Patients With Type 2 Diabetes More Likely To Have Signs of Cognitive Impairment

SAN FRANCISCO, CA - Women with type 2 diabetes are about 30 percent more likely than non-diabetics to show cognitive impairment in short-term memory tasks, and the risk increases in patients with untreated disease or in those who have had diabetes for 15 years or more, according to a new study presented here in April at the AAN Annual Meeting.

The risk is subtle but there. Cognitive impairment is an intermediate step between normal cognitive function and dementia, and it's important for physicians to take note of it, said principal investigator Giancarlo Logroscino, MD, in a phone interview following his presentation. Dr. Logroscino is a neuroepidemiologist and Associate Professor of Epidemiology at Harvard Medical School in Boston, MA.

We need to include cognitive evaluations in the assessment of women with type 2 diabetes, particularly in women who have had the disease for at least 15 years. For example, he said, it may be worthwhile to consider testing women with type 2 diabetes in their 60s who complain of losing car keys, forgetting appointments, and struggling to remember the names of familiar objects.

One encouraging finding from the study was that women receiving oral therapy for diabetes have risks that are comparable to those without diabetes. It may be that diabetes treatment is important not only for metabolic health but for cognitive health as well, Dr. Logroscino said.

DATA FROM NURSES' HEALTH STUDY

In this study funded by the NIH, Dr. Logroscino and his co-investigators sought to identify whether there was a link between type 2 diabetes and poor cognitive performance and cognitive decline over a two-year follow-up period.

He noted that type 2 diabetes affects older women disproportionately and is a stronger risk factor for cardiovascular disease in women than men. He added that few studies have ascertained whether different diabetic treatments ameliorate the negative effect that type 2 diabetes has on cognitive function.

Figure. Dr. Giancarl...

Investigators contacted 18,999 women who participate in the Nurses' Health Study, an ongoing prospective cohort of US women who are registered nurses. The study has included biennial questionnaires that capture information on type 2 diabetes and numerous covariates. From 1995 to 2001, Dr. Logroscino and his co-investigators conducted interviews by phone to assess cognitive function. The 18,999 subjects were 70 to 81 years old, and 90 percent of them responded to follow-up interviews two years later. The tests were designed to assess such cognitive functions, as verbal recall, attention, and short-term memory.

After making a multivariate adjustment, the investigators found that the diabetic women were 25 to 35 percent more likely than non-diabetic women to have a poor baseline score on several tests, which they defined as the lowest 10th percentile. Women who had had diabetes 15 years or more were 50 percent more likely to have a poor baseline performance. Those who had had any metabolic complications, such as ketoacidosis or coma, were 2.5 times more likely to have a poor baseline score. Even if those complications are transient, they have implications for cognitive capacity for many years, said Dr. Logroscino.

Figure. Dr. Charles ...

BENEFIT OF DIABETES MEDICATIONS

Women who reported that they were taking no medication for their diabetes were 1.7 times more likely to have a poor baseline performance than were non-diabetics. However, women who did receive such treatment had similar risks as non-diabetic women, Dr. Logroscino said.

The investigators found that these risks were similar in the two-year follow-up interviews. Women with diabetes were 26 percent more likely to have experienced substantial cognitive decline, which the investigative team defined as the worst 10th percentile of the distribution of decline. Those with diabetes for 15 years or more were 64 percent more likely to have such levels of decline.

EXPERTS COMMENT

Commenting on the study by phone, one expert in Alzheimer disease, who was not involved with the study, noted that the baseline cognitive problems and subsequent declines that the investigators documented were subtle and did not constitute full dementia. The take-home message is that glucose control is important for cognitive health, said Charles DeCarli, MD, Professor of Neurology at the University of California-Davis, where he is Director of the Alzheimer's Disease Center and Imaging of Dementia and Aging Laboratory.

The challenge is to find at-risk people and treat them before the symptoms of decline occur, he said. The affected patients may not be people who come into your clinic complaining of cognitive impairment, he said. The neurologist in the clinic probably won't be asked to evaluate these people.

He suggested that neurologists evaluate diabetics with cognitive problems as thoroughly as they would other patients. If someone comes in with diabetes and high glucose and has cognitive impairment, it may be the consequence of long-term poor glucose control and therefore too late to reverse, he said. Don't get sucked into thinking that all it takes is to get the glucose under control. At the same time, neurologists should not think, because a patient has high glucose and clinical dementia, that you just need to fix the glucose. You still need to identify the cause of the dementia in order to manage the patient properly, he said.

This study shows that we need to be aware of the cognitive risks of diabetes and treat it before these patients develop cognitive problems, Dr. DeCarli said. He added that he would like to see the researchers continue to follow the women who show decline and see how they fare, and also to treat those with untreated diabetes and see if there are any changes due to treatment. It may be that these subtle impairments could reverse early on before permanent damage sets in, he said.

ARTICLE IN BRIEF

✓ Using the Nurses Health Study as a population base, Dr. Giancarlo Logroscino and colleagues reported that women with type 2 diabetes are about 30 percent more likely than non-diabetics to show cognitive impairment in short-term memory tasks, and that women who had had diabetes 15 years or more were 50 percent more likely to have a poor baseline performance on tests.