Objectives: To distinguish patients (pts) with enthesitis having spondyloarthritis (SpA) from pts with enthesitis without SpA by ultrasound (US) enthesitis score.
Methods: The study sample included 127 pts with enthesitis (76 pts with SpA, 26 pts with rheumatoid arthritis, 25 pts with mechanically-related enthesitis). The entheses of plantar fascia, Achilles, patellar, quadriceps and common extensor tendon on lateral epicondyle were examined by US. Two operators, blinded to clinical diagnosis and enthesitis symptoms, assessed enthesis thickness, echogenicity, enthesophytes, power Doppler signal and erosions. Logistic regression and receiver operating characteristic (ROC) curve analysis were used to determine the predictive value of each enthesitis lesion for diagnosis of SpA. The best predictive value for SpA was accomplished when absence and presence of increased thickness, hypoechogenicity and enthesophytes were scored as 0 and 1; absence and presence of PD and erosions were scored as 0 and 4. Belgrade Ultrasound Enthesitis Score (BUSES) represents a cumulative score of derived enthesitis lesion scores at examined entheses. Independent-samples t-test was used for BUSES comparison between pts with and without SpA. Validity of BUSES for SpA diagnosis was evaluated by sensitivity and specificity. Cut-off point was chosen as the smallest value with specificity of at least 90%. The reliability was analysed by intra-class-correlation coefficient (ICC).
Results: BUSES was 9.9 ± 12.4 (mean ± SD) in SpA pts and 3.1 ± 4.2 in pts without SpA (p<0.001). BUSES cut-off point ≥ 7 achieved excellent specificity (90.2%) and fair sensitivity (47.4%). ICC was 0.99.
Conclusions: BUSES is highly specific, valid and reliable to identify patients with SpA.