
414
M. Sengupta et al.: High-signal osteophytes and knee pain
joint line tenderness such as the capsule and synovium. Thus, both through statistical and clinical associations, marginal osteophytes may be associated with pain. Not only marginal osteophytes are associated with pain, but bone lesions with high signal on fat-suppressed T2 weighted MRIs (so-called bone marrow lesions (BMLs)) are linked to knee pain7. Given the proven relation of osteophytes to pain and the association of BMLs with pain, we hypothesized that osteophytes which show ‘edema’ changes on MRI would be especially strongly related to knee pain. Given these findings we took advantage of an MRI natural history study of persons with symptomatic OA to determine if the presence of high-signal osteophytes on MRI was associated with pain presence, location or severity.
Design and methods STUDY POPULATION
The subjects in this study were a subsample chosen from the Boston Osteoarthritis of the Knee Study, a natural history study of symptomatic knee OA. The subsample chosen here was randomly selected from the whole sample (365 persons). The recruitment methods for the whole sample are described in detail elsewhere7. Briefly, men were recruited from Veterans Administration (VA) hospital records and outpatient clinics. The women were recruited from VA hospital records and outpatient clinics, Boston Medical Center outpatient clinics and newspaper advertisements. The study included a baseline examination and follow-up examinations at 15 and 30 months. At baseline, patients who did not have contraindications to MRIs had an MRI of the more symptomatic knee. MRIs of the same knee were also performed at 15 and 30 months follow-up visits.
Patients were also weighed with shoes off on a balancebeam scale, and height was assessed. The institutional review boards of Boston University Medical Center and the Veterans Administration Boston Health Care System approved the examinations. Symptomatic knee OA was defined as knee pain, aching or stiffness on most days plus the presence of a definite osteophyte in the symptomatic knee on knee X-ray. Persons with inflammatory arthritis were excluded. Subjects were men aged 45 years and older and women aged 50 years and older. MRI MEASUREMENTS
All studies were performed with a Signa 1.5 T MRI system (General Electric Corp., Milwaukee, Wisconsin) using a phased-array knee coil. A positioning device was used to ensure uniformity among patients. Coronal, sagittal, and axial images were obtained. Fat suppressed spinecho (FSE) proton density and T2-weighted images (repetition time, 2200 ms; echo time, 20/80 ms) with a slice thickness of 3 mm, a 1-mm interslice gap, one excitation, a field of view of 11e12 cm, and a matrix of 256 128 pixels were obtained. The T2 weighted sagittal sequence on which the majority of the scoring was performed was 7 min and 20 s in duration. Osteophyte size, BMLs, synovosis and attrition were assessed using a semiquantitative, multi-feature scoring method, whole-organ magnetic resonance imaging score (WORMS), for whole-organ evaluation of the knee that is applicable to conventional MRI techniques8. Osteophytes along the anterior (a), central weight bearing (c) and posterior (p) margins of the femoral condyles and tibial plateaus, and along the medial (M) and lateral (L) margins of the patella were graded from 0 to 7 using
Fig. 1. Osteophyte signal within a medial tibial marginal osteophyte.
415
Osteoarthritis and Cartilage Vol. 14, No. 5 the following scale: 0 ¼ none; 1 ¼ equivocal; 2 ¼ small; 3 ¼ small-moderate; 4 ¼ moderate; 5 ¼ moderate-large; 6 ¼ large; and 7 ¼ very large. Osteophytes were also assessed for increased signal intensity on fat suppressed images in the axial, sagittal and coronal planes in locations analogous to where they are scored for size. MRIs of all patients were reviewed for the presence of osteophytes and their size and location. Then the slice of MRI ( one slice) which had the largest osteophyte was scored for signal intensity (see Fig. 1). A scale of 0e2 was used in terms of signal intensity (0 ¼ normal intensity, 1 ¼ possible high-intensity, and 2 ¼ probable highintensity). The osteophyte signal measures anywhere within one given knee were summed, creating an osteophyte signal aggregate. The intra-observer correlation coefficient (ICC) for osteophyte signal aggregate was 0.74. For this analysis the predictor (osteophyte signal) was read at baseline.
Patellofemoral Compartment
Medial Tibiofemoral Compartment
PAIN MEASUREMENTS
Pain was assessed in three ways.
