Functional Imaging of Craving : Part 3

by National Institute of Health

(Page 3 of 6)

When a particular brain region is being used, the small arteries that supply the region with oxygen and other nutrients expand. As a result, the amount of oxygenated hemoglobin flowing into that region increases. In fact, the amount of oxygenated hemoglobin flowing into the tissue far exceeds the amount of deoxygenated hemoglobin produced by any increased metabolism in the active brain tissue. Thus, the ratio of oxygenated to deoxygenated hemoglobin - and, therefore, the BOLD signal - increases in the activated tissue, a change that can be measured by fMRI.

fMRI offers many advantages over earlier functional imaging techniques. First, fMRI does not expose subjects to radioactive substances, allowing researchers to study each subject on multiple occasions. Second, investigators can collect numerous images during a single imaging session, thereby allowing statistical analysis of data obtained from a single subject. This ability to analyze data from individual subjects means that the investigator can avoid the reduction in accuracy that occurs when brain images from many subjects are combined. Third, fMRI measurements represent brain activity during only a few seconds, enabling researchers to detect rapid changes in brain activity (enhancing temporal resolution). (PET and SPECT cannot measure brain activity during such a short interval.) Finally, the spatial resolution of fMRI exceeds that of the other modalities.

A Preliminary Model of the Brain Regions Involved in Craving

Modell and colleagues developed the first model of neural systems that may play a role in alcoholism and particularly in craving. The model is based on the observation that the emotions associated with craving (as determined by patients' self-reports) are similar to the compulsions described by patients with obsessive-compulsive disorder (OCD). Because OCD appears to be associated with dysfunction in a brain area called the orbital cortex, Modell and colleagues argued that a similar dysfunction could underlie craving for alcohol. Specifically, the investigators postulated a dysfunction in a type of brain circuit called a striatal-thalamocortical loop. The particular circuit proposed to be affected connects the orbital cortex, ventral striatum (including the nucleus accumbens and portions of the medial caudate nucleus), and thalamus. Failure of this circuit may cause poor control of urges to drink. Striatal-thalamocortical loops are found in all terrestrial vertebrates. In higher mammals, these circuits augment activity in small regions of the cortex that are involved in cognition, emotion, memory, or some specific actions, simultaneously inhibiting other nearby cortical regions. Striatalthalamocortical loops receive nerve signals (input) from a set of cortical regions with related functions. This input serves to monitor the current behavioral context (ongoing behavior, motivation, and memories of similar circumstances) and provides the basis for selecting which small cortical regions, or circuits, need to be activated. In the human brain, multiple variants of these circuits affect different regions of the cortex and function in parallel. The particular circuit proposed to be relevant to alcohol craving involves the orbital cortex.

The orbital cortex is part of the limbic system, the brain system that is thought to deal with emotions. One of the orbital cortex's functions is to evaluate the motivational significance of stimuli that predict reward. In addition, the orbital cortex inhibits actions that are inappropriate for the person's current situation (context-inappropriate actions). Damage to the orbital cortex can lead to antisocial behavior. Thus, it is reasonable to expect that a circuit serving this region may be involved in craving.

Craving Studies in Alcoholics

Although several imaging studies have examined the effects of craving among cocaine abusers, only one such study has successfully induced and assessed craving among alcoholics. In that study, the investigators used SPECT technology to examine changes in cerebral blood flow in nine alcoholics2 who had ingested either water or a low dose of alcohol. The alcohol dose was too small to produce a direct pharmacological effect, but the researchers hypothesized that the dose would induce a desire to drink more alcohol. The subjects, who knew that they were receiving either water or alcohol but did not know which one they were receiving on a given day, always received the water on the first day of scans and alcohol on the second day of scans. Self-ratings obtained following the SPECT scans confirmed that the low alcohol dose did, in fact, induce a desire to drink in most subjects.

To facilitate the analysis, the investigators conducted a region-of-interest analysis in which they measured changes in blood flow only in a few brain regions that they expected to be activated during craving. Those ROIs, which included the right and left caudate nucleus, the thalamus, and the orbital frontal cortex, were chosen based on the neural systems model of alcoholism discussed in the previous sections of this article. Of these regions, only the right caudate nucleus showed a significant increase in blood flow during the craving condition. In addition, the extent of the subjects' self-reported desire for alcohol predicted the magnitude of the increase in caudate blood flow following alcohol administration.

Modell and Mountz interpreted these results as confirming Modell and colleagues' neural systems model of alcoholism. The lack of activation in the other components of the proposed circuit (the thalamus and the orbital cortex), however, weakens that interpretation. Furthermore, the results are suggestive at best, because the study included no nonalcoholic control subjects. Accordingly, it is uncertain whether the increased caudate activation resulted from alcohol administration itself or from the craving elicited by alcohol. The use of control subjects would have enabled researchers to distinguish between the two interpretations; the control subjects presumably would have exhibited similar blood flow changes if these changes had resulted from alcohol's pharmacological effects but would not have experienced changes resulting from craving. Despite these limitations, however, the correlation between craving score and magnitude of blood flow change suggests that the activation of the right caudate nucleus may be related to craving.

Only one other published imaging study has analyzed brain regions potentially associated with craving in alcoholics. In that study, Hommer and colleagues attempted to induce craving for alcohol by injecting 18 hospitalized alcoholics3 and 12 control subjects with m-chlorophenylpiperazine (mCPP), an agent that has been used to induce craving among alcoholics. Widely used in psychiatric research, mCPP has been a probe for the activity of neuronal systems that use the neurotransmitter serotonin to relay signals from one neuron to the next. For this signal transmission, serotonin released by the signal-emitting neuron must interact with docking molecules (receptors) on the surface of the signal-receiving neuron. Several different serotonin receptors exist. The agent mCPP primarily mimics serotonin's effects at one particular serotonin receptor while blocking serotonin's effects at other receptors. Alcohol administration also results in serotonin release and activation of the 5HT2C receptor.

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.