Background: Accurate determination of progression of lower-extremity arterial occlusive disease (LEAOD) is required for natural history studies and evaluation of therapies for atherosclerosis.
Objective: To determine if changes in the ankle-brachial index (ABI) correlated with progression of LEAOD as determined by arteriography or duplex scanning.
Design: In patients with prior suprainguinal or infrainguinal lower-extremity revascularization, progression of LEAOD in native arteries was determined by comparing a preoperative (baseline) arteriogram with late follow-up arteriography or duplex scanning. Superficial femoral and popliteal arteries were graded as having less than 50% stenosis, 50% to 99% stenosis, or as being occluded. Tibial arteries were graded as continuously patent or occluded. Operated and nonoperated extremities were included in the study. The baseline ABI was performed postoperatively and repeated at follow-up arteriography or duplex scanning. Progression of LEAOD by the ABI was defined as a decrease in the ABI of 0.15 or greater. Progression of LEAOD by imaging studies was defined as an increase in 1 category of stenosis. Extremities with suprasystolic pressures were excluded.
Setting: Tertiary vascular surgical service. EXTREMITIES AND PATIENTS: One hundred ninety-three extremities were studied in 114 patients during a mean follow-up of 3.3 years.
Results: Seventy-two lower extremities (37.3%) showed progression of atherosclerosis by late follow-up arteriography or duplex scanning. Using the imaging studies as the criterion standard, the ABI had 102 true negatives, 29 true positives, 42 false negatives, and 20 false positives (sensitivity, 41%; specificity, 84%; positive predictive value, 59%; negative predictive value, 71%; and accuracy, 68%) for determining the progression of LEAOD.
Conclusions: The ABI is relatively insensitive in identifying the progression of LEAOD as demonstrated by the use of imaging studies. In studies of natural history or therapy for atherosclerosis, imaging studies should be used in preference to the ABI to evaluate the progression of LEAOD accurately.