Identification of orthostatic postural control strategies using the three-dimensional statokinesigram in young and elderly healthy subjects

Abstract

Introduction: The most common measures of postural sway use a force platform to quantify the displacement of the center of pressure in anteroposterior and mediolateral directions. Few studies have been performed using the three-dimensional statokinesigram. However, the association between the spatial distribution and sensory conditions of the stabilometric test in different age groups remains unclear. Objective: The objective of this study was to propose a method for identification of the spatial distribution of the center of pressure during quiet standing by the three-dimensional statokinesigram and test the association between these patterns and sensory information in a sample of young and elderly healthy volunteers. Methods: A quantitative method for generation, processing and analysis of the three-dimensional statokinesigram was proposed basead on a nonparametric analysis by percentil sweep of the bidimensional histogram. Spatial patterns of the center of pressure were inferred from known postural control strategies. Thirty-five young healthy volunteers (28 women; 27.0±7.7 years; 64.3±13.4 kg; 1.64±0.09 m; 24.0±4.4 kg/m2) and 38 elderly healhy subjects (36 women; 67.3±8.7 years; 70.0±14.0 kg; 1.55±0.07 m; 29.3±5.3 kg/m2) were instructed to keep a quiet orthostatic position during all trials with either eyes opened or closed and feet together or apart. Other 11 healthy young volunteers (5 women, 26.0±7.7 years, 74.9±21.0 kg, 1.71±0.10 m, 25.4±5.8 kg/m2) performed 4 tests lasting 60s each, under conditions of open and closed eyes and on the same positioning of the feet. Results: Both groups presented a moderate significant association (young: Goodman-Kruskal γ=0.531, P<0.031, γ*2≈15%; elderly: γ=0.582, P<0.027, γ*2≈19%) between the frequency distribution of spatial patterns and sensory conditions of the trials. All tested stabiloletric parameters were significantly different (P<0.05) when grouped by the spatial patterns. Post-hoc tests indicated that significant higher values were observed for some parameters of those subjects presenting with double-centered pattern compared to those with single-centered ones, while almost all parameters from the multi-centered pattern were significantly higher (P<0.01) compared to both single-centered and double-centered patterns. Robustness analysis revealed that the clinical results is almost invariable to the tested combinations of kernel estimator (mean and median), kernel size (3x3 and 5x5 bins), and interpolation method (bicubic and bilinear). Test-retest analyses showed no statistically significant differences (P>0.05) for the variables of the stabilograms, statokinesigram, and the amount of regions identified in the three-dimensional statokinesigram. Intraclass correlation coefficient values for all parameters analyzed showed statistical significance (P<0.05) and were found in range (minimum, maximum) 0.565 to 0.985. Conclusion: The proposed method allowed the objective evaluation of spatial patterns on the three-dimensional statokinesigram and it was robust to changes in parameters related its processing. The method was sensitive to the association between the spatial patterns of the center of pressure and the available sensory conditions of the trials in both young and elderly, healthy subjects.