Reproducibility of the angiographic left ventricular ejection fraction in patients with coronary artery disease

Abstract

Quantitative analysis of the single and repeated cine left ventriculogram was performed in 20 patients with coronary artery disease to determine both the intrinsic variance of individual beats separated by different time intervals and variance between analyses of different observers. In addition, ventriculograms obtained from left ventricular injections of contrast medium prior to coronary arteriography were compared to ventriculograms obtained from either left ventricular or pulmonary artery injections after arteriography. The time period between studies varied from 30 minutes to 90 minutes to four days. Analysis of the same ventriculogram by different observers resulted in an average difference in ejection fraction of 0.05 (pNS). The average difference in ejection fraction was 0.02 between two early beats of the same ventriculogram (pNS). The average difference between sequential ventriculograms was 0.07 (pNS), but individual variations greater than 0.10 were not uncommon, particularly between studies done before and after arteriography, or several days apart. Patients exhibiting wide variance in ejection fractions between two studies either had wide variance in other hemodynamic measurements or degree of asynergy, or both. This study provides a frame of reference for analysis of sequential ventriculograms in patients with coronary artery disease, especially in evaluating changes in the state of the disease or the effects of therapy. It is especially important that: (1) standard hemodynamic measurements be made before ventriculography, (2) the same radiographic techniques repeated whenever possible, and (3) the same person analyze the two ventriculograms.