Are ultrasonographic signs of inflammation predictors for response to intra-articular glucocorticoids in knee osteoarthritis?

OBJECTIVES: To investigate the predictive value of ultrasound (US) characteristics for the effect of intra-articular glucocorticoids in knee osteoarthritis (OA).

METHODS: In this prospective cohort study, 62 patients with symptomatic knee OA (clinical knee OA criteria, pain>4 on a Numerical Rating Scale (NRS; 0-10)) received an intra-articular glucocorticoid injection (40 mg triamcinolone acetonide). Patients with NRS pain ≤4 at 4 weeks were defined as responders. On inclusion, demographics, clinical data (body mass index, local swelling) knee x-rays and knee injury and Osteoarthritis Outcome Score (KOOS) questionnaire were collected. Six US features were assessed including: effusion, synovial hypertrophy, Baker's cyst, infrapatellar bursitis, meniscal protrusion and cartilage thickness. Stepwise multiple logistic regression analyses with forward selection were conducted to identify possible predictors.

RESULTS: At 4 weeks, 42% of the study participants reached a NRS ≤4; an effect comparable to existing literature. Regression analyses showed that patients who used analgesics at baseline were less likely to have a good response. The small proportion of patients with infrapatellar bursitis was more likely to respond to the injection.

CONCLUSIONS: No patient, disease or US characteristic of inflammation, turned out to be a reliable and clinically meaningful predictor for the effect of intra-articular glucocorticoids after four weeks in knee OA.