Evaluation of cardiac output by thoracic electrical bioimpedance during exercise in normal subjects

We compared cardiac output determined simultaneously by two methods, the CO2 rebreathing technique and the thoracic electrical bioimpedance method (Bomed NCCOM-3 equipment). The studies were performed in duplicate in 11 healthy male subjects at rest and during three levels of steady-state exercise on a cycle ergometer at 60, 120, and 180 W. Cardiac output at 60 and 120 W was slightly lower (p less than 0.01) by the thoracic impedance method (12.2 +/- SE 2.2 and 15.7 +/- SE 3.5 L/min, respectively) than by the CO2 rebreathing method (14.0 +/- SE 2.1 and 17.9 +/- SE 3.0 L/min, respectively), suggesting a systematic bias between the two methods of measurement. However, if allowance is made for that bias, there would be acceptable agreement between the two methods at 60 and 120 W. Although the results were not significantly different between the two methods at rest and at 180 W, there was no acceptable agreement between the two methods probably because the CO2 rebreathing method at rest was more liable to show error due to the small arteriovenous CO2 difference, while the impedance method was less reliable at 180 W. Cardiac output by both methods correlated with O2 consumption, with the correlation being higher for cardiac output by the rebreathing method (r = 0.94) than for thoracic impedance (r = 0.88). The results suggest that the thoracic electrical bioimpedance method can be used for determination of cardiac output during mild or moderate levels of exercise in normal subjects.