Interobserver variability in coronary angiography

Abstract

Four experienced coronary angiographers (two radiologists and two cardiologists) independently assessed the location and degree of coronary artery stenosis, and the location and degree of left ventricular wall motion abnormalities in 20 coronary angiograms. Marked interobserver variability was noted in quantifying percent coronary artery stenosis and degree of left ventricular wall motion abnormalities. For example, in only 13/20 (65%) of the coronary angiograms did all observers agree about the significance of a stenosis (defined as greater than 50% in diameter luminal narrowing) in the proximal or mid left anterior descending coronary artery. In 3/20 (15%) angiograms there was disagreement by at least one observer about the significance of lesions noted in the main left coronary artery. The ventricle was divided into five segments and the degree of wall motion abnormality graded into six categories of increasing severity from normal to dyskinesis. There was a 42% mean disagreement among all four observers where a disagreement between observers was defined as any difference in grading wall motion abnormalities. Interobserver variability reveals a significant limitation of coronary angiography.