Correlation between estimated ejection fraction measured by echocardiography with ejection fraction estimated by cardiac catheterization in patients awaiting cardiac transplantation

BACKGROUND: Ejection fraction measurement is an integral part of pre-heart transplant evaluation. In the clinical arena, the correlation and agreement between ejection fraction (EF) measured by echocardiography versus catheterization in pre-heart transplant patients has not been studied. METHODS: Data on all patients evaluated for heart transplantation at our program between 2003 and 2005 (n ≤ 64) were retrospectively reviewed. Patients with both transthoracic echocardiography (with interpretable images) and left heart catheterization with ventriculography were included (n≤63, mean time difference 2.2 days, median 2 days). We correlated reported echocardiographically estimated EF with that estimated EF by left heart catheterization. RESULTS: Mean estimated EF by ventriculography was 21.2 ± 10.0 in comparison to 22.8 ± 10.5 by echocardiography. The correlation between echocardiographic and measured EF was poor in the total population (r2 ≤ 0.36, p < 0.001) and in patients with non-ischemic cardiomyopathy (r2 ≤ 0.23). However, limiting this statistic to patients with ischemic cardiomyopathy only, we found strong correlation between these two methods (r2 ≤ 0.75, p < 0.001). CONCLUSIONS: Among patients referred for heart transplant evaluation, there is a poor agreement and correlation between echocardiographically estimated EF and estimated EF by ventriculography. However, this correlation is strong in patients with ischemic cardiomyopathy. The cause of this discrepancy is not known and warrants further investigation.