Doppler extremity arterial diagnosis for optimization of treatment in the emergency department (DEAD FOOT).

Abstract

Outcomes in acute limb ischemia (ALI) depend on early recognition. Rapid evaluation methods in the emergency department (ED) include physical exam, hand-held Doppler (HH), or spectral Doppler (SD) using point-of-care ultrasound (POCUS). This study aims to estimate performance characteristics of HH versus emergency medicine (EM)-performed SD compared to angiography along with clinician confidence when evaluating for ALI. This was a prospective observational pilot study. A convenience sample of patients presenting to an urban, academic ED with concern for ALI who underwent angiography were eligible. The primary aim evaluated diagnostic performance of HH and SD in correctly classifying flow in posterior tibial (PT) and dorsalis pedis (DP) arteries in patients with 100% angiographic occlusion of the proximal vasculature. Binomial exact 95% confidence intervals (CIs) and Fisher's exact tests were used. Twenty-six patients were enrolled. Three cases (11.5% of 26) had normal angiography, five (19.2%) had partial occlusion, and 18 (69.2%) had 100% occlusion. For cases with 100% occlusion, HH always classified PT or DP as abnormal (sensitivity 100, 95% CI 81.5-100%). SD also showed high sensitivity (94.4, 95% CI 72.7-99.9%). Limited utility of specificity calculations was due to low true negatives (n = 3), but these measures were low for HH (66.7, 95% CI 9.4-99.2%) and SD (33.3, 95% CI 0.8-90.6%). There was no association (p = .305) between sonographer confidence and correct diagnostic classification for either HH or SD. In a patient population with suspected ALI, EM-performed HH and SD performed extremely well at identifying abnormal PT or DP flow in patients with complete (100%) angiographic occlusion. These results are useful to power larger trials to determine the role that SD may play in complementing HH evaluation for ALI.