RESUMO
Fat infiltration of skeletal muscle (myosteatosis) is a characteristic of reduced muscle quality. Objectives of this retrospective, observer agreement, methods comparison, cross-sectional study were to compare epaxial myosteatosis of dogs with surgically confirmed intervertebral disc extrusion (IVDE) using a visual semi-quantitative method and quantitative 2-point Dixon fat-water chemical-shift separation MRI method. Thoracolumbar MRI studies of 86 dogs were reviewed. Assessed locations were the site of extrusion and the adjacent intervertebral disc space immediately cranial. A visual fat score (VFS) was assigned to fat infiltration of the epaxial muscles. Quantitative assessment of fat infiltration was determined by calculating a muscle-fat fraction (MFF). There was moderate agreement between the two reviewers for VFS assessment (weighted Kappa 0.60, 95%CI 0.53-0.66) with moderate to strong correlation between scoring methods (P ≤ .001). Both VFS (P = .023, P = .02) and MFF (P = .041, P = .043) were significantly correlated with neurologic grade at presentation for the erector spinae muscle group at both sites. Higher VFS in the multifidus muscle increased the odds of a successful outcome when assessed at the extrusion site (odds ratio and 95%CI 4.3 (1.02-78.7), P = .0461). In summary, the visual and Dixon method of epaxial muscle-fat assessment are feasible in dogs with IVDE with moderate to strong correlation. Increased VFS of myosteatosis in the multifidus muscle at the extrusion site significantly increased the odds of a successful outcome. Further studies are indicated to assess the utility of myosteatosis as an MRI imaging biomarker in the preoperative imaging assessment of dogs with IVDE.