Despite biological support for a role of angiotensin converting enzyme (ACE) in Alzheimer's disease (AD), studies assessing the ACE I/D polymorphism in AD are conflicting. We re-evaluated this association in the Rotterdam Study, a population-based cohort study. The mechanism of association was further explored by adjusting for vascular factors, and by analysing atrophy, white matter lesions and infarcts on MRI in non-demented individuals. Genotypes were available for 6488 participants. During average follow-up of 6 years 250 subjects developed AD. MRI data were available for 494 non-demented participants. Homozygosity for the I-allele conferred a slightly increased risk of AD compared to carrying a D-allele (RR 1.12 (95% CI 0.99-1.25)). This increase was only significant in women, and independent of vascular factors (RR 1.39 (95% CI 1.14-1.69)). Non-demented women with the II genotype had smaller hippocampal and amygdalar volumes. Vascular pathology was not significantly associated with ACE. This suggests a modest but significant increase in risk of AD and early AD pathology in women homozygous for the ACE I-allele independent of vascular factors.