Incongruity of Geometric and Spectral Markers in the Assessment of Body Sway

Abstract

Different measurements of body oscillations in the time or frequency domain are being employed as markers of gait and balance abnormalities. This study investigates basic relationships within and between geometric and spectral measures in a population of young adult subjects. Twenty healthy subjects stood with parallel feet on a force platform with and without a foam pad. Adaptation effects to prolonged stance were assessed by comparing the first and last of a series of eight successive trials. Centre of Foot Pressure (CoP) excursions were recorded with Eyes Closed (EC) and Open (EO) for 90s. Geometric measures (Sway Area, Path Length), standard deviation (SD) of the excursions, and spectral measure (mean power Spectrum Level and Median Frequency), along the medio-lateral (ML) and antero-posterior (AP) direction were computed. Sway Area was more strongly associated than Path Length with CoP SD and, consequently, with mean Spectrum Level for both ML and AP, and both visual and surface conditions. The squared-SD directly specified the mean power Spectrum Level of CoP excursions (ML and AP) in all conditions. Median Frequency was hardly related to Spectrum Level. Adaptation had a confounding effect, whereby equal values of Sway Area, Path Length, and Spectrum Level corresponded to different Median Frequency values. Mean Spectrum Level and SDs of the time series of CoP ML and AP excursions convey the same meaning and bear an acceptable correspondence with Sway Area values. Shifts in Median Frequency values represent important indications of neuromuscular control of stance and of the effects of vision, support conditions, and adaptation. The Romberg Quotient EC/EO for a given variable is contingent on the compliance of the base of support and adaptation, and different between Sway Area and Path Length, but similar between Sway Area and Spectrum Level (AP and ML). These measures must be taken with caution in clinical studies, and considered together in order to get a reliable indication of overall body sway, of modifications by sensory and standing condition, and of changes with ageing, medical conditions and rehabilitation treatment. However, distinct measures shed light on the discrete mechanisms and complex processes underpinning the maintenance of stance.

Keywords: CoP excursion; Path Length; Sway Area; adaptation; frequency spectrum; stance; support surface; vision.

Figure 1

CoP excursion and power spectrum in a representative subject (top panels, Foam; bottom panels, Solid). The CoP excursions along the ML (red traces) and AP (blue traces) directions are reported for the four conditions of interest [(A), EC Foam; (C), EO Foam; (G), EC Solid; (I), EO Solid]. The CoP excursions in the horizontal plane and the 95% prediction ellipses are shown for EC (B,H) and EO (D,J) conditions, both with Foam (B,D) and Solid (H,J) supports. The corresponding power spectra are reported in (E,F,K,L) for both ML (red) and AP (blue) directions.

Figure 2

Mean Sway Area (A) and Path Length (B). Sway Area and Path Length are greater in Foam (EC, red; EO, green) than in Solid support condition (EC, yellow; EO, blue). Vision has a greater effect with Foam than Solid support for both Sway Area and Path Length. (C) the relationship between Sway Area and Path Length is shown for all visual and support conditions. Each dot corresponds to a subject. Asterisks indicate significant differences (**p < 0.01, ***p < 0.001).

Figure 3

CoP SD (top panels) and mean Spectrum Level (bottom panels). The mean CoP SDs across subjects are reported for the ML (A) and AP (B) directions for all visual (EC, red and yellow; EO green and blue) and support (Foam, red and green; Solid, yellow and blue) conditions. The mean Spectrum Level is reported for both ML (D) and AP (E) directions [same colour code as in (A,B)]. The CoP SD and mean Spectrum Level are greater with Foam than Solid support. Vision removal affects SD and Spectrum Level mainly with Foam support (red and yellow). (C) shows the relationship between AP and ML SDs. Both CoP SDs increased with a broadly similar pattern from the more- to the less-stable conditions. (F) shows the relationship between AP and ML mean Spectrum Level. Much as for the SDs, there was a proportionality between AP and ML Spectrum Level. Each dot in C and F corresponds to a subject. Asterisks indicate significant differences (*, p < 0.05; ***, p < 0.001).

