Arthritis & Rheumatism, Volume 62,
November 2010 Abstract Supplement

Abstracts of the American College of
Rheumatology/Association of Rheumatology Health Professionals
Annual Scientific Meeting
Atlanta, Georgia November 6-11, 2010.


The Importance of Methodological Quality for the Association between Radiographic and Clinical Features of Hip and Knee Osteoarthritis: A Systematic Review.

Kinds2,  M. B., Vignon5,  E. P., Bijlsma4,  J. W. J., Viergever1,  M. A., Marijnissen4,  A. C. A., Lafeber4,  F. P. J. G., Welsing3,  P. M. J.

Image Sciences Institute, University Medical Center Utrecht, The Netherlands
Rheumatology & Clinical Immunology, Image Sciences Institute, University Medical Center Utrecht, The Netherlands
Rheumatology & Clinical Immunology, Julius Center for Health Sciences & Primary Care, University Medical Center Utrecht, The Netherlands
Rheumatology & Clinical Immunology, University Medical Center Utrecht, The Netherlands
Rheumatology, University Hospital Lyon-Sud Pierre-Benite, France

Background:

Still there is debate on the presumed association between radiographic and clinical features of osteoarthritis (OA). Inconsistency in reported associations might be caused by different definitions of clinical OA, and by different protocols and scoring methods for radiographic damage. Objective of this review was to evaluate whether there is an association between radiographic OA and clinical OA of hip and knee, accounting for the importance of disease definition, radiographic protocol, and standardized outcome measures.

Methods:

A systematic literature search was performed with the keywords 'OA', 'hip', 'knee', 'radiographic', and 'clinical'. For comparison of study results, seven classification criteria for general study quality and methodological quality were developed. All studies were evaluated for an association between radiographic and clinical OA. Associations were classified as 'present' when statistically significant, as 'absent' when not statistically significant, and as 'non-evident' when not all performed comparisons were statistically significant. The influence of classification criteria on the association was investigated, by classifying both the number and the specific criteria fulfilled.

Results:

The literature search resulted in 47 studies describing associations between radiographic and clinical OA. When all studies were evaluated, associations were present in 15%, absent in 19%, and non-evident in 66%. Associations were strongest in the six studies fulfilling all classification criteria; present in 50%, absent in 17%, and non-evident in 33% of studies. The frequency of studies with present associations significantly (p<0.01) diminished when the number of fulfilled criteria decreased.

Frequency of associations: + (association), +/- (non-evident association), and - (no association) in 6, 8, 17, and 16 studies fulfilling 7 (all), 6, 5, and <=4 classification criteria respectively. Linear regression analysis with number of criteria as independent variable and association + vs association +/- and - as dependent variable: p<0.01.

Specifically the criteria for radiographic protocol, and standardized outcome measures were important. Associations were present in 35%, absent in 12%, and non-evident in 53% of the 17 studies in which the radiographic protocol was adequate. In the 30 studies not fulfilling this criterion, associations were significantly less present (p=0.012): associations were non-evident in 74% of studies, present in only 3%, and absent in 23% of studies. Considering standardized outcomes, associations were present in 26%, absent in 37%, and non-evident in 37% of studies when the criterion was fulfilled. When the criterion was not fulfilled, associations were significantly less present (p=0.002): associations were non-evident in 86%, present in only 7%, and absent in 7% of studies.

Conclusion:

Radiographic OA and clinical OA were more commonly associated when studies fulfilled criteria for study quality and methodological quality. Specifically the radiographic protocol and standardized outcome measures are important for future research regarding the association between radiographic and clinical OA.

Sponsor:

Dutch Arthritis Association.

To cite this abstract, please use the following information:
Kinds, M. B., Vignon, E. P., Bijlsma, J. W. J., Viergever, M. A., Marijnissen, A. C. A., Lafeber, F. P. J. G., et al; The Importance of Methodological Quality for the Association between Radiographic and Clinical Features of Hip and Knee Osteoarthritis: A Systematic Review. [abstract]. Arthritis Rheum 2010;62 Suppl 10 :952
DOI: 10.1002/art.28719

Abstract Supplement

Meeting Menu

2010 ACR/ARHP