Statins for the prevention of dementia.
Level 1 - systematic review of randomized trials
systematic review of double-blind randomized placebo-controlled trials
PubMed 19370582 · doi:10.1002/14651858.CD003160.pub2
What was done
An updated systematic review searched major databases (including MEDLINE, EMBASE, PsycINFO, CINAHL, and the Cochrane Dementia and Cognitive Improvement Group register) through October 2007 for double-blind, randomized placebo-controlled trials evaluating statins for the prevention of dementia or Alzheimer's disease in at-risk individuals. Two independent reviewers extracted data and assessed trial outcomes and adverse effects.
What was found
Two randomized trials met inclusion criteria: HPS 2002 and PROSPER 2002, totaling 26,340 participants aged 40–82 years with mean follow-ups of 5.0 and 3.2 years, respectively. Simvastatin and pravastatin reduced LDL cholesterol by 1.0 mmol/L and 1.02 mmol/L compared to placebo. In HPS 2002, incident dementia was identical between arms (31 cases in simvastatin vs. 31 cases in placebo), and cognitive impairment rates on the modified Telephone Interview for Cognitive Status were similar (23.7% in simvastatin vs. 24.2% in placebo), with no differences by age or cerebrovascular history. In PROSPER 2002, cognitive function declined at identical rates in both arms across the Mini Mental State Examination, Stroop, picture word learning, and letter digit coding tests. Statins showed no detrimental effects on cognition. Meta-analysis could not be performed due to differences in cognitive scales and testing schedules.
Why it matters
This review provides high-quality randomized evidence showing that statin therapy initiated in late life does not reduce the risk of dementia or cognitive decline, refuting earlier observational associations likely confounded by indication bias.
Limits
Only two trials were eligible, and heterogeneity in cognitive testing instruments and timing prevented quantitative meta-analysis. Statin initiation was studied exclusively in late life (ages 40–82, heavily weighted toward ≥70 years) over relatively short follow-up periods (3.2 to 5 years), leaving mid-life or longer-term effects unaddressed.
Cited by
- contradicts Failing to lower LDL cholesterol below 100 mg/dL or 70 mg/dL places individuals at higher risk for dementia.