Risk factors for peripheral arterial disease incidence in persons with diabetes: the Atherosclerosis Risk in Communities (ARIC) Study

Abstract

Background: Some risk factors for peripheral arterial disease (PAD) have been identified, but little information is available on PAD risk factors in individuals with diabetes. Methods: Using data from the Atherosclerosis Risk in Communities (ARIC) Study, we assessed the relation of traditional and non-traditional risk factors with the risk of PAD in 1651 participants with diabetes, but not PAD, at baseline. Incident PAD was defined as an ankle–brachial index (ABI) < 0.9 assessed at regular examinations; hospital discharge codes for PAD, amputation, or leg revascularization; or claudication assessed by annual questionnaire. Results: Over a mean of 10.3 years of follow-up, 238 persons developed incident PAD identified, yielding a PAD event rate of 13.9 per 1000 person years. Adjusted for sex, age, race, and center, the risk of developing PAD was increased 1.87-fold (95% confidence interval (95% CI): 1.36–2.57) in persons who were current smokers versus non-smokers, 2.27-fold (95% CI: 1.57–3.26) for baseline coronary heart disease (CHD) versus no baseline CHD, and 1.75-fold (95% CI: 1.18–2.60) for the highest quartile versus lowest quartile of triglycerides. We found no evidence of an association with other blood lipids or hypertension. Compared with the lowest quartiles, comparably-adjusted relative risks for the highest quartiles were 1.60 (95% CI: 1.10–2.33) for waist-to-hip ratio, 2.52 (95% CI: 1.70–3.73) for fibrinogen, 1.70 (95% CI: 1.17–2.47) for factor VIII, 1.73 (95% CI: 1.18–2.54) for von Willebrand factor, 2.15 (95% CI: 1.43–3.24) for white blood cell count, 1.81 (95% CI: 1.19–2.74) for serum creatinine, 0.55 (95% CI: 0.37–0.83) for serum albumin, and 2.73 (95% CI: 1.77–4.22) for carotid intima-media thickness. Persons who had a prior history of diabetes and were taking insulin had a relative risk of 1.97 (95% CI: 1.35–2.87) for future PAD events, compared with those with newly identified diabetes at baseline. In our final multivariable model, current smoking, prevalent CHD, elevated fibrinogen and carotid IMT, and a prior history of diabetes with insulin treatment were independently associated with greater PAD incidence. Conclusion: These markers might be useful to identify individuals with diabetes at particular risk for PAD.