Objective: To evaluate the reliability of the Thompson system for use in grading the gross pathological changes of intervertebral disk (IVD) degeneration in dogs and to investigate the agreement between gross pathological findings and low-field (0.2-T) magnetic resonance imaging (MRI) findings.
Sample: Vertebral columns from cadavers of 19 dogs of various ages, breeds, and origins.
Procedures: 182 intervertebral segments were collected from 19 canine cadavers. Sagittal T2-weighted MRI of the T11 through S1 portion of the vertebral column was performed within 24 hours after the dogs were euthanized. The vertebral columns were subsequently divided in the midsagittal plane, and high-resolution photographs were obtained of each intervertebral segment (end plate-disk-end plate). The MRI images and photographs were graded separately in a blinded manner by 4 observers who used both Pfirrmann and Thompson grading criteria.
Results: The interobserver agreement for Thompson scores ranged from 0.76 to 0.88, and the intraobserver agreement ranged from 0.88 to 0.94 (Cohen weighted κ analysis). Agreement between scores for the Pfirrmann and Thompson grading criteria was κ = 0.70.
Conclusions and clinical relevance: Grading of IVD degeneration in dogs by use of the Thompson system resulted in high interobserver and intraobserver agreement, and scores for the Thompson system had substantial agreement with low-field MRI findings graded by use of the Pfirrmann system. This suggested that low-field MRI can be used to diagnose IVD degeneration in dogs.