Clinical validation of the non-invasive cardiac output monitor USCOM-1A in critically ill patients

Abstract

Cardiac output is frequently monitored to maintain and improve cardiac function with the primary goal of adequate tissue perfusion. The pulmonary artery catheter is considered to be the gold standard although several non-invasive devices are being introduced and gaining attention. To evaluate the accuracy of the ultrasonic cardiac output monitor (USCOM)-1A (Pty Ltd, Coffs Harbour, NSW, Australia), a non-invasive cardiac output device including its capability to differentiate between different shock states in haemodynamically unstable ICU patients was used in this single-centre, prospective, observational study. Cardiac output was measured with a pulmonary artery catheter and transcutaneously via a suprasternal approach with the USCOM-1A by continuous-wave Doppler ultrasound in 25 adult patients in a mixed medical and surgical ICU in a major teaching hospital in the Netherlands. A total of 1315 USCOM-1A cardiac output measurements were performed. In order to reduce time-variability, the mean of five consecutive USCOM-1A measurements was calculated. Total 263 values were compared with 263 thermodilution cardiac output measurements performed with a pulmonary artery catheter. Data were analysed for systematic error, precision and correlation. Systematic and random errors were found. On average USCOM-1A values were 12% lower than thermodilution measurements (systematic error), while the random error was 17% (coefficient of variation). The error comprised an inter-operator variability of 3%, an inter-patient variability of 11% and residual variability of 15%. The correlation coefficient of the calculated cardiac index with the USCOM-1A and the pulmonary artery catheter was r = 0.8024 and 0.6438, respectively. Temperature and gender did not influence correlations. The learning curve for USCOM-1A skill acquisition was steep. The correlation between the two techniques was acceptable, although relevant systematic and variable errors were detected. USCOM-1A provided adequate data to distinguish non-invasively different shock types in ICU patients.