Lateral Tibiofemoral Compartment
(1) Presence/absence: Participants were evaluated using the frequent knee symptoms question, ‘‘During the past 30 days, have you had pain, aching, or stiffness in or around your right knee on most days?’’ A similar question was used to assess frequent knee symptoms in the left knee. (2) Severity: All patients were evaluated for the Western Ontario McMasters Osteoarthritis Index (WOMAC), an outcome measure validated for the evaluation of pain in patients with knee OA9, recommended by the Osteoarthritis Research Society in clinical trials evaluating symptoms related to knee OA10. (3) Location: Location of self-reported pain in both knees was recorded dichotomously as present or absent based on locations identified on a standardized diagram (Fig. 2).
Statistical analysis We examined the relation of high-signal osteophytes to the pain status (presence or absence), severity of knee pain, and location of knee pain, respectively. The details were described below.
1. Pain status: The osteophyte signal measures anywhere within one given knee were summed, creating an osteophyte signal aggregate. The osteophyte signal aggregate was used to create quartiles of osteophyte signal aggregate. Knee pain presence was assessed using the frequent knee symptom question (on most days, do you have pain, aching or stiffness in your x joint?), as a dichotomous outcome; pain present/absent. Logistic regression was conducted with quartile of osteophyte signal aggregate as the independent predictor and frequent knee symptom question as the dependent outcome adjusted for gender, body mass index (BMI), and age. 2. Pain severity: Knee pain severity was assessed using the WOMAC pain subscale. Association between quartile of osteophyte signal aggregate and pain severity was assessed using a linear regression with analyses adjusting for covariates including attrition, osteophyte size, synovitis, BMI, and age.
Fig. 2. Diagram used for location of self-reported pain.
3. Pain regional location: The third analysis assessed if region-specific self-reported knee pain location was associated with osteophyte signal on MRI. Compartment-specific knee pain (i.e., patello-femoral (PF), medial tibio-femoral (TF), and lateral TF), was assessed based on a self-reported pain location identified on a knee diagram. The scores for all high-signal osteophytes in each compartment of the knee (i.e., PF compartment [medial/lateral anterior femur/tibia and patella], medial TF compartment [central/posterior medial femur and central/posterior medial tibia], and lateral TF compartment [central/posterior lateral femur and central/posterior lateral tibia]) were summed. Logistic regression was used to evaluate the association between compartment-specific high-signal osteophytes aggregates (independent variable) and compartment-specific knee pain (dependent variable). Analyses were adjusted for covariates including attrition, osteophyte size, synovitis, gender, BMI, and age. Table I Characteristics of study population (n ¼ 217) Characteristics Age (years): mean SD % Male % K&L grade > 2 Proportion of subjects with osteophytes (Y/N) BMI: mean SD Percentage with knee pain WOMAC pain: mean SD Osteophyte signal aggregate: mean SD
67.3 9.1 70.0 74.5 94.4 31.0 5.1 89.8 7.3 3.9 5.5 5.2
SD: Standard Deviation; K&L: Kellgren and Lawrence.
416
M. Sengupta et al.: High-signal osteophytes and knee pain Table II Association between osteophyte signal quartile and knee pain presence Osteophyte signal aggregate quartile
Osteophyte aggregate Number of subjects Percentage with pain PR (95% CI)
0
1
2
3
0e1 58 94.83 1.0
2e4 56 83.93 0.89 (0.78e1.00)
5e8 47 91.49 0.97 (0.86e1.10)
9e25 55 89.09 0.94 (0.83e1.06)
PR: Prevalence ratio; CI: Confidence Interval; Adjusted for age, sex, and BMI.
