Measuring ejection fraction-a comparison of two methods

ABSTRACT

Coronary angiography is used as a gold standard for documentation of much prevalent coronary artery disease. Amongst other data, an estimate of ejection fraction is almost always acquired by some method to quantitate left ventricular function, which is one of the most powerful prognostic factors. Planimeterically measured ejection fraction has come to be accepted as a gold standard. We have compared the digitally acquired planimeteric ejection fraction with visual estimates of experienced observers on 50 patients. Our data shows that ejection fraction by eyeballing usually underestimates the true EF significantly. In this study eyeballing differed from digital EF by a mean value of 13.47 percent, an SD of 10.23 percent and a range of+12.33 percent to -31.74 percent. There is no linear relationship between eyeballed and digital EF. There is considerable inter-observer variation. The inaccuracy in visually assessed EF is unrelated to ventricle size and presence of dyskinetic segments. The degree of inaccuracy is greater with poor left ventricular function. There are significant inter-observer differences in visual estimation of LV size as well as assessment of LV function. It is recommended that objective methods should be used as much as possible in reporting coronary angiograms