In conflict with a number of recent studies, six-year data from the Rotterdam study of aging failed to show a link between dietary fat and risk of dementia. The results are reported in today’s Neurology.

Drawing on disparate lines of evidence, some researchers have been building a case that dietary fat intake affects dementia risk (see related news item). For example, higher dietary cholesterol increased cerebral Aβ deposition in some animal experiments, and certain polyunsaturated fats may have antiinflammatory properties, which might counteract the alleged inflammatory contribution to dementia. Some, but not all, epidemiologic studies have linked higher serum levels of cholesterol (mostly during midlife) with higher dementia risk, and cholesterol-lowering drugs such as statins appear to reduce the risk of Alzheimer's dementia. Indeed, earlier results from the Rotterdam study have contributed to this theory (see Kalmijn et al., 1997), finding that high intake of total fat and low intake of fish (with its high levels of cis n-3 polyunsaturated fatty acids) were significantly associated with risk of dementia at the two-year follow-up in the Rotterdam study.

Now Monique Breteler and her colleagues in the Netherlands at the Erasmus Medical Center and at the University of Utrecht report the findings of six years of follow-up in the study of elderly residents (55 years or older at baseline) of a suburb of Rotterdam. Of the 5,395 subjects whose baseline and follow-up exams included complete dietary data, 197 (seven percent) had developed dementia (146 with AD; 29 with vascular dementia; and 22 with other dementias).

The researchers found no contribution to dementia risk from the intake of total fat, saturated fatty acids, or trans-unsaturated fatty acids, or from dietary cholesterol (adjusted for age, gender, total energy intake, and vitamin E intake). Conversely, dietary intake of the so-called "good" fatty acids, the cis mono- or polyunsaturated fatty acids, failed to reduce AD risk. These findings held true for all dementias, vascular dementia, and AD, and when additional lifestyle elements were factored in (e.g., smoking, alcohol, fruit and vegetables, dietary supplements, body mass index). Similarly, excluding patients using lipid-lowering drugs, or those with myocardial infarction or diabetes, did not change the results.

"Given the limited number of studies on fat intake and risk of dementia, we think it is premature to conclude from our observational study that cholesterol or cholesterol-affecting fats are not associated with risk of dementia or its subtypes. Larger, prospective studies with longer follow-up periods are needed to confirm our findings," the authors caution in their conclusion.—Hakon Heimer


See ARF related live discussion.


  1. See comment by Benjamin Wolozin
    See reply by Monique Breteler

    Comment by Benjamin Wolozin: Thinking about brain and serum cholesterol metabolism in Alzheimer’s disease.—Posted 24 December 2002.
    The recent study by Engelhart and colleagues investigates the relationship of serum cholesterol and serum polyunsaturated fatty acids (PUFA) to the incidence of Alzheimer’s disease in the cohort of patients present in the Rotterdam study. Engelhart et al. do not observe any clear relationship between the intake of cholesterol, total PUFA, n-3 PUFA or n-6 PUFA and the incidence of Alzheimer’s disease.

    The literature on the relationship between cholesterol and Alzheimer’s disease is quite confusing, and at times contradictory. Studies in cell culture, transgenic mice, and rabbits indicate that production of Aβ is sensitive to cellular cholesterol. Reducing cholesterol, by treating with β-methylcyclodextran, statins, or BM15.766 (a cholesterol-lowering drug that acts distal to HMG CoA reductase) reduces the production of Aβ and accumulation of neuritic plaques (Fassbender et al., 2001; Refolo et al., 2001; Simons et al., 1998). Human subjects treated with simvastatin or lovastatin also have less Aβ (Friedhoff et al., 2001; Simons et al., 2002). Several retrospective studies and one prospective study indicate that there is a lower prevalence of Alzheimer’s disease or delayed progression of Alzheimer’s disease among subjects taking statins (Jick et al., 2000; Rockwood et al., 2002; Simons et al., 2002; Wolozin et al., 2000; Yaffe et al., 2002). Together, these studies suggest some link between cholesterol and Alzheimer’s disease.

