|
|
|
Reproducibility of 4D-MSPECT for Quantitative Gated SPECT
Sestamibi Perfusion Analysis.
Objective: The purpose of this study was to measure
patient specific reproducibility of quantitative gated SPECT Sestamibi
perfusion data acquired on different imaging systems and processed with
4D-MSPECT.
Methods: A total of 79 patients with no history of prior
myocardial infarction were studied. 14 of these patients had
angiographic data, all having confirmed coronary heart disease. All
patients followed either a rest Tl-201/stress Tc-99m Sestamibi or one
day rest/stress Sestamibi protocol. An average dose of 1.1 GBq Tc-99m
Sestamibi was administered for gated SPECT stress imaging. Imaging was
performed 15-30 minutes post treadmill exercise or 60 minutes post
pharmacologic stress. Gated SPECT Sestamibi perfusion data for all
patients was acquired on the Siemens E.CAM and Picker 3000XP SPECT
systems. Camera selection for the first image set was randomized to
camera availability with a mean interval of 25 minutes prior to the
second set of images. 1800 gated short axis data sets with matching 6.3
mm pixel size and slice thickness were generated from the 16 frame raw
data on the ICON and Odyssey workstations. All data were evaluated for
left ventricular ejection fractions (LVEF), end-diastolic volumes and
end-systolic volumes on an Odyssey FX workstation with 4D-MSPECT and
QGS software. 4D-MSPECT's interactive user interface was used to
optimize basal plane selection. The slice differences between end
diastole and end systole were matched for each pair of studies to avoid
operator influence on volumetric determination with MSPECT.
Results: Ejection fraction values for the 4D-MSPECT user
optimized and QGS auto-processed data sets were graphed and compared
for reproducibility within the applications. Results from 4D-MSPECT
yielded y = 0.914x + 9.21 and an r value of 0.886. The QGS automated
results yielded y = 0.905x + 9.935 and an r value of 0.879.
Conclusions: 4D-MSPECT exhibits excellent patient specific
reproducibility of quantitative gated SPECT perfusion data acquired on
different imaging systems with slope values near unity for LVEF values.
|
