A Folate, Vitamin B12 and Homocysteine-related dietary pattern and risk of subclinical atherosclerosis: findings from the MRC National Survey of Health and Development (NSHD)

Elevated homocysteine concentrations have been shown to increase the risk of stroke. However, the relationship between dietary patterns (DP) associated with homocysteine and associated nutrients (i.e. folate and vitamin B12) with subclinical atherosclerosis, an early marker of stroke, remains unknown. This study identifies a DP characterised by plasma homocysteine, dietary folate and vitamin B12, and investigates its association with carotid intima-media thickness (cIMT).

Food consumption in the NSHD, a British birth cohort, was assessed using 5-day estimated food diaries at 36 years (y), 43y, 53y and 60–64y. High resolution cIMT in the right and/or left common carotid artery was measured amongst 1565 participants at 60–64 years. Reduced rank regression was used to identify DPs based on plasma homocysteine (μmol/L), dietary folate (ug/kcal) and vitamin B12 (ug/kcal) as response variables. Multiple linear regression models examined cross-sectional and prospective associations between DP z-scores.

The DP explaining the maximum amount of total variation (4.5%) in all response variables was correlated with high intakes of folate (r=0.96) and vitamin B12 (r=0.27) and weakly correlated with homocysteine (r =0.10). The DP was characterised by high intakes of vegetables, fruit and breakfast cereal and low intakes of processed meat, white bread, sugar and preserves. No cross-sectional association was observed between DP z-scores and left (0.34% 95% CI −1.09%, 1.79%) or right cIMT (0.07%, 95%CI −1.28% to 1.45%), adjusting for age, gender, dietary misreporting, socioeconomic status, BMI, smoking, physical activity, lipid-lowering/antihypertensive medication and diabetes. Similarly, there was no association between DP z-scores measured at the ages of 36y, 43y, 53y and cIMT at 60–63y.

No cross-sectional or prospective associations were observed between cIMT and a DP high in dietary folate and vitamin B12 in this study. However, the observed DP explained only a small amount of variation in response variables and food intake. Further examination of DPs explaining greater amounts of variation in alternative response variables and their relationships with cIMT is warranted.