Independent Relevance of Different Measures of Adiposity for Carotid Intima-Media Thickness in 40 000 Adults in UK Biobank

Abstract

Background Uncertainty persists about carotid intima-media thickness (CIMT) as a marker of subclinical atherosclerosis and the independent relevance of different measures of adiposity for CIMT. We assessed the independent relevance of general adiposity (body mass index), central adiposity (waist circumference), and body composition (fat mass index and fat-free mass index) with CIMT among adults in the United Kingdom. Methods and Results Multivariable linear regression of cross-sectional analyses of UK Biobank assessed the mean percentage difference in CIMT associated with equivalent differences in adiposity measures. To assess independent associations, body mass index and waist circumference were mutually adjusted, as were fat mass index and fat-free mass index. Among 39 367 participants (mean [SD] age 64 [8] years, 52% female, 97% White), median (interquartile range) CIMT was 0.65 (0.14) mm in women and 0.69 (0.18) mm in men. All adiposity measures were linearly and positively associated with CIMT after adjusting for confounders. Fat-free mass index was most strongly associated with CIMT after adjustment for fat mass index (% difference in CIMT: 1.23 [95% CI 0.93-1.53] women; 3.44 [3.01-3.86] men), while associations of fat mass index were attenuated after adjustment for fat-free mass index (0.28 [-0.02, 0.58] women; -0.59 [-0.99, -0.18] men). After mutual adjustment, body mass index remained positively associated with CIMT, but waist circumference was completely attenuated. Conclusions Fat-free mass index was the adiposity measure most strongly associated with CIMT, suggesting that CIMT may reflect vascular compensatory remodeling rather than atherosclerosis. Hence, screening for subclinical atherosclerosis should evaluate carotid plaques in addition to CIMT.

Keywords: adiposity; carotid intima‐media thickness atherosclerosis; fat mass; fat‐free mass.

Figure 1. Mean percentage difference in CIMT per specified unit increase in adiposity measures, by sex.

All models adjusted for age, ethnicity, Townsend deprivation, smoking status, alcohol intake, physical activity, and family history of cardiovascular disease. Group‐specific estimates are plotted as squares, with the size of each square proportional to the amount of statistical information. Vertical lines represent group‐specific 95% CIs. CIMT indicates carotid–intima media thickness.

Figure 2. Mean percentage difference in CIMT per specified unit increase in adiposity measures after basic adjustment and mutual adjustment for adiposity measures, by sex.

Analyses adjusted for age, ethnicity, Townsend deprivation, smoking status, alcohol intake, physical activity, and family history of cardiovascular disease with further mutual adjustment for other adiposity measures. Point estimates are plotted as squares, with the size of each square proportional to the amount of statistical information. Horizontal lines represent 95% CIs. BMI indicates body mass index; and CIMT, carotid–intima media thickness.

Figure 3. Mean percentage difference in CIMT per specified unit increase in adiposity measures, by age group and sex.

All models adjusted for age, ethnicity, Townsend deprivation, smoking status, alcohol intake, physical activity, and family history of cardiovascular disease. Point estimates are plotted as squares, with the size of each square proportional to the amount of statistical information. Horizontal lines represent 95% CIs. CIMT indicates carotid–intima media thickness.

Figure 4. Mediation analyses showing the mean percentage difference in CIMT per specified unit increase in adiposity measures, by sex.

Basic adjustment was made for age, ethnicity, Townsend deprivation, smoking status, alcohol intake, physical activity, and family history of cardiovascular disease. Point estimates are plotted as squares, with the size of each square proportional to the amount of statistical information. Horizontal lines represent 95% CIs. BMI indicates body mass index; CIMT, carotid–intima media thickness; and SBP, systolic blood pressure.