An external spirometric method using a differential linear transformer was used to measure tidal volume (VT) and to determine factors influencing CO2elimination and HFJV-induced “PEEP effect” in 15 critically ill patients under HFJV. VTincreased with increasing driving pressure (DP) and decreasing frequency (f) and was influenced little by changes in I/E ratio. CO2elimination, as reflected by the measurement of PaCO2, was mainly influenced by the absolute level of VTrather than by the product VTX frequency (PaCO2= 5715/VT, r = 0.75,P< 0.05). The primary phenomenon explaining HFJV-induced “PEEP effect” was intrapulmonary gas trapping due to incomplete exhalation of the first VTadministered: the spontaneous relaxation times of these first VTwere longer than expiratory time allotted to the ventilatory settings. HFJV-induced “PEEP effect” increased with I/E ratio, DP, and f and was markedly influenced by the mechanical properties of the total respiratory system. At given ventilatory settings, HFJV-induced “PEEP effect” was greater in patients with a normal or elevated time constant of the total respiratory system (τ RS) than in patients with a low τRS. These results suggest that HFJV should not be used in patients with chronic obstructive pulmonary disease and asthma, and should be preferentially administered to patients having stiff lungs or decreased chest wall compliance.