Introduction:

The Cardiovascular Health Diet Index (CHDI) is a newly proposed diet quality score for cardiovascular health, adapted to the Brazilian food culture, and the first to include ultra-processed foods as one of the components.

Objectives:

Our aim was to prospectively investigate the association between the CHDI and its components with subclinical atherosclerosis.

Methods:

Data from the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil) cohort, consisting of 7,653 participants without prior cardiovascular disease and with plausible energy intake were used. Subclinical atherosclerosis was assessed by measuring carotid intima-media thickness (cIMT) at baseline (2008-2012) and at the third wave (2017- 2019). The CHDI (0-110 points) and its 11 components (fruits, vegetables, fish and seafood, sugar-sweetened beverages, whole grains, nuts, legumes, processed meat, red meat, dairy, and ultra-processed foods), were applied to dietary data obtained from a food frequency questionnaire at baseline. Linear mixed-effects models were used to assess the association between the CHDI total score and its components at baseline and changes in cIMT over time.

Results:

The average CHDI total score was 56.8 points (95% CI 56.6; 57.0). After a median 8-year follow-up period, a 10-point increase in the CHDI total score was associated with a decrease in cIMT of -0.0029 mm (-0.0051; -0.0008). Higher consumption of whole cereals (β -0.0088 mm, 95% CI -0.0017; -0.0001) and nuts (-0.0013 mm, -0.0022; -0.0005) led to a reduction in cIMT, while lower consumption of of processed meat (-0.0007 mm, -0.0015; -0.0001), sugar-sweetened beverages (-0.0013 mm, -0.0023; -0.0004), and ultra-processed foods (-0.0017 mm, -0.0027; -0.0007) also contributed to a reduction in cIMT, after controlling for various factors.

Conclusion:

In this Brazilian large cohort with individuals without overt CVD, higher scores in the CHDI and in whole cereals, nuts, sugar-sweetened beverages, processed meat and ultra-processed foods components were prospectively associated with decreased subclinical atherosclerosis after an 8-year follow-up period.