Published in:
01-05-2008 | Original article
Absolute quantitation of myocardial blood flow with 201Tl and dynamic SPECT in canine: optimisation and validation of kinetic modelling
Authors:
Hidehiro Iida, Stefan Eberl, Kyeong-Min Kim, Yoshikazu Tamura, Yukihiko Ono, Mayumi Nakazawa, Antti Sohlberg, Tsutomu Zeniya, Takuya Hayashi, Hiroshi Watabe
Published in:
European Journal of Nuclear Medicine and Molecular Imaging
|
Issue 5/2008
Login to get access
Abstract
Purpose
201Tl has been extensively used for myocardial perfusion and viability assessment. Unlike 99mTc-labelled agents, such as 99mTc-sestamibi and 99mTc-tetrofosmine, the regional concentration of 201Tl varies with time. This study is intended to validate a kinetic modelling approach for in vivo quantitative estimation of regional myocardial blood flow (MBF) and volume of distribution of 201Tl using dynamic SPECT.
Methods
Dynamic SPECT was carried out on 20 normal canines after the intravenous administration of 201Tl using a commercial SPECT system. Seven animals were studied at rest, nine during adenosine infusion, and four after beta-blocker administration. Quantitative images were reconstructed with a previously validated technique, employing OS-EM with attenuation-correction, and transmission-dependent convolution subtraction scatter correction. Measured regional time-activity curves in myocardial segments were fitted to two- and three-compartment models. Regional MBF was defined as the influx rate constant (K
1) with corrections for the partial volume effect, haematocrit and limited first-pass extraction fraction, and was compared with that determined from radio-labelled microspheres experiments.
Results
Regional time-activity curves responded well to pharmacological stress. Quantitative MBF values were higher with adenosine and decreased after beta-blocker compared to a resting condition. MBFs obtained with SPECT (MBFSPECT) correlated well with the MBF values obtained by the radio-labelled microspheres (MBFMS) (MBFSPECT = −0.067 + 1.042 × MBFMS, p < 0.001). The three-compartment model provided better fit than the two-compartment model, but the difference in MBF values between the two methods was small and could be accounted for with a simple linear regression.
Conclusion
Absolute quantitation of regional MBF, for a wide physiological flow range, appears to be feasible using 201Tl and dynamic SPECT.