Determination of ejection fraction and left ventricular fraction using isotopic ventriculography and bidimensional echocardiography. Comparison with contrast ventriculography

OBJECTIVES: The aim of this study was to determine the correlation and agreement between the values of left ventricular ejection fraction and volumes assessed by echocardiography and radionuclide ventriculography with the results obtained by contrast angiography, as well as the variability of each method in these measurements. PATIENTS AND METHODS: In a group of 59 patients with a first acute myocardial infarction we have determined left ventricular ejection fraction and volumes by two-dimensional echocardiography, equilibrium radionuclide ventriculography and contrast angiography initially and six months after myocardial infarction. We also assess the variability in the determinations in these three methods. RESULTS: We found significant correlations in ejection fraction and volumes by radionuclide ventriculography and echocardiography with contrast angiography. The correlation was higher in ejection fraction and end-systolic volume by radionuclide ventriculography (r = 0.88 and r = 0.73) than by echocardiography (r = 0.55 and r = 0.63; p < 0.01), whereas the correlation of end-diastolic volume was moderate by both methods (r = 0.58 and r = 0.47), without significant differences. The agreement between contrast angiography and radionuclide ventriculography was higher, with narrower limits of agreement than between contrast angiography and echocardiography in ejection fraction as well as in ventricular volumes. We have found high and significant correlations between two determinations by each method in all parameters, although they were higher in ejection fraction by contrast angiography (r = 0.96) and radionuclide ventriculography (r = 0.98) than by echocardiography (r = 0.70; p < 0.01). The limits of agreement were always wider in echocardiography, narrower in contrast angiography and the narrowest in radionuclide ventriculography, showing its superior reproducibility. CONCLUSIONS: In this group of myocardial infarction patients, the variability in the measurements was lower by radionuclide ventriculography than by echocardiography, this could be the reason for overall better results found in correlation and agreement between radionuclide ventriculography and contrast angiography than between echocardiography and contrast angiography in the assessment of left ventricular ejection fraction and volumes.