Noninvasive cardiac output (CO) measurement can be useful in many clinical settings where invasive monitoring is not desired. Bioimpedance (intrabeat measurement of changes in transthoracic voltage amplitude in response to an injected high-frequency current) has been explored for this purpose but is limited in some clinical settings because of inherently low signal-to-noise ratio. Since changes in fluid content also induce changes in thoracic capacitive and inductive properties, we tested whether a noninvasive CO measurement could be obtained through measurement of the relative phase shift of an injected current (i.e., bioreactance). We constructed a prototype device that applies a 75-kHz current and determines the relative phase shift (dΦ/d t) of the recorded transthoracic voltage. CO was related to the product of peak dΦ/d t, heart rate, and ventricular ejection time. The preclinical study was done in nine open-chest pigs put on right heart bypass so that CO could be varied at known values. This was followed by a feasibility study in 27 postoperative patients who had a Swan-Ganz catheter (SGC). The measurements of noninvasive CO measurement and cardiopulmonary bypass pump correlated to each other ( r = 0.84) despite the large variation in CO and temperatures. Similarly, in patients, mean CO values were 5.18 and 5.17 l/min as measured by SGC and the noninvasive CO measurement system, respectively, and were highly correlated over the range of values studied ( r = 0.90). Preclinical and clinical data demonstrate the feasibility of using blood flow-related phase shifts of transthoracic electric signals to perform noninvasive continuous CO monitoring.