Diagnostic performance of long head of biceps tendon tears on MRI: systematic review and meta-analysis.

Abstract

The purpose of this meta-analysis was to determine the diagnostic performance of conventional MRI and MR arthrography for tendinosis, and partial and complete tears of the long head of the biceps tendon (LHBT) using arthroscopy as the reference standard. A systematic review was performed using predefined data fields in PubMed, and all articles published from January 2000 up to April 2022 were retrospectively pooled and reviewed. Six MRI studies on complete tear (n = 555) and ten studies on partial tear/tendinosis (n = 2487) were included in the analysis. Two of the included studies in each group investigated the use of MR arthrography. The data sets were analyzed using a univariate approach with the DerSimonian and Laird random effects model and the proportional hazards model. MRI shows high specificities in diagnosing complete tears of the LHBT ranging from 93.0 to 99.0%. Diagnostic sensitivity was more heterogeneous ranging from 55.9 to 90.0%. The overall negative likelihood ratio was 0.29 (95% CI: 0.17-0.50) and the overall positive likelihood ratio was 37.3 (95% CI: 11.9-117.4). The mean sensitivity in diagnosing partial tear/tendinosis of the LHBT was 67.8% (95% CI: 54.3-78.9%) and the specificity was 75.9% (95% CI: 63.6-85.0%), resulting in a balanced accuracy of 71.9%. The overall negative likelihood ratio was 0.44 (95% CI: 0.32-0.59) and the overall positive likelihood ratio was 2.64 (95% CI: 1.91-3.65). MRI is highly specific for the diagnosis of complete tears of the LHBT, whereas diagnostic sensitivity was more heterogeneous. The diagnosis of partial tears and/or tendinosis of the LHBT remains challenging on MRI, which may warrant complementary clinical examination or other imaging modalities to increase diagnostic confidence in equivocal cases. Conventional MRI and MR arthrography have high diagnostic performance for complete tendon tear when compared to arthroscopy. The diagnosis of tendinosis/partial tears remains challenging and may require comparison with clinical tests and other imaging modalities. •There is no clear consensus regarding the primary imaging modality for the evaluation of LHBT disorders. •Conventional MRI and MR arthrography are highly specific in diagnosing complete tears of the LHBT. •Diagnosis of partial tears/tendinosis of the LHBT on conventional MRI and MR arthrography remains a diagnostic challenge.