, new findings suggest.
Moreover, the new data from the prospective Atherosclerosis Risk in Communities (ARIC) cohort also suggest that the previously identified increased risk for dementia among people with prediabetes appears to be entirely explained by the subset who go on to develop type 2 diabetes.
“Our findings suggest that preventing prediabetes progression, especially in younger individuals, may be an important way to reduce the dementia burden,” wrote PhD student Jiaqi Hu of Johns Hopkins University, Baltimore, and colleagues. Their article was published online in Diabetologia.
The result builds on previous findings linking dysglycemia and cognitive decline, the study’s lead author, Elizabeth Selvin, PhD, of the Bloomberg School of Public Health at Johns Hopkins, said in an interview.
“Our prior work in the ARIC study suggests that improving glucose control could help prevent dementia in later life,” she said.
Other studies have also linked higher A1c levels and diabetes in midlife to increased rates of cognitive decline. In addition, Dr. Selvin noted, “There is growing evidence that focusing on vascular health, especially focusing on diabetes and blood pressure, in midlife can stave off dementia in later life.”
This new study is the first to examine the effect of diabetes in the relationship between prediabetes and dementia, as well as the age of diabetes onset on subsequent dementia.
Prediabetes linked to dementia via diabetes development
Of the 11,656 ARIC participants without diabetes at baseline during 1990-1992 (age 46-70 years), 20.0% had prediabetes (defined as A1c 5.7%-6.4% or 39-46 mmol/mol). During a median follow-up of 15.9 years, 3,143 participants developed diabetes. The proportions of patients who developed diabetes were 44.6% among those with prediabetes at baseline versus 22.5% of those without.
Dementia developed in 2,247 participants over a median follow-up of 24.7 years. The cumulative incidence of dementia was 23.9% among those who developed diabetes versus 20.5% among those who did not.
After adjustment for demographics and for the Alzheimer’s disease–linked apolipoprotein E (APOE) gene, prediabetes was significantly associated with incident dementia (hazard ratio [HR], 1.19). However, significance disappeared after adjustment for incident diabetes (HR, 1.09), the researchers reported.
Younger age at diabetes diagnosis raises dementia risk
Age at diabetes diagnosis made a difference in dementia risk. With adjustments for lifestyle, demographic, and clinical factors, those diagnosed with diabetes before age 60 years had a nearly threefold increased risk for dementia compared with those who never developed diabetes (HR, 2.92; P < .001).
The dementia risk was also significantly increased, although to a lesser degree, among those aged 60-69 years at diabetes diagnosis (HR, 1.73; P < .001) and age 70-79 years at diabetes diagnosis (HR, 1.23; P < .001). The relationship was not significant for those aged 80 years and older (HR, 1.13).
“Prevention efforts in people with diabetes diagnosed younger than 65 years should be a high priority,” the authors urged.
Taken together, the data suggest that prolonged exposure to hyperglycemia plays a major role in dementia development.
“Putative mechanisms include acute and chronic hyperglycemia, glucose toxicity, insulin resistance, and microvascular dysfunction of the central nervous system. ... Glucose toxicity and microvascular dysfunction are associated with increased inflammatory and oxidative stress, leading to increased blood–brain permeability,” the researchers wrote.
Dr. Selvin said that her group is pursuing further work in this area using continuous glucose monitoring. “We plan to look at ... how glycemic control and different patterns of glucose in older adults may be linked to cognitive decline and other neurocognitive outcomes.”
The researchers reported no relevant financial relationships. Dr. Selvin has reported being on the advisory board for Diabetologia; she had no role in peer review of the manuscript.
A version of this article first appeared on Medscape.com.