Cardiac output (CO) is a principal determinant of perfusion in many critically ill patients. The objectives of this study were to determine whether physicians' estimates of CO, or cardiac index (CI), are accurate compared with CO/CI measured by esophageal doppler, and to estimate the physician time necessary for Emergency Department (ED) CO/CI measurement. We prospectively evaluated a convenience sample of critically ill, adult ED patients. Based on all available clinical information, residents and emergency medicine attendings estimated patients' CI as being high, normal, or low. A blinded investigator measured CO/CI using an esophageal doppler probe. Times to achieve optimal doppler signal were recorded. Agreement between physician CI estimates and measured CI values was assessed using the weighted kappa statistic. Thirty-three patients were evaluated. There was no agreement beyond chance between physicians' estimates of CI and measured CI. The mean time for optimal doppler signal was 5.7+/-4.3 min. Physicians' estimates of CI were inaccurate compared with measured CI. Esophageal doppler measurement of CO/CI appears to be practical from a physician time standpoint.