Non-invasive measurements of cardiac output in atrial fibrillation: Inert gas rebreathing and impedance cardiography

Abstract

Background. Atrial fibrillation (AF) is associated with significant morbidity and mortality. To test the effect of interventions, knowledge of cardiac output (CO) is important. However, the irregular heart rate might cause some methods for determination of CO to have inherent weaknesses. Objective. To assess the validity of these methods in AF, a new inert gas rebreathing device and impedance cardiography was tested with echocardiography as reference. Methods. Using a cross-sectional design, 127 patients with AF and 24 in SR were consecutively recruited. Resting CO was measured using inert gas rebreathing (n = 62) or impedance measurement of intrathoracic blood flow (n = 89) in separate studies with echocardiographic measurement as reference. Results. CO determined with impedance cardiography was mean 4.77 L/min ± 2.24(SD) compared to 4.93 L/min ± 1.17 by echocardiography (n = 89, n.s.) in patients with AF. CO by inert gas rebreathing was 4.98 L/min ± 2.49(SD) compared to 5.70 L/min ± 2.49 by echocardiography (n = 62, n.s.) in patients with AF and SR (AF 5.42 ± 2.9 vs. 6.27, n.s. and SR 4.09 ± 1.08 vs. 4.35 ± 0.86, n.s.). Mean bias between impedance cardiography and echocardiography was 0.14 ± 0.95 L/min and −0.13 ± 0.98 L/min between inert gas rebreathing and echocardiography. Inert gas rebreathing showed larger intra-patient variation than impedance cardiography (0.11 vs. 0.054). Correlation between inert gas rebreathing and echocardiography was r = −0.060 and between impedance cardiography and echocardiography was r = 0.128. Impedance cardiography and inert gas rebreathing both underestimated CO compared to echocardiography. Conclusion. Variation between the inert gas rebreathing and the reference method for AF patients was less than desired. Impedance cardiography was superior to inert gas rebreathing and showed acceptable agreement with echocardiography and variability similar to echocardiography.

Key Words::

• Atrial fibrillation
• cardiac output
• echocardiography
• cardiography
• impedance

Acknowledgements

We wish to acknowledge the valuable assistance of Bo Rasmussen, MD and Ulrik Dixen, MD, PhD who performed echocardiographic studies on many of the patients. We thank biostatistician Steen Ladelund for evaluating the statistics.

The study was supported by grants from Hvidovre Hospital Research Foundation, Danish Medical Association Research Foundation and the Lundbeck Foundation.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.