Validation of Epworth Richmond’s Echocardiography Education Focused Year

Abstract

College of Intensive Care Medicine mandatory training in echocardiography requires a minimum 30 echocardiograms. Trainee echocardiography accuracy following mandatory training is unknown. Our objective was to determine the reliability of ICU trainee echocardiography after our yearlong College of Intensive Care Medicine accredited course. Single-center, prospective observational study. Trainee echocardiograms were compared to independent, blindedexpert echocardiograms, with a target of 37 matched scans per trainee. Trainees were eligible to participate on completion of our in-house, mandatory echocardiography training. Epworth Richmond is in an Australian, private academic hospital with a critical care echocardiography teaching program. None. Reliability was assessed on nine measurements. The primary outcome was agreement on left ventricular function; secondary outcomes included qualitative and quantitative assessments. Agreement was assessed using Cohen’s kappa, Fleiss-Cohen weighted kappa, Lin’s concordance correlation coefficient, or calculation of sensitivity and specificity as appropriate. Seven trainees performed a total of 270 matched scans. There was excellent agreement between experts and trainees for left ventricular function (Kappa, 0.86; 95% CI, 0.81–0.91). Secondary outcome measures (95% CI) were ventricular size ratio Kappa 0.76 (0.58–0.94), sensitivity 92.3% (64.0–99.8%), specificity 97.2% (94.9–99.1%), pericardial effusion Kappa 0.37 (0.13–0.60), sensitivity 33.3% (13.3–59.0%), specificity 97.6% (94.8–99.1%). Concordance coefficients (95% CI) for the remaining outcomes were left ventricular outflow tract velocity time integral 0.79 (0.74–0.84), left ventricular internal diameter in diastole 0.82 (0.78–0.86), left ventricular outflow tract diameter 0.61 (0.53–0.69), tricuspid annular plane systolic excursion 0.71 (0.64–0.78), tricuspid regurgitation maximum velocity 0.55 (0.44–0.65), and inferior vena cava diameter 0.60 (0.49–0.72). ICU trainees showed very high overall agreement with experts on left ventricular function and substantial agreement for most secondary outcomes. This is the first study to assess echocardiography accuracy in Australian ICU trainees.