ROC analysis of noninvasive tests for peripheral arterial disease

The purpose of this study is to evaluate the diagnostic accuracy of selected noninvasive tests for assessing peripheral arterial disease. The ankle/brachial index (ABI) and the femoral and popliteal pulsatility indices (PI), and combinations of these tests, were evaluated using receiver operating characteristic (ROC) analysis to determine their diagnostic accuracy depending on the localization of the disease. Verification bias, introduced by the preferential selection of patients for angiography based on the noninvasive test results, was examined. This study suggests that, in assessing whether a patient has significant peripheral arterial disease (lesions > or = 50%), determining an ABI is justified (ROC area 0.95 +/- 0.02). For disease localized to the aortoiliac segment, performing a single test, the femoral PI, is sufficient (ROC area 0.80 +/- 0.04). For disease including the femoropopliteal and infrapopliteal segments, a combination of tests is necessary. Utilized threshold values need to be adjusted for verification bias.