Introduction: Monitoring cardiac output in the paediatric cardiac intensive care unit is difficult. Monitoring techniques are often too invasive, not precise enough or can only be performed sporadically and under adverse conditions. The use of bioimpedance as a non invasive technique has shown contradictory results. New mathematical algorithms for bioimpedance seem to be promising. The aim of this study was to compare a recent bioimpedance technique against echocardiographical findings.

Methods: In newborns (n = 20) after switch operation for TGA, stroke volume (SV) and cardiac output (CO) were measured with the AESCULON® (Osypka) bioimpedance monitor. The measurements (n = 240) were accompanied by echocardiographical measuring of SV and CO across the aortic valve (Vivid 7, GE). The patients were monitored during their whole stay in the ICU under typical post surgical conditions, including general anaesthesia, artificial ventilation, temporary cardiac stimulation and inotropic therapy. Bioimpedance and echocardiographical findings were compared by Bland-Altman, Pearson’s, linear regression and Rank-Sum tests.

Results: For SV, good correlation (r = 0.86) could be found between bioimpedance (median 3.9 ml) and echocardiography (median 3.2 ml) as well as for CO (r = 0.84, median 0.61 l/min versus 0.50 l/min,
p < 0.05). Bias was – 0.46 for the stroke volume and – 0.07 for the cardiac output measures.

Conclusions: New bioimpedance algorithms seem to reflect cardiac output correctly and with good correlation to “classical” techniques. They thus can open a perspective to a continuous and reliable monitoring of critically ill patients in the paediatric cardiac ICU.