Quantitative Dynamic PET Imaging of the Heart: From Mouse to Man

Positron emission tomography (PET) imaging is an excellent tool to evaluate molecular, metabolic, physiologic and pathologic conditions of the body for diagnosis, therapy planning and research. It is based on annihilation coincidence detection technique, wherein a positron annihilates with an electron to generate two 511 keV photons that can be detected simultaneously. Several factors such as the detector element size, positron range, and noncollinearity of annihilation photons together limit the system spatial resolution of a PET system. Quantification of PET data is thus very challenging due to the limited spatial resolution. Tracer kinetic modeling is commonly used for quantitative evaluation of dynamic PET data. The image derived time-resolved curves are susceptible to incomplete radioactivity recovery called partial volume (PV) effect, and cross-contamination between adjacent regions called spillover (SP) effect, thereby image derived data do not reflect the true activity concentration in the region of interest. The objective of this dissertation is to develop tracer kinetic models for evaluation of metabolism and perfusion in the heart using dynamic PET imaging in vivo, accounting for PV and SP corrections.
Using Siemens microPET Focus F120 small animal scanner, efforts were devoted into the development and validation of an in vivo quantification methodology for 2-[18F] fluoro-2deoxy-D-glucose (FDG) dynamic PET images of mouse heart, which can simultaneously give an accurate estimation of metabolic kinetic parameters, and model corrected blood input function with PV and SP corrections in a 3-compartment glucose transport model. This advanced method not only improves quantitation but is also repeatable. Using the developed kinetic model, this dissertation also discussed the non-invasive measurement of the myocardial glucose uptake rate ki in mouse heart subjected to transverse aortic constriction (TAC) surgery induced pressure overload left ventricle hypertrophy (LVH), to test the hypothesis that metabolic changes precede and possibly trigger cardiac dysfunction in the pressure overloaded heart.
Adapting the quantification methodology developed in small animals to clinical practice, a first-in-human study was carried out on a Siemens Biograph mCT PET/CT clinical scanner using dynamic FDG PET imaging, to test the hypothesis that significant changes in glucose metabolism precede any structural and functional changes of the heart in patients with hypertension.
Also, this dissertation explored quantitative assessment of absolute blood flow in mouse heart using dynamic 13N-ammonia PET imaging in a 3-compartment tracer kinetic model. Larger positron range of isotope 13N leads to further degradation of the spatial resolution, thereby making quantification more challenging than that in FDG PET imaging. Poor image quality, small size of the mouse heart and rapid heart rate results in significant PV effect and SP contamination. A quantitative 3-compartment tracer kinetic model was developed to measure blood flow in mouse heart in vivo. A correlation between blood flow and metabolism was established in the stressed mouse heart, for the first time, using dynamic 13N-ammonia and FDG PET imaging respectively.