WASHINGTON — Cognitive changes in Parkinson's disease can be seen on positron emission tomography imaging and correlate well with psychological tests, according to data reported in a poster at the World Parkinson Congress.
Dr. David Eidelberg, director of the Center for Neurosciences, and his colleagues at The Feinstein Institute for Medical Research at the North Shore-Long Island Jewish Health System in Manhasset, New York, studied 47 Parkinson's disease (PD) patients (31 men and 16 women) using 18fluorodeoxyglucose PET imaging to identify metabolic patterns associated with cognitive function in PD. In addition to affect, four domains—memory/verbal learning, attention/executive function, visuospatial function, and general cognitive function—were assessed with neuropsychological testing. “We looked for patterns in the brain that correlated with their psychologic performance,” Dr. Eidelberg said.
The patients were an average age of 58 years and had PD for an average of 12 years. General cognitive function was assessed using the Mini-Mental State Examination. PET scans were analyzed using network analysis that isolates different aspects of neural circuits that correlate with a person's cognitive function. Analysis revealed a significant pattern of covarying metabolic reductions in the parietal cortex, anterior cingulate area, and medial frontal lobe (see illustration in which the left and right panels represent the right and left hemispheres, respectively). This pattern correlated negatively with performance on the California Verbal Learning test, Stroop test, digit symbol test, and Hooper Visual Organization test, and positively with the Trail Making test, but not with affect. The researchers also validated this PD cognitive-related pattern (PDCP) in 21 patients with PD, who were scanned twice over a 2-month period. Comparison of the test-retest results showed that PDCP was highly reliable as a predictor of psychological performance. “What makes this appealing is that there is a way to measure cognitive function indirectly,” Dr. Eidelberg said.
The researchers also investigated two clinical applications for the technique. First, they computed the PD-related pattern (motor function) and PDCP in 15 early-stage patients, who were scanned at baseline and at 2 and 4 years. PDCP expression was significantly elevated at the third time point with respect to both baseline and the second time point. “The motor pattern is higher typically and progresses like a straight line. The PDCP starts slower and may not be a straight line in terms of its evolution,” he said.
In PD the motor and cognitive changes have different time courses. Cognitive changes have a later evolution in the course of the disease. “Treating the motor component of Parkinson's disease, which is what we all do, does not appear to really do much for cognition and at times makes it worse,” Dr. Eidelberg said.
The researchers also prospectively computed PDCP expression for PD patients on and off treatment during consecutive days. “While the levodopa and deep brain stimulation were very helpful in the elevated PD-related pattern network, the same interventions in the same people had no real effect on PDCP,” he said. This finding implies that new therapies are needed to treat cognitive changes.
The data validate PDCP as a stable and reproducible imaging marker of cognitive function in PD. Unlike the PD motor-related pattern, the nonmotor pattern evolves slowly over time and its expression is not altered by therapeutic interventions targeting the motor manifestation of PD.
Metabolic decreases in the prefrontal/frontal cortex (in green; Brodmann areas 9, 46) and parietal cortex (Brodmann 7, 39, 40) typify PDCP for mild/moderate cognitive impairment. Courtesy Dr. David Eidelberg