Abstract
Objectives:
Subscapularis (SSP) tendon injuries are increasingly being diagnosed by arthroscopy. Preoperative diagnosis remains a challenge with clinical examination and magnetic resonance imaging (MRI) lacking sensitivity. The purpose of the study was to evaluate a new MRI finding in the coronal plane, the ‘Digitation Sign’.
Methods:
A consecutive series of 197 patients who underwent surgery for arthroscopic rotator cuff repair between January 2020 and December 2020 was reviewed. Inclusion criteria were: availability of a preoperative MRI less than 18 months old and age of more than 18 years. SSP injuries were described intraoperatively according to the Lafosse et al. and the French Arthroscopy Society classifications. Of the 197 patients, 44 had a preoperative arthroscan, 21 did not have an available DICOM file, and 1 patient had a history of repair on the side being examined. We included 132 shoulders for analysis. MRI scans were reviewed by 3 independent observers. The SSP was assessed on axial, sagittal, and coronal sections according to Adams’ criteria. The ‘Digitation Sign’ was assessed in the coronal plane on the T2 sequences. The statistical analysis determined the inter-observer agreement, and sensitivity and specificity calculations.
Results:
The inter-observer analysis detected substantial agreement (κ = 0.640). Of the 132 patients included, 74 (56%) had an arthroscopically confirmed SSP injury. 55 (74.3%) of the SSP injuries were correctly diagnosed by the ‘Digitation Sign’ alone, 62 (83.8%) with Adams’ criteria alone and 66 (89.2%) with a combination of both. 49 (84.5%) of the intact SSP were confirmed with the ‘Digitation Sign’ alone, 45 (77.6%) with Adams’ criteria alone and only 39 (67.2%) with a combination of both. The sensitivity of the ‘Digitation Sign’ combined with Adams’ criteria was statistically better than the sensitivity of Adams’ criteria alone (p = 0.046). The ‘Digitation Sign’ was more specific than Adams’ criteria (p = 0.014). The effectiveness of the ‘Digitation Sign’ combined with Adams’ criteria increased with the severity of the injury (grade III or IV). SSP injuries were significantly associated with biceps injuries (p <0.0001).
Conclusion:
Preoperative MRI analyzed according to a protocol that combines Adams’ criteria and the ‘Digitation Sign’ is simple, reproducible and sensitive in the diagnosis of SSP injuries. The specificity of the ‘Digitation Sign’ enables quick verification of the integrity of the subscapularis in the first reading of standard coronal sections of cuff tear analysis.