OBJECTIVE To assess the oxygen cost of breathing with either pressure-support ventilation (PSV) or biphasic intermittent positive airway pressure ventilation (BIPAP). METHODS Prospective, randomized, crossover study. METHODS Medical intensive care unit of a university hospital. METHODS Twenty clinically stable and spontaneously breathing patients after long-term mechanical ventilation. METHODS Patients were randomized to start on either PSV or BIPAP, and measurements were performed after an adaptation period of 30 mins. Immediately after, the ventilatory mode was changed and after another 30-min adaptation period, the same measurements were performed. RESULTS Indirect calorimetry was performed during each ventilatory mode for a period of 30 mins. Oxygen consumption, energy expenditure, CO2 production, and respiratory quotient did not differ significantly between the two ventilatory modes, regardless of the patients' randomization. There were no statistically significant differences with regard to respiratory rate, minute volume, and blood gas analysis. All patients tolerated both ventilatory modes without any signs of discomfort. CONCLUSIONS Pressure support ventilation and BIPAP are both used for weaning patients gradually from the ventilator. BIPAP may be advantageous in patients not breathing sufficiently with PSV, since no patient effort is necessary with use of this ventilatory mode.