Comparison between high-field 3 Tesla MRI and computed tomography with and without arthrography for visualization of canine carpal ligaments: A cadaveric study.

Abstract

To compare the quality of visualization of canine carpal ligaments by using computed tomography (CT), MRI, CT arthrography (CTA), and magnetic resonance arthrography (MRA). Prospective descriptive study. Cadavers from dogs weighing more than 20 kg. A 16-slice CT scanner and a 3 Tesla MRI were used for the investigation. A dilute contrast medium was injected into the middle carpal and radiocarpal joints under fluoroscopic control, and CTA and MRA images were acquired. To evaluate the difference between imaging modalities, 3 observers graded carpal ligaments of clinical interest using a scale from 0 to 4 for their quality of visualization. Data were analyzed by using a random-effect ordinal logistic regression with Bonferroni adjustment. The interobserver agreement was calculated by using the weighted Cohen's κ. Normal carpal joints (n = 9) were investigated. Magnetic resonance arthrography improved visualization of the majority of carpal ligaments compared with MRI (P < .05) and offered the best visualization overall. Magnetic resonance imaging and MRA offered better visualization compared with both CT and CTA (P < .05). There was no difference between CT and CTA. Interobserver agreement was discrete (0.2 < κ ≤ 0.4) for all observers. Arthrography improved the capabilities of MRI but not of CT for visualization of the canine carpal ligaments. Magnetic resonance arthrography was particularly useful for evaluation of the stabilizers of the antebrachiocarpal joint. 3 Tesla MRA and MRI allow excellent visualization of the ligamentous morphology and may be helpful in the diagnostic process of carpal sprains in dogs.