Increased white matter signal hyperintensities in long-term abstinent alcoholics compared with nonalcoholic controls

Abstract

Background: The harmful effects of alcohol dependence on brain structure and function have been well documented, with many resolving with sufficient abstinence. White matter signal hyperintensities (WMSH) are thought to most likely be consequences secondary to the vascular (i.e., hypertension and atherosclerosis) effects of AD. We hypothesized that such effects would persist into long-term abstinence, and evaluated them in middle-aged long-term abstinent alcoholics (LTAA) compared with age and gender comparable nonalcoholic controls (NAC).

Methods: Ninety-seven participants (51 LTAA and 46 NAC) underwent cognitive, psychiatric, and structural brain magnetic resonance image evaluations. WMSH were identified and labeled as deep or periventricular by an automated algorithm developed in-house. WMSH volumes were compared between groups, and the associations of WMSH measures with demographic, alcohol use, psychiatric, and cognitive measures were examined within group.

Results: Long-term abstinent alcoholics had more WMSH than NAC. There was a significant group by age interaction, with WMSH increasing with age in LTAA, but not in NAC. Within LTAA, WMSH load was independently positively associated with alcohol burden and with age. No associations were evident between WMSH volumes and abstinence duration, family drinking history, years of education, or psychiatric or cognitive variables.

Conclusion: The magnitude of alcohol abuse was related to increased WMSH volume. The presence of an age effect in the LTAA but not the controls indicates a synergistic effect wherein alcohol advances the onset of aging-related WMSH formation. The increased WMSH load did not appear to have any significant clinical correlates, indicating that the white matter lesions in our sample may not have been severe enough to manifest as cognitive deficits. A limitation of the study is that we did not have data on the presence or severity of lifetime or current indices of vascular risk factors such as hypertension, smoking, or diabetes.

Figure 1

This figure shows the results of the WMSH delineation for a mid-ventricular slice in a 54-year-old male alcoholic, abstinent 10.7 years. Left: The original FLAIR image. Right: The final segmentation. The dark gray areas represent normal white matter, the medium gray areas represent deep WMSH, and the white areas represent periventricular WMSH.

Figure 2

WMSH volumes (adjusted for between subject differences in total white matter volume) vs. age in control subjects. R²_{periventricular WMSH} = 0.00297 and R²_{deep WMSH} = 0.0113.

Figure 3

WMSH volumes (adjusted for between subject differences in total white matter volume) vs. age in LTAA subjects. R²_{periventricular WMSH} = 0.161 and R²_{deep WMSH} = 0.201.