BOSTON – Older adults with type 2 diabetes who were taking metformin had a significantly lower 5-year risk for dementia than did peers who were taking other oral antidiabetic agents, investigators reported at the Alzheimer’s Association International Conference 2013.
A retrospective study of nearly 15,000 adults aged 55 years and older who started on single-agent drug therapy for type 2 diabetes between 1999 and 2001 showed that 5 years after starting therapy, patients who started on metformin had a 23% lower 5-year risk for dementia compared with patients who started on a thiazolidinedione (TZD) such as rosiglitazone (Avandia), reported Rachel Whitmer, Ph.D., an investigator in the research division of Kaiser Permanente Northern California, Oakland.
Metformin users also had a 20% lower risk for any dementia compared with those on a sulfonylurea such as glyburide. Among patients with a diabetes duration of less than 5 years, metformin users had a 40% lower dementia risk than did sulfonylurea users, and among those with a diabetes duration of at least 10 years, the risk for metformin users was 19% lower.
It is estimated that 12%-25% of people over age 65 have type 2 diabetes, putting them at about a 50% increase in risk for dementia, compared with nondiabetic people of the same age. - Dr. Rachel Whitmer
The differences between the drug types held up after adjustment for race, sex, education, and hemoglobin A1c (HbA1c) levels, Dr. Whitmer said.
It is estimated that 12%-25% of people over age 65 have type 2 diabetes, putting them at about a 50% increase in risk for dementia, compared with nondiabetic people of the same age. Nearly all elderly patients with type 2 diabetes will be started on some antidiabetic agent at some point, she noted.
"A lot of work out there [indicates] that insulin, and specifically insulin resistance, may play a role in Alzheimer’s disease, and there has been some nice work from animal models and cell-culture studies looking at metformin specifically," Dr. Whitmer said.
Metformin appears to lessen neuronal insulin resistance and is associated with neurogenesis. In addition, because it is associated with reduced risk of myocardial infarction and stroke, metformin may protect against vascular dementias, she added.
Dr. Whitmer and her colleagues delved into the Kaiser Permanente database to identify 14,891 patients aged 55 years and up with no history of dementia who were started on new diabetes pharmacotherapy between October 1999 and November 2001.
In all, 9.9% of patients in the sample were diagnosed with dementia during 5 years of follow-up, including 9.5% of 8,528 patients on metformin, 19.9% of 3,383 patients on a sulfonylurea, 9.8% of 2,095 patients on a TZD, and 11.1% of 905 patients on insulin.
In multivariate models adjusting for age, race, education, and diabetes duration, metformin was associated with a hazard ratio (HR) for any dementia of 0.77, compared with a TZD. Neither sulfonylureas nor insulin was associated with a significant reduction in risk. When glycemic control (HbA1c) was added to the model, metformin remained significantly better (HR, 0.84) at protecting against dementia, whereas the other agents were not significantly better than a TZD.
Similarly, in a model with sulfonylureas as the referent, metformin was significantly better, with adjusted HRs of 0.79 and 0.80 in models without and with HbA1c, respectively.
An analysis of effect by dementia subtype also showed significant benefit for metformin vs. sulfonylurea for protection against Alzheimer’s disease (HR, 0.70) and vascular dementia (HR, 0.75).
The investigators plan to examine relationships between dose and dementia, and to look at the effects of combination therapies.
The study was supported by the National Institute of Diabetes and Digestive and Kidney Diseases, and by Kaiser Permanente. Dr. Whitmer is employed by Kaiser Permanente Northern California.