Validation of an ultrasound dilution technology for cardiac output measurement and shunt detection in infants and children.

Objective: To validate cardiac output measurements by ultrasound dilution technology (COstatus monitor) against those obtained by a transit-time ultrasound technology with a perivascular flow probe and to investigate ultrasound dilution ability to estimate pulmonary to systemic blood flow ratio in children. Design: Prospective observational clinical trial. Setting: Pediatric cardiac operating theater in a university hospital. Material and Methods: In 21 children (6.1 2.6 kg, mean sd) undergoing heart surgery, cardiac output was simultaneously recorded by ultrasound dilution (extracorporeal arteriovenous loop connected to existing arterial and central venous catheters) and a transit-time ultrasound probe applied to the ascending aorta, and when possible, the main pulmonary artery. The pulmonary to systemic blood flow ratio estimated from ultrasound dilution curve analysis was compared with that estimated from transit-time ultrasound technology. Results: Bland-Altman analysis of the whole cohort (90 pairs, before and after surgery) showed a bias between transit-time ultrasound (1.01 +/- 0.47 L/min) and ultrasound dilution technology (1.03 +/- 0.51 L/min) of -0.02 L/min, limits of agreement -0.3 to 0.3 L/min, and percentage error of 31%. In children with no residual shunts, the bias was -0.04 L/min, limits of agreement -0.28 to 0.2 L/min, and percentage error 19%. The pooled co efficient of variation was for the whole cohort 3.5% (transit-time ultrasound) and 6.3% (ultrasound dilution), and in children without shunt, it was 2.9% (transit-time ultrasound) and 4% (ultrasound dilution), respectively. Ultrasound dilution identified the presence of shunts (pulmonary to systemic blood flow 1) with a sensitivity of 100% and a specificity of 92%. Mean pulmonary to systemic blood flow ratio by transit-time ultrasound was 2.6 +/- 1.0 and by ultrasound dilution 2.2 +/- 0.7 (not significant). Conclusion: The COstatus monitor is a reliable technique to measure cardiac output in children with high sensitivity and specificity for detecting the presence of shunts.