Abstract
Background: Perioperative interventions, targeted to increase global blood flow defined by explicit measured goals, reduce postoperative complications. Consequently, reliable non-invasive estimation of the cardiac output could have far-reaching benefit. Methods This study compared a non-invasive Doppler device – the ultrasonic cardiac output monitor (USCOM) – with the oesophageal Doppler monitor (ODM), on 25 patients during major abdominal surgery. Stroke volume was determined by USCOM (SVUSCOM) and ODM (SVODM) pre and post fluid challenges. Results: A ≥ 10% change (∆) SVUSCOM had a sensitivity of 94% and specificity of 88% to detect a ≥ 10% ∆ SVODM; the area under the receiver operating curve was 0.94 (95% CI 0.90–0.99). Concordance was 98%, using an exclusion zone of <10% ∆ SVODM. 135 measurements gave median SVUSCOM 80 ml (interquartile range 65–93 ml) and SVODM 86 ml (69–100 ml); mean bias was 5.9 ml (limits of agreement –20 to +30 ml) and percentage error 30%. Conclusions Following fluid challenges SVUSCOM showed good concordance and accurately discriminated a change ≥ 10% in SVODM.