There are also a paucity of large-scale studies of cholesterol levels and dementia. In a cross-sectional analysis in the French Three-City Study of 9294 older subjects , high total cholesterol levels were related to increased risk of dementia after adjustment for numerous potential confounding factors (OR = 1.76, 95% CI 1.05–2.96), and findings were similar after exclusion of subjects with vascular disease (OR = 1.88). In an analysis of cases with dementia caused by AD (n = 112), this risk was attenuated (OR = 1.18, 95% CI 0.62–2.23); however, those results are difficult to interpret due to the wide confidence interval.
In a small prospective study, amongst 1026 participants in the Framingham Study (n = 77 AD cases) , there was no association between total or HDL cholesterol levels and AD risk (RR = 0.95, 95% CI 0.87–1.04, per each 10 mg dL−1 increase in total cholesterol; RR = 1.10, 95% CI 0.93–1.31, for increasing HDL). In a larger study of 3933 Japanese-American men (n = 215 dementia cases, n = 82 AD cases) , each 1 SD increase in total cholesterol was associated with a nonsignificant 10% increase in risk of dementia (95% CI 0.95–1.26); results were not provided separately for AD, although the investigators reported that there were no associations – again, however, the number of AD cases was small. In the WHICAP study of Caucasian, Hispanic and African-American subjects , both cross sectional and prospective analyses of AD were conducted. In the cross-sectional analyses, including 244 AD cases, HDL levels were associated with a reduced risk of AD (OR = 0.52, 95% CI 0.34–0.80; OR = 0.61, 95% CI 0.39–0.94; OR = 0.74, 95% CI 0.49–1.13, for second, third and fourth quartiles compared with the bottom quartile respectively). There was no association with total cholesterol or non-HDL cholesterol. In prospective analyses, including less than half the number of AD cases (n =119), there was no relation between HDL or non-HDL cholesterol and AD risk, but there was a significant trend of increasing risk of AD with increasing level of total cholesterol (RR = 0.48, 95% CI 0.26–0.86 for top versus bottom quartile, P = 0.04). Overall, these studies are inconsistent and do not conclusively support or refute a relation between cholesterol levels and AD risk; however, the studies are limited, and the existing data are generally based on investigations with relatively few AD cases, and thus fairly low statistical power to detect effects.
Two studies which reported a consistent relation between cholesterol and AD examined mid-life cholesterol levels. In one of these, the sample also was small with only 48 cases of AD . High levels of total cholesterol at mid-life were significantly related to an increased risk of AD an average 21 years later (RR = 2.8, 95% CI 1.2–6.7). The second study was much larger, with 721 cases of dementia diagnosed amongst 8845 subjects enrolled in the Kaiser Permanente healthcare system of Northern California . Those with high total cholesterol at mid-life had a 42% increased risk of dementia an average of 27 years later (95% CI 1.22–1.66); however, analyses of AD cases were not presented separately. As with hypertension (see above), it is possible that cholesterol levels earlier in life may have the largest impact on dementia development.