The haemodynamic changes of the prone position were investigated in 40 ASA I-II patients undergoing lumbar spine surgery. Patients were randomly assigned, following propofol intravenous induction, to receive maintenance of anaesthesia using either isoflurane 1-1.2% in air or target controlled propofol 3 microg.ml(-1) infusion. Measurements of non-invasive blood pressure, heart rate and cardiac output were made in the supine position. The patient was then turned prone onto a Montreal pattern mattress and measurements repeated. Cardiac output measurements were made using a non-invasive cardiac output monitor. We found a significant reduction in cardiac index in both groups and a significantly greater change with propofol compared to isoflurane on turning supine to prone (CI change 0.4 vs 0.7 l.min(-1).m(-2) p = 0.001 and SVRI change 89 vs 177 dyne.s(-1).cm(-5), p = 0.041). We conclude that turning healthy patients prone produces a clinically significant reduction in cardiac output, the change being greater during maintenance of anaesthesia using propofol compared to isoflurane.