Dynamic Characteristics of Ventilatory and Gas Exchange during Sinusoidal Walking in Humans

Abstract

Our present study investigated whether the ventilatory and gas exchange responses show different dynamics in response to sinusoidal change in cycle work rate or walking speed even if the metabolic demand was equivalent in both types of exercise. Locomotive parameters (stride length and step frequency), breath-by-breath ventilation (V̇E) and gas exchange (CO2 output (V̇CO2) and O2 uptake (V̇O2)) responses were measured in 10 healthy young participants. The speed of the treadmill was sinusoidally changed between 3 km·h-1 and 6 km·h-1 with various periods (from 10 to 1 min). The amplitude of locomotive parameters against sinusoidal variation showed a constant gain with a small phase shift, being independent of the oscillation periods. In marked contrast, when the periods of the speed oscillations were shortened, the amplitude of V̇E decreased sharply whereas the phase shift of V̇E increased. In comparing walking and cycling at the equivalent metabolic demand, the amplitude of V̇E during sinusoidal walking (SW) was significantly greater than that during sinusoidal cycling (SC), and the phase shift became smaller. The steeper slope of linear regression for the V̇E amplitude ratio to V̇CO2 amplitude ratio was observed during SW than SC. These findings suggested that the greater amplitude and smaller phase shift of ventilatory dynamics were not equivalent between SW and SC even if the metabolic demand was equivalent between both exercises. Such phenomenon would be derived from central command in proportion to locomotor muscle recruitment (feedforward) and muscle afferent feedback.

Fig 1. Scheme for setting work rate at the equivalent metabolic demand between walking and cycling.

First, the mean values of V̇O₂ at two constant treadmill speed of 3 km·h^-1 and 6 km·h^-1 (upper panel) should be obtained. Second, the regression line between work rate of cycling and V̇O₂ was calculated, we selected three step work rates of cycling between 20 to 100 watts. At the speeds of 3 km·h^-1 and 6 km·h^-1, corresponding work rates were calculated from this regression line (lower panel).

Fig 2. Time course of the ventilatory and gas exchange responses at different oscillation periods.

Example in a representative subject of the ventilatory and gas exchange variables of ventilation (V̇_E), O₂ uptake (V̇O₂), CO₂ output (V̇CO₂), and end-tidal PCO₂ (P_ETCO₂) responses against four different oscillation periods during SW at T = 1, 2, 5, and 10 min. Oscillating line is the superimposed gas exchange variables data. Smooth line is the sine-wave fundamental component of these dynamics.

Fig 3. Comparisons of A and PS of ventilatory and gas exchange dynamics between SW and SC.

Comparisons of A and PS between SW (solid line) and SC (dotted line) as a function of the periods of the sinusoidal changes in the treadmill speed and the cycling work rate at the equivalent metabolic demand. #,##; p < 0.05, 0.01 vs. cycling. Data are shown in mean ± SE.

Fig 4. Time course of the locomotive responses at different time periods.

Example of a representative subject of the locomotion pattern divided between the step frequency (left panels) and the stride length (right panels) during SW for all periods from 1 min to 10 min (sinusoidal speed change was from 3 km·h⁻¹ to 6 km·h⁻¹). Oscillating line is the superimposed step frequency and stride length. Smooth line is the sine-wave fundamental component of the dynamics (A). The A of the step frequency and stride length remained almost unchanged (B). The PS for the step frequency and stride length were quite small and virtually close to zero at each period (C). * p < 0.05 vs. T:1min, ○ p < 0.05 vs. T:2min. Data are shown in mean ± SE.

Fig 5. Relationship between the A ratio of locomotive and gas exchange parameters during SW and SC.

The A ratio of the V̇_E against V̇O₂ (A), V̇CO₂ (B), and P_ETCO₂ (C) during SW and SC, respectively. Locomotive responses (step frequency (D) and stride length (E)) were presented. Note that the A ratio for the V̇_E correlated to the A ratio of the V̇CO₂ closer during SW than SC. In contrast, the A ratio for P_ETCO₂ was not tightly related to the V̇_E dynamics. The slope of the regression lines of the V̇_E-V̇CO₂ relationship was steeper during SW than SC (p < 0.05). SL; y = 1.160x + 0.028, r = 0.722 (p < 0.01), SC; y = 0.804x + 0.031, r = 0.782 (p < 0.01).