    However, studies that examine the relationship between serum cholesterol and Alzheimer’s disease have produced mixed results. Engelhart’s study is the most recent of several studies demonstrating the lack of any linkage between serum cholesterol and AD (Prince et al., 2000; AbstractYoshitake et al., 1995). Some recent studies did observe a possible effect of serum cholesterol and AD (Kivipelto et al., 2002; Kivipelto et al., 2001; Notkola et al., 1998).

    Analysis of the biology of cholesterol suggests a potential reason explaining why measurements of serum cholesterol might not show a strong association with AD. Alzheimer’s disease is thought to be caused, at least in part, by the accumulation of aggregated Aβ in the brain. If true, this suggests that the most important site for regulating Aβ production is also in the brain. This logic is important because cholesterol in the brain is synthesized de novo. Little, if any, cholesterol comes from the blood. The independence of brain and serum cholesterol is reflected by the different half-lives of cholesterol in the two organs. The half-life of cholesterol in the rodent brain is six months (although there is likely to be a large difference between turnover of cholesterol in myelin and in neurons), and probably longer in humans (AbstractAndersson et al., 1990; Bogdanovic et al., 2001; Dietschy and Turley, 2001). The half-life of cholesterol in serum is hours. The independence of serum and brain cholesterol could readily account for the lack of any clear association between serum cholesterol and Alzheimer’s disease.

    If cholesterol in the brain is the critical issue, how might it affect Aβ? There are several possible scenarios. One possibility is that reducing neuronal cholesterol reduces the production of Aβ because the γ- or β-secretase complex requires a high-cholesterol environment to be active. Studies of the enzymes that produce Aβ indicate that these proteins reside in cholesterol-rich regions of the membrane, termed lipid rafts (Wahrle et al., 2002). A second possibility is that statins or cholesterol-lowering diets (in mice) lower apolipoprotein E levels. This could occur because statins prevent isoprenylation of ras, rac and rho, which is required for activation of rho and stimulation of apolipoprotein E production (Martin et al., 2001; Takemoto and Liao, 2001). Isoprenylation is also important in inflammation, and statin-mediated inhibition of the inflammatory response might also be beneficial in AD (Takemoto and Liao, 2001).

    Whether cholesterol is truly important in AD, and whether statins are actually beneficial as an anti-Alzheimer’s therapy remains to be determined. Alzheimer’s research is littered with theories and medications that look promising in the lab or in epidemiological studies, but have proved to be disappointing in the clinic. Although a small preliminary study suggested that statins might prevent the progression of dementia in patients in early stages of Alzheimer’s disease, a much larger study has failed to show a beneficial effect of statins on dementia (Shepherd et al., 2002). On the other hand, molecular genetic studies increasingly suggest a linkage between cholesterol and AD because AD has been associated with two genes related to cholesterol metabolism—apolipoprotein E and cholesterol 24 hydroxylase (Cyp46), and there might also be a link between AD and α2-macroglobulin, which is also important for cholesterol metabolism (Blacker et al., 1998; Corder et al., 1993; Kolsch et al., 2002). The putative linkage between cholesterol 24 hydroxylase and AD is particularly notable because this enzyme does not exist outside the brain, which emphasizes the potential importance of cerebral cholesterol metabolism in AD (Lund et al., 1999). Strong genetic validation of the putative link between brain cholesterol metabolism and the pathophysiology of AD would provide important support for the concept of modulating brain cholesterol metabolism to treat AD, whether or not statins prove to be the best tool for this approach.


    . Rates of cholesterol, ubiquinone, dolichol and dolichyl-P biosynthesis in rat brain slices. FEBS Lett. 1990 Aug 20;269(1):15-8. PubMed.

    . Alpha-2 macroglobulin is genetically associated with Alzheimer disease. Nat Genet. 1998 Aug;19(4):357-60. PubMed.

