Detection of soft tissue foreign bodies by nurse practitioner-performed ultrasound

BACKGROