Accuracy of impedance cardiography for evaluating trends in cardiac output: a comparison with oesophageal Doppler

Background. Impedance cardiography (ICG) enables continuous, beat-by-beat, non-invasive, operator-independent, and inexpensive cardiac output (CO) monitoring. We compared CO values and variations obtained by ICG (Niccomo (TM), Medis) and oesophageal Doppler monitoring (ODM) (CardioQ (TM), Deltex Medical) in surgical patients. Methods. This prospective, observational, single-centre study included 32 subjects undergoing surgery with general anaesthesia. CO was measured simultaneously with ICG and ODM before and after events likely to modify CO (vasopressor administration and volume expansion). One hundred and twenty pairs of CO measurements and 94 pairs of CO variation measurements were recorded. Results. The CO variations measured by ICG correlated with those measured by ODM [r = 0.88 (0.82-0.94), P<0.001]. Trending ability was good for a four-quadrant plot analysis with exclusion of the central zone (<10%) [95% confidence interval (CI) for concordance (0.86; 1.00)]. Moderate to good trending ability was observed with a polar plot analysis (angular bias: -7.2 degrees; 95% CI -12.3 degrees; -2.5 degrees; with radial limits of agreement -38 degrees; 24 degrees). After excluding subjects with chronic obstructive pulmonary disease, a Bland-Altman plot showed a mean bias of 0.47 litre min(-1), limits of agreements between -1.24 and 2.11 litre min(-1), and a percentage error of 35%. Conclusion. ICG appears to be a reliable method for the non-invasive monitoring of CO in patients undergoing general surgery.