    . On the turnover of brain cholesterol in patients with Alzheimer's disease. Abnormal induction of the cholesterol-catabolic enzyme CYP46 in glial cells. Neurosci Lett. 2001 Nov 13;314(1-2):45-8. PubMed.

    . Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer's disease in late onset families. Science. 1993 Aug 13;261(5123):921-3. PubMed.

    . Cholesterol metabolism in the brain. Curr Opin Lipidol. 2001 Apr;12(2):105-12. PubMed.

    . Diet and risk of dementia: Does fat matter?: The Rotterdam Study. Neurology. 2002 Dec 24;59(12):1915-21. PubMed.

    . Simvastatin strongly reduces levels of Alzheimer's disease beta -amyloid peptides Abeta 42 and Abeta 40 in vitro and in vivo. Proc Natl Acad Sci U S A. 2001 May 8;98(10):5856-61. PubMed.

    . Treatment with controlled-release lovastatin decreases serum concentrations of human beta-amyloid (A beta) peptide. Int J Neuropsychopharmacol. 2001 Jun;4(2):127-30. PubMed.

    . Statins and the risk of dementia. Lancet. 2000 Nov 11;356(9242):1627-31. PubMed.

    . Apolipoprotein E epsilon4 allele, elevated midlife total cholesterol level, and high midlife systolic blood pressure are independent risk factors for late-life Alzheimer disease. Ann Intern Med. 2002 Aug 6;137(3):149-55. PubMed.

    . Midlife vascular risk factors and late-life mild cognitive impairment: A population-based study. Neurology. 2001 Jun 26;56(12):1683-9. PubMed.

    . Polymorphism in the cholesterol 24S-hydroxylase gene is associated with Alzheimer's disease. Mol Psychiatry. 2002;7(8):899-902. PubMed.

    . cDNA cloning of cholesterol 24-hydroxylase, a mediator of cholesterol homeostasis in the brain. Proc Natl Acad Sci U S A. 1999 Jun 22;96(13):7238-43. PubMed.

    . Statin-induced inhibition of the Rho-signaling pathway activates PPARalpha and induces HDL apoA-I. J Clin Invest. 2001 Jun;107(11):1423-32. PubMed.

    . Serum total cholesterol, apolipoprotein E epsilon 4 allele, and Alzheimer's disease. Neuroepidemiology. 1998;17(1):14-20. PubMed.

    . The association between APOE and dementia does not seem to be mediated by vascular factors. Neurology. 2000 Jan 25;54(2):397-402. PubMed.

    . A cholesterol-lowering drug reduces beta-amyloid pathology in a transgenic mouse model of Alzheimer's disease. Neurobiol Dis. 2001 Oct;8(5):890-9. PubMed.

    . Use of lipid-lowering agents, indication bias, and the risk of dementia in community-dwelling elderly people. Arch Neurol. 2002 Feb;59(2):223-7. PubMed.

    . Pravastatin in elderly individuals at risk of vascular disease (PROSPER): a randomised controlled trial. Lancet. 2002 Nov 23;360(9346):1623-30. PubMed.

    . Cholesterol depletion inhibits the generation of beta-amyloid in hippocampal neurons. Proc Natl Acad Sci U S A. 1998 May 26;95(11):6460-4. PubMed.

    . Treatment with simvastatin in normocholesterolemic patients with Alzheimer's disease: A 26-week randomized, placebo-controlled, double-blind trial. Ann Neurol. 2002 Sep;52(3):346-50. PubMed.

    . Pleiotropic effects of 3-hydroxy-3-methylglutaryl coenzyme a reductase inhibitors. Arterioscler Thromb Vasc Biol. 2001 Nov;21(11):1712-9. PubMed.

    . Cholesterol-dependent gamma-secretase activity in buoyant cholesterol-rich membrane microdomains. Neurobiol Dis. 2002 Feb;9(1):11-23. PubMed.

