The Effect of Examiner Variability on Multiple Canine Stifle Kinematic Gait Collections in a 3-Dimensional Model.

OBJECTIVE: To evaluate examiner variability in a superficial skin marker model of canine stifle kinematics. STUDY DESIGN: Experimental. ANIMALS: Six clinically normal dogs. METHODS: Dogs had 11 retroreflective markers fixed to the skin on the right hindlimb. Dogs were trotted 5 times through the calibrated testing space and this was repeated on 4 different testing days. Examiner A applied all markers to a dog and collected 6 good trials for analysis. The markers were then removed and Examiner B immediately repeated the process on the same dog. This was repeated for each dog on the 4 testing days. The dogs were trotted at a velocity of 1.70-2.10 m/s through the testing space to obtain the dynamic data sets. Comparisons were performed with Fourier analysis and Generalized Indicator Function Analysis (GIFA). Significance was set at P < .05 for all comparisons. RESULTS: Fourier analysis and GIFA found differences within and between examiners. Fourier analysis found no differences in sagittal and transverse planes for the experienced (A) and novice examiner (B), respectively. Fourier analysis detected fewer differences for the experienced examiner (A). CONCLUSION: Variability occurs within and between examiners using the same kinematic model. Transverse and frontal plane kinematics produce variable results between examiners. Prior experience with the model reduces the amount of variability and results in consistent and repeatable sagittal plane kinematic data collection.

The effect of marker location variability on noninvasive canine stifle kinematics.

OBJECTIVE: Evaluate the effect of marker placement on kinematics of the canine stifle in 3 distinct hindlimb models. STUDY DESIGN: In vivo biomechanical study. ANIMALS: Normal adult mixed-breed dogs (n=5). METHODS: Ten retroreflective markers were affixed to the skin on the right rear leg of each dog to establish normal stifle kinematics. Four additional markers were placed around the greater trochanter (GT), 2 m cranial, caudal, dorsal, and ventral to evaluate single marker placement variability on kinematic model data. Dogs were walked and trotted 5 times through the calibrated space. Sagittal flexion and extension angle waveforms were acquired during each trial with 3 models that were produced simultaneously during each gait. The GT marker was reassigned to 1 of the 4 additional locations (cranial, caudal, dorsal, and ventral) to alter the kinematic model. Comparison of sagittal flexion and extension angle waveforms was performed with Generalized Indicator Function Analysis. RESULTS: Each model provided consistent equivalent sagittal flexion-extension data. Analysis revealed statistically significant differences between all GT locations. The differences were greatest in the cranial and caudal locations for all models. CONCLUSIONS: Deviation of the GT marker in the cranial/caudal direction from an anatomically normal position produces a greater degree of difference than deviation in a dorsal/ventral direction.

Comparison of canine stifle kinematic data collected with three different targeting models.

OBJECTIVE: To model the kinematics of the canine stifle in 3 dimensions using the Joint Coordinate System (JCS) and compare the JCS method with linear and segmental models. STUDY DESIGN: In vivo biomechanical study. ANIMALS: Normal adult mixed breed dogs (n=6). METHODS: Dogs had 10 retroreflective markers affixed to the skin on the right pelvic limb. Dogs were walked and trotted 5 times through the calibrated space and the procedure was repeated 5 days later. Sagittal flexion and extension angle waveforms acquired during each trial with all 3 models (JCS, Linear, and Segmental) were produced simultaneously during each gait. The JCS method provided additional internal/external and abduction/adduction angles. Comparison of sagittal flexion and extension angle waveforms was performed with generalized indicator function analysis (GIFA) and Fourier analysis. A normalization procedure was performed. RESULTS: Each model provided consistent equivalent sagittal flexion-extension data. The JCS provided consistent additional internal/external and abduction/adduction. Sagittal waveform differences were found between methods and testing days for each dog at a walk and a trot with both GIFA and Fourier analysis. After normalization, differences were less with Fourier analysis and were unaltered with GIFA. CONCLUSIONS: Whereas all methods produced similar flexion-extension waveforms, JCS provided additional valuable data. CLINICAL RELEVANCE: The JCS model provided sagittal plane flexion/extension data as well as internal/external rotation and abduction/adduction data.