Dementia and its associations in type 2 diabetes mellitus: The Fremantle Diabetes Study

https://doi.org/10.1016/S0168-8227(01)00266-2Get rights and content

Abstract

Background: recent studies indicate that diabetes is an important risk factor for dementia in older patients, but the cause remains unknown. Objectives: to determine whether vascular or diabetes-related risk factors predict the development of dementia in older subjects with diabetes. Patients: 63 patients with type 2 diabetes of mean age 75.3 years. Methods: Subjects were screened for cognitive impairment using the Mini-Mental State Examination (MMSE) and informants who knew the subjects answered the Informant Questionnaire for Cognitive Decline in the Elderly (IQCODE). Probable dementia was diagnosed using highly specific criteria based on the combination of both tests. Potential risk factors for dementia obtained at the time of the cognitive test and annually from a median of 3.2 years previously were examined using univariate methods and simple general linear modelling. Results: since there was a significant association between MMSE and English speaking ability, ten non-Anglo–Celt patients were excluded from the analysis. Probable dementia was diagnosed in six of the remaining 53 subjects (11%). Dementia was significantly and independently associated with higher diastolic and mean arterial blood pressure measurements over the 3 years before assessment. Blood pressure declined over this time in the patients with probable dementia and was similar to that in non-demented subjects at the time of cognitive assessment. Conclusions: these preliminary data suggest that the high rate of dementia found in older people with diabetes may be explained by the high rate of hypertension in this population.

Introduction

There is a progressive increase in the number of people reaching old age. Dementia and diabetes mellitus are both common in later life and often co-exist. Three recent prospective studies have established that diabetes increases the risk of dementia in general and Alzheimer's disease (AD) in particular [1], [2], [3], whilst another large prospective study reported that diabetes was associated with cognitive decline in elderly women [4]. Based on these studies, the prevalence of dementia in diabetic patients is approximately twice that in the normal population [1], [2], [3].

Studies of the aetiology of dementia have been hampered by the exclusive criteria used to differentiate AD from vascular dementia [5]. A more pragmatic approach is to consider dementia by risk factors rather than aetiological sub-type [5]. Recent evidence indicates that, apart from diabetes, conventional risk factors for cerebrovascular disease are also risk factors for both vascular dementia and AD [5], [6], [7]. Since hypertension and dyslipidaemia are over-represented in patients with diabetes, this could help to explain their increased risk of dementia. Nevertheless, diabetes-specific variables may also contribute. In one study, the highest risk of dementia was in insulin-treated subjects [1] whilst, in another, diabetes duration was associated with cognitive decline [4]. A detailed assessment of diabetes-specific factors contributing to dementia is, however, hampered by incomplete diagnostic and clinical information in previously published studies [1], [2], [3], [4].

In the present study, we have piloted a method for identifying older diabetic subjects with cognitive problems who are enrolled in a longitudinal observational study of diabetes in an urban Australian community setting. We found a strongly significant, independent association between antecedent hypertension and probable dementia in these patients.

Section snippets

Study subjects

We assessed a convenience sample of 63 type 2 diabetic subjects who were at least 70 years of age. Each was a participant in the Fremantle Diabetes Study (FDS), details of which have been published previously [8]. Briefly, the FDS is a prospective community-based study of diabetes care, control and complications in a representative sample of patients living within the primary catchment area of Fremantle Hospital, a postcode-defined population of 120 097 people. We identified 2277 patients from

Data analysis

Mean arterial pressure (MAP) was calculated using the formula: ((2×diastolic pressure)+systolic pressure)/3). Three-year mean values were calculated by averaging the four measurements of each variable of interest made at the time of the cognitive test and at each of the three prior annual assessments. Statistical analysis was performed using spss for Windows (SPSS Inc, Chicago, IL). Two-sample comparisons for normally distributed variables were by Student's t-test. The Wilcoxon–Mann–Whitney

Results

The present sample contained 63 patients (25 females and 38 males) who were aged between 70 and 88 years. The median duration of diabetes was 9.7 years (range 3.5–29.7 years). The majority (62%) were taking oral hypoglycaemic agents, 25% were on diet alone and 13% were taking insulin with or without other blood glucose lowering therapies. Most of the patients were Anglo–Celt (70%) and the next largest sub-group were Southern Europeans (16%). MMSE data were available for all 63 subjects. In two

Discussion

In the context of an ageing population and limited health care resources, diabetes and dementia are two conditions that will have increasing epidemiological and clinical importance. We investigated the prevalence and causes of probable dementia in a sample drawn from a larger community-based cohort of people with diabetes. The prevalence of probable dementia in this sample with a mean age of 75 years was 11%. It is noteworthy that a greater percentage of cases had borderline results on the

Acknowledgements

We acknowledge the dedication and hard work of Denise Jackson, Giovanna Stuccio, Mary Balme and Dorothy Ridley in the collection of these data. We thank the Biochemistry Department at Fremantle Hospital for performing the biochemical tests and Dr John Beilby for apoliprotein E genotyping. This project was funded by the Novo Nordisk Regional Diabetes Research Support Scheme. The Fremantle Diabetes Study as a whole has been supported by the Raine Foundation/University of Western Australia and the

References (22)

  • P.F Malloy et al.

    Cognitive screening instruments in neuropsychiatry: a report of the Committee on Research of the American Neuropsychiatric Association

    J. Neuropsychiatry Clin. Neurosci.

    (1997)
  • Cited by (42)

    • The "test Your Memory" test performs better than the MMSE in a population without known cognitive dysfunction

      2013, Journal of the Neurological Sciences
      Citation Excerpt :

      Importantly, however, our primary aim was to compare the TYM to the MMSE in measuring variation in cognitive functioning in non-demented people, rather than validation of the cut off scores for the detection of dementia. In many studies the MMSE is used to give a global measure of cognitive functioning when examining the relation between risk factors and cognition or investigating the effects of treatment on cognition [27–29]. A meta-analysis, examining the performance of the MMSE, found that the MMSE has limitation when used for this purpose [30].

    • Human cerebral neuropathology of Type 2 diabetes mellitus

      2009, Biochimica et Biophysica Acta - Molecular Basis of Disease
    • Impact of impaired fasting glucose and other metabolic factors on cognitive function in elderly people

      2007, Nutrition, Metabolism and Cardiovascular Diseases
      Citation Excerpt :

      Amongst these, cognitive impairment and diabetes are becoming highly prevalent in old people with deleterious effects on their quality of life [1,2]. Several studies have shown that subjects with diabetes are at higher risk of dementia and cognitive decline [3–11], raising the possibility that the metabolic abnormalities of diabetes, i.e., hyperglycaemia and insulin resistance (IR), may play a role in the genesis of cognitive decline in elderly subjects. The association between IR, glucose metabolism, and cognitive performance in a non-diabetic elderly population has been previously examined.

    • Psychosocial and Psychiatric Challenges of Diabetes Mellitus

      2006, Nursing Clinics of North America
      Citation Excerpt :

      Recent studies have shown that diabetes is associated with cognitive impairments leading to a higher incidence of dementia and Alzheimer's disease. The increased risk of dementia in the diabetes population has been documented by several studies, however, specific risk factors involved are not known [79,80]. Although acute hypoglycemia is associated with cognitive dysfunction [81,82], recurrent hypoglycemia may not have a lasting effect on cognitive functioning in adults [83,84].

    View all citing articles on Scopus
    View full text