Comparison and reproducibility of visual echocardiographic and quantitative radionuclide left ventricular ejection fractions

• Published in: The American journal of cardiology Vol. 77; no. 10; pp. 843 - 850
• Main Authors: VAN ROYEN, N, JAFFE, C. C, KRUMHOLZ, H. M, JOHNSON, K. M, LYNCH, P. J, NATALE, D, ATKINSON, P, DEMAN, P, WACKERS, F. J. T
• Format: Journal Article
• Language: English
• Published: New York, NY Elsevier 15-04-1996; Elsevier Limited
• Subjects: Adult; Aged; Aged, 80 and over; Biological and medical sciences; Cardiovascular system; Echocardiography; Female; Gated Blood-Pool Imaging; Heart; Humans; Investigative techniques, diagnostic techniques (general aspects); Male; Medical research; Medical sciences; Middle Aged; Reproducibility of Results; Retrospective Studies; Stroke Volume; Ultrasonic imaging; Ultrasonic investigative techniques; Ventricular Function, Left; Radionuclide study; Sonography; Human; Echocardiography; Exploration; Scintigraphy; Comparative study; Left ventricle; Ejection fraction
• ISSN: 0002-9149; 1879-1913
• DOI: 10.1016/S0002-9149(97)89179-5

Left ventricular (LV) ejection fraction (EF) is commonly assessed by equilibrium radionuclide angiography and echocardiography. These methods are presumed to be interchangeable for this purpose. This study (1) compares quantification of LVEF by equilibrium radionuclide angiography with visual estimation of LVEF by echocardiography, (2) determines the reproducibility of both methods, and (3) evaluates whether differences in determinations of LVEF are of clinical relevance. Seventy-three clinically stable patients had both equilibrium radionuclide angiography and echocardiography performed within a 4-day period. LVEF by both techniques was compared after blinded analysis by 3 echocardiographers and 3 nuclear technologists. Reproducibility was assessed by blinded repeat analysis after a 1-week interval. The frequency of differences in repeat assessments of EF that the authors considered to be of potential clinical relevance (i.e., difference > or = 10% EF units) was assessed for both techniques. Correlation of LVEF determined by both methods was good (r = 0.81, SEE = 3.5) but with substantial differences in individual patients (limits of agreement, 23.6%). Intra- and inter-observer reproducibility was good for both methods, but better for radionuclide LVEF than for echocardiographic LVEF. Limits of agreement were substantially better for radionuclide LVEF than for echocardiographic LVEF (1.8% to 3.6% versus 13.4% to 17.4%, respectively). Clinically relevant differences did not occur on repeat processing of equilibrium radionuclide angiography. In contrast, potentially clinically relevant differences occurred in 8% to 26% of studies on repeat analysis of echocardiography. Thus, LVEF determined by equilibrium radionuclide angiography and echocardiography show good agreement. Both methods provide clinically valuable measurements for LV function. However, when a precisely reproducible measurement is required for patient management decisions, equilibrium radionuclide angiography is the method of choice.