The 1990s have witnessed major advances in impedance cardiography tech
nology. Problems existed with the methods used to calculate cardiac ou
tput. Excessive lung fluid, as often found in critically ill patients,
may also invalidate measurements. The signal processing and measureme
nt techniques used in older systems were deficient. The newer systems,
of which there are at least six, incorporate novel and improved signa
l processing techniques. They also offer analog visual displays, perso
nal computer interfacing, sophisticated analytical software and haemod
ynamic patient management systems. Evaluation of these systems is diff
icult because no true 'gold standard' method of cardiac output measure
ment exists. When compared with thermodilution techniques, limits of a
greement of +/- 20-30% seem acceptable. These Limits can be achieved i
n normal subjects but not in critically ill patients. Validation data
are available for only half of the new systems. Until. recently, the m
ain application for impedance cardiography has been research but impro
ved accuracy should lead to increased clinical usage.