MEG May Provide Objective Diagnosis of Posttraumatic Stress Disorder
By Janis C. Kelly
Medscape Medical News
January 27, 2010 — Magnetoencephalography (MEG) may have the potential to objectively diagnose posttraumatic stress disorder (PTSD) — something conventional brains scans, including computed tomography and magnetic resonance imaging, have failed to do.
New research, led by Apostolos P. Georgopoulos, director of the Brain Sciences Center, Veterans Affairs Medical Center, Minneapolis, Minnesota, suggests that MEG can successfully differentiate PTSD patients from healthy control subjects.
Dr. Georgopoulos told Medscape Psychiatry that the results of this study are likely to change basic understanding of PTSD. "First, it establishes PTSD as a brain disease with a biological substrate. Then, with work on [validation], it can bring new understanding, guidance, and prediction to several PTSD dimensions," he said.
However, independent imaging and PTSD experts are taking a more cautious approach to the study findings, saying it is premature to declare MEG as a definitive diagnostic test for PTSD.
The study was published online January 20 in the Journal of Neural Engineering.
Robust Differences
Potential subjects were identified through a clinical records review of US veterans to identify those with a likely current PTSD diagnosis using a structured clinical diagnostic interview. A total of 74 veterans with PTSD who had served in World War II, Vietnam, Afghanistan, or Iraq and a control group of 250 healthy controls recruited from the general public participated in the study.
Imaging was performed while subjects engaged in a fixation task: fixing their eyes on a spot about 645 cm in front of them for 60 seconds. The results of this synchronous neural interactions (SNI) test, which assesses the functional interactions among neural populations, were then compared in the PTSD and control groups.
The researchers report that they were able to differentiate PTSD patients from healthy control subjects with more than 90% accuracy and that there was a "significant association found between the certainty of PTSD patient classification based on the SNI and their symptom severity."
"Altogether, these findings document robust differences in brain function between the PTSD and control groups that can be used for differential diagnosis and which posses the potential for assessing and monitoring disease progress and effects of therapy," the study authors write.
If so, this technique would represent a breakthrough in the search for a biomarker for PTSD, but other experts in PTSD imaging are more cautious.
Caution Warranted
Commenting on the findings, Iris-Tatjana Kolassa, MD, University of Konstanz, Germany, who has also used MEG-based source imaging to map abnormal brain activity in survivors of torture and war with PTSD, said the findings add to the literature but cautions that further research needs to be done to demonstrate its utility in diagnosing the condition.
"We know from a number of studies, including those from our own lab, that measures of neural oscillations and synchrony are sensitive indicators of the brain's architecture, also in relation to mental disorders," Dr. Kolassa told Medscape Psychiatry.
Both Dr. Kolassa and Alexander Neumeister, MD, Mount Sinai School of Medicine, New York City, who has also commented on the study, were concerned by the composition of the subject and control groups.
"The control group was recruited from the general public, not from the same population as the patients. Quite a number of factors therefore differentiate the 2 groups. These might range from general education, socioeconomic status, to patient status, etc. Thus, the observed differences may not arise from PTSD but from unspecific variables or from frequently comorbid symptoms, such as depressive symptoms, somatization, and especially drug abuse or history of medication.
"Many variables differentiate between healthy and sick people, and it may not be too difficult to differentiate such variations with a post hoc measure," said Dr. Kolassa.
Dr. Neumeister, who recently reported on brain function differences in patients with PTSD and comorbidities vs PTSD alone, said the study provides "interesting preliminary data but not more than that."
He noted that it was a small study and that interpretation is complicated by the lack of information on medications subjects may have been taking.
"Many of these were older patients, who might be expected to have hypertension, cardiovascular problems, diabetes, or other conditions requiring treatment. Since we know that even just drinking coffee vs not drinking coffee can affect MEG, it would be important to account for medications," he said.
"PTSD also goes along with depression and other problems, and there is no mention of whether these patients were receiving treatments, such as selective serotonin reuptake inhibitors, which completely mess up MEG readings," he added.
"This paper has some limitations. It is interesting, but people should not conclude that this is a way to diagnosis PTSD," Dr. Neumeister concluded. He also would like to see this study replicated by researchers who have no financial interest in the outcome.
Dr. Georgopoulos remains optimistic that the study has documented "robust differences in brain function between the PTSD and control groups that can be used for differential diagnosis and that possess the potential for assessing and monitoring disease progression and effects of therapy." His group has previously reported similar MEG "signatures" for other brain diseases, such as Alzheimer's disease and multiple sclerosis. They are moving forward with larger studies in PTSD.
The study was supported by the US Department of Veterans Affairs and the American Legion Brain Sciences Chair. Dr. Georgopoulos has a financial interest in the technology used in this study. Dr. Neumeister has disclosed financial relationships with Eli Lily, Pfizer, and Ortho-McNeil-Janssen Pharmaceuticals. Dr. Kolassa has disclosed no relevant financial relationships.
