Diagnostic accuracy of ultrasound compared to magnetic resonance imaging for rotator cuff tears: a systematic review

Abstract

Objective: This systematic review aimed to conduct a comprehensive comparative analysis of the diagnostic performance of magnetic resonance imaging (MRI) and ultrasound (US) in detecting rotator cuff tears (RCTs). Methods: A systematic literature search was conducted in the PubMed, Scopus, and Embase databases from January 2021 to March 2025 for original studies on adults (>= 18 years) with suspected RCTs undergoing both US and MRI. Thirteen studies involving 792 participants met the inclusion criteria. Results: US was better than MRI for full-thickness tears, with a sensitivity of 100 (especially for supraspinatus), a specificity of 81-100, and a kappa of up to 1.0. For partial-thickness tears, the results varied (sensitivity 72-94, accuracy 76-94). Non-supraspinatus tendons were less reliable, and the results were affected by the operator's skill, the patient's sex (worse in females), and the side of the body. MRI was better for partial tears, complicated cases, and planning surgery. Conclusion: The findings demonstrate that US can serve as a valuable auxiliary imaging modality for rotator cuff evaluation, particularly excelling in detecting full-thickness tears when MRI is contraindicated, unavailable, or in resource-limited settings. While US shows excellent diagnostic performance for full-thickness supraspinatus tears, MRI remains the primary diagnostic standard in clinical practice for comprehensive rotator cuff assessment, particularly for partial tears, complex pathologies, and detailed pre-operative planning.