Haemodynamic changes during induced hypotension depend upon the hypotensive agent used. We investigated if, using transoesophageal echocardiography (TEE), we could identify the haemodynamic differences between trimethaphan and prostaglandin E1. Twenty-nine patients undergoing total hip replacement were selected for study. Hypotension was induced to a mean arterial pressure of 8.0-9.3 kPa with either trimethaphan (5-20 micrograms.kg-1.min-1) or prostaglandin E1 (0.5-2.0 micrograms.kg-1.min-1). The left atrial dimension, cardiac output, fractional shortening, pulmonary venous flow and mitral valve flow were evaluated using TEE. During induced hypotension, left atrial dimension decreased in both trimethaphan and prostaglandin E1 groups (P < 0.05). In the trimethaphan-treated patients systolic velocity in pulmonary venous flow decreased from 41.9 +/- 4.8 cm.sec-1 before induced hypotension to 27.8 +/- 4.2 cm.sec-1 by 30 min after stable hypotension had been established (P < 0.01). The late/early ratio of peak velocity in mitral blood flow decreased in prostaglandin E1 treated patients. Cardiac output increased from 4.2 +/- 0.5 L.min-1 to 5.3 +/- 0.4 L.min-1 during 30 min hypotension with prostaglandin E1 administration (P < 0.05), but cardiac output decreased from 5.0 +/- 0.5 to 3.5 +/- 0.4 L.min-1 with trimethaphan (P < 0.01). The differences in haemodynamic variables could be attributed to the venule dilatation effect of trimethaphan. We conclude that it was possible to detect the haemodynamic differences between trimethephan and prostaglandin E1 using TEE.