CHICAGO — Hypoglycemic episodes increase the risk of dementia in elderly patients with diabetes, according to the first study to evaluate the association in this older patient population.
The findings suggest that possible benefits of tight glycemic control in the elderly should be weighed against potentially negative consequences on the aging brain, Dr. Rachel A. Whitmer and associates reported in a poster at the annual meeting of the American Academy of Neurology.
“We are just starting to learn that having people very tightly controlled, if they are elderly, may not be the best thing,” she said in an interview.
The analysis looked at 16,806 elderly patients with type 1 and type 2 diabetes from the Kaiser Permanente of Northern California Diabetes Registry. The patients' mean age was 66 years, 55% were male, 60% were white, and 16,667 had type 2 diabetes.
The researchers identified 1,510 patients from this larger cohort who were hospitalized at least once for hypoglycemia from 1998 to 2003. The same patient records were checked again between 2003 and Jan. 15, 2007, and 1,837 patients (11% of the total) had a diagnosis of dementia.
Compared with those with no episodes, patients with at least one hypoglycemic episode had a 45% increased risk of dementia (hazard ratio 1.45), whereas those with at least two episodes had a twofold increased risk (HR 2.27), and those with at least three episodes had a threefold increased risk of dementia (HR 3.52), after adjusting for age, body mass index, race, education, gender, and duration of diabetes.
The effect remained strong after further adjustment of the data for hypertension, stroke, cardiovascular disease and end-stage renal disease (HR 1.42 for at least one event, 1.91 for at least two events, and 2.98 for at least three events); and after final adjustments for glycosylated hemoglobin levels and diabetes treatment (HR 1.36, 1.81, and 2.20 for at least one, two and three events, respectively), said Dr. Whitmer, an investigator with the research division, Kaiser Permanente, Oakland, Calif.
Of the 1,510 patients with reported hypoglycemic events, 589 were receiving only insulin, 446 only oral hypoglycemic agents, 358 both insulin and oral hypoglycemic agents, and 117 were on a diabetes diet only.
To address the potential for reverse causality, in which patients might have early dementia not diagnosed in their charts that could cause them to stop taking their diabetes medications and develop hypoglycemia, the investigators evaluated 929 cases in which dementia was diagnosed only after 2005. The effect of severe hypoglycemia on dementia was even more robust in the 2-year lag model, with hazard ratios of 1.22 (for at least one episode), 1.93 (for at least two episodes), and 2.85 (for at least three episodes), after adjustment for all variables.
“[Elderly diabetics] are a more vulnerable group… we need to have a better understanding of how glycemic control could affect brain health,” said Dr. Whitmer.”