Hypoglycemic episodes increase dementia risk in seniors with type 2 diabetes

Hypoglycemic (low blood sugar) episodes that require a visit to the hospital are associated with increased risk of dementia in elderly patients with type 2 diabetes mellitus, according to a Kaiser Permanente Division of Research study in the Journal of the American Medical Association.

Though not a randomized clinical trial, this study provides additional information for the debate about how tightly blood sugar should be controlled in patients with type 2 diabetes, particularly in elderly patients, say the researchers. The study appears in the current diabetes mellitus themed issue of JAMA.

While several studies have shown low blood sugar to affect cognitive function in children with type 1 diabetes, this is the first study to evaluate the association in older patients with diabetes.

"We know that the brain becomes more vulnerable with age, and we need a better understanding of how glycemic control can affect brain health over the long term," said the study's principal investigator, Rachel A. Whitmer, Ph.D., a research scientist with the Kaiser Permanente Division of Research in Oakland, Calif. "This study adds to the evidence base that perhaps we should rethink the notion of very tight glycemic control for our elderly patients with diabetes mellitus."

The study looked at 16,667 elderly patients with type 2 diabetes from the Kaiser Permanente Northern California Diabetes Registry. The average age for the study population was 66 years; 55 percent were male and 60 percent were white. The researchers identified 1,465 patients in the cohort that were hospitalized or had an emergency room admittance at least once for hypoglycemia from 1980 to the end of 2002.

Compared to those with no hypoglycemic episodes, patients with one hypoglycemic episode had a 45 percent increased risk of being diagnosed with dementia after 2003. Those with two episodes had a 115 percent increased risk, and those with three or more episodes had a 160 percent increased risk of dementia. Adjustments were made for age, body mass index, race, education, gender, and duration of diabetes. The effect remained after further adjustments for hypertension, stroke, cardiovascular disease, end-stage renal disease, glycosylated hemoglobin levels, and treatment for diabetes.

Researchers were concerned that cognitive impairment from early dementia could be contributing to an increased likelihood of having a hypoglycemic episode. To address this, the researchers introduced various "lag" periods between the occurrence of hypoglycemic episodes and the initial diagnosis of dementia. In effect, researchers considered only hypoglycemic episodes in midlife (at an age where dementia is virtually impossible). Even when only considering hypoglycemic episodes in midlife (mean age 52), there remained a significant association between hypoglycemic episodes occurring before 1985 and onset of dementia in 2003 or later. Compared to those with no hypoglycemic episodes, those with one or more episode had a 32 percent greater risk of dementia (adjusted for age, body mass index, education, gender, duration of diabetes, co-morbidities, glycosylated hemoglobin and treatment for diabetes).

"Our findings suggest that pursuit of 'tight' glycemic control (i.e. to hemoglobin A1c levels less than 7 percent) may be inadvisable in older patients with type 2 diabetes if required treatment is causing hypoglycemia," explained Joe Selby, MD, MPH, a co-author on this study and the director of the Kaiser Permanente Division of Research.

Funded by the National Institutes of Health, this study adds additional information to a growing evidence base on this topic. Three recent trials -- the ACCORD Study, the ADVANCE Study and the VADT Study -- have shown clearly that tight glycemic control does not benefit all patients and may cause harm to some, especially to patients with more longstanding diabetes, with previous cardiovascular disease, and possibly to those with a history of severe hypoglycemia, Selby said. For these patients, the harms of tight glycemic control, in the form of cardiovascular disease and death, may outweigh any possible benefits, he added.

In view of these three trials, recent recommendations from the American Heart Association and the American College of Cardiology have urged that treatment be individualized and that caution be exercised to prevent severe hypoglycemia. The guidelines further emphasize the critical importance of blood pressure control, lipid or cholesterol control, smoking cessation, and use of aspirin, said Selby.

Additional researchers on the Kaiser Permanente study include: Andrew Karter, Ph.D., and Charles P. Quesenberry Jr., Ph.D., both with the Kaiser Permanente Division of Research; and Kristine Yaffe, MD, with geriatric psychiatry at the UCSF-affiliated Veterans Affairs Medical Center.

About the Kaiser Permanente Division of Research (http://www.dor.kaiser.org/)

The Kaiser Permanente Division of Research conducts, publishes, and disseminates epidemiologic and health services research to improve the health and medical care of Kaiser Permanente members and the society at large. It seeks to understand the determinants of illness and well-being and to improve the quality and cost-effectiveness of health care. Currently, DOR's 400-plus staff is working on more than 250 epidemiological and health services research projects.