Results Two hundred and seventeen subjects were included in this analysis. They are predominantly males reflecting the VA community from which they were drawn (see Table I). Thirty-five percentage of knees had isolated radiographic TF ROA, 8% knees had isolated radiographic PF ROA, 65% knees had radiographic PF ROA (isolated or with radiographic TF OA), 75% of knees had radiographic TF OA (isolated or with radiographic PF OA), and 57% knees had both radiographic TF and radiographic PF ROAs. We did not find any association of high-signal osteophytes (subjects were divided into quartiles of osteophyte signal aggregate) with the presence of pain (Table II). Despite relatively small numbers of subjects without pain on most days there was no apparent trend observed. Similarly there was no association between high-signal osteophytes (subjects were divided into quartiles of osteophyte signal aggregate) and pain severity (Table III). It was expected that patients with increasing pain severity (measured by the WOMAC scale) would have higher numbers of high-signal osteophytes. Although there seemed to be a trend towards increasing WOMAC scores in the highest quartile of osteophyte signal, this was not statistically significant. There was no association found between high-signal osteophytes (subjects were divided into tertiles of compartment-specific osteophyte signal aggregate) and pain location (Table IV). There was no statistically significant association found between the self-reported location of pain and the presence of high-signal osteophytes. In the PF joint there was a significant association found in the reverse direction to what we anticipated. The analyses were also conducted without adjusting for BMLs with similar results.
Discussion We have examined whether the presence of high-signal osteophytes on MRI was associated with pain presence,
location or severity. Our research demonstrates that highsignal osteophytes are not associated with pain presence, location or severity. The structural determinants of pain in knee OA are not well understood, but probably involve a multitude of interactive pathways. The current practice of monitoring only a few of these features (usually radiographic joint-space narrowing and osteophytes) provides only a keyhole view of the disease process and heretofore has only been weakly related to symptoms in persons with knee OA. Because OA is a disease of all the tissues in the joint11, measurements of structure need to be seen broadly and capture a broad number of important anatomic features, such as osteophytes, effusions, meniscal tears, subchondral bone architectural changes or ligamentous instability, in addition to cartilage loss. Most of these structures cannot be seen on plain radiography, whereas they are clearly visualized on MRI. This is a critical area for further research, as it is likely that future treatments in OA will target structural abnormalities known to have pain/function consequences. Recent MRI data have been insightful in providing information on the etiology of pain in knee OA7,12e15. Whilst osteophytes are considered to be a defining structural feature of knee OA4, and have previously been shown to be correlated with knee pain on plain radiographs5, high signal within osteophytes does not appear to be associated with pain. The fact that BMLs are associated with pain and high-signal osteophytes are not may be the result of the osteophyte signal being much smaller than the BMLs associated with pain7. The comparative histology of BMLs and high signal within osteophytes is unknown and whilst they are both high-signal lesions on MRI they may be very different from a histopathologic perspective. The findings in the PF joint are interesting and difficult to explain. If it were true it would suggest that the presence of osteophyte signal is inversely associated with pain and contrary to our hypothesis when we undertook this investigation. Moreover they are borderline statistically significant. There are a number of limitations of the current investigation that warrant mentioning. This is a cross-sectional study
Table III Association between osteophyte signal quartile and knee pain severity Osteophyte signal aggregate quartiles
Osteophyte signal aggregate WOMAC pain subscale: mean (SD) B-coefficient *(P value)
0
1
2
3
0e2 7.39 (3.84) Ref.
2e5 6.67 (3.82) 0.777 (0.73)
5e10 7.02 (4.02) 0.140 (0.82)
10e25 8.16 (3.79) 0.556 (0.88)
Adjusted for sex, gender, BMI, BML, attrition, synovosis, and osteophyte size. *B-coefficient represents the difference in WOMAC pain score (0e20 scale) between the quartile and the referent. Numbers less than 0 represent lower WOMAC scores than the lowest quartile.
417
Osteoarthritis and Cartilage Vol. 14, No. 5 Table IV Association between osteophyte signal quartile and knee pain location Location of self-reported pain
Tertiles 0
1
2
PF
Osteophyte signal aggregate No. of subjects Percentage with pain PR (95% CI)
0 76 52 Ref.
1e2 71 46 0.83 (0.59, 1.18)
3e12 86 34 0.62 (0.43, 0.89)
Medial TF
Osteophyte signal aggregate No. of subjects Percentage with pain PR (95% CI)
0 89 35 Ref.
1e2 54 33 0.91 (0.57, 1.45)
3e8 79 36 0.70 (0.43, 1.13)
Lateral TF
Osteophyte signal aggregate No. of subjects Percentage with pain PR (95% CI)
0 140 25 Ref.
