Hemodialysis patients experience a variety of hemodynamic abnormalities that contribute to cardiovascular disease mortality which is the leading cause of death in these patients. Impedance cardiography has been utilized in order to monitor cardiac hemodynamics with lower cost and inconvenience, but it has not been appropriately validated in the hemodialysis population. We repeatedly used impedance cardiography to assess short- (48 hours) and long-term (15 days) reproducibility of cardiac output measurements and we compared baseline impedance cardiography measurements with echocardiographic measurements. We studied 109 stable hemodialysis patients, aged 59.70 +/- 11.97 years being on hemodialysis for 67.59 +/- 40.15 months, on a non-dialysis day. Cardiac output was obtained with the BioZ impedance cardiography system (Cardiodynamics, San Diego, Ca, USA). Baseline echocardiography was performed using a Hewlett-Packard Sonos 2500 (Andover, Mass., USA). The values of impedance cardiography derived cardiac output were 5.28 +/- 0.79, 5.27 +/- 0.75 and 5.25 +/- 0.74 l/min at baseline (107 patients), 48 hours (107 patients) and 15 days (98 patients) respectively, showing high reproducibility. Bland and Altman analysis estimated that bias at 48 hours and at 15 days were: -0.013 (95% confidence intervals = -0.045 to 0.019) and 0.028, (95% confidence intervals = -0.044 to 0.101), respectively. In addition baseline impedance cardiography derived cardiac output was significantly correlated with the echocardiographic derived cardiac output (r = 0.9, p < 0.0001). Impedance cardiography is a simple non invasive technique for cardiac output estimation in hemodialysis patients which has high reproducibility when performed under controlled conditions, and is closely correlated with echocardiographic measurements of cardiac output.