A preliminary evaluation of the reliability of a modified functional scoring system for assessing neurologic function in ambulatory thoracolumbar myelopathy dogs

BACKGROUND The objective of this study was to develop and assess the reliability of a modified scoring system for evaluating the function of the two pelvic limbs separately, in ambulatory thoracolumbar myelopathy dogs. A previously established neurologic score scale for dogs with T3-L3 lesions was modified in order to provide a separate score for each pelvic limb. RESULTS Seventeen ambulatory dogs with thoracolumbar myelopathies were evaluated. Using the new scale, two observers independently performed 22 observational gait analyses (OGAs) in ten dogs without videotape. Another 18 OGAs were performed in seven dogs by watching videotapes of them ambulating. There was poor agreement (concordance correlation coefficient, 0.87) between the two observers for all 40 OGAs. When stratified, the agreement was moderate (concordance correlation coefficient, 0.90) in the OGAs without videotaping and poor (concordance correlation coefficient, 0.80) for the OGAs based on videotapes. For the decision regarding which pelvic limb was more severely affected, a fair agreement (kappa value, 0.30) between the two observers was noted. Without videotape there was only slight agreement (kappa value, 0.05), but with videotape there was moderate agreement (kappa value, 0.56). CONCLUSIONS The modified scoring system in this study provides moderate reliability in assessing the functional neurologic status of each pelvic limb, by OGA without videotape, in canine T3-L3 patients. Further development of this scoring system is required. However, imperfect agreement when visually quantifying neurological deficits is not unexpected.