Patterns of abdominal fat distribution: the Framingham Heart Study

Abstract

Objective: The prevalence of abdominal obesity exceeds that of general obesity. We sought to determine the prevalence of abdominal subcutaneous and visceral obesity and to characterize the different patterns of fat distribution in a community-based sample.

Research design and methods: Participants from the Framingham Heart Study (n = 3,348, 48% women, mean age 52 years) underwent multidetector computed tomography; subcutaneous adipose tissue (SAT) and visceral adipose tissue (VAT) volumes were assessed. Sex-specific high SAT and VAT definitions were based on 90th percentile cut points from a healthy referent sample. Metabolic risk factors were examined in subgroups with elevated SAT and VAT.

Results: The prevalence of high SAT was 30% (women) and 31% (men) and that for high VAT was 44% (women) and 42% (men). Overall, 27.8% of the sample was discordant for high SAT and high VAT: 19.9% had SAT less than but VAT equal to or greater than the 90th percentile, and 7.9% had SAT greater than but VAT less than the 90th percentile. The prevalence of metabolic syndrome was higher among women and men with SAT less than the 90th percentile and high VAT than in those with high SAT but VAT less than the 90th percentile, despite lower BMI and waist circumference. Findings were similar for hypertension, elevated triglycerides, and low HDL cholesterol.

Conclusions: Nearly one-third of our sample has abdominal subcutaneous obesity, and >40% have visceral obesity. Clinical measures of BMI and waist circumference may misclassify individuals in terms of VAT and metabolic risk.

Figure 1

Prevalence of high SAT (A) and high VAT (B) by age-group in women (▪) and men (□). Error bars represent SE. For SAT, the linear trends were not significant (P = 0.76 for men and P = 0.11 for women). For VAT, the linear trends were significant for both men (P < 0.0001) and women (P < 0.0001).

Figure 2

Prevalence of high SAT or high VAT by BMI category in women (A) and men (B) and by waist circumference category in women (C) and men (D). Error bars represent SE. A and B: □, normal weight; , overweight; ▪, obese. C and D: □, normal waist circumference; ▪, high waist circumference.

Figure 3

Prevalence of metabolic risk factors (MetS) and CVD by SAT/VAT concordant and discordant categories in women (A) and men (B). Age-adjusted P < 0.0001 across all four categories for each risk factor except cardiovascular disease (P = 0.01 for men and P = 0.05 for women). *P < 0.01 for the low SAT/high VAT and high SAT/low VAT comparisons. Error bars represent upper one-sided 97.5% CIs. HTN, hypertension; IFG, impaired fasting glucose; TG, triglyceride. , low SAT and VAT; □, high SAT and low VAT; , low SAT and high VAT; ▪, high SAT and VAT.