Send to

Choose Destination
Radiology. 2014 Jan;270(1):205-11. doi: 10.1148/radiol.13130743. Epub 2013 Oct 28.

Increased number of microinfarcts in Alzheimer disease at 7-T MR imaging.

Author information

From the C.J. Gorter Center for High-Field-Strength MRI (S.v.R., A.M.v.O., M.J.V., A.G.W., M.A.v.B., J.v.d.G.), Department of Radiology (S.v.R., A.M.v.O., M.J.V., A.G.W., M.A.v.B., J.v.d.G.), and Department of Gerontology and Geriatrics (G.J.B.), Leiden University Medical Center, PO Box 9600, C3-Q, 2300 RC Leiden, the Netherlands; Alzheimer Center and Department of Neurology, Neuroscience Campus Amsterdam, Amsterdam, the Netherlands (J.D.C.G., W.M.v.d.F., P.S.); Departments of Radiology (F.B.) and Epidemiology and Biostatistics (W.M.v.d.F.), VU University Medical Center, Amsterdam, the Netherlands; and Department of Gerontology and Geriatrics, Bronovo Hospital, Den Haag, the Netherlands (G.J.B.).



To assess the prevalence and number of cortical microinfarcts in patients with Alzheimer disease (AD) by using a 7-T magnetic resonance (MR) imaging system, to assess the independent association of cortical microinfarcts with cognitive dysfunction, and to investigate potential confounding effects of the coexisting presence of cerebral amyloid angiopathy (CAA).


The local institutional review board approved this study. In all cases, informed consent was obtained. High-spatial-resolution fluid-attenuated inversion recovery and T2*-weighted images were acquired in 14 AD patients and 18 control subjects to assess the presence of microinfarcts and microbleeds. Presence of CAA was assessed according to the Boston criteria. Image analysis was performed independently by two reviewers. Mann-Whitney U test was performed to assess differences in number of microinfarcts between groups. Negative binomial regression models were used to assess the association between diagnosis of AD and diagnosis of CAA and number of microinfarcts, between diagnosis of AD and number of microbleeds and number of microinfarcts, and between cognitive function and number of microinfarcts, all corrected for age and sex.


Interobserver agreement was excellent for detecting microinfarcts (κ = 0.91) (P < .001). Patients with AD demonstrated higher number (P = .005) of microinfarcts (mean, 7.2) compared with control subjects (mean, 1.8). Negative binomial regression models showed an independent association between AD and number of microinfarcts (P = .006) and a trend for CAA and microinfarcts (P = .052). A negative correlation was found between cognitive function and the number of microinfarcts (P = .009).


Patients with AD show more microinfarcts than do control subjects, the number of microinfarcts correlates with global cognitive performance, and the presence of microinfarcts was mainly AD rather than CAA related.

[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Atypon
Loading ...
Support Center