1e2 45 26 1.07 (0.60, 1.93)
3e6 39 28 0.93 (0.41, 2.07)
Adjusted for BMI, osteophyte size, BML, synovosis, attrition, and age. PR: Prevalence ratio.
and we have not evaluated the relation between change in signal and change in pain. There were relatively few asymptomatic subjects recruited into this study making the exploration of the association with pain presence somewhat tenuous. This study began in 1997, and the MRI pulse sequences used at that time do not have the same resolution and signal to noise ratio as pulse sequences used in current investigation that may have limited our ability to measure osteophyte signal. In summary high-signal osteophytes detected on MRI are not associated with the presence of pain, pain severity or the self-reported location of pain.
Acknowledgments We would like to thank the participants and staff of the BOKS study, without whom this research would not have been possible. WORMS MRI reading performed by OARG with assistance from John Lynch. Arthritis Foundation Clinical Sciences Grant. National Institutes of Health (AR47785).
References 1. McAlindon TE, Cooper C, Kirwan JR, Dieppe PA. Knee pain and disability in the community. Br J Rheumatol 1992;31(3):189e92. 2. Creamer P, Hunt M, Dieppe P. Pain mechanisms in osteoarthritis of the knee: effect of intraarticular anesthetic. J Rheumatol 1996;23(6):1031e6. 3. Creamer P, Lethbridge-Cejku M, Hochberg MC. Where does it hurt? Pain localization in osteoarthritis of the knee. Osteoarthritis Cartilage 1998;6(5):318e23. 4. Spector TD, Hart DJ, Byrne J, Harris PA, Dacre JE, Doyle DV. Definition of osteoarthritis of the knee for epidemiological studies. Ann Rheum Dis 1993;52(11): 790e4. 5. Cicuttini FM, Baker J, Hart DJ, Spector TD. Association of pain with radiological changes in different compartments and views of the knee joint. Osteoarthritis Cartilage 1996;4(2):143e7.
6. McAlindon TE, Watt I, McCrae F, Goddard P, Dieppe PA. Magnetic resonance imaging in osteoarthritis of the knee: correlation with radiographic and scintigraphic findings. Ann Rheum Dis 1991;50(1):14e9. 7. Felson DT, Chaisson CE, Hill CL, Totterman SM, Gale ME, Skinner KM, et al. The association of bone marrow lesions with pain in knee osteoarthritis [see comments]. Ann Intern Med 2001;134(7):541e9. 8. Peterfy CG, Guermazi A, Zaim S, Tirman PF, Miaux Y, White D, et al. Whole-Organ Magnetic Resonance Imaging Score (WORMS) of the knee in osteoarthritis. Osteoarthritis Cartilage 2004;12(3):177e90. 9. 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(12):1833e40. 10. Pham T, van der Heijde D, Altman R, Anderson J, Bellamy N, Hochberg M, et al. OMERACT-OARSI Initiative: Osteoarthritis Research Society International set of responder criteria for osteoarthritis clinical trials revisited. Osteoarthritis Cartilage 2004;12(5):389e99. 11. Martin JA, Buckwalter JA. Roles of articular cartilage aging and chondrocyte senescence in the pathogenesis of osteoarthritis [Review] [34 Refs]. Iowa Orthop J 2001;21:1e7. 12. Bhattacharyya T, Gale D, Dewire P, Totterman S, Gale ME, McLaughlin S, et al. The clinical importance of meniscal tears demonstrated by magnetic resonance imaging in osteoarthritis of the knee [comment]. J Bone Joint Surg Am 2003;85-A(1):4e9. 13. Felson DT, McLaughlin S, Goggins J, LaValley MP, Gale ME, Totterman S, et al. Bone marrow edema and its relation to progression of knee osteoarthritis. Ann Intern Med 2003;139(5 Pt 1):330e6. 14. Gale DR, Chaisson CE, Totterman SM, Schwartz RK, Gale ME, Felson D. Meniscal subluxation: association with osteoarthritis and joint space narrowing. Osteoarthritis Cartilage 1999;7(6):526e32. 15. Hill CL, Gale DG, Chaisson CE, Skinner K, Kazis L, Gale ME, et al. Knee effusions, popliteal cysts, and synovial thickening: association with knee pain in osteoarthritis. J Rheumatol 2001;28(6):1330e7.