To compare a new transport ventilator to manual ventilation in terms of maintaining the respiratory and hemodynamic levels of critically ill patients. Prospective, randomized, single-center study. ICU in a university hospital. A total of 16 patients (22 transports) who were spontaneously breathing and required ventilatory assistance on excursions from the ICU. For each transport, the patient was randomly assigned to receive either manual ventilation (group M) or mechanical ventilation (group V). For transports in group V, the ventilators were set the same as in the ICU. Respiratory and hemodynamic variables were measured 30 min before transport (T(0)), on arrival at the site of procedure (T(1)), on return to the ICU (T(2)), and 30 min after return the ICU (T(3)). After transport, five patients in group M showed a significant deterioration in PaO(2)/fraction of inspired oxygen ratio, while one patient in group V showed deterioration (p = 0.056). The mean (+/- SD) respiratory rate in group M at T(2) (32 +/- 9 breaths/min) was significantly higher (p < 0.001) than at T(0) (19 +/- 6 breaths/min) and also was higher (p < 0.01) than in group V at T(2) (19 +/- 6 breaths/min). The mean tidal volume and positive end-expiratory pressure in group M at T(2) showed significantly larger variation (p < 0.05 and p < 0.001, respectively) than in group V. The transport ventilator that was recently approved by the US Food and Drug Administration reliably provides more stable ventilatory support than does manual ventilation. Generally, the use of this transport ventilator for intrahospital transport is preferable to manual ventilation.