Anaemia increases the risk of dementia in cognitively intact elderly
Introduction
Anaemia is very common after age 65 years, accounting for 4.2% [22] to 28% [14] when WHO criteria [35] are applied in different gender and age groups. In addition, anaemia incidence has been reported to increase with age [2]. Dementia and Alzheimer disease (AD) are also frequent occurrences in the elderly, and both prevalence [20] and incidence [12] increase exponentially with age. Although a complete blood count is recommended [27] to identify “potentially reversible cognitive impairment” [7], apart from pernicious anaemia which is a well-known cause of “reversible dementia” [26], the role of anaemia on cognition, independent of Vitamin B12 levels, is controversial. A study based on informants’ reports, found a higher, although not statistically significant, risk of Alzheimer disease (AD) in subjects with a history of anaemia [6]; a case–control study from the Mayo Clinic [5] showed that anaemia in the year prior to dementia diagnosis, was related to incident AD but the association was not confirmed prospectively. Finally, an association between anaemia and vascular dementia has been reported by a cross-sectional study in Australia, but no association was found with AD [21].
The current study used data from the Kungsholmen Project, a longitudinal population-based study carried out in Stockholm, Sweden, to evaluate the role of anaemia on the development of dementia. Since cerebral hypo perfusion has been suggested to be involved in neurodegeneration [32], and an abrupt reduction in brain oxygenation is associated with lower cognitive performance [8], we hypothesized that a chronic anaemic status might increase the risk of developing dementia. To exclude the effect of other possible confounders, we tested this hypothesis in three models where the effect of different covariates was taken into account separately. The first model included chronic diseases, as some of them have been related to development of dementia, and they might also lower haemoglobin concentration. The second model took into account markers of inflammation, as this has been related to dementia [28] and low haemoglobin concentration might be due to an inflammatory state (“anaemia of inflammation”) [23]. The third model included nutritional indicators, due to the fact that weight loss has been associated with development of dementia [34], and malnutrition, which is frequent in the elderly, might lead to low haemoglobin concentration. Finally, all the previous variables were considered together in a fully adjusted model.
Section snippets
Study population
We used data from baseline and 3-year follow-up examination of the Kungsholmen Project, a longitudinal, population-based study on aging and dementia. Details of the study design have been already reported [11]. Briefly, all inhabitants living in Kungsholmen area (Stockholm, Sweden) aged 75 years and above in October 1987 (baseline) were asked to take part in the project. Of the 2368 eligible subjects, 1810 (76.4%) participated in the screening phase, which included an interview by trained
Results
Table 1 shows the main features of the study population according to different haemoglobin concentration cut offs: lower haemoglobin concentration was associated with older age (p < 0.01) and lower score on baseline MMSE (p < 0.05).
According to WHO criteria, the prevalence of anaemia in the dementia free cohort of the Kungsholmen Project was 15.7% (n = 54) in men and 7.3% (n = 75) in women (p < 0.01) (Fig. 1).
The follow-up examination took place approximately 3 years after baseline (mean follow-up period
Discussion
In our prospective study, anaemic subjects with good baseline cognitive performance had a two-fold higher risk of developing dementia 3 years later than persons without anaemia. The association was still substantial and significant after adjustments for a number of potential confounders such as history of chronic diseases, inflammatory markers, or indicators of malnutrition. The hazard ratios of incident dementia was higher when lower haemoglobin cut offs were used to define anaemia, suggesting
Acknowledgments
The authors thank all workers of the Kungsholmen Project for data collection and management and the Swedish Research Council. This study was supported by grants (Borsa di studio di Perfezionamento all’ Estero) from the University of Ferrara (Italy).
References (36)
- et al.
Risk of Alzheimer disease among elderly patients with anemia: population-based investigation in Olmsted county, Minnesota
Ann Epidemiol
(1997) - et al.
Potentially reversible cognitive impairment in patients presenting to a memory disorders clinic
J Clin Neurosci
(2000) - et al.
Jugular bulb saturation and cognitive dysfunction after cardiopulmonary bypass
Ann Thorac Surg
(1994) - et al.
“Mini-Mental State”. A pratical method for grading the cognitive state of patients for the clinician
J Psychiatr Res
(1975) Demonstrating the case that AD is a vascular disease: epidemiologic evidence
Ageing Res Rev
(2002)- et al.
Cognitive performance, hyperoxia, and heart rate following oxygen administration in healthy young adults
Physiol Behav
(1999) - et al.
Alzheimer disease and angiogenesis
Lancet
(2003) Diagnostic and statistical manual of mental disorders
(1987)- et al.
Incidence of anemia in older people: an epidemiologic study in a well defined population
J Am Geriatr Soc
(1997) Epidemiology of anaemia in the elderly: information on diagnostic evaluation
J Am Geriatr Soc
(2003)