Ankle--brachial index, vascular calcifications and mortality in dialysis patients
Nephrology, Dialysis and Transplantation
Background. The ankle-brachial index (ABI) is a noninvasive method to evaluate peripheral artery disease (PAD). ABI <0.9 diagnoses PAD; ABI >1.3 is a false negative caused by noncompressible arteries. The aim of this study is to evaluate the association between ABI with vascular calcifications (VC) and with mortality, in haemodialysis (HD) patients. Methods. We studied 219 HD patients (60% male; 20% diabetic). At baseline, ABI was evaluated by a Doppler device. VCs were evaluated by two
... ated by two methods: the abdominal aorta calcification score (AACS) in a lateral plain X-ray of the abdominal aorta and the simple vascular calcification score (SVCS) in plain X-rays of the pelvis and hands. VC were also classified by their anatomical localization in main vessels (aorta and iliac-femoral axis) and in peripheral or distal vessels (pelvic, radial or digital). The cutoff values for the different VC scores in relation with ABI were determined by receiver operating characteristic curve analysis. Biochemical parameters were time averaged for the 6 months preceding ABI evaluation. Results. An ABI <0.9, an ABI >1.3 or a normal ABI were found, respectively, in 90 (41%), in 42 (19%) and in 87 (40%) patients. AACS !6 and SVCS >3 were found, respectively, in 98 (45%) and 95 (43%) patients. The adjusted odds ratio (OR) for having an ABI <0.9 was 2.5 (P ¼ 0.007) for AACS !6 and 4.5 (P < 0.001) for iliac-femoral calcification score (CS) !2. The adjusted OR for having an ABI >1.3 was 4.2 (P ¼ 0.003) for pelvic CS and 3.7 (P ¼ 0.006) for hand CS !2. During an observational period of 28.9 months, all-cause and cardiovascular mortality occurred, respectively, in 50 (23%) and in 29 (13%) patients. Adjusting for age, diabetes, P levels, HD duration and cardiovascular disease at baseline, an ABI <0.9 [hazard ratio (HR) ¼ 3.9, P < 0.001] and an ABI >1.3 (HR ¼ 2.7, P ¼ 0.038) were associated with allcause mortality; an ABI <0.9 (HR ¼ 7.2, P ¼ 0.002) and an ABI >1.3 (HR ¼ 5.1, P ¼ 0.028) were associated with cardiovascular mortality. Conclusions. Both low and high ABI were independent predictors of all-cause and cardiovascular mortality. VC in main arteries were associated with an ABI <0.9. VC in peripheral and distal arteries were associated with an ABI >1.3. ABI is a simple and noninvasive method that allows the identification of high cardiovascular risk patients.