Malnutrition and Risk of Structural Brain Changes Seen on Magnetic Resonance Imaging in Older Adults

OBJECTIVES: To study the associations between protein energy malnutrition, micronutrient malnutrition, brain atrophy, and cerebrovascular lesions. DESIGN: Cross-sectional. SETTING: Geriatric outpatient clinic. PARTICIPANTS: Older adults (N = 475; mean age 80 +/- 7). MEASUREMENTS: Nutritional status was assessed using the Mini Nutritional Assessment (MNA) and according to serum micronutrient levels (vitamins B1, B6, B12, D; folic acid). White matter hyperintensities (WMHs), global cortical brain atrophy, and medial temporal lobe atrophy on magnetic resonance imaging (MRI) were rated using visual rating scales. Logistic regression analyses were performed to assess associations between the three MNA categories (< 17, 17-23.5, = 23.5) and micronutrients (per SD decrease) and WMHs and measures of brain atrophy. RESULTS: Included were 359 participants. Forty-eight participants (13%) were malnourished (MNA < 17), and 197 (55%) were at risk of malnutrition (MNA = 17-23.5). Participants at risk of malnutrition (odds ratio (OR) = 1.93, 95% confidence interval (CI) = 1.01-3.71) or who were malnourished (OR = 2.80, 95% CI = 1.19-6.60) had a greater probability of having severe WMHs independent of age and sex than those with adequate nutritional status. Results remained significant after further adjustments for cognitive function, depressive symptoms, cardiovascular risk factors, history of cardiovascular disease, smoking and alcohol use, and micronutrient levels. Lower vitamin B1 (OR = 1.51, 95% CI = 1.11-2.08) and B12 (OR = 1.45, 95% CI = 1.02-2.04) levels were also related to greater risk of severe WMHs, independent of age and sex. Results remained significant after additional adjustments. MNA and vitamin levels were not associated with measures of brain atrophy. CONCLUSION: Malnutrition and lower vitamin B1 and B12 levels were independently associated with greater risk of WMHs. Underlying mechanisms need to be further clarified, and whether nutritional interventions can modify these findings also needs to be studied.