Does diabetes trigger Alzheimer's disease?

Summarized by Robert W. Griffith, MD
March 24, 2000 (Reviewed: November 11, 2002)

Introduction

Type 2 diabetes is a known risk factor for thrombotic stroke, so that a relationship to vascular dementia might be presumed. Indeed, it has been shown to be associated with an increased incidence of multi-infarct dementia in previous studies. A relationship between diabetes and Alzheimer disease has not been established, to date. The authors of the article summarized here used data from the Rotterdam Study to explore this possibility.

Method

All persons 55 or older living in a suburb of the Dutch city in 1990-1993 were invited to participate in the Rotterdam Study. Approximately 8,000 (78%) accepted, and 7,046 of these participants who were free of dementia at baseline (and therefore at risk of developing it) were enrolled.¹

Examinations for diabetes were done at baseline on 6,370 of these subjects, who formed the population for further study. After appropriate blood sampling for glucose analyses, diabetes was defined as use of antidiabetic medication, or a random or post-glucose load (75 g) serum glucose level above 11 mmol/L (200 mg/dL).

The duration of follow-up ranged from 4 days to 6 years - the average follow-up was 2.1 years. Follow-up assessment of incident dementia was done using the same tests as at baseline on 5,232 of the participants, and dementia was assessed on data from family practitioners and medical records in the remaining 1,138.

Possibly confounding variables measured at baseline included educational background, height, weight, body fat distribution, smoking habits, alcohol consumption, blood pressure, antihypertensive medication, and resting ECG.

The relative risk of dementia in association with diabetes was calculated using a Cox proportional hazard regression model, which permitted adjustment for gender, age, and the other confounding variables mentioned above.

Results

At baseline, 692 participants had diabetes, of whom 70 were being treated with insulin. Apart from educational level and the presence or absence of atrial fibrillation, all the examined variables differed significantly in frequency between the participants with diabetes and those without. Women comprised 59% of the participants; the mean age was 69 years at baseline.

Over the follow-up period 126 (2%) of the subjects developed dementia, 89 (71%) of whom had Alzheimer disease, and 18 (14%) had vascular dementia. The prevalence of diabetes, adjusted for age and gender, was significantly higher in those persons who developed dementia.

Analyses showed that the presence of diabetes almost doubled the risk of dementia - the relative risk (RR) adjusted for age and gender was 1.9 (95% CI, 1.3-2.8). The RR for Alzheimer disease in diabetics was also 1.9 (95% CI, 1.2-3.1). In patients taking insulin the RR for dementia was 4.3 (95% CI, 1.7-10.5). Diabetics with cerebrovascular disease developed Alzheimer disease with a RR of 3.0 (95% CI, 1.0-9.3), while those without cerebrovascular disease had a RR of1.8 (95% CI, 1.1-3.0). There were no significant effects of other possible confounders when they were added to the regression model.

Comment

The results of this large prospective population-based cohort study confirm those of smaller studies demonstrating an association between existing diabetes and the development of dementia. A new finding in this study was the increased relative risk of Alzheimer disease - previous work had shown an increased risk of vascular dementias in diabetics, but the reverse effect regarding Alzheimer's. The association with vascular dementia is to be expected, because of the known effects of diabetes on micro- and macro-vasculature. However, the increased risk of Alzheimer disease is less readily explained.

Recent reports, from the same authors and from other investigators, support the view that Alzheimer disease has a partly vascular pathogenesis. Alternatively, diabetes may itself contribute to cognitive impairment, possibly due to accumulation of hyperglycemic-driven protein glycosylation end products.

Source

• Diabetes mellitus and the risk of dementia A. Ott, RP. Stolk, F. Van Harskamp, The Rotterdam Study. Neurology, 1999, vol. 53, pp. 1937--1942

Footnotes
1. Dementia was assessed using the Mini-Mental State Examination (MMSE) and the Geriatric Mental State Schedule (GMS-A). Dementia was diagnosed by a panel using DSM-III-R criteria, and classified as Alzheimer disease (including a subgroup with cerebrovascular disease), and vascular dementia, using well-established criteria.

Please take a moment to give us your comments. For questions about Health matters you may check our "Questions & Answers" Portal and Service.