3D imaging has good specificity but poor sensitivity for the diagnosis of pathologies of the long head of the biceps: a systematic review and meta-analysis

Purpose To systematically review and meta-analyse the literature to determine which three-dimensional (3D) imaging modality provides the best diagnostic accuracy to detect pathologies of the long head of the biceps tendon (LHBT). Materials and methods A search was performed on PubMed, Embase(R), and Cochrane. Studies that compared the diagnostic accuracy of 3D imaging modalities versus arthroscopy for the assessment of LHBT pathologies were included. Studies assessing superior labral anterior posterior (SLAP) lesions were excluded. Results Fifteen studies were included; nine were eligible for meta-analysis. Six studies on instability indicated a sensitivity of 0.68 (CI 0.46-0.84) and specificity of 0.76 (CI 0.68-0.82). Four studies on full-thickness tears indicated a sensitivity of 0.56 (CI 0.28-0.81) and specificity of 0.97 (CI 0.93-0.99). Four studies on partial-thickness tears indicated a sensitivity of 0.52 (CI 0.20-0.82) and specificity of 0.64 (CI 0.25-0.91). Two studies on any tear indicated a sensitivity of 0.58 (CI 0.28-0.83) and specificity of 0.99 (CI 0.93-1.00). Only one study on other pathologies indicated a sensitivity of 0.61 and specificity of 0.84. Conclusion To diagnose LHBT pathologies, 3D imaging modalities overall have low-to-moderate sensitivity, but high-to-excellent specificity. The consistency in reported sensitivity is generally poor, while the consistency and reported specificity is good for the detection of instability, full-thickness tears and any tear, but poor for the detection of partial-thickness tears. 3D imaging may be adequate to rule out LHBT pathologies, but are not sufficiently reliable to confirm the presence of such pathologies.