Objective: Several studies report a higher prevalence of peripheral arterial disease (PAD) in women and among blacks. These studies based their PAD definition on an ankle-brachial index (ABI) <0.90. We hypothesized that there is an inherent contribution of gender and ethnicity to normal ABI values, independent of biologic and social disparities that exist between gender and ethnic groups. Consequently, an ABI threshold that disregards these fundamental gender-related and ethnicity-related differences could partly contribute to reported prevalence differences.
Methods: A cross-sectional study was designed as part of the Multi-Ethnic Study of Atherosclerosis (MESA), a multicenter United States population study. We selected a subgroup of participants with unequivocally normal ABIs (1.00 to 1.30), and additionally excluded participants with any major PAD risk factor (smoking, diabetes, dyslipidemia, hypertension). In a linear model with ABI as the dependent variable, demographic, clinical, biologic, and social variables were introduced as independent factors.
Results: Among 1775 healthy participants, there was no association between ABI level and subclinical cardiovascular disease (coronary calcium or carotid plaque). Male gender, weight, and high education level were positively correlated with ABI, whereas black race, triglycerides, pack-years (in past smokers), and pulse pressure were negatively correlated. In the fully adjusted model, women had about 0.02 lower ABI values than men, and blacks showed ABI values about 0.02 lower than non-Hispanic whites.
Conclusion: These data suggest intrinsic ethnic and gender differences in ABI. Such differences, although small in magnitude, are highly significant and can distort population estimates of disease burden.