INTRODUCTION: The presence of cerebrovascular pathology may increase the risk of clinical diagnosis of Alzheimer's disease (AD).

METHODS: We examined excess risk of incident clinical diagnosis of AD (probable and possible AD) posed by the presence of lacunes and large infarcts beyond AD pathology using data from the Statistical Modeling of Aging and Risk of Transition study, a consortium of longitudinal cohort studies with more than 2000 autopsies. We created six mutually exclusive pathology patterns combining three levels of AD pathology (low, moderate, or high AD pathology) and two levels of vascular pathology (without lacunes and large infarcts or with lacunes and/or large infarcts).

RESULTS: The coexistence of lacunes and large infarcts results in higher likelihood of clinical diagnosis of AD only when AD pathology burden is low.

DISCUSSION: Our results reinforce the diagnostic importance of AD pathology in clinical AD. Further harmonization of assessment approaches for vascular pathologies is required.

Document Type


Publication Date


Notes/Citation Information

Published in Alzheimer's & Dementia, v. 13, issue 6.

© 2016 the Alzheimer's Association

© 2016. This manuscript version is made available under the CC-BY-NC-ND 4.0 license https://creativecommons.org/licenses/by-nc-nd/4.0/.

The document available for download is the authors' post-peer-review final draft of the article.

Digital Object Identifier (DOI)


Funding Information

This study was supported by National Institute of Aging Funding including: R01 AG038651, P30 AG008017, P01 AG043362, K25 AG043546, P30 AG028383, P30 AG010161, R01 AG015819, R01 AG017917, P50 AG005681, R01 AG034119, P50AG005681, P01AG003991 and P30 AG053760P30.

NIHMS831451-supplement-01.pdf (190 kB)
Supplemental Table 1. Results of Cox Proportional Hazard Model with Pathology Patterns as Independent Variables with Outcomes being Incidence of Clinical AD using autopsied cases after 2006 (N=542)