Background and Purpose—Increasing attention has been paid to associations between cognitive dysfunction and brain microbleeds (MBs). Because all previous studies have investigated patients with neurological disorders, we examined subjects without neurological disorder in order to clarify pathogenic relationships.

Methods—A total of 518 consecutive adults without neurological disorder who had undergone health-screening tests of the brain were studied prospectively. Gradient-echo T2*-weighted MRI using a 1.5-T system was used to detect MBs. The Mini-Mental State Examination (MMSE) was administered to determine cognitive functions. MMSE scores <27 or >1.5 SDs below the age-related mean were regarded as subnormal.

Results—MBs were found in 35 subjects (6.8%). MMSE score <27 was found in 25 subjects (4.8%), with MMSE score >1.5 SDs below the age-related mean in 34 subjects (6.6%). Univariate analysis showed presence and number of MBs, short duration of education, and severe white matter hyperintensities as significantly associated with subnormal scores. In logistic regression analysis, presence of MBs (odds ratio [OR], 5.44; 95% CI, 1.83 to 16.19) and number of MBs (OR, 1.32; 95% CI, 1.04 to 1.68) still displayed significant associations with MMSE score <27. Logistic regression analysis revealed a significant relationship between presence (OR, 3.93; 95% CI, 1.44 to 10.74) and number (OR, 1.26; 95% CI, 1.01 to 1.59) of MBs and MMSE score >1.5 SDs below the age-related mean. Among MMSE subscores, "attention and calculation" was significantly lower in MB-positive subjects (P=0.017).

Conclusions—MBs appear to be primarily associated with global cognitive dysfunction.