Figure 4

Relationship between CoP variance (SD²) and Spectrum Level. The values of the CoP SD² are plotted against the corresponding values of the mean Spectrum Level for both ML (A) and AP (B) directions and both visual and surface conditions (EC Foam, red; EC Solid, yellow; EO Foam, green; EO Solid, blue). Each dot corresponds to a subject. The slope of the lines best fitting the relationship between CoP SD² and mean Spectrum Level is identical between visual and support conditions and between ML and AP directions.

Figure 5

Relationship between geometric measures and variance of the CoP excursions. Sway Area (A,B) and Path Length (C,D) are plotted against the CoP variance (SD²) for both ML (A,C) and AP directions (B,D) for all visual and support conditions. Each dot corresponds to a subject. The association with CoPSD² is stronger for Sway Area than for Path Length.

Figure 6

Median Frequency and Spectrum Level. The mean Median Frequency (average of all subjects) is reported for both ML (A) and AP (B) directions, both visual and both support conditions (red, EC Foam; green, EO Foam; yellow, EC Solid; blue, EO Solid). With Foam, the Median Frequency is higher than with Solid support for both ML and AP directions. With EO, the Median Frequency is smaller than that with EC, especially with Foam, for both ML and AP directions. The relationship between mean Spectrum Level and Median Frequency is reported in the lower panels for both ML (C) and AP (D) directions, both visual and both support conditions. Each dot corresponds to a subject. The Spectrum Level is not related to the value of the Median Frequency, both in ML and AP directions and with both visual and support conditions. Asterisks indicate significant differences (***p < 0.001).

Figure 7

Effect of adaptation on mean Spectrum Level (upper panels) and Median Frequency (lower panels). In the ML direction (A), the mean Spectrum Level is similar between the non-adapted (filled columns) and the adapted trials (empty columns). In the AP direction (B), the mean Spectrum Level increases with adaptation, except for the EO Solid condition. The Median Frequency (C,D) moves towards smaller values with adaptation, particularly with EC, for both Foam and Solid supports (red and yellow, respectively). Asterisks indicate significant differences (*, p < 0.05; **, p < 0.01; ***, p < 0.001).

Figure 8

Differences in the power spectrum between the non-adapted and adapted trials. (A–D) refer to EC Foam, (E–H) to EC Solid condition. In (A,B,E,F) the mean Spectrum profiles of the non-adapted (black) and adapted trials (red) are reported for ML (A and E) and AP directions (B,F). In (C,D,G,H), the difference (adapted minus non-adapted) between the spectrum profiles of the non-adapted and adapted trials is shown. Red areas indicate a larger amplitude of the spectrum in the adapted than the non-adapted trial, oppositely the blue areas. With adaptation, the Spectrum Level increases at low frequencies (<0.3 Hz) and decreases at high frequencies (>0.3 Hz). This effect is limited to the low frequency range on a Solid support. The ordinates are capped in all the panels.

Figure 9

Sway Area (top panels) and Path Length (bottom panels) vs. Median Frequency for non-adapted (filled symbols) and adapted trials (empty symbols). For each visual condition and support surface, the mean value of Sway Area or Path Length is plotted against the corresponding mean value of the Median Frequency for both ML (A,C) and AP (B,D) directions. Note that the values of Sway Area and Path Length are the same for the left and right panels. Filled symbols refer to the non-adapted, open symbols to the adapted trial. With adaptation, both Path Length and Median Frequency decrease, especially in the EC Foam condition. Sway Area, except for the EC Foam, tended to increase with adaptation, even if not significantly. The Table inset summarises the effect of adaptation on the considered variables (= indicates no significant effect of adaptation; ↓ points to a significant diminution in the value of the variable).

Figure 10

Romberg Quotients (RQ) of geometrical and spectral variables. The top panels show the RQs for the non-adapted (A) and adapted (B) trials and for the Foam (pink) and Solid (grey) support conditions. Asterisks indicate significant differences (*, p < 0.05; **, p < 0.01; ***, p < 0.001). The middle panels show the relationship between RQs of AP Spectrum Level and RQs of Sway Area for both non-adapted (C) and adapted trials (D) (Foam, pink; Solid, grey). Each circle corresponds to a subject. The bottom panels show the relationship between RQs of AP Median Frequency and RQs of Path Length for both non-adapted (E) and adapted trials (F). The apparent proportionality between the RQs calculated on Median Frequency and Path Length is related to the overall shift to the left of the Path Length dots in Solid support condition.