Longitudinal ambulatory measurements of gait abnormality in dystrophin-deficient dogs

Abstract

Background: This study aimed to measure the gait abnormalities in GRMD (Golden retriever muscular dystrophy) dogs during growth and disease progression using an ambulatory gait analyzer (3D-accelerometers) as a possible tool to assess the effects of a therapeutic intervention.

Methods: Six healthy and twelve GRMD dogs were evaluated twice monthly, from the age of two to nine months. The evolution of each gait variable previously shown to be modified in control and dystrophin-deficient adults was assessed using two-ways variance analysis (age, clinical status) with repeated measurements. A principal component analysis (PCA) was applied to perfect multivariate data interpretation.

Results: Speed, stride length, total power and force significantly already decreased (p<0.01) at the age of 2 months. The other gait variables (stride frequency, relative power distributions along the three axes) became modified at later stages. Using the PCA analysis, a global gait index taking into account the main gait variables was calculated, and was also consistent to detect the early changes in the GRMD gait patterns, as well as the progressive degradation of gait quality.

Conclusion: The gait variables measured by the accelerometers were sensitive to early detect and follow the gait disorders and mirrored the heterogeneity of clinical presentations, giving sense to monitor gait in GRMD dogs during progression of the disease and pre-clinical therapeutic trials.

Figure 1

Evolution of the different gait variables with age . Each GRMD dog is represented using its own symbol (see table 1). The healthy population is represented by a grey zone, covering the mean (grey line) ± 1 SD (black lines). A: evolution of the motor score in GRMD dogs; B: evolution of the speed normalized by the height at withers (HW); C: evolution of the stride length normalized by the height at withers (HW); D: evolution of the stride frequency; E: evolution of the total power; F: evolution of the force; G: evolution of the medio-lateral part of the power; H: evolution of the regularity.

Figure 2

Using principal component analysis (PCA) for the follow-up of individuals . The active individuals, used to define the plane, were eight healthy adult dogs represented by empty circles, and eleven GRMD adult dogs represented by grey points. 94.53% of the total variance is explained by the two first components, component 1 along the X axis and component 2 along the Y axis. The projection of the individuals on the PCA plane following these components are represented in panels A to D, and the projection of the variables in panel E. The healthy and GRMD dogs of the present study were used as supplementary individuals. Each GRMD dog is represented using his own symbol (see table 1). The healthy growing dogs are represented by black points. A: projection of the 2 months-old dogs on the PCA plane; B: projection of the 4 months-old dogs on the PCA plane; C: projection of the 6 months-old dogs on the PCA plane; D: projection of the 9 months-old dogs on the PCA plane. E: projection of the seven variables. The first component explains 72.61% of the total variance and is mainly explained by the stride frequency, the stride length, the total power and the medio-lateral part of the power. The second component explains 21.92% of the total variance and is mainly explained by the cranio-caudal and the dorso-ventral part of the power. F Evolution of the distance of each GRMD dog from the centre of gravity of the six age-matched healthy controls. The grey zone represents the mean (grey line) ± 1SD (black lines) of this distance in healthy dogs.