In 25 cardiac surgical patients, right ventricular ejection fraction was continuously measured with a new pulmonary artery catheter and transoesophageal echocardiography, scanning the 'fractional area change' in a standardised transatrial cross section area. Measurements were recorded at three predefined time points (pre-, intra-, and postoperatively). Both methods were compared using the Bland-Altman analysis. Comparing right ventricular ejection fraction values obtained from the pulmonary artery catheter with those assessed by transoesophageal echocardiography, bias was -3.7%, with a precision of 30.9%. Bias and precision significantly improved when the heart rate was less than 100 beats.min(-1), pulmonary artery pressures were low and cardiac performance adequate. In conclusion, the new continuous pulmonary artery catheter system appears to be a valid and useful bedside monitoring device in the haemodynamic management of critically ill patients.