Familial Alzheimer's disease and vitamin B12 deficiency

Age and Ageing, July, 1994 by A. McCaddon, C.L. Kelly

Summary

A kindred with autosomal dominant familial Alzheimer's disease in which a mutation results in a valine to glycine substitution at amyloid precursor protein codon 717 has recently been described. Individuals in this pedigree were studied retrospectively and prospectively, to evaluate the relationship, if any, between Alzheimer's disease and vitamin [B.sub.12] deficiency. The presence of Alzheimer's disease was found to be associated with lower serum vitamin [B.sub.12] values compared with unaffected family members. There were no significant differences between macrocytosis, the presence of anaemia, serum folate or red cell folate between affected and unaffected family members. The implications of this finding are discussed with regard to previous descriptions of an association, and in relation to the clinical features of this particular kindred.

Introduction

Recent reports suggest that an association may exist between Alzheimer's disease (AD) and vitamin [B.sub.12] deficiency(1)(2)(3)(4)(5)(6)(7). [B.sub.12] levels have been found to be lower in patients with AD than in patients with multi-infarct dementia and this deficiency appears to be independent of nutritional intake(4). One possible criticism of this relationship, however, is that AD is essentially a histopathological diagnosis, and patients with Alzheimertype dementia and low [B.sub.12] may represent a previously undetected sub-group of patients with AD.

Familial Alzheimer's disease (FAD) is a better model to study this association further, since affected patients conform to the NINCDS/ADRDA criteria for definite AD(8). Three distinct point mutations have now been found in the amyloid precursor protein gene at codon 717 that co-segregate with affected individuals in chromosome 21 linked families(9)(10)(11). One such kindred has been well documented with regard to the clinical features of the disease(12). The diagnosis has also been confirmed histopathologically in one individual(13). The index case in this family was noted to have a profound [B.sub.12] deficiency at initial presentation, which prompted us to review the records of affected family members for evidence of [B.sub.12] deficiency. In addition, it was possible to obtain blood samples from other affected and unaffected members after obtaining informed consent. These were assayed for routine haematological indices, serum [B.sub.12], serum folate, and red cell folate. The aims were to assess the degree of correlation, if any, between the presence of AD and levels of serum [B.sub.12] in order to explore further the validity of an association between the two diseases.

Material and Methods

A genealogical history was constructed from available medical records with the assistance of the patients' family. Records of affected siblings were obtained, where possible, and reviewed for documented evidence of cognitive impairment, relevant investigations including assay of [B.sub.12] and folate, and evidence of a formal diagnosis of Alzheimer's disease.

Blood samples were taken from affected and unaffected individuals after obtaining informed consent. In the case of affected subjects consent was also obtained from the next of kin. All samples were taken within a 2-month period in 1991 under the auspices of the Alzheimer Disease Research Group, London, who kindly agreed to provide serum for this study. Cognitive impairment was assessed using the Mini-Mental State Examination(14), and some members of the family were offered a more detailed clinical and neuro-psychological examination by the Research Group, St Mary's Hospital London, the results of which are presented elsewhere(12). Samples were analysed by Coulter counter for routine haematological indices. Serum [B.sub.12], serum folate, and red cell folate were assayed by radio-immunoassay.

Results were divided into two main groups, confirmed Alzheimer's Disease (group 1) and unaffected family members (group 2). The second group included both unaffected family members (generation III) and 'at risk' members (generation IV). The 'at risk' members are not depicted for ethical reasons. In view of the inevitably small sample sizes, results were analysed using Student's t test.

Results

The genealogical chart is shown in the Figure. Haemoglobin (Hb), haematocrit (Hct), mean corpuscular volume (MCV), serum [B.sub.12] ([B.sub.12]), serum folate (fol), and red cell folate (RCF) for the two groups are presented in Tables I and II. The Tables are incomplete owing to difficulty in obtaining medical records, lack of measuring or recording of [B.sub.12] values in the records of affected members, or reluctance to donate blood samples.

[CHART OMITTED]

Table I. Group 1: confirmed AD

Code     Hb     Hct   MCV   B12   fol   RCF
II.2
II.3
III.3                 91    163   3.2
III.6                       430
III.8
III.9
III.11                      *     *
III.14   15.2         102    50   3.9   293
III.15   13.8   0.4    93   156   5.6   118
III.16   13.7   0.4    87   181   2.4   180