Brain Variation May Predict OCD Surgery Success

Deborah Brauser

December 24, 2014

How well a patient responds to a surgical procedure for refractory obsessive-compulsive disorder (OCD) may depend on whether or not they have certain neuroanatomic brain variations in structure and connectivity, new imaging research suggests.

Dr Sameer Sheth

A small retrospective study showed that patients with severe OCD who responded to dorsal anterior cingulotomy were more likely to have less gray matter signaling in the right anterior cingulate cortex than nonresponders. Response was defined on the basis of OCD score changes.

In addition, treatment response was significantly correlated with connectivity between the eventual lesion and the caudate, putamen, thalamus, pallidum, and hippocampus ― especially on the right side.

"We're beginning to understand more about the brain networks that are involved so that we can do a better job of predicting, even before we do a procedure, who might respond," principal investigator Sameer A. Sheth, MD, PhD, assistant professor of neurosurgery at Columbia University Medical Center and from New York Presbyterian Hospital, in New York City, told Medscape Medical News.

"This is important because when you're thinking about undergoing a surgical procedure for any condition, you want to know what the chances are of actually benefiting," said Dr Sheth, adding that this may eventually lead to individualizing surgical therapy to a patient's symptoms.

The study was published online December 23 in JAMA Psychiatry.

Mechanism for Response Variability?

The investigators note that in roughly 10% to 20% of patients, OCD symptoms do not respond to pharmacologic or cognitive-behavioral therapies. Neurosurgical treatments have been shown to be successful in some but not all of these patients.

"The mechanisms underlying this response variability are poorly understood," they write. The surgical approach itself is a "highly consistent, stereotyped procedure." So they sought to examine whether preoperative imaging could help identify markers for response to surgical treatment.

The researchers retrospectively examined preoperative T1 and diffusion MRI scans from 15 adults (60% men; mean age, 37 years) who underwent dorsal anterior cingulotomy at Massachusetts General Hospital to treat refractory OCD.

The procedure "involves lesioning the dorsal anterior cingulate cortex (dACC), a region believed to play a role in the pathogenesis of the neural network that causes OCD," write the investigators.

There were eight responders to the procedure and seven nonresponders, as measured on the Yale-Brown Obsessive-Compulsive Scale (Y-BOCS). In addition, voxel-based morphometry (VBM) and diffusion tensor imaging were used to identify structural and connectivity brain variants.

Results from VBM showed that a poor response to treatment was significantly correlated with signal strength in a gray matter cluster centered in the right anterior cingulate cortex (P = .017). On the flip side, decreased gray matter volume in this region predicted improved response.

The mean gray matter partial volume in this region for responders was 0.47 vs 0.66 for nonresponders. There was also a mean decrease in OCD scores on the Y-BOCS of 60% vs 11% for each group, respectively.

During white matter analysis, the investigators found that responders showed significantly different connectivity between the eventual lesion and the thalamus, putamen, pallidum, and hippocampus vs the nonresponders (P = .001 for all).

"Greater right-side connectivity between the lesioned areas in the dACC and these 4 target areas correlated with better response to cingulotomy," write the investigators.

Interestingly, when the analysis was repeated using a standardized lesion area, so that evaluations could also be performed before surgical lesions were created, "4 seed-target [laterality metric] pairs" again showed significant differences between the patient groups in the thalamus (P = .001), pallidum (P = .002), and hippocampus (P = .002), as well as in the caudate nucleus (P = .008), but not in the putamen. And, again, greater connectivity on the right side predicted better response on the Y-BOCS.

Overall, "these variations may allow us to predict which patients are most likely to respond to cingulotomy, thereby refining our ability to individualize this treatment for refractory psychiatric disorders," write the investigators.

"In general, understanding more about how the brain works is important. We realized a while ago that psychiatric disease is just like neurological disease and based in the brain. Better understanding of brain circuits should lead to better treatment," added Dr Sheth.

"Serious Business"

"Lesioning the brain is serious business," Odile A. van den Huevel, MD, PhD, from the Department of Psychiatry at VU University Medical School in Amsterdam, the Netherlands, and from the OCD Team at Haukeland University Hospital in Bergen, Norway, writes in an accompanying editorial.

"The risks must be placed in the context of alternatives and balanced against the risks of no treatment," adds Dr van den Huevel.

She notes that the current study's findings warrant replication in a larger trial and called the predicted value remarkable. "Precise targeting will benefit the outcome of neurosurgery." However, Dr van den Huevel writes that the retrospective design of this and other recent studies is a limitation.

"The burning issue remains whether these brain imaging markers can predict treatment response prospectively at the individual level."

The editorialist notes that a consensus guideline on neurosurgery in psychiatry has recently been proposed by representative members of international psychiatric and neurosurgical societies.

"The consensus guideline...stresses the importance of being careful not to prematurely designate an investigational intervention as the standard of care. The field may benefit from small pilot studies to optimize the targets for surgical interventions," she writes, adding that comparative studies are urgently needed.

"If reliable predictive markers are identified, invasive ablative treatments might be offered only to patients with predicted good outcome, thereby preventing unnecessary costs and iatrogenic damage in the remaining patients," she concluded.

The study authors have reported no relevant financial relationships. Dr van den Heuvel reports no conflicts with regard to this editorial but has received speaker honorarium from Lundbeck and research grants from a number of companies and organizations, including PhotoPharmics and Parkinson Vereniging. A full list can be found in the original article.

JAMA Psychiatry. Published online December 23, 2014. Abstract, Editorial


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: