Effect of prior metabolic rate on the kinetics of oxygen uptake during moderate-intensity exercise.

Pulmonary oxygen uptake (VO2) dynamics during moderate-intensity exercise are often assumed to be dynamically linear (i.e. neither the gain nor the time constant (tau) of the response varies as a function of work rate). However, faster, slower and unchanged VO2 kinetics have been reported during work-to-work transitions compared to rest-to-work transitions, all within the moderate-intensity domain. In an attempt to resolve these discrepancies and to improve the confidence of the parameter estimation, we determined the VO2 response dynamics using the averaged response to repeated exercise bouts in seven healthy male volunteers. Each subject initially performed a ramp-incremental exercise test for the estimation of the lactate threshold (thetaL). They then performed an average of four repetitions of each of three constant-work-rate (WR) tests: (1) between 20 W and a work rate of 50% (WR50) between 20 W and 90% thetaL (step 1-->2), (2) between WR50 and 90% thetaL (step 2-->3), and (3) between 20 W and 90% thetaL (step 1-->3); 6 min was spent at each work rate increment and decrement. Parameters of the kinetic response of phase II VO2 were established by non-linear least-squares fitting techniques. The kinetics of VO2 were significantly slower at the upper reaches of the moderate-intensity domain (step 2-->3) compared to steps 1-->2 and 1-->3 [group mean (SD) phase II tau: step 1-->2 25.3 (4.9) s, step 2-->3 40.0 (7.4) s and step 1-->3 32.2 (6.9) s]. The off-transient values of tau were not significantly different from each other: 36.8 (16.3) s, 38.9 (11.6) s and 30.8 (5.7) s for steps 1-->2, 2-->3 and 1-->3, respectively. Surprisingly, the on-transient gain (G, deltaVO2/deltaWR) was also found to vary among the three steps [G = 10.56 (0.42) ml x min(-1) W(-1) 11.85 (0.64) ml x min(-1) W(-1) and 11.23 (0.52) ml x min(-1) W(-1) for steps 1-->2, 2-->3 and 1-->3, respectively]; the off-transient G did not vary significantly and was close to that for the on-transient step 1-->3 in all cases. Our results do not support a dynamically linear system model of VO2 during cycle ergometer exercise even in the moderate-intensity domain. The greater oxygen deficit per unit power increment in the higher reaches of the moderate-intensity domain necessitates a greater transient lactate contribution to the energy transfer, or a greater phosphocreatine breakdown, or possibly both.