Creeping towards broader clinical application of PET myocardial blood flow quantification

Abstract

The value of an imaging test is assessed according to diagnostic accuracy, improved risk stratification, and downstream management implications. In clinical practice, the enthusiasm for incorporating measurements of absolute myocardial blood flow (MBF) with dynamic cardiac PET is based on the excellent performance within these categories. Specifically, myocardial flow reserve (MFR) has consistently provided incremental risk stratification beyond gated and relative perfusion variables. More recently, PET MBF has also been used to diagnose and predict adverse events related to coronary microvascular dysfunction. In addition, visual interpretation of both SPECT and PET uses normalized or relative perfusion images that are limited by artifacts and less accurate in the presence of balanced or severe diffuse disease. Finally, preserved MFR can exclude high-risk CAD and may therefore aid in appropriate patient selection for subsequent invasive testing. In part due to the added value of MBF measurements, rest-stress myocardial perfusion PET may be a preferred, first-line test in certain clinical scenarios. Despite all of these favorable characteristics, and having been performed for over 30 years, the technique has been limited to research applications and is employed at a limited number of hospital based, predominately academic medical centers where it may not be included in the final report. The majority of published data are predominantly retrospective and from single-centers. As MBF assessments have become more widespread, potential factors to consider and pitfalls in data acquisition and analysis have been highlighted and are summarized in Table 1. Indeed, even before attempting to ascribe value to an imaging test, the implicit assumption is that the test is highly accurate with robust reproducibility. Given the number of options, factors to consider, and the potential errors that can be introduced, agreement on what to measure and normal values have not been defined. In the current issue, Koenders et al address a basic and important point related to accuracy and reproducibility, myocardial creep. Within this context, myocardial creep refers to altered myocardial blood flow measurements due to changes in diaphragmatic position after vasodilator administration. Specifically, the authors defined myocardial creep as misalignment of at least one-third of the width of the left ventricle in 2 time frames including filling of the left ventricle during the first pass phase. If myocardial creep was identified, manual re-alignment of the contours was performed, and the outcome was a difference in MBF or MFR of at least 10% between corrected and uncorrected scans. According to this definition, in this retrospective cohort study of 104 patients, approximately half demonstrated myocardial creep during the stress acquisition. Although the magnitude of motion was not quantified, the high prevalence of visually significant creep is an obstacle in applying MBF in clinical practice. As the authors demonstrate, myocardial creep is especially a concern in the RCA distribution where misalignment can lead to inappropriate counts in the inferior wall when the signal is actually from the left ventricular cavity. Their results are consistent with this observation where the mean stress MBF in the RCA distribution decreased from 4.0 to 2.7 mL/min/g. Reprint requests: Paul C. Cremer , MD, Department of Cardiovascular Imaging, Heart and Vascular Institute, Cleveland Clinic, Cleveland, OH 44195; [email protected] J Nucl Cardiol 2019;26:735–7. 1071-3581/$34.00 Copyright 2019 American Society of Nuclear Cardiology.