A comparison of transesophageal and transthoracic echocardiographic assessment of left ventricular function in pediatric patients with congenital heart disease

Objective: To determine the quantitative utility of transesophageal echocardiographic assessments of left ventricular function in pediatric patients with congenital heart disease by evaluating the variability between observers and between echocardiographic windows. Design: Retrospective, blinded analysis. Setting: University-associated pediatric hospital. Participants: Transthoracic and transesophageal echocardiographic images of 25 pediatric patients with congenital heart disease were reviewed. Interventions: None. Measurements and Main Results: End-diastolic area, end-systolic area, and fractional area change were measured from short-axis images of the left ventricle at the midpapillary level by two separate investigators. These measurements were compared by the method of Bland and Altman and Sheiner and Beal. Significant differences in measurements of end-diastolic and end-systolic area by different observers were noted, but they were systematic. A similar situation was noted for the comparison of transthoracic and transesophageal measurements of end diastolic and end-systolic area. In the comparison of fractional area change between observers or windows, bias and absolute prediction error were lower, with 95% confidence limits of bias or absolute prediction error of 10% or less. Conclusions: The potential error in the measurement of fractional area change is 10% under optimal conditions. This would suggest that the assessment of ventricular function in the operating room or intensive care unit, under less than optimal conditions, should be viewed as a qualitative, rather quantitative, measurement. There may be significant interobserver and interwindow variability. (C) 1995 by W.B. Saunders Company