ObjectivesTo analyze the impact of the position of the thoracic external electrodes on the values of cardiac output measured by electrical bioimpedance and to compare the results obtained by bioimpedance with those values determined by thermodilution in critically ill patients. DesignOpen, prospective, comparative trial. SettingICU of a teaching hospital. PatientsTwenty healthy volunteers and ten critically ill patients. InterventionsMeasurements of cardiac output by bioimpedance at rest and after physical activity in normal volunteers and after changing the neck or xiphoid electrodes. Comparisons of cardiac output obtained by thermodilution and bioimpedance with internal and external electrodes in patients. Measurements and Main ResultsMean ± SD values are presented. Cardiac output values at rest and after exercise were 6.7 ± 1.3 and 10.8 ± 2.6 L/min at rest and after exercise, respectively (p < .001). Displacement of the xiphoid electrodes 3 cm in the caudal direction was accompanied by a decrease of the mean cardiac output from 7.1 ± 1.2 to 5.8 ± 1.3 L/min (p < .001) and displacement 3 and 6 cm cranially was accompanied by increases in cardiac output from 7.1 ± 1.2 to 8.1 ± 1.4 L/min (p < .001) and 8.6 ± 1.5 L/min (p < .001), respectively. In the ten patients, cardiac output measurements were virtually identical when results obtained by thermodilution (6.7 ± 3.1 L/ min) were compared with those results obtained by bioimpedance using internal esophageal (6.6 ± 3.1 L/min), but not external (4.7 ± 1.6 L/min) electrodes. Conclusionsa) The values of cardiac output derived from measurements obtained by bioimpedance using internal electrodes were comparable with those values derived from thermodilution. b) Values of cardiac output from bioimpedance studies with external electrodes were dependent on the position of the xiphoid electrodes. (Crit Care Med 1992; 20:62)