Detecting Subtle Brain Injuries

Continued from page 1

A second variation of MRI, known as magnetic resonance spectroscopic imaging (MRSI), can analyze the spectral frequencies of chemicals in the body. Andrew Maudsley and his colleagues at the University of Miami have used new advances in MRI technology, including higher-power magnets, to develop MRSI methods that can measure concentrations of two chemicals in the brain: n-acetylaspartate (NAA), a marker of white-matter density, and choline, which has been linked to injury. Previous MRSI methods yielded information only about specific brain regions, but the new technique can measure chemical concentrations across the whole brain. The researchers found decreases in NAA, possibly due to damaged axons, and increases in choline in a group of 25 patients with traumatic brain injury. "We see widespread metabolic changes, even in those with the mildest injuries," says Maudsley, who presented the work at the conference.

A third study presented at the conference found that changes to slow-wave activity, which have been previously linked to traumatic brain injury, are likely caused by damage to the white matter. Mingxiong Huang and his colleagues used magnetoencephalography (MEG), which measures the magnetic fields produced by the electrical activity of nerve cells, to pinpoint the source of abnormal brain activity, and they discovered that it often overlapped with the location of damage detected using DTI.

While the research is promising, moving these new technologies into clinical use is likely to be a challenge. "The bar for clinical diagnosis of individual patients is different than for measuring a group effect," says David Moore, a neurologist at Walter Reed Army Medical Center. Physicians would need to be able to detect brain changes characteristic of injury on an individual level.

Both DTI and MRSI can be performed using most standard MRI machines, but they require much more extensive data analysis than most medical imaging, something that radiologists aren't used to providing. "It is computationally and analytically intensive," says Brody. MEG, which is used to pinpoint seizures in epilepsy patients, is even more complicated, and the machines are still quite rare in clinical centers.

In addition, it's not yet clear how soon after injury these approaches can identify patients likely to suffer long-term problems. While no protective treatments for brain injury yet exist, they are under development, and they would need to be delivered immediately.