Evaluation of the repeatability of kinetic and temporospatial gait variables measured with a pressure-sensitive treadmill for dogs

OBJECTIVE To evaluate intrasession and intersession repeatability of measurements for temporospatial and kinetic variables obtained with a pressure-sensitive treadmill designed for gait analysis of dogs. ANIMALS 16 client-owned dogs. PROCEDURES The influence of treadmill speed on accuracy of ground reaction force (GRF) measurements was assessed by simulated gait analysis at 0 to 7.5 km/h with a custom test device. A similar test was performed with 1 client-owned dog ambulating on the treadmill at 5 speeds (3 to 7 km/h) for GRF calculations. Fifteen client-owned dogs were then walked on the treadmill at 3 km/h for collection of temporospatial and kinetic data. Intrasession repeatability was determined by comparing 2 sets of measurements obtained ≤ 2 hours apart. Intersession repeatability was determined by comparing the first set of these measurements with those for a second session ≥ 4 days later. Intraclass correlation coefficients (ICCs; consistency test) and difference ratios were calculated to assess repeatability. RESULTS Increases in treadmill speed yielded a mean 9.1% decrease in weight-normalized force data at belt speeds of up to 7.5 km/h for the test device, compared with the value when the treadmill belt was stationary. Results were similar for the dog at increasing treadmill speeds (mean decrease, 12.4%). For temporospatial data, intrasession ICCs were > 0.9 and intersession ICCs ranged from 0.75 to 0.9; for GRFs, intrasession and intersession ICCs ranged from 0.68 to 0.97 and from 0.35 to 0.78, respectively. CONCLUSIONS AND CLINICAL RELEVANCE Repeatability of temporospatial data for healthy dogs was good to excellent; results for kinetic data varied. Further research is needed to investigate use of this system for gait analysis with larger samples of dogs and dogs with lameness