When it comes to a neuroanatomical understanding of psychiatric symptoms, OCD is considered the superstar of the field. A wealth of data has implicated the involvement of a specific neuroanatomical loop in OCD. This loop begins in the frontal lobe (primarily in the orbital portion of the frontal lobe which sits just above the eyes), moves to the cingulate gyrus (a fold tucked away in the medial frontal lobe), then plunges down into subcortical areas, including the caudate nucleus (part of the basal ganglia of Parkinsonism fame), and the thalamus (the sensory switchboard).

If you find yourself puzzled by why the basal ganglia, a control center for movement, is associated with OCD, you're not alone. As yet there is no clear explanation, but there is clearly a connection between OCD and movement disorders. For example, the post-encephalitis Parkinson's patients made famous by Oliver Sacks in Awakenings had movement problems, and often exhibited compulsive rituals as well. Similarly, 70% of patients with the motor and phonic tics of Tourrette's Syndrome also have OCD symptoms.

OCD is probably a disease of the “frontostriatothalamic loop.” SSRIs may work best in those whose caudates are particularly revved up.

The most riveting evidence for this "frontostriatothalamic loop," as it has been called, is far more recent, and comes from PET (positron emission tomography) and fMRI (functional magnetic resonance imaging) studies, techniques which we reviewed in the June 2004 issue of TCR. The landmark study of neuroimaging in OCD, published in Archives of General Psychiatry in 1987, was a PET study of 14 patients with OCD who were compared with control groups of both normal subjects and depressed patients. (Recall that PET technology allows you to measure metabolic activity in specific brain regions.) These investigators discovered that OCD patients were hypermetabolic in the left orbital frontal gyri and in both sides of the caudate nuclei (Arch Gen Psychiatry 1987; 44:211-218). Five years later, this same group of researchers published another paper, to even greater acclaim, showing that treatment--whether with SSRIs or CBT (cognitive-behavior therapy)-- reduced the caudate hypermetabolism toward normal levels (Arch Gen Psychiatry 1992; 49:681-689). This led to one of those feel-good moments in psychiatry when psychopharmacologists and therapists looked at one another with mutual respect.

Once researchers had worked out some of the basics of OCD neurobiology, they began to focus on distinguishing the brain correlates of specific OCD symptoms. Since it is hard to locate a large group of OCD patients who report only a single type of symptom (e.g., pure washers or pure checkers), researchers have taken to using methods of provoking specific obsessions while scanning OCD patients' brains. In a recent study, researchers exposed patients to pictures of dirty bathrooms (for washing obsessions), open car doors (for checking), and old newspapers and magazines (for hoarding). Patients were asked to imagine, for example, that they were in the dirty bathroom and were not able to wash. Their brains were then imaged using fMRI (Arch Gen Psychiatry 2004; 61:564- 576.)

In the washing experiment, brain regions most activated were those involved in the processing of emotion and disgust--the anterior cingulate and orbitofrontal gyri. In the checking experiment, as you might expect, the most active regions were those related to paying attention and reflexive bodily movements--parts of the thalamus, the basal ganglia, and the cortical motor strip.

What about the hoarders? In these patients, the results weren't as neatly distinct, with hypermetabolism in a variety of frontal regions. In fact, hoarding may be a different disease altogether, as is hinted at by the fact that these patients are devilishly hard to treat successfully. A recent study used PET scanning to evaluate the neural correlates of hoarding (Am J Psychiatry 2004; 161:1038-1048). In this study, 45 OCD patients were scanned: 12 of them had primarily hoarding compulsions, while the remaining 33 reported the standard array of symptoms. While the non-hoarders showed the wellestablished pattern of hypermetabolism, especially in the basal ganglia and the thalamus, the hoarders showed decreased metabolic activity in various areas, most conspicuously in the cingulate gyrus portion of the frontal lobe.

What's so special about the cingulate cortex? In most studies, OCD patients are overly revved up in this area, and this may account for the dramatic success of the cingulotomy, a neurosurgical procedure used for the most treatment-refractory cases of OCD. It works about 50% of the time, with few cognitive side effects (Acta Psychiatr Scand. 2003; 107:283-90). It is thought to work by interrupting the obsessive loop, which must pass through the cingulate cortex as it travels from the frontal lobe to the caudate. Hoarders tend to respond poorly to cingulotomies, which is not surprising given their hypometabolism in this region.

Interesting stuff, to be sure, but does it help us make treatment decisions while we are sitting in the office with our patients? Maybe. One study, for example, found that PET was helpful in predicting which patients with OCD and depression would respond to Paxil (paroxetine). Hypermetabolism in the right caudate predicted response among OCD patients, while lower hypometabolism in the amygdala and thalamus predicted response in depression (Am J Psychiatry 2003; 160:522-532).

Stay tuned--we're getting there!

TCR VERDICT
The OCD loop: Frontal lobe to striatum to thalamus and back again.