In the current issue of the New England Journal of Medicine, a consortium of scientists from Seattle and Boston reports on their large longitudinal study on the relationship between blood glucose and dementia risk (Crane et al., N Engl J Med 2013; 369: 540-548). Some – but not all – studies on the association between diabetes mellitus and dementia suggest that diabetes increases the risk for developing dementia. The present study examined for the first time whether an elevated blood glucose also increases the risk for dementia, even in the absence of diabetes mellitus.
The participants in the study were recruited from the Adult Changes in Thought (ACT) study. This is a study in non-demented people over the age of 65 years, who underwent an examination every two years to determine signs of (incipient) dementia. At least five blood glucose concentrations (or hemoglobin A1c or results of an earlier assay that were converted to Hb A1C values) had to be available from the participants in the two years prior to study inclusion. Finally, 2067 subjects were included in the study. The dementia screening was performed using the Cognitive Abilities Screening Instrument. When abnormalities were found in this test, a more extensive diagnostic procedure was initiated. Hb A1c values were converted to mean blood glucose levels using the following formula:
Mean daily blood glucose = (28.7 × hemoglobin A1c) – 46.7
The presence of diabetes was defined according to the prescription of antidiabetic agents. A participant who received at least two prescriptions for antidiabetic drugs per year was assumed to suffer from diabetes.
35264 blood glucose values and 10208 values for hemoglobin A1c (or its older analogue) were available from the 2067 study participants. 232 of the participants had diabetes at baseline. During the five years prior to the study, the median blood glucose in people without diabetes was 101 mg/dL, and for those with diabetes it was 175 mg/dL. The duration of follow-up was 6.8 years on average.
524 (25.4%) of the 2067 study participants developed dementia. 403 participants suffered from a possible or probable Alzheimer’s disease, 55 from vascular dementia and 66 from dementia due to other causes. For the participants who did not have diabetes at baseline, an average glucose concentration of 115 mg/dL compared to the reference value of 100 mg/dL increased the risk for dementia by nearly 20% (hazard ratio 1.18, confidence interval 1.04 to 1.33; p = 0.01). For the participants who had diabetes at baseline, 160 mg/dL were considered as the reference value. An increase in mean glucose concentration to 190 mg/dL increased the risk of dementia by 40% (hazard ration 1.40, confidence interval 1.12 to 1.76, p = 0.002).
The authors conclude from their results that any increase in blood sugar, not just clearly pathological values as measured in diabetes mellitus, increases the risk for dementia. They emphasize the importance of obesity and diabetes in most societies, and the need for glucose lowering interventions. Unfortunately, the publication lacks any statement on the important question of whether the presence of diabetes mellitus increased the risk for dementia. Interestingly, at the end of the observation period, 343 participants (232 at baseline) were suffering from diabetes, and of these, 74 were suffering from dementia, which was a proportion of 21.6%. Of the 1724 participants who were not suffering from diabetes at the end of follow-up, 450 were suffering from dementia, which was a proportion of 26.1%. This result is not being further discussed or at least analyzed. Because it does not fit into the story?
This post is also available in: German