    . Decreased prevalence of Alzheimer disease associated with 3-hydroxy-3-methyglutaryl coenzyme A reductase inhibitors. Arch Neurol. 2000 Oct;57(10):1439-43. PubMed.

    . Serum lipoprotein levels, statin use, and cognitive function in older women. Arch Neurol. 2002 Mar;59(3):378-84. PubMed.

    . Incidence and risk factors of vascular dementia and Alzheimer's disease in a defined elderly Japanese population: the Hisayama Study. Neurology. 1995 Jun;45(6):1161-8. PubMed.

  2. This is a report of a large cohort study finding no association between dietary fat consumption and risk of dementia. The large size and long follow-up of the cohort, and the use of validated measures of dietary intake and diagnosis of dementia support the validity of the findings. There is one large omission in the study, however. The effect of the ApoE-ε4 allele on the association between dietary fats and incident dementia is not considered. Previous studies have found an association between high intake of fats and incident Alzheimer's disease in individuals with the ApoE-ε4 allele (Neurobiol Aging. 2000;21 (suppl):S246. Abstract 1124.; Arch Neurol. 2002 Aug;59(8):1258-63), and the ApoE gene has been reported to modulate the effects of dietary exposures (Am J Clin Nutr. 2001;73:669-70). Judgment on the findings of this study by Engelhart et al. should be withheld until analyses stratified by the presence of the ApoE-ε4 allele are reported.


    . Caloric intake and the risk of Alzheimer disease. Arch Neurol. 2002 Aug;59(8):1258-63. PubMed.

    . The APOE gene and diets--food (and drink) for thought. Am J Clin Nutr. 2001 Apr;73(4):669-70. PubMed.

  3. This report is one of nearly 40 others on Alzheimer's disease by the same research group using the Rotterdam study data. The article sets out to test the hypothesis that "cholesterol increases and n-3 PUFA reduce dementia risk." The results do not prove the hypothesis, and the article describes the study limitations and makes readers aware that "larger, prospective studies with longer follow-up periods are needed to confirm our findings." I see additional flaws in this study. First, it excludes data and discussion of ApoE allele variants despite their obvious importance in lipid metabolism. The same investigators reported on this significance earlier this year in another report on diet and AD on the same study population (see Engelhart, 2002). Second, I remain unconvinced by the authors' discussion of the discrepancy between the results of their previous article (Kalmijn, 1997) on the identical subject and this current article. In spite of this, the current paper is an important proof of principle that the involvement of cholesterol and other lipids in neurodegeneration in general, and in Alzheimer's disease in particular, is complex and extends beyond the role of cholesterol in amyloid β generation. Correspondingly, insufficient understanding of the use of statins (or lipid-modulating lipid ) to treat Alzheimer's disease may cause problems rather then a cure. We discussed our vision of the role of brain fat (particularly cholesterol) as a primary etiologic factor in AD and cautioned against oversimplification. Other cautions are provided in three other Neurology reports published this year. Based on the above, I do not think that the Rotterdam study questions links between fat and dementia risk.

    See also:
    Koudinov AR, Koudinova NV, Beisiegel U. Cholesterol homeostasis failure at neuromuscular junctions and CNS synapses: a unifying cause of synaptic degeneration ? Letter to Neurology (2002a).

    Koudinov AR, Koudinova NV. Alzheimer‚s anti-amyloid vaccination and statins: two approaches, one dogma. The time for change. Letter to the British Medical Journal (2002). Abstract

    Koudinov AR, Koudinova NV. Brain Cholesterol Pathology is the Cause of Alzheimer's Disease. ARF hypotheses. Posted 19 August 2002.

    Alzheimer's research forum. Cholesterol and Alzheimer's-Charging Fast but Still at a Distance From Solid Answers. ARF live discussion held 19 November 2002.


    . Cholesterol and Alzheimer's disease. Biochem Soc Trans. 2002 Aug;30(4):525-9. PubMed.

