Assessment of left ventricular ejection fraction with quantitative gated SPECT: Accuracy and correlation with first-pass radionuclide angiography

Citation
E. Vallejo et al., Assessment of left ventricular ejection fraction with quantitative gated SPECT: Accuracy and correlation with first-pass radionuclide angiography, J NUCL CARD, 7(5), 2000, pp. 461-470
Citations number
40
Categorie Soggetti
Cardiovascular & Respiratory Systems
Journal title
JOURNAL OF NUCLEAR CARDIOLOGY
ISSN journal
10713581 → ACNP
Volume
7
Issue
5
Year of publication
2000
Pages
461 - 470
Database
ISI
SICI code
1071-3581(200009/10)7:5<461:AOLVEF>2.0.ZU;2-T
Abstract
Background. Quantitative gated single photon emission computed tomography ( SPECT [QGS]) software is widely used for the assessment of left ventricular ejection fraction (LVEF). Potentially confounding variables that may affec t the accuracy of quantitative analysis of LVEF remain undefined. This stud y evaluated the accuracy of QGS as a means of determining LVEF in a wide ra nge of LVEF values; evaluated the effect of extracardiac activity, count st atistics, heart size, and perfusion defects on the accuracy of QGS LVEF; an d compared QGS LVEF obtained at rest with that obtained after stress. Methods and Results. QGS-derived LVEF was compared with rest first-pass rad ionuclide angiography (FPRNA) LVEF in 400 electrocardiographic-gated SPECT studies. The overall correlation between QGS and FPRNA LVEF was only fair ( r = 0.66, SEE = 11.85%). In 35 of the patient studies (9%) with high extrac ardiac activity, the automated software failed, and no correlation was obta ined. In the remaining 365 patient studies (91%), left ventricular contours were successfully identified. In these studies, correlation was better (r = 0.74, SEE = 9.77%). Agreement was better for images with high counts (r = 0.81, SEE = 8.66%) than for images with low counts (r = 0.61, SEE = 11.17% ). Patient studies with abnormal LVEF had better correlation (r = 0.77, SEE = 6.4%) than studies with normal LVEF (r = 0.46, SEE = 10.2%). Agreement b etween QGS LVEF and FPRNA LVEF was better in hearts with large end diastoli c volumes (> 104 mL) than in hearts with small volumes. Overall, mean QGS L VEF was lower than mean FPRNA LVEF (54% +/- 14% vs 58% +/- 14%, P < .0001). There was no difference between mean rest and stress QGS LVEF in the same patients, even in patients with stress-induced ischemia. Conclusions. QGS is a valuable method for assessing resting LVEF. However, QGS LVEF is often lower than FPRNA LVEF. Accuracy is affected by high extra cardiac activity, low count density, and small size of the left ventricle.