Data are lacking regarding age-related modifications of phase I (PhI) of pulmonary Vo(2) on-kinetics during moderate-intensity exercise. We studied three groups (aged 20-30, 40-50, and 60-70 years) of 10 normal subjects, who underwent one incremental and four below-gas exchange threshold constant-power cardiopulmonary exercise tests. Data from constant-power tests were time-aligned and averaged, and the PhI-phase II transition (PhI-IItr) determined when a sharp decrease from baseline of respiratory exchange ratio occurred. The Vo(2) phase II time constant (tau) was obtained by an exponential fitting starting 1) from PhI-IItr ("experimental" fitting strategy) and 2) after 20 s from exercise onset ("fixed-duration" fitting strategy). Assuming estimated arterial-venous O(2) concentration difference not to change with respect to resting value, cardiac output (CO) values at rest and PhI-IItr were obtained according to Fick's principle. Average pulmonary flow acceleration (AFA) during PhI was calculated as the ratio between CO increase during PhI and PhI duration. PhI duration was related to age (r = 0.74, P < 0.0001), increasing from 21 +/- 3 s to 27 +/- 3 s to 32 +/- 4 s in the 20-30, 40-50, and 60-70 age groups, respectively, and to AFA (r = -0.60, P < 0.001), but not to CO increase during PhI. With respect to the experimental fitting strategy, the fixed-duration strategy overestimated Vo(2) phase II tau the more the higher the subject's age, with a lower goodness of fit in the 60-70 group (SE 0.035 vs. 0.056, P < 0.01). In conclusion, PhI duration is related to age in healthy male humans and is linked to CO acceleration-rather than to increase-during PhI. A significant overestimation of phase II tau thus may occur in healthy elderly subjects and patients with a pathologically induced longer PhI duration when fitting data where the PhI-PhIItr was not experimentally determined but assumed to be a set value (i.e., 20 s).