Purposes: Our aim was to determine the diagnostic accuracy of native MRI regarding different ligamentous lesions of the wrist and to analyze the influence of technical characteristics, such as field strength, application of fat saturation, 3D sequences, and wrist coils.
Methods: The systematic search was performed using MEDLINE, Embase, and CENTRAL databases. Studies that were published before the 12th of February 2024 were included. All studies comparing the diagnostic accuracy of native wrist MRI to that of wrist arthroscopy for suspected ligamentous lesions were included. Results were analyzed by anatomic localization and technical aspects of the MRI. To assess the quality of included studies, we used the revised QUADAS-2 tool.
Results: The systematic search revealed 5,181 articles. Thirty-seven studies, reporting 3893 ligamentous lesions, were eligible for inclusion. The studies displayed heterogeneity in terms of technical conditions, such as field strength, the use of wrist coils, the application of 3D sequences and fat saturation. Research methods also varied. Overall sensitivity and specificity were 0.78 (0.66 - 0.86) and 0.81 (0.70 - 0.89) for 1.5T MRI, while sensitivity was 0.73 (0.68 - 0.78) and specificity was 0.90 (0.59 - 0.98) for 3T MRI. There was no significant difference between the two subgroups (p=0,3807 and p=0,4248). Sensitivity was 0.82 (0.75 - 0.87) for triangular fibrocartilage complex (TFCC) lesions, 0.63 (0.50 - 0.74) for scapholunate ligament (SL) tears, and 0.41 (0.25 - 0.60) for lunotriquetral ligament (LT) lesions. Specificity for TFCC lesions was 0.82 (0.73 - 0.89), for SL tears 0.86 (0.73 - 0.93), and for LT lesions 0.93 (0.81 - 0.98).
Conclusion: The sensitivity and specificity of MRI is influenced by the anatomic location of the lesion and technical conditions. In terms of diagnostic accuracy, no significant difference was found between 1.5T and 3T MRI.
Level of evidence: III. Systematic review of Level II. - III.
Copyright © 2024. Published by Elsevier Inc.