Two forms of dementia that are often confused with each other can be distinguished with a high degree of accuracy using a special positron emission tomography (PET) scan that measures the brain's sugar uptake, according to research by a University of Utah dementia expert.

The scan, called an FDG-PET, involves injecting fluorodeoxyglucose, a short-lived radioactive form of sugar, into a PET scan subject to map activity levels in different parts of the brain. The two types of dementia — Alzheimer's and frontotemporal dementia (FTD) — are often hard to differentiate. But while Alzheimer's affects the back part of the brain, impacting memory and spatial thinking, FTD more often affects the front of the brain and language.

With the FDG-PET scan, six doctors from three national Alzheimer's disease centers, including the U.'s Center for Alzheimer's Care, Imaging and Research, could correctly identify the type of dementia in almost 90 percent of cases — as much as a 14 percent improvement over what they did using clinical diagnostic methods alone. The study appeared online in the journal Brain.

It's important to distinguish between Alzheimer's and FTD — two of the three most common of more than 70 types of dementia — because their natural courses can differ, and so can symptoms. As more treatments are developed — FTD treatment just deals with symptoms, there are few effective treatments for Alzheimer's and there is no cure for either — knowing which type of dementia a person has will be even more important, said Dr. Edward Y. Zamrini, associate professor of neurology and director of clinical trials for the U.'s Alzheimer's center.

"It makes sense to treat the right disease with the right medication," he said.

Alzheimer's typically affects people after age 65, although a subset have early onset. For every five years over age 65, the risk of developing it doubles. FTD can affect people in their 40s and 50s, although it's less common overall than Alzheimer's. And FTD is much more likely to be familial, an important fact for people who want to know their risk.

Areas of the brain that are thick or that have lost connections are not going to use up as much of the glucose, which will be visible on the scan.

The study's lead author, Dr. Norman L. Foster, professor of neurology and director of the U.'s Alzheimer's center, called the scan a powerful tool for early treatment of FTD, which besides striking younger people is marked by behavioral changes and trouble with language. It, too, can take years to develop.

"Early diagnosis of FTD can have a tremendous impact on the treatment for patients and their family members. Many patients are misdiagnosed and may be hospitalized and receive drugs for the wrong disease," Foster said in a release announcing the research. "Accurate diagnosis bypasses the costs, side-effects and frustration of misguided care. Furthermore, one-third of FTD patients have a family history of a similar disorder, and family members need to know if they are at increased risk of the disease."

The scan is meant to be used along with medical history and psychometric testing, the tools now used to diagnose dementia.

Cases of dementia are expected to increase significantly in the future — nowhere faster than the Intermountain West and Utah. One barrier to PET scan use, however, is that not all insurance companies will cover it, although Medicare will to evaluate dementia.

The researchers looked at medical records and the FDG-PET scans of 45 patients who were later autopsied. Microscopic examination found 31 had Alzheimer's and 14 had FTD. The researchers also summarized the clinical course each patient's dementia had followed, then gave the information to neurologists at the centers, each with between 10 and 25 years of experience, who had to decide who had which disease.

Those using only the clinical methods were correct in 76 percent to 79 percent of cases. Scan alone yielded a correct dementia diagnosis in 85 percent to 89 percent of cases. And combining the information from the scan and clinical notes raised the correct diagnosis rate to 90 percent. It also increased the experts' confidence in a correct diagnosis or made them question and sometimes change a wrong one in 42 percent of the cases, the release said.

The study was funded by the National Institute on Aging in the National Institutes of Health.