Combination of Brain Scans and Word-Recall Test Best for Predicting Dementia, but It's Expensive
WebMD Health News
Latest Alzheimer's News
- Sudoku, Crosswords Make Your Brain Years Younger
- Exercise, Healthy Eating Can Reduce Dementia Risk
- 'Robopets' Bring Companionship to Nursing Homes
- Hormone Therapy for Prostate Cancer, Dementia Risk
- Alzheimer's Spread Like an Infection Through Brain
- Want More News? Sign Up for MedicineNet Newsletters!
Reviewed By Laura J. Martin, MD
June 30, 2010 -- About half of older people with memory loss who meet the clinical definition of mild cognitive impairment will develop Alzheimer's disease within five years, but predicting who will and will not progress to dementia remains a challenge.
Genetic, biochemical, and brain imaging analyses all have some value in measuring cognitive decline, but the predictive value of these tests for identifying patients who will progress to Alzheimer's disease has not been clear until now.
In one of the largest studies ever to compare the tests, researchers from the University of California-Berkeley found that word-recollection memory testing combined with PET scans of the brain was best able to predict who would develop Alzheimer's disease.
Study participants whose PET scans and memory tests were abnormal were nearly 12 times more likely to develop Alzheimer's disease than people whose scores on both tests were normal.
"These two tests were highly predictive of progression to Alzheimer's disease in people with mild cognitive impairment," researcher Susan M. Landau, PhD, of the University of California-Berkeley, tells WebMD.
PET Predicts Alzheimer's Progression
Landau and colleagues analyzed data from a large trial known as the Alzheimer's Disease Neuroimaging Initiative, designed to evaluate the various brain imaging tests now being evaluated for diagnosing Alzheimer's disease.
The study included 85 mostly elderly people with mild cognitive impairment, a medical term used to describe a state of memory loss or other cognitive decline that is greater than would be expected with normal aging but has not yet progressed to dementia.
All the study participants had five different tests:
- An "episodic memory test," which involved recalling a list of words
- A blood test to identify a variant of the ApoE gene that has been linked to Alzheimer's disease
- An MRI brain scan to measure the size of the area of the brain associated with memory and learning, known as the hippocampus
- A PET scan of the brain to detect metabolic changes thought to be associated with Alzheimer's
- A spinal tap performed to measure proteins found in spinal fluid that are believed to play a key role in Alzheimer's disease
"These tests have all shown promise for predicting Alzheimer's disease, but they have never been compared to each other before," Landau says.
All Tests Were Predictive
The study participants were followed for nearly two years, during which time 28 received a diagnosis of Alzheimer's disease.
All the tests were found to be significant predictors of progression to Alzheimer's disease, but the combination of word-memory testing and PET scan was more predictive than any single test or other test combination.
The study appears in the June 30 online issue of Neurology. It was jointly funded by the National Institutes of Health and a consortium of pharmaceutical companies.
'PET Cost Will Limit its Use'
Several previous studies have also concluded that PET scanning may be the best single test for predicting progression to Alzheimer's. But the cost of the imaging test will limit its use, neurologist Carol F. Lippa, MD, writes in an editorial accompanying the study.
The cost of a PET scan is anywhere from $3,000 to $6,000 -- about twice the cost of MRI brain scans.
"One wonders if such extensive testing is realistic for clinical care when physicians are struggling to get their patients' basic medications covered by insurance," she writes.
Lippa tells WebMD the goal is to develop less costly predictors of progression to Alzheimer's that are as good as or better than PET.
"We know this is a really good test," she says. "But we also know that it is way too expensive to use to screen everybody over a certain age. The main value of PET may be as a reference to compare to new strategies."
Health Solutions From Our Sponsors
Susan M. Landau, PhD, research scientist, University of California-Berkeley and Lawrence Berkeley National Laboratory, Berkeley, Calif.
Carol F. Lippa, MD, department of Neurology, Drexel University College of Medicine, Philadelphia.
News release, American Academy of Neurology.
Grumman, M. Archives of Neurology, 2004; vol 61: pp 56-66.
©2010 WebMD, LLC. All Rights Reserved.