Background. The purpose of this study was to evaluate the possibility of measuring cardiac output during positron emission tomography (PET) examination of myocardial perfusion with nitrogen 13 ammonia. Methods and results. In 7 patients undergoing right-sided cardiac catheterization for evaluation of heart failure and 6 patients who had undergone heart transplantation, a thermodilution catheter for continuous measurement of cardiac output was inserted. An N-13 ammonia scan of the heart was subsequently performed, and with use of factor analysis, the time-activity curve from the right ventricle was derived from the dynamic image sequence. The PET-derived cardiac output was subsequently obtained according to the Stewart-Hamilton principle as the amount of injected tracer divided by the area under the time-activity curve. PET-acquired cardiac output measurements correlated closely with the invasively determined values for a wide range of cardiac output values (P < .0001). The mean difference was 0.12 L/min, with an SD of 0.74 L/min. The interobserver variation was low, with a mean difference of 0.06 L/min and an SD of 0.46 L/min. Conclusions. Cardiac output determination with N-13 ammonia and PET appears to be both accurate and precise and can be performed simultaneously with measurement of myocardial perfusion.