This study was a prospective, randomized design to compare oxygenation and pulmonary hemodynamics between inhaled nitric oxide (NO) and inhaled prostacylcin (PGI2), and between inhaled and i.v. PGI2in acute respiratory failure with pulmonary hypertension. Acute respiratory failure with pulmonary hypertension was induced in 12 piglets weighing 9–12 kg by repeated lung lavages and a continuous infusion of the stable endoperoxane analogue of thromboxane. Thereafter the animals were randomly assigned either for NO or PGI2application. All animals were treated with different concentrations of NO or different doses of PGI2applied i.v. and inhaled in random order. Continuous monitoring included ECG, central venous pressure (CVP), mean pulmonary artery pressure (MPAP), mean arterial pressure (MAP), artertial oxygen saturation (SaO2), and mixed venous oxygen saturation (SvO2) measurements. NO inhalation of 10 ppm resulted in a significant increase in Pao2/fraction of inspired oxygen (FiO2) from 7.8 ± 1.34 kPa to 46.1 ± 9.7 kPa. MPAP decreased significantly from 5.1 ± 0.26 kPa to 3.7 ± 0.26 kPa during inhaled NO of 40 ppm; i.v. infusion of PGI2slightly increased oxygenation parameters. A significant increase in Pao2/FiO2up to 32.4 ± 3.1kPa was observed during PGI2aerosol delivery (p< 0.01); i.v. PGI2decreased MAP from 11.5 ± 0.39 kPa to 9.8 ± 0.66 kPa (p< 0.05) and MPAP from 5.8 ± 0.53 kPa to 4.5 ± 0.66 kPa, respectively (p< 0.05). PGI2aerosol delivery significantly decreased the MPAP to 3.7 ± 0.53 kPa (p< 0.05) without influencing the MAP. It was concluded that inhaled NO and inhaled PGI2act as selective pulmonary vasodilators in acute respiratory failure with pulmonary hypertension resulting in improved oxygenation mainly due to improved mismatch of pulmonary perfusion and ventilation. Intravenous PGI2improves oxygenation and pulmonary hemodynamics to a lesser extent than aerosolized PGI2and has the risk of systemic hypotension at a higher dose.