Assessment of Diastolic Function Parameters with ECG Gated Myocardial Perfusion Spect: Comparison of Two Algorithms

Abstract

Background: To compare Cedars-Sinai QGS and Michigan University Corridor4DM algorithms
for determination peak filling rate (PFR), time to peak filling rate (TPFR) and mean filling rate in
first third of diastole (MFR3) using 16 frames gated myocardial perfusion single photon emission
computed tomography (SPECT). To determine inter-observer reproducibility of Cedars QGS and
Michigan University Corridor4DM for determination PFR, TPFR and MFR3 using 16 frames gated
myocardial perfusion SPECT.
Methods: Forty patients (28 males and 12 females) with age range 35-70 years (mean
58.85±8.82) referred for assessment of left ventricular perfusion and function were included in the
99m study. All patients were injected 1100 Mega Becquerel (MBq) of freshly prepared Tc Sestamibi.
One hour later, patients underwent gated myocardial perfusion SPECT on Siemens ecam® dual
head variable angle gamma camera using 16 frames per cardiac cycle. Data were reconstructed
using filter back projection and re-orientated to generate short axis slices. Short axis slices were
processed with QGS and Corridor 4DM for assessment of PFR, TPFR and MFR3 by two
observers. Data from both observers were compared to determine inter-observer reproducibility
of both methods. Observer1 PFR, TPFR and MFR3 values derived from QGS and Corridor4DM
were compared and correlated.
Results: Peak filling rate values determined with Cedars QGS program were not significantly
different from those determined with Corridor4DM (p= 0.564). Good correlation was found
2
between QGS and 4DM measured PFR values (R =0.6698). TPFR values determined with QGS
program were not significantly different from those determined with Corridor 4DM program (p=
2
0.615). However, there was poor correlation between these two methods with R value =0.0382.
MFR3 values determined with QGS were not statistically different from those derived from 4DM
2
(p=0.587). However, there was poor correlation between these values R = 0.0174. Cedars QGS
2
algorithm was highly reproducible for determination of PFR, TPFR and MFR/3 with R values of
0.9922, 0.9874 and 0.9932 respectively. PFR, TPFR and MFR3 derived from Corridor4DM were
2
also highly reproducible with R values of 0.7775, 0.8381 and 0.456 respectively.
Conclusions: Both Cedars QGS and Michigan University Corridor 4DM programs are robust for
determination of PFR, TPFR and MFR3 diastolic function parameters. There is good correlation
between QGS and 4DM derived PFR measurements. However, there is poor correlation between
QGS and 4DM derived TPFR and MFR3 values