Accuracy of asymmetry indices of ground reaction forces for diagnosis of hind limb lameness in dogs

Laurent Fanchon Unité de Médecine de l’Elevage et du Sport, Ecole Nationale Vétérinaire d’Alfort, 94700 Maisons, Alfort, France.

Search for other papers by Laurent Fanchon in
Current site
Google Scholar
PubMed
Close
 DVM, MS
and
Dominique Grandjean Unité de Médecine de l’Elevage et du Sport, Ecole Nationale Vétérinaire d’Alfort, 94700 Maisons, Alfort, France.

Search for other papers by Dominique Grandjean in
Current site
Google Scholar
PubMed
Close
 DVM, PhD

Abstract

Objective—To determine the accuracy of asymmetry indices of ground reaction forces (GRF) for diagnosis of hind limb lameness in dogs.

Animals—36 healthy dogs and 13 dogs with naturally acquired cranial cruciate ligament rupture or hip dysplasia.

Procedures—Lameness for affected dogs ranged from not detectable to minor and constant. While dogs trotted on an instrumented treadmill, GRF variables were recorded and analyzed with asymmetry indices. Each index was tested for its ability to discriminate between healthy and affected dogs. Combinations of several indices were also assessed.

Results—Vertical force variables had better accuracy than craniocaudal force variables. Peak vertical force was the most accurate variable. Partial asymmetry during trotting was detected in healthy dogs. A multivariate approach that used peak vertical force and maximal rising slope yielded the optimum combination to distinguish between healthy and affected dogs. In addition, sensitivity of 92% or specificity of 95% may be achieved with 2 cutoff values while simultaneously maintaining specificity or sensitivity, respectively, at > 85%.

Conclusions and Clinical Relevance—Asymmetry indices of GRFs were accurate for detection of hind limb lameness in dogs. This is particularly relevant for study designs in which only a single gait evaluation is possible.

Abstract

Objective—To determine the accuracy of asymmetry indices of ground reaction forces (GRF) for diagnosis of hind limb lameness in dogs.

Animals—36 healthy dogs and 13 dogs with naturally acquired cranial cruciate ligament rupture or hip dysplasia.

Procedures—Lameness for affected dogs ranged from not detectable to minor and constant. While dogs trotted on an instrumented treadmill, GRF variables were recorded and analyzed with asymmetry indices. Each index was tested for its ability to discriminate between healthy and affected dogs. Combinations of several indices were also assessed.

Results—Vertical force variables had better accuracy than craniocaudal force variables. Peak vertical force was the most accurate variable. Partial asymmetry during trotting was detected in healthy dogs. A multivariate approach that used peak vertical force and maximal rising slope yielded the optimum combination to distinguish between healthy and affected dogs. In addition, sensitivity of 92% or specificity of 95% may be achieved with 2 cutoff values while simultaneously maintaining specificity or sensitivity, respectively, at > 85%.

Conclusions and Clinical Relevance—Asymmetry indices of GRFs were accurate for detection of hind limb lameness in dogs. This is particularly relevant for study designs in which only a single gait evaluation is possible.

All Time Past Year Past 30 Days
Abstract Views 176 0 0
Full Text Views 12524 11881 127
PDF Downloads 781 279 48
Advertisement