The validity of trans-esophageal Doppler ultrasonography as a measure of cardiac output in critically ill adults

Objective: To determine the validity of the esophageal Doppler monitor (EDM) and echo-esophageal Doppler (Echo-ED) in measuring cardiac output in the critically ill. Design: Systematic search of relevant international literature and data synthesis. Search strategy: Literature search (1989-2003) using Ovid interface to Medline, Embase and Cochrane databases aimed at finding studies comparing EDM or Echo-ED cardiac output with that derived from simultaneous pulmonary artery thermodilution (PAC(TD)) with Bland Altman measures of validity. Patients: Critically ill adults in operating departments or intensive care units. Data synthesis: Summary validity measures synthesized from Bland Altman analyses included pooled median bias and the median percentage of clinical agreement (PCA) derived from the limits of agreement. Main results: Eleven validation papers for EDM (21 studies) involving 314 patients and 2,400 paired measurements. The pooled median bias for PAC(TD) versus EDM was 0.19 l/min (range -0.69 to 2.00 l/min) for cardiac output (16 studies), and 0.6% (range 0-2.3%) for changes in cardiac output (5 studies). The pooled median percentage of clinical agreement for PAC(TD) versus EDM was 52% (interquartile range 42-69%) for cardiac output and 86% (interquartile range 55-93%) for changes in cardiac output. These differences in PCA were significant (p=0.03 Mann-Whitney) for bolus PAC(TD) as the clinical "gold standard". We found an insufficient number of studies (2 papers) to assess the validity of Echo-ED. Conclusions:The esophageal Doppler monitor has high validity (no bias and high clinical agreement with pulmonary artery thermodilution) for monitoring changes in cardiac output.