The next step in improving (semi-)quantitative MPI PET
- PDF / 254,966 Bytes
- 3 Pages / 593.972 x 792 pts Page_size
- 41 Downloads / 208 Views
,a,c J. A. van Dalen, PhD,b and J. D. van Dijk, MSc, PhD
a
Department of Nuclear Medicine, Isala Hospital, Zwolle, The Netherlands Department of Medical Physics, Isala Hospital, Zwolle, The Netherlands Technical Medical Center, University of Twente, Enschede, The Netherlands
Received Aug 14, 2020; accepted Aug 14, 2020 doi:10.1007/s12350-020-02346-3
See related article, https://doi.org/10.10 07/s12350-020-02266-2. BACKGROUND Myocardial blood flow (MBF) quantification with PET results in an improved risk assessment of coronary artery disease (CAD) in addition to semi-quantitative myocardial perfusion imaging.1 A decrease in the myocardial flow reserve (MFR, MBF in stress/MBF in rest) or an increasing area of perfusion deficits results in a higher risk on cardiac mortality, as shown in Figure 1. To be able to quantify MBF, tracer uptake in the left ventricle and the myocardium during the scan has to be calculated from the imaging data. However, in the process of data acquisition, reconstruction, post-processing and interpretation of the data, there are several pitfalls that one should be aware of as these pitfalls can result in unreliable uptake calculations and hence MBF measurements.2 Multiple pitfalls for MBF quantification and corresponding solutions have been identified in the last decade. First, it is important that the PET scanner has a sufficient count-rate capability to prevent detector saturation during the first-pass phase which can lead to artificially high MBF values.3,4 Secondly, to increase the reproducibility of MBF measurements, a constant activity infusion profile is required.5 Yet not all Strontium82-Rubidium-82 generators are able to produce such a constant activity bolus. Third, misregistration of PET
Reprint requests: J. D. van Dijk, MSc, PhD, Department of Nuclear Medicine, Isala Hospital, PO Box 104008000 GK Zwolle, The Netherlands; [email protected] J Nucl Cardiol 1071-3581/$34.00 Copyright Ó 2020 American Society of Nuclear Cardiology.
with CT data that is used for attenuation correction can result in altered MBF measurements.6 Fourth, reconstruction settings and post-processing software packages can result in a MBF bias as well.7,8 Although the reproducibility is often unaffected when changing reconstruction settings or shifting to different software, a bias may occur. Fifth, patient motion and repositioning of the heart after administration of a stress agent can also negatively affect the reliability of MBF measurements.9,10 However, this influence can be limited by correcting the dynamic time-frames for possible motion and ‘‘myocardial creep’’.11 Last, the test-retest reliability of MBF quantification is relatively large with a typical uncertainty of 21%.12 This implies that there is a large ‘grey area’ in distinguishing reduced and normal MBF values and physicians should be aware of this during interpretation. Although our knowledge of and solutions to pitfalls increases, several issues remain. For example, how to prevent scans in which there is not a well-defined activity bolus
Data Loading...
