Abstract P395: Statin Use in Late Life Does Not Impact Risk of Dementia or Cognitive Change in the Next Few Years: Results of a Systematic Review and Methodological Commentary

Abstract

Introduction: The recently published American College of Cardiology and American Heart Association guidelines on the management of cholesterol expanded the population recommended to receive statins by approximately 13 million United States adults. Statins are hypothesized to reduce risk of dementia through treatment of hyperlipidemia, through anti-oxidant and anti-inflammatory effects, or through influence on brain amyloid-beta or tau metabolism. However, firm conclusions about whether statin use impacts cognitive decline and dementia remain elusive. Existing reviews have focused solely on trials, did not systematically discuss study quality, or discussed and meta-analyzed all observational studies as a group, which may be inappropriate when differences in study design or analyses yield non-comparable effect estimates. Hypothesis: The purpose of this systematic review is to summarize findings from randomized controlled trials and observational cohort studies, study designs that are most useful for evaluating the putative causal effect of statin use on cognition. We hypothesized that the literature would support a protective effect of statin use on cognitive decline and dementia. METHODS: Following a systematic literature search, we grouped eligible reports by study design and statistical approach, and provide specific commentary on findings, as well as the potential for and direction of bias. RESULTS AND CONCLUSIONS: While observational studies of statin use at or near the time of dementia diagnosis suggest a protective effect, these findings may be attributable to reverse causation. Randomized controlled trials and well-conducted observational studies of baseline statin use and subsequent cognition do not support a causal preventive effect of statin use in late-life on cognitive decline or dementia. This of course does not undermine existing recommendations regarding statin use for preventing cardiovascular disease. Important questions remain unanswered, including questions about the impact of midlife or long-term statin use. Much future human research on statins and cognition will be observational; careful study design and analysis will be required to avoid common sources of bias. We advocate continuing systematic review with attention to study quality and the likelihood of bias; the AlzRisk database entry on this topic, which is currently limited to studies reporting on Alzheimer’s disease, will be available shortly and will be updated over time (see www.alzrisk.org for reviews of 11 other potential non-genetic risk factors for Alzheimer’s disease).