Radiology Feature: MRI may visualize cognitive decline
 |
Patients with mild cognitive impairment present microbleeds at baseline, independent of subsequent cognitive decline, as well as an altered iron distribution in subcortical nuclei, contrary to healthy control subjects, explained Sven Haller, MD, of the department of diagnostic and interventional neuroradiology at the University Hospitals of Geneva, and colleagues.
“Not all patients with mild cognitive impairment develop Alzheimer's disease, nor do they decline at identical rates,” Haller told Health Imaging News. “A substantial proportion of patients remains stable for several years or can even improve, so you want to be able to identify those who will progress and, ideally, at an early stage.”
The researchers recruited 38 women and 31 men, including 35 healthy control subjects (average age, 63.7 years) and 69 patients (average age, 65 years) with mild cognitive impairment for the study, and baseline visits began between July 2007 and September 2008. The 69 patients with mild cognitive impairment underwent neuropsychologic testing, which was repeated after a follow-up period of one year on 67 of the patients in order to determine if their disease was stable or progressing. At baseline, each patient underwent susceptibility-weighted MR imaging for the observation of blood vessels in the brain and the presence of microbleeds.
The scans were analyzed independently by two neuroradiologists with five and six years of experience, respectively, who were blinded to each other’s findings. Iron deposition in deep gray matter was also assessed. Individual classification of mild cognitive impairment was analyzed by using an artificial intelligence technique, a support vector machine.
At the one-year follow up, the authors determined that 40 of the 67 patients with mild cognitive impairment had stable disease, presenting with microbleeds in 33 percent of patients and 27 had progressive disease, with microbleeds in 54 percent of patients. Additionally, 14 percent of the control participants had microbleeds.
At baseline, the number of microbleeds was significantly higher in the mild cognitive impairment group than in the control group, but was not found to differ between the patients with stable and those with progressive mild cognitive impairment, the authors offered. However, iron concentration via the support vector machine helped to discriminate patients with mild cognitive impairment from the healthy control subjects (accuracy, 84 percent; sensitivity, 89 percent; and specificity, 85 percent) and those with stable from those with progressive mild cognitive impairment, they wrote.
The researchers found no significant difference in age or sex between the control, stable mild cognitive impairment and progressive mild cognitive impairment groups.
“Analysis of iron deposition at baseline performed with a support vector machine might help identify individual patients with mild cognitive impairment at risk for cognitive decline,” explained Haller, who noted that because every patient undergoes MRI for dementia work up, combining the support vector machine method and MRI can offer better discrimination between stable and progressive disease groups. “This is an interim result,” he said. “The next step in my opinion is to combine modalities.”
Eventually, the identification of patients with stable cognitive impairment and those with progressive disease may help physicians in medication administration, because due to the expense and side effects of the medications, it should not be administered to every patient, noted Haller. “If we could tell if someone is at risk for progression or will remain stable, or if the decrease in memory function is different from normal cognitive decline associated with aging, we could then discuss the application of Alzheimer’s-related medications,” he said.