Neurology -- Correspondence for Gustafson et al., 63 (10) 1876-1881

Correspondence published:

A 24-year follow-up of body mass index and cerebral atrophy: Frank-Erik de Leeuw, Wiesje M van der Flier and Philip Scheltens (2 February 2005)

Reply to de Leeuw et al: Deb Gustafson (2 February 2005)

A 24-year follow-up of body mass index and cerebral atrophy 2 February 2005

Frank-Erik de Leeuw,
Department of Neurology
UMC St. Radboud, Dept Neurology (HP326). PO Box 9101, 6500HB Nijmegen, The Netherlands,
Wiesje M van der Flier and Philip Scheltens
Send Correspondence to journal:
Re: A 24-year follow-up of body mass index and cerebral atrophy

h.deleeuw{at}neuro.umcn.nl Frank-Erik de Leeuw, et al.

We read with great interest the article by Gustafson et al on the relation between body mass index (BMI) and cerebral atrophy. There is an increasing interest regarding the role of vascular risk factors in the etiology of AD and the observation presented by the authors will possibly add another “weighty” underlying mechanism by which vascular risk factors may ultimately lead to Alzheimer’s disease.
It is still unclear whether representing a compound score for a cardiovascular risk profile is independently related to cerebral, and especially temporal, atrophy. It is possible that the relation is confounded by vascular cerebral lesions, including cerebral white matter lesions that are established risk factors for dementia. [1]
White matter lesions can act as a classical confounder since their presence is related both to the determinant the authors have investigated [2]. It is also related to cortical atrophy and used as an outcome measure by the authors. Their finding of a relation between BMI and temporal atrophy is interesting since recent studies show a relation between the degree of white matter lesions and the degree of atrophy of the medial temporal lobe. [3,4]
This could suggest that white matter lesions underlie the association between BMI and temporal atrophy. To qualify for a variable (before qualifying as a confounder), the variable should intermediate between body mass index and cortical atrophy. However, there is evidence that cortical neurons are damaged or lost via transneuronal degeneration, secondary to subcortical white matter lesions. [5] This suggests that white matter lesions could function as an intermediate between BMI and cortical atrophy.
This may be clarified by rating the white matter lesions and subsequently making the appropriate adjustments in the analysis. Another possibility for making these adjustments would be investigating the relation between BMI and cortical atrophy stratified on the absence or presence of white matter lesions.
References
1. Vermeer SE, Prins ND, den Heijer T, Hofman A, Koudstaal PJ, Breteler MMB. Silent brain infarcts and the risk of dementia and cognitive decline. N Engl J Med. 2003;348:1215-1222
2. Gustafson DR, Steen B, Skoog I. Body mass index and white matter lesions in elderly women. An 18-year longitudinal study. Int Psychogeriatr. 2004;16:327-336.
3. De Leeuw FE, Barkhof F, Scheltens Ph. White matter lesions and hippocampal atrophy in Alzheimer’s disease. Neurology 2004; 62:310-312.
4. Van der Flier WM, Middelkoop HA, Weverling-Rijnsburger AW, et al. Interaction of medial temporal lobe atrophy and white matter hyperintensities in AD. Neurology. 2004;62:1862-1864.
5. Schuff N, Capizzano AA, Du AT, et al. Different patterns of N- acetylaspartate loss in subcortical ischemic vascular dementia and AD. Neurology. 2003;61:358-364.

Reply to de Leeuw et al 2 February 2005

Deb Gustafson,
Department of Psychiatry, Sahlgrenska University Hospital
SE-413 45 Göteborg, Sweden
Send Correspondence to journal:
Re: Reply to de Leeuw et al

Deb Gustafson [Deb.Gustafson{at}neuro.gu.se] Deb Gustafson

The comments by de Leeuw et al on our article [1] are insightful and point to one mechanism whereby a high BMI may be related to temporal atrophy - via white matter lesions (WML). We did evaluate the relationship between BMI and WML using the CT data collected in this sample of women, and there was no relationship. Thus, WML were not included in multivariate analyses on the relationship between BMI and temporal atrophy.
However, based on the suggestion of de Leeuw et al, the analyses were run again after stratification for the presence or absence of WML and with age adjustment, the major confounder in our analysis. Interestingly, we found that the relationship between a higher mean BMI and temporal atrophy was only observed among women without WML.
Among 204 women without WML, the mean BMI in 95 women with and 109 women without temporal lobe atrophy was 24.9 +/- 4.0 vs 23.6 +/- 3.5 kg/m2 in 1968 (age- adjusted p=0.031), 25.2 +/- 3.8 vs 23.8 +/- 3.6 kg/m2 in 1974 (age- adjusted p=0.029), 26.0 +/- 4.1 vs 24.1 +/- 3.7 kg/m2 in 1980 (age- adjusted p=0.004), and 27.2 +/- 5.0 vs 25.5 +/- 3.9 kg/m2 in 1992 (age- adjusted p=0.005), respectively.
In addition, multivariate logistic regression models predicting temporal atrophy showed no interaction between BMI and the presence of WML at any examination. Therefore, our data do not support the premise that WML are intermediate between a high BMI and atrophy of temporal lobe.
There are many reasons for discrepant findings in two samples of Swedish women – one showing a relationship between BMI and WML [2], and the aforementioned showing a relationship between BMI and temporal atrophy. [1] Firstly, most studies show that the relationship between vascular risk factors and dementia vary over the adult lifespan, as they do for other chronic diseases, such as cardiovascular disease. That a high BMI is related to WML in late life, as we showed in a cross-sectional sample of 85-year old women, but not in a sample of women during midlife who were age 46-60 at baseline, may be due to a variety of factors including the presence of other pathologies in the brain or that the 85-year-old women were a group of surviving ‘elite’ with high levels of overweight and obesity and therefore having a different picture of dementia-related susceptibility.
Second, whether WMLs are an intermediate in the pathway between a high BMI and cerebral atrophy is controversial. There are numerous pathologies that co-exist in the brains of older persons with and without dementia, which cannot be disentangled from each other in terms of which pathology came first or to what extent they influence each other, and as can be observed using CT data.
We appreciate the comments of de Leeuw et al, as they highlight our need for a greater understanding of the progression and temporality of risk factors for dementia and dementia-related processes throughout the lifespan.
References
1. Gustafson D, Lissner L, Bengtsson C, Björkelund C, Skoog I. A 24- year follow-up of overweight and risk for temporal atrophy, Neurology, 2004;63:1876-1881.
2. Gustafson DR, Steen B, Skoog I. Body mass index and white matter lesions in elderly women. An 18-year longitudinal study. International Psychogeriatrics, 2004;16:327-336.