Towards developing diagnostic criteria for early knee osteoarthritis: data from the CHECK study
J Runhaar1, M Kloppenburg2, M Boers3
1Department of General Practice, Erasmus MC University Medical Center Rotterdam, Rotterdam.
Rheumatology (Oxford, England)
|November 27, 2020
Summary
Developing diagnostic criteria for early knee osteoarthritis (OA) is crucial. This study created predictive models with fair ability to identify knee OA in primary care patients, requiring further validation.
Area of Science:
- Orthopedics and Sports Medicine
- Rheumatology
- Epidemiology
Background:
- Early diagnosis and treatment of knee osteoarthritis (OA) are increasingly recognized as vital.
- Currently, validated diagnostic criteria for early knee OA are lacking, hindering timely intervention.
- This study addresses the need for developing such criteria.
Purpose of the Study:
- To initiate the development of diagnostic criteria for early knee osteoarthritis (OA).
- To identify predictive baseline factors for clinically relevant knee OA in primary care.
Main Methods:
- Utilized data from 761 individuals (1185 knees) from the CHECK study, followed for 10 years.
- 36 experts evaluated medical records to diagnose knee OA, serving as the gold standard.
- Backward selection methods were employed to build predictive models using baseline data (history, physical exam, imaging, blood tests).
Main Results:
- The prevalence of clinically relevant knee OA during follow-up was 37%.
- Developed predictive models included 7-11 baseline factors.
- These models achieved an area under the curve (AUC) between 0.746 and 0.764, indicating 'fair' predictive ability.
Conclusions:
- The developed diagnostic models demonstrate fair predictive capability for early knee OA in primary care settings.
- Further research is necessary to refine these models and validate the identified predictive factors.
- The ultimate goal is to establish clinically feasible and relevant diagnostic criteria for early knee OA.
