Dynamic SPECT: evolution of a widely available tool for the assessment of coronary flow reserve.

Abstract

Myocardial perfusion imaging (MPI) SPECT is well established in the diagnosis, monitoring response to treatment and risk stratification in patients with known or suspected coronary artery disease (CAD). PET enables quantitative assessment of myocardial blood flow (MBF, in millilitres per gram per minute) and coronary flow reserve (CFR), and quantification with O-water, N-ammonia and recently Rb has been validated over a wide range of blood flows in animal models and humans [1–4]. Quantitative assessment of MBF has been shown to improve the diagnostic accuracy of conventional MPI with SPECT or PET, to improve cardiac risk assessment and to predict outcome [5–7]. Quantitation of MBF enables absolute assessment of myocardial flow and vasodilator reserve without the assumption of a normal reference region [8]. Therefore, the limitation of conventional MPI (underestimation of the extent and severity of multivessel CAD, when tracer uptake in the best-perfused myocardial region does not represent normally perfused myocardium) can be overcome by the use of absolute quantitation [9]. Whereas PET is very costly and complex, SPECT systems are widely used for the assessment of myocardial perfusion in patients for the diagnosis and management of CAD. However, quantitative assessment with SPECT has been limited. To enable quantitation with SPECT a multidetector system is required to permit fast acquisition of dynamic data in 5 – 10 s, and a suitable SPECT tracer is necessary. Transmission imaging for attenuation correction will allow accurate quantitation. Quantification of myocardial perfusion reserve has been attempted using SPECT and Tl in dogs [10] and Tc-labelled tracers [10–12]. Dynamic SPECT imaging using multidetector SPECT systems and kinetic modelling of Tc-teboroxime has shown good correlation with microsphere-determined blood flow. However, limitations in detector sensitivity and temporal resolution of conventional SPECT systems prohibit further assessment [10, 11]. Another SPECT technique based on first-pass planar imaging followed by conventional SPECTMPI has been used to estimate a retention index of MBF and CFR [12]. This technique has shown a generally good correlation with PETmeasured flow, but CFR is underestimated at high flow rates [13]. The use of a retention index to estimate CFR using this method compared to absolute MBF from PET results in an underestimation of CFR values in the SPECT-based technique, since tracer retention decreases with increasing blood flow [14]. Spatial and timing resolution are poorer with SPECT and the tracer retention index underestimates CFR compared to quantitative PET. SPECT is indeed simpler than PET but this technique, unlike PET, does not include dynamic acquisition of tomographic data. In addition, the technique works only for tracers that act like microspheres, showing a constant extraction over a large range of flow rates and showing no washout from the time of injection to the time of measurement. Conventional SPECT systems are limited in the dynamic collection of tomographic data. These systems consist of slowly rotating cameras with large detectors. The detectors’ orbit is limited by mechanical as well as safety factors and the angular projections obtained are inconsistent, resulting in blurred images and possible bias in the estimated kinetic parameters. In addition, conventional detector crystals suffer S. Ben-Haim Institute of Nuclear Medicine, University College London, University College Hospital, London, UK