Abstract:OBJECTIVE To assessthe relationship between body fat distribution and incident lower-extremity arterial disease (LEAD). RESEARCHDESIGN AND METHODS Weincluded 155,925 postmenopausal women with anthropometric measures from theWomen’s Health Initiative who had no known LEAD at recruitment. A subset of 10,894 participants had body composition dataquantified by dual energy X-rayabsorptiometry (DXA). Incident cases ofsymptomatic LEAD were ascertained and adjudicated via medicalrecord review.RESULTS We identified 1152 incident LEAD cases during amedian 18.8 years follow-up. After multivariable adjustment and mutualadjustment, waist and hipcircumference was positively and inversely associated with risk of LEAD, respectively (both P-trend values<0.0001). In a subset (n = 22,561) where various cardiometabolic biomarkerswere quantified, a similar positive association of waist circumference withrisk of LEAD was eliminatedafter adjustment for diabetes and HOMA-IR (P-trend = 0.89), whereas hipcircumference remained inversely associated with the risk after adjustment for majorcardiometabolic traits (P-trend = 0.0031). In the DXA subset, higher trunk fat (P-trend = 0.0081)and higher leg fat (P-trend <0.0001) was associated with higher and lowerrisk of LEAD, respectively. Furtheradjustment for diabetes, dyslipidemia, and blood pressure diminished theassociation for trunk fat (P-trend = 0.49), yet the inverse association for legfat persisted (P-trend = 0.0082).CONCLUSIONSAmong USpostmenopausal women, a positive association of upper-body fat with risk of LEAD appeared to be attributable totraditional risk factors especially insulin resistance. Lower-body fat wasinversely associated with risk of LEADbeyond known risk factors.

Body Fat Distribution, Cardiometabolic Traits, and Risk of Major Lower-Extremity Arterial Disease in Postmenopausal Women