Will 3-dimensional PET-CT enable the routine quantification of myocardial blood flow?

Quantification of myocardial blood flow (MBF) and flow reserve has been used extensively with positron emission tomography (PET) to investigate the functional significance of coronary artery disease. Increasingly, flow quantification is being applied to investigations of microvascular dysfunction in early atherosclerosis and in nonatherosclerotic microvascular disease associated with primary and secondary cardiomyopathies. Fully three-dimensional (3D) acquisition is becoming the standard imaging mode on new equipment, bringing with it certain challenges for cardiac PET, but also the potential for MBF to be measured simultaneously with routine electrocardiography (ECG)-gated perfusion imaging. Existing 3D versus 2D comparative studies support the use of 3D cardiac PET for flow quantification, and these protocols can be translated to PET-CT, which offers a virtually noise-free attenuation correction. This technology combines the strengths of cardiac CT for evaluation of anatomy with cardiac PET for quantification of the hemodynamic impact on the myocardium. High throughput clinical imaging protocols are needed to evaluate the incremental diagnostic and prognostic value of this technology.

Introduction

Following the development of combined positron emission tomography (PET)–computed tomography (CT) technology for whole-body imaging¹ and the rapid growth of cardiac CT applications, the use of PET-CT for myocardial perfusion imaging (MPI) is increasing. Traditionally, cardiac PET imaging has been performed in so-called 2-dimensional (2D) mode, where collimating septa were used between detector rings to reduce interplane scatter. Attenuation correction was performed with isotope transmission scanning, but this has been replaced by x-ray transmission scanning exclusively in the current generation of PET-CT systems. All current commercial PET systems now provide a 3-dimensional (3D) imaging mode, and 2 vendors have moved to 3D-only systems. The advantages of 3D PET are well established in quantitative brain studies, where increased scanner sensitivity can improve image quality or lower injected activity versus the 2D mode. The use of 3D PET is also widely accepted for whole-body PET oncology imaging, but the requirements for image quantification are less demanding than for neurology. Likewise in cardiology, 3D PET has the important potential to improve image quality and reduce patient doses by lowering injected tracer activities. Care must be taken to ensure that quantitative accuracy is maintained versus established 2D methods, because MPI and the assessment of absolute myocardial blood flow (MBF) and myocardial flow reserve (MFR) rely more heavily on quantification than tumor imaging with whole-body PET.

A unique feature of PET that distinguishes it from other imaging modalities is its ability to quantify molecular function in the living body. High sensitivity and accurate attenuation correction enable precise quantification of extremely low activity (Bq/mL) and molecular (fmol/mL) concentrations.² The ability to label tracer doses of organic compounds, combined with the quantitative nature of volumetric PET imaging, makes this technology ideally suited to study the molecular biology and physiology of many organ systems in vivo with high sensitivity and specificity. Among these is myocardial perfusion, which is central to the diagnosis and management of ischemic heart disease.

Clinical MPI to assess the relative distribution of blood flow has been performed traditionally with single-photon methods (single photon emission computed tomography [SPECT]), but PET imaging is increasingly recognized for its superior accuracy,³ particularly in patients with body habitus or body shapes where SPECT imaging is potentially less specific or less conclusive.⁴ In addition to relative MPI, dynamic imaging of tracer uptake and clearance kinetics with PET can be used to quantify MBF in absolute terms (mL min⁻¹ g⁻¹). Serial rest and stress imaging protocols are used to measure MFR (stress/rest MBF ratio) for assessment of the vasodilator response of the coronary vascular bed from the epicardial coronary arteries to the microcirculation.

PET imaging is now recognized as the most accurate noninvasive means by which to quantify MBF and flow reserve, which are independent of relative perfusion measurements. MBF quantification has been used widely to understand heart diseases, including coronary artery disease, microvascular disease, and severe ischemic heart disease, and to evaluate new therapies.⁵ However, these studies have generally used 2D PET technology. With the advent and widespread adoption of PET/CT imaging, there has been a metamorphosis from 2D to 3D imaging, both in terms of clinical use and in terms of new scanner availability. Hence 2 questions arise: (1) Can 3D imaging be applied for MBF quantification? (2) Will the use of cardiac PET/CT yield more widespread clinical application of blood flow quantification?

The ideal requirements for clinical 3D cardiac PET are shown in Table 1. In this review we first discuss the key elements of tracer characteristics, scanner data acquisition, attenuation correction, and scatter correction, which are relevant issues for both relative MPI and absolute MBF imaging. Because these have also been discussed in a recent comprehensive review,⁶ we focus subsequently on the methods of MBF quantification with 3D PET imaging, where there is more limited literature describing its advantages and challenges. Finally, we will discuss clinical applications and the types of research that need to be completed for wider implementation.