What Functional MRI Finds with Anorexia

In yesterday’s post, functional MRI showed how rapidly anti-TNF agents can improve pain response in patients with Crohn’s disease.  A more complete description of this study is available from the AGA Blog: This is Your Brain on Anti-TNF Therapy. This link also includes access to a video abstract discussion with the author.

Another intriguing use of this technology provides insight into why Anorexia is difficult to treat.  The study was summarized in the NY Times:

Anorexia May Be Habit, Not Willpower  Here’s an excerpt:

The study’s findings may help explain why the eating disorder, which has the highest mortality rate of any mental illness, is so stubbornly difficult to treat. But they also add to increasing evidence that the brain circuits involved in habitual behavior play a role in disorders where people persist in making self-destructive choices no matter the consequences, like cocaine addiction or compulsive gambling

The researchers used a brain scanning technique to look at brain activity in 21 women with anorexia and 21 healthy women while they made decisions about what foods to eat…

As expected, both the anorexic and the healthy women showed activation in an area known as the ventral striatum, part of the brain’s reward center. But the anorexic women showed more activity in the dorsal striatum, an area involved with habitual behavior, suggesting that rather than weighing the pros and cons of the foods in question, they were acting automatically based on past learning…

My take: This study shows an association between food selection and differences in brain activity between women with anorexia and in controls.  These changes do not prove a causal association but provide an important piece of information about what might be going wrong.

Atlanta Botanical Garden, Bruce Munro Exhibit

Atlanta Botanical Garden, Bruce Munro Exhibit

Seeing is Believing

As noted in a recent blog (Food Marketing Detectable on Functional MRI | gutsandgrowth), functional MRI is being studied for a number of applications.  Now, more data has emerged that a “pain signature” can be identified with this technology (NEJM 2014; 368: 1388-97).

Using a series of experiments, the authors enrolled 114 healthy participants and ultimately identified an imaging signature that was associated with heat-induced pain and increased nonlinearly with increasing stimulus intensity.  The first part of the study involved a machine-learning analyses after inducing physical pain by applying heat to the forearm of the participants.  The sensitivity and specificity were 94% or more in discriminating painful heat from nonpainful warmth, pain anticipation, and pain recall.  In the fourth part of the study, the authors showed that the signature response was reduced when an opiod analgesic (remifentanil) was administered.

Because this study enrolled otherwise healthy patients, the results cannot be extrapolated to other populations.  Nevertheless, it is likely that other painful conditions will have unique functional MRI signatures.

Pain is not easy to ascertain and obtaining functional MRIs is not likely to have a role in the near future as a clinical tool.  The concept of identifying a measurable pain biomarker though has been strengthened by this study.

Related blog entry:

Pain changes brain | gutsandgrowth