Recently, several people with OCD have written to me to ask: "Should I get a brain scan?" It's not an uncommon question. Is my brain normal? Would a brain scan help with my diagnosis? With my treatment?
OCD was one of the first psychiatric disorders in brain scans showed evidence of abnormal brain activity in specific regions. The earliest work was from UCLA in the 1980s, by Baxter and colleagues. They used PET imaging to measure brain activity (actually brain metabolism and blood flow, but those are thought to correlate pretty closely with activity). These PET scans produce images of the brain, like this (From Baxter et al, 1987):
The top row of brains here is from an individual without any diagnosis; the bottom row is from a patient with OCD. The images are reconstructed horizontal slices through the brain; the front of the brain is at the top, as if the person was lying on his or her back. The colors correspond to brain activity (if we allow a few technical assumptions that need not trouble us here). The warmer colors - reds and yellows - correspond to brain regions with higher activity; the cooler ones - blues and greens - are brain areas with less activity. (This is at rest - these folks aren't doing anything while their brains are imaged; they're just lying there in the PET scanner.) By comparing the images in the top row with those in the bottom, you can see at a glance that, while the overall patterns are quite similar, a few brain regions markedly more active in the OCD brain. The two most prominent are theorbitofrontal cortex (OFC), at the front of the brain just above the eyeballs, and thecaudate nucleus, a component of the basal ganglia deep within the brain. Remember those terms - we're going to come back to them.
The details here aren't that important for our discussion; but you can see how the cortex (at the top) projects to the striatum (the caudate and related structures), down to the thalamus, and back again to the cortex. It's like a positive feedback loop. And, somehow, this loop is in overdrive in OCD. At least, that's how it looks from the neuroimaging data.
A few decades of subsequent work have refined but largely supported this picture. Additional areas of hyperactivity have been identified, such as the anterior cingulate cortex (ACC), the anterior thalamus, and, in some studies, the insula. Increased activity in these regions has been found to worsen when symptoms are provoked by, for example,showing patients images that trigger their symptoms. Similar patterns have been shown using other brain imaging techniques, such as functional magnetic resonance imaging(fMRI). Hyperactivity in these regions correlates with symptom severity, predicts treatment response, and normalizes with successful medication therapy orpsychotherapy. Very recently, new ways of using fMRI to measure how different parts of the brain are functionally connected to one another have revealed patterns of abnormality in the OCD brain that center on - you guessed it - the OFC and the basal ganglia. (My own group has recently done a study on fMRI measures of functional connectivity in the brain in OCD that I'm rather proud of; but we'll save that for another day.)
It's a little more complicated than that, of course. In the brain, it's a safe bet that things are always more complicated than you think! There are other brain regions involved, too; and the precise pattern of brain activity may be different depending on what type of OCD symptoms an individual person has. Still, the overall picture is clear: there are functional abnormalities in the brains of people with OCD, within the cortex-basal ganglia circuitry, and they almost certainly contribute to symptoms.
Which brings us back to the original, clinical, question: is there any utility in a brain scan for a person with OCD? Will it help clarify diagnosis? Will it help us choose the best treatment?
I dream of the day when I can say 'yes' to those questions. It's a Holy Grail of psychiatry research - to be able to run a brain scan (or any other quantitative, objective test -genetics; blood testing; EEG; whatever) and have it tell us what's going to work for an individual patient. But today? Not so much.
The problem is, even when we are pretty sure there are differences between brain activity in folks with OCD and those without, on average, that doesn't necessarily help us make predictions about a given individual. It's kind of like the relationship of height to sex. On average, men are taller than women. But if I just tell you someone's height, you're not going to be very accurate in guessing their sex. And if I tell you whether someone is male or female, you're not going to be able to guess their height with any great precision. There is an indubitable relationship between height and sex, on average, but it doesn't do you a whole lot of good in an individual case.
Similarly, we're pretty sure that there are relationships between abnormal activity in the OFC, basal ganglia, and other brain structure and the symptoms of OCD, but this doesn't do us much good in an individual case. It's an important scientific insight, but it's not a diagnostic test. The fact is, the vast majority of the time, a brain scan in someone with OCD looks completely normal.
What about outlier cases? If I tell you that someone is 6 foot 8, you're going to have a pretty good guess as to their sex. (You might be wrong, of course; but your guess would be reasonable.) Is there a parallel to this in the brains of individuals with OCD? Is there any particular brain abnormality that might, in an extreme case, provide us with some guidance? Unfortunately, the answer is still 'no'. We just don't know enough yet. If the only thing we know about someone is that their OFC is thin or has a particularly high level of activity, I suppose that might increase the odds that they have OCD. But we know that with much more confidence through a much lower-tech approach: talking to them. And, at least today, that is a far more sure guide to a diagnosis than any brain scan.
It's a frustrating answer: no, getting a brain scan won't help us diagnose you, and it won't help us choose the best treatment. Some day, I hope! But not today.