Axis-dependent relationships and agreement between whole-body and limb-specific center of pressure measurements in dogs

Static posturography is widely used to assess postural stability in dogs. Considerable methodological variation exists regarding whether center of pressure (COP) parameters are derived from whole-body, forelimb-, or hindlimb-based measurements and whether values are expressed as absolute or base of support (BOS)-normalised parameters. The aims were to examine the associations between whole-body, forelimb-, and hindlimb-based COP measurements in standing dogs and to evaluate the effect of BOS normalisation on agreement between these measurement approaches. Static posturography was performed in clinically healthy adult dogs using a pressure measurement plate. Craniocaudal displacement, mediolateral displacement, support surface, statokinesiogram length, and average speed were analysed using whole-body, forelimb-, and hindlimb-based COP calculations and BOS-normalised parameters. Associations between measurement approaches were assessed using Spearman's rank correlation coefficients, while agreement was evaluated using Bland-Altman analyses. Positive correlations were observed between measurement approaches, although their strength and significance varied among COP parameters. Mediolateral displacement demonstrated stronger associations between measurement approaches than craniocaudal displacement, while statokinesiogram length, average speed, and support surface showed moderate to strong correlations across most comparisons. However, Bland-Altman analyses revealed systematic differences and, for several parameters, magnitude-dependent patterns, indicating limited agreement despite significant correlations. BOS normalisation altered the magnitude of several COP parameters but did not consistently improve agreement or eliminate systematic differences between measurement approaches. These findings demonstrate that whole-body, forelimb-, and hindlimb-based COP measurements quantify related but not interchangeable aspects of postural stability. Consequently, measurement approach should be considered when interpreting stabilographic outcomes and comparing results across studies.