Pulse Pressure and Prevalence of Peripheral Arterial Disease: The Atherosclerosis Risk in Communities (ARIC) Study
Recent reports have shown that pulse pressure (PP) is a meaningful independent predictor of some types of cardiovascular disease. However, the relationship between PP and peripheral arterial disease (PAD) is largely unexplored. We examine data from a large biracial cohort in order to determine whether PP has an independent association with PAD. A total of 10,592 subjects were examined from the ARIC study (20.5% African American) in 1987-1989. Participants taking anti-hypertension medications were excluded but those with untreated elevated (>=140/90) blood pressure (12.3%) were not. The presence of PAD was indicated by an ankle-brachial index <=0.9. Clinical predictors of PAD and potential confounders were examined individually for associations with PAD, after controlling for age. In a bivariate model, it was found that subjects in the highest quartile of PP (>52 mm Hg) were 2.27 times more likely to have PAD than those in the lowest quartile (<37 mm Hg). Controlling for age, sex, race, mean arterial pressure (MAP), diabetes mellitus, body mass index, high-density lipoprotein, cholesterol level, and smoking status in a multivariate logistic regression model, PP had a significant independent association with PAD. Every 10 mm Hg difference in PP was associated with a 41% increase in prevalence of PAD. PP was also shown to have a stronger association with PAD than either the non-significant systolic blood pressure (SBP) or non-significant diastolic blood pressure (DBP). Each of the three blood pressure components was analyzed independently in the above multivariate model, controlling for MAP. When SBP and DBP were included separately in the multivariate model (excluding MAP), every 10 mm Hg difference in PP was associated with a 73% and 31% increase in prevalence of PAD, respectively, whereas both SBP and DBP were not statistically significant. Conclusion: PP was independently associated with PAD in this population-based sample of subjects who were not being treated for hypertension.