Prognostic model to predict the incidence of radiographic knee osteoarthritis
Authors' conclusion
A novel prognostic model based on common clinical variables and protein biomarkers was developed and externally validated to predict rKOA incidence over a 96-month period in individuals without any radiographic signs of disease.
Study context
- Population
- Subjects without radiographic signs of knee OA (Kellgren and Lawrence (KL)=0 in both knees), OA initiative (OAI) cohort (development) and Prospective Cohort of A Coruña (PROCOAC) (external validation)OAI development dataset
- Participants
- 282 subjects
- Comparison
- Clinical model vs clinical model plus serum protein biomarkers
- Outcome
- Incidence of radiographic knee OA
- Follow-up
- 96 months
Methodological detailsinternal cross-validation, external validation, nomogram
- Internal cross-validation
- External validation
- Nomogram
Performance
As reported in the abstract
Prediction of radiographic KOA (rKOA) incidence over 96 months (development dataset, OAI)
Model with demographic, anthropometric and clinical data (age, sex, BMI, WOMAC pain score): AUC
0.702
Inclusion of ZA2G, A2AP and APOA1 data significantly improved sensitivity and predictive performance: AUC
0.831
Simplest model (clinical covariates plus ZA2G and A2AP serum levels): AUC
0.826
Further findings
External validation in PROCOAC (n=100), AUC
- AUC
- 0.713
- Independent dataset of 100 individuals
Original publication
Prognostic model to predict the incidence of radiographic knee osteoarthritis. Ann Rheum Dis. Published 11 April 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →