Methodological Considerations for PET Studies in Alcoholics

by National Institute of Health

(Page 6 of 6)

Developing Models for Interpreting Findings of PET Analyses

The data obtained in alcoholics with functional imaging techniques such as PET typically must undergo a set of processing steps to yield information that is useful to researchers. For example, researchers must develop mathematical representations of physiological processes such as the metabolism of neurotransmitters or their receptors. With these models, investigators can develop a mathematical equation describing the tissue response curve expected in the measurements. The tissue-response curve plots the radioactivity of specific parts of the brain before, during, and after the injection of the radiotracer. Thus, the tissue-response curve correlates with the amount of the chemical identified by the radiotracer in the regions of interest. By performing the scans on groups of people with and without alcoholism, the increases and decreases of chemicals in the brains of alcoholics can be identified. By identifying those variables in the model that give the best agreement between the expected and measured values, one can quantify the physiological process.

To develop appropriate models and provide a basis for interpreting the measured behavior of the tracer, all available qualitative information about the physiology and biochemistry of the tracer is collected. For example, it is important to know how fast and to what extent the tracer is transported from the site of the injection in the bloodstream to the tissue being analyzed. The basic steps of this transport are as follows: the tracer is transported by the blood to the small blood vessels (capillaries) in the brain. The tracer moves across the capillary wall into the fluid-filled spaces between the brain cells. The tracer crosses the membrane surrounding the cells or binds to neurotransmitter receptors in the synaptic clefts between neurons. If it enters the cells, the tracer participates in various biochemical reactions.

PET can follow the progress of the tracer by measuring the amounts of radioactivity in different areas of the brain as well as the tracer concentrations in the blood. To interpret the data obtained in this way, investigators can use a variety of mathematical or statistical modeling methods (the compartment model, graphical model, and tissue input graphical model approaches). In many cases, researchers attempt to simplify their models by making assumptions about the processes involved in the model (about how easily the tracer can cross the capillary walls). To make sure these assumptions are reasonable or correct, however, the simplified model must first be validated. To this end, the investigators must make sure that the model yields reasonable values for the variables tested and that it can distinguish the disease state (conditions in an alcoholic) from the healthy state (conditions in a nonalcoholic).

Correcting for Partial Volume Errors

Compared with structural imaging techniques (MRI), PET images are blurred because of the limited resolution of the PET scanners (their limited ability to distinguish closely spaced regions of small dimensions). This limited resolution has two potential consequences: an apparent loss (or "spill-out") of radioactivity signals from a small region of interest into the adjacent tissues owing to the size of the small brain region compared with the spatial resolution of the scanner. A "spill-in" of radioactive signals into the region of interest from adjacent brain areas with different radioactive tracer concentrations.

These effects, which are known as "partial volume errors," are more pronounced in alcoholics with alcohol-related brain shrinkage, where loss of signal because of partial volume errors could be confounded with an actual loss of tissue function.

Several methods are available to correct for this problem. The most common approach is to perform both an MRI scan and a PET scan of a patient's brain and then to combine the images using several available methods. Computer simulations then are used to mimic the effect of limited spatial resolution to characterize the partial volume effects for each brain region. With this information, investigators then can apply correction factors to obtain more accurate estimates of the actual regional activity (regional blood flow or glucose metabolism).

The Future of Pet Studies in Alcoholism

Although researchers have been employing PET and other functional imaging techniques in the analysis of alcohol-induced changes in brain functioning, the full potential of these approaches has not yet been realized. For example, it might be useful to correlate functional imaging data with information on demographic traits of the subjects as well as with behavioral measures, including questionnaires addressing psychological traits and the desire for alcohol. Demographic and psychological traits may identify biological subtypes of alcoholism detected by PET. Questionnaires to identify behavioral data could function as biological markers for the presence of distinct biological variants of alcoholism and could help identify the effects of potential therapeutic interventions targeted to the distinct variant. Other possible applications of PET include the following: studies in humans and animals to characterize neurochemical processes associated with alcohol reinforcement and/or craving, such as the production, release, and transport of neurotransmitters or changes in receptor concentrations. Identification of neural circuits that play a role in the cognitive deficits associated with acute alcohol intake as well as elucidation of the pathways through which functional deficits in specific neural circuits and the resulting cognitive deficits may contribute to excessive alcohol intake. Analyses of neurobiological markers of vulnerability to alcohol abuse. Combination with structural imaging techniques to obtain a fused image automatically. Development of new pharmaceutical agents to prevent and treat alcoholism.

Summary

PET allows researchers to visualize in living human beings the damage to the brain that results from chronic excessive alcohol consumption. This technology has been used to analyze alcohol's effects on various neurotransmitter systems as well as on glucose metabolism and regional blood flow in the brain. Such analyses have detected deficits in alcoholics, particularly in the frontal lobes, which control numerous cognitive functions, and in the cerebellum, which controls voluntary movements. In addition, PET is a promising tool for monitoring the effects of alcoholism treatment and abstinence on damaged portions of the brain. Finally, PET may be able to help researchers develop new medications targeted at correcting the chemical deficits found in the brains of people with alcohol dependence and alcohol abuse.

About the Author

NIH is the nation's medical research agency - making important medical discoveries that improve health and save lives. The National Institutes of Health (NIH), a part of the U.S. Department of Health and Human Services, is the primary Federal agency for conducting and supporting medical research.