Hyperhomocysteinemia and risk of incident cognitive outcomes: An updated dose-response meta-analysis of prospective cohort studies.
Level 3 - non-randomized controlled study
Systematic review and meta-analysis of prospective cohort studies (Level 3 evidence)
PubMed 30826501 · doi:10.1016/j.arr.2019.02.006
What was done
A systematic review and dose-response meta-analysis of prospective cohort studies identified via PubMed and EMBASE up to June 18, 2018. The authors evaluated the relationship between blood homocysteine concentrations and incident all-cause dementia, Alzheimer-type dementia, vascular dementia, and cognitive impairment without dementia (CIND). Pooled relative risks (RRs) and 95% confidence intervals (CIs) were estimated using fixed- or random-effects models.
What was found
Across 28 prospective cohort studies with 28,257 participants (follow-up 2.7 to 35 years), there were 2,557 incident cognitive outcome cases (1,035 all-cause dementia, 530 Alzheimer's disease, 92 vascular dementia, and >900 CIND). A linear dose-response relation was found for Alzheimer-type dementia (P > 0.05 for non-linearity), with a pooled RR of 1.15 (95% CI: 1.04 to 1.26; I2 = 56.6%, n = 5) per 5 μmol/L increase in blood homocysteine. Begg's and Egger's tests showed no clear evidence of publication bias for Alzheimer dementia (P = 0.806 and P = 0.084). Blood homocysteine was not appreciably associated with all-cause dementia, vascular dementia, or CIND due to publication bias and low statistical power.
Why it matters
The findings identify a modest linear increase in Alzheimer-type dementia risk with rising homocysteine concentrations. However, prospective evidence remains insufficient to establish links with other dementia subtypes or general cognitive impairment.
Limits
The dose-response finding for Alzheimer dementia is based on only 5 studies with moderate heterogeneity (I2 = 56.6%). Analyses for vascular dementia and CIND were limited by low case numbers (e.g., 92 vascular dementia cases) and publication bias. As a meta-analysis of observational cohorts, unmeasured confounding cannot be ruled out and causality cannot be determined.
Cited by
- supports Elevated homocysteine is associated with brain dysfunction and dementia.