Sensitivity and specificity of 3 Tesla magnetic resonance imaging and multidetector computed tomographic tenography to identify artificially induced soft tissue lesions in the equine cadaveric digital flexor tendon sheath

• Published in: Equine veterinary education 2023-07, Vol.35 (7), p.e507-e516
• Main Authors: Aßmann, Anton D., Ohlerth, Stefanie, Torgerson, Paul R., Bischofberger, Andrea S.
• Format: Article
• Language: English
• Subjects: Abnormalities; Cadavers; computed tomographic tenography; Computed tomography; Diagnosis; digital flexor tendon sheath; horse; Horses; Lesions; Magnetic resonance imaging; Medical imaging; MRI; Pressure loss; Sensitivity; Sheaths; Soft tissues; Statistical analysis; Tendons
• ISSN: 0957-7734; 2042-3292
• DOI: 10.1111/eve.13751

Summary Background A common cause of lameness in horses is tenovaginitis of the digital flexor tendon sheath (DFTS). Ultrasound and contrast tenography have become frequently used diagnostic tools for abnormalities in the DFTS, but are considered insufficiently accurate in describing specific lesions. Objectives To determine the sensitivity and specificity of computed tomographic tenography (CTT) and 3 Tesla magnetic resonance imaging (MRI) to detect artificial lesions within the DFTS. Study design Blinded laboratory study. Methods Using an arthroscopic hook knife, lesions (10–20 mm long and 3–5 mm deep; n = 52) were created tenoscopically in the superficial digital flexor tendon (SDFT), deep digital flexor tendon (DDFT), manica flexoria (MF) and proximal scutum in 19 distal limb specimens. MRI and CTT were performed and images were reviewed. Sensitivity and specificity were calculated for each modality and compared. Results Computed tomographic tenography and MRI identification of SDFT and MF lesions showed similar sensitivity (75% vs. 85%, respectively; p = 1) and specificity for MF lesions (96% for both). SDFT lesions specificity was similar for CTT versus MRI (85% vs. 77%, respectively; p = 0.88). For DDFT lesions, MRI sensitivity (62%) was higher compared to CTT (38%), although not statistically significant (p = 0.58). MRI specificity (92%) was lower than CTT (96%, p = 1). Lesions in the proximal scutum were more frequently identified by MRI compared to CTT (sensitivity: 93% vs. 57%, p = 0.17; specificity: 96% vs. 100%, p = 1). Main limitations Artificial, small lesions were assessed. An incomplete arthroscopic portal seal resulted in pressure loss and contrast leakage. Conclusions The sensitivity and specificity of both modalities are good to excellent for the diagnosis of most artificially created lesions. CTT performed similarly to MRI in detecting SDFT and MF lesions. MRI showed a higher sensitivity for the diagnosis of DDFT and proximal scutum lesions. The results indicate a future application of CTT for the diagnosis of DFTS pathologies, whenever suspecting MF tears.