Anaemia defined as a haemoglobin level <13 g/dl in men and <12 g/dl in women is common in older people and associated with numerous health consequences. The aim of this study was to systematically review all published data from the past 30 years that studied the association between anaemia and cognitive performance in people aged 65 years and over. An English and French Medline and Cochrane Library search ranging from 1979 to 2011 indexed under the Medical Subject Heading (MeSH) terms 'haemoglobin' or 'anaemia' combined with the terms 'dementia' or 'cognition disorders' or 'memory disorders' or 'orientation' or 'executive functions' or 'attention' or 'brain' or 'neuropsychological tests' was performed. Ninety-eight studies were selected. The following specific conditions were excluded: cancer, chronic kidney diseases, chronic heart disease and post-operative cognitive dysfunction. Five observational studies and six prospective cohort studies were included in the final analysis. According to the studies, the number of participants ranged from 302 to 2250 community-dwelling older people aged 55 years or over. Four studies considered the association between haemoglobin concentration and global cognitive functions, another three examined the association between haemoglobin concentration and the incidence of dementia, and four studies evaluated some specific aspects of cognition. A significant positive association was shown between anaemia and global cognitive decline as well as the incidence of dementia. A significant association was also shown between anaemia and executive functions. This systematic review shows a probable association between anaemia and cognitive performances, particularly with executive functions.

BackgroundThe natural history and disease mechanisms of Alzheimer’s disease and related disorders (ADRD) are still poorly understood. Very few resources are available to scrutinise patients as early as needed and to use integrative approaches combining standardised, repeated clinical investigations and cutting-edge biomarker measurements.MethodsIn the nationwide French MEMENTO cohort study, participants were recruited in memory clinics and screened for either isolated subjective cognitive complaints (SCCs) or mild cognitive impairment (MCI; defined as test performance 1.5 SD below age, sex and education-level norms) while not demented (Clinical Dementia Rating [CDR] <1). Baseline data collection included neurological and physical examinations as well as extensive neuropsychological testing. To be included in the MEMENTO cohort, participants had to agree to undergo both brain magnetic resonance imaging (MRI) and blood sampling. Cerebral 18F-fluorodeoxyglucose positon emission tomography and lumbar puncture were optional. Automated analyses of cerebral MRI included assessments of volumes of whole-brain, hippocampal and white matter lesions.ResultsThe 2323 participants, recruited from April 2011 to June 2014, were aged 71 years, on average (SD 8.7), and 62% were women. CDR was 0 in 40% of participants, and 30% carried at least one apolipoprotein E ε4 allele. We observed that more than half (52%) of participants had amnestic mild cognitive impairment (17% single-domain aMCI), 32% had non-amnestic mild cognitive impairment (16.9% single-domain naMCI) and 16% had isolated SCCs. Multivariable analyses of neuroimaging markers associations with cognitive categories showed that participants with aMCI had worse levels of imaging biomarkers than the others, whereas participants with naMCI had markers at intermediate levels between SCC and aMCI. The burden of white matter lesions tended to be larger in participants with aMCI. Independently of CDR, all neuroimaging and neuropsychological markers worsened with age, whereas differences were not consistent according to sex.ConclusionsMEMENTO is a large cohort with extensive clinical, neuropsychological and neuroimaging data and represents a platform for studying the natural history of ADRD in a large group of participants with different subtypes of MCI (amnestic or not amnestic) or isolated SCCs.Trial registrationClinicaltrials.gov, NCT01926249. Registered on 16 August 2013.Electronic supplementary materialThe online version of this article (doi:10.1186/s13195-017-0288-0) contains supplementary material, which is available to authorized users.

No abstract

No abstract