Positron emission tomography of brain glucose metabolism with [18F]fluorodeoxyglucose in humans

Abstract

The practice of neuroimaging, first by SPECT and PET, and then by magnetic resonance imaging (MRI), greatly contributes to the fundamental understanding of neuroanatomical correlates of brain function. It reveals novel treatment options in disciplines such as neurology, neurosurgery, and neuropsychiatry. The new opportunities afforded by neuroimaging yield images not only of brain tissue structure of an ever-increasing power of resolution but also, and perhaps more importantly, of the basic organization of brain work. Understanding brain work requires insight into the roles of regional cerebral blood flow and energy metabolism, obtained by measures of oxygen and glucose consumption rates, neuronal network and neurotransmitter system activity, and most recently the abnormal deposition of amyloid-beta in brain tissue and the resulting abnormalities of second messenger cascades that promise to reveal the origins of neurodegeneration. The quantification of the brain images, acquired by means of the different methods of neuroimaging, is vital to the improved understanding and interpretation of experimental and clinical findings that are reported at a significant and ever-increasing rate. While many brain-imaging agents, such as markers of amyloid-beta in dementia, are studied with the ultimate goal of application to clinical prognostication and differential diagnosis, other agents are fundamental research tools for understanding new drugs, such as antipsychotics, antidepressants, and anxiolytics, as well as drugs for the relief of devastating neurological disorders such as multiple sclerosis, stroke, and dementia. Some of the earliest markers were tracers of blood flow used with autoradiography, SPECT, and PET, but in 1977, Sokoloff et al. published the seminal description of the use of labeled 2-deoxyglucose to trace the rate of glucose phosphorylation in brain, initially by autoradiography but soon after also by positron emission tomography of the brain uptake of the glucose analog 2-fluoro-2-deoxyglucose, labeled with the positron emitter fluorine-18 (FDG). The basis for the use of this tracer ex as well as in vivo is a deceptively simple model of brain glucose metabolism that includes only the steps of the bidirectional transport of glucose and glucose analogs across the blood-brain barrier imposed by the tight junctions between the endothelial cells of the brain’s capillaries, and the step of phosphorylation enabled by the presence of the enzyme hexokinase in the cells of the brain. This chapter provides a brief explanation of the quantitative method of PET imaging with FDG used by neuroscientists for the last 40 years to quantify the uptake and metabolism of this tracer in terms of the absolute rate of glucose phosphorylation in brain tissue during the period following the tracer administration. The chapter also highlights the issues of relative precision and accuracy of the method applied to high-resolution research tomography. It includes a description of the basic elements of quantification, and, in particular, of the necessary mathematical modeling of the dynamic brain records of the uptake of the tracer, both to justify the role of such modeling in study design and to validate some of the simplifications that are necessary in some clinical settings. As fundamental tools of neuroimaging, quantification and kinetic modeling are as important as image reconstruction and structural identification of regions of interest. The quantitative methods presented here continue to underpin the routine approaches to measures of brain glucose consumption rates in different regions of the brain and hence matter to most clinicians and clinician scientists involved in the neuroimaging practice of regional glucose phosphorylation rates.