Physical examination to screen for peripheral artery disease in a defined Primary Care population: A diagnostic accuracy study

INTRODUCTION Peripheral arterial disease (PAD) is an underdiagnosed prevalent disease which implies high cardiovascular risk. Professionals usually depend on physical examination to screen for PAD. OBJECTIVE To assess the diagnostic accuracy of physical examination to screen for PAD in a rural Primary Care population and to evaluate the nurse-physician level of agreement in pedal pulse palpation. METHODS Diagnostic accuracy study in which two experienced professionals (physician-nurse) prospectively performed pedal pulse palpation (grading as absent, reduced, normal, or bounding), femoral bruit auscultation and calf circumference (index tests) comparing with Doppler ABI (reference test, positive cut-off: 0.9 ≥ ABI ≥ 1.4) in 158 consecutive subjects. INCLUSION CRITERIA presence of diabetes, dyslipidaemia, hypertension, smoking habit (current or former), or age ≥ 65. RESULTS Of 315 legs included, PAD was confirmed in 38 (12.1%) legs. Absent dorsalis pedis (DP) and posterior tibial (PT) pulses were found in 37 (11.7%) and 67 (21.3%) legs, respectively. Regarding nurse evaluation, when a positive test was set if DP or PT were absent (more sensitive cut-off), sensitivity was = 86.8 (95% CI: 74.8-98.9), specificity = 82.7 (95% CI: 78.0-87.3), likelihood ratio+ = 5.01 (95% CI: 3.77-6.67), likelihood ratio- = 0.16 (95% CI: 0.07-0.36), and diagnostic odds ratio (dOR) = 31.5 (95% CI: 11.7-84.8). Age, diabetes, and calcification (ABI ≥ 1.4) influenced the rate of a false negative finding in pedal palpation. Physician-nurse weighted kappa coefficient was = 0.649 (95% CI: 0.599-0.699). The presence of a femoral bruit auscultation had a dOR = 3.8 (95% CI: 1.1-13.1), and a calf circumference