    . A fluid connection: cholesterol and Abeta. Proc Natl Acad Sci U S A. 2001 May 8;98(10):5371-3. PubMed.

    . Statins and risk of polyneuropathy: a case-control study. Neurology. 2002 May 14;58(9):1333-7. PubMed.

    . Are users of lipid-lowering drugs at increased risk of peripheral neuropathy?. Eur J Clin Pharmacol. 2001 Mar;56(12):931-3. PubMed.

    . Cholesterol and Alzheimer's disease: is there a link?. Neurology. 2002 Apr 9;58(7):1135. PubMed.

  4. We appreciate the comment by Jose Luchsinger and Richard Mayeux. We did not include the results of our analyses stratified according to apoE genotype but we agree that this is relevant additional information. We found no relation between the various measures of dietary fat intake and risk of dementia and subtypes of dementia, and this was similar across strata of ApoE genotype.

    Benjamin Wolozin lists several factors that are related to cholesterol and possibly to risk of Alzheimer’s disease (statin use, serum cholesterol levels, dietary fat intake, brain cholesterol metabolism). Obviously, this does not imply that if they are indeed related to risk of Alzheimer's disease this is all through the same mechanism, nor that this is always through a direct effect of cholesterol.

    The studies that did observe a relation between plasma cholesterol levels and risk of AD evaluated mid-life cholesterol levels in relation to late-life dementia risk (Kivipelto et al., 2002; Kivipelto et al., 2001; Notkola et al., 1998). It is well conceivable that this relation is mediated by a detrimental effect of high cholesterol levels on the vasculature (Breteler, 2000). Interestingly, studies evaluating late-life cholesterol levels do not find a relation with increased risk of Alzheimer’s disease (Slooter, 2000). This suggests that the claimed effect of statin use on risk of Alzheimer’s disease, if real, is not through lowering of peripheral cholesterol levels. This fits with our current finding that late-life dietary fat intake is not related to risk of Alzheimer’s disease within six years. Several other mechanisms can be conceived through which statins might affect Alzheimer risk, and these should be evaluated. We emphasize, though, that convincing evidence for an effect of statins on risk of Alzheimer’s disease is still lacking. The observational studies thus far were all potentially biased and their findings, though hypothesis-raising, cannot be considered evidence. The fact that they showed largely similar results may just as well reflect that they are biased in an identical way as it could mean that there is indeed a beneficial effect of statins on disease risk.

    The notion that cholesterol and cerebral cholesterol metabolism are critical in the development of Alzheimer's disease remains an extremely interesting hypothesis in need of further investigation. Basic scientists, epidemiologists and clinical researchers may provide complementary evidence to further our understanding of this possible mechanism.


    . Apolipoprotein E epsilon4 allele, elevated midlife total cholesterol level, and high midlife systolic blood pressure are independent risk factors for late-life Alzheimer disease. Ann Intern Med. 2002 Aug 6;137(3):149-55. PubMed.

    . Serum total cholesterol, apolipoprotein E epsilon 4 allele, and Alzheimer's disease. Neuroepidemiology. 1998;17(1):14-20. PubMed.

    . Vascular risk factors for Alzheimer's disease: an epidemiologic perspective. Neurobiol Aging. 2000 Mar-Apr;21(2):153-60. PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Fat and Calories Mean Higher AD Risk

Paper Citations

  1. . Dietary fat intake and the risk of incident dementia in the Rotterdam Study. Ann Neurol. 1997 Nov;42(5):776-82. PubMed.

Other Citations

  1. live discussion

Further Reading


  1. . Calorie restriction extends Saccharomyces cerevisiae lifespan by increasing respiration. Nature. 2002 Jul 18;418(6895):344-8. PubMed.

Primary Papers

  1. . Diet and risk of dementia: Does fat matter?: The Rotterdam Study. Neurology. 2002 Dec 24;59(12):1915-21. PubMed.