AIM: To evaluate the diagnostic accuracy of point-of-care (POC) ultrasound (US) in detecting joint changes in haemophilic arthropathy compared to comprehensive US, physical examination and MRI. METHODS: Single-centre, prospective study examining 22 ankles, eight knees and nine elbows of 28 subjects (median/range, 16/6-27 years) with inherited bleeding disorders. Same-day physical and imaging joint assessment was performed using Haemophilia Joint Health Score (HJHS) 2.1, POC-US, comprehensive US and MRI. Two readers reviewed the imaging studies using standardized scoring systems. The inter-reader agreement was assessed with weighted kappa statistics. Diagnostic test performance of POC-US and comprehensive US was benchmarked against MRI findings using sensitivity, specificity, predictive values and accuracy. Spearman correlation was used to correlate physical examination and imaging findings. RESULTS: Moderate inter-reader agreement was observed for soft-tissue changes (kappa = 0.58), which was lower than comprehensive US (kappa = 0.76) and MRI (kappa = 0.77). Agreement for osteochondral lesions was poor (κ 0.09), trailing behind comprehensive US (kappa = 0.55) and MRI (kappa = 0.75). POC-US showed reduced accuracy across all joint parameters in the elbows and ankles compared to comprehensive US, but comparable diagnostic performance in the knees. A good correlation was noted between total POC-US scores and HJHS (r = 0.68, p <
0.0002), similar to comprehensive US (r = 0.66, p <
0.0002). CONCLUSION: POC-US is valuable for assessing soft-tissue changes in haemophilic arthropathy, correlates satisfactorily with physical examination, but its accuracy is inferior to comprehensive US for ankles and elbows. Both POC-US and comprehensive US should be employed cautiously for assessing osteochondral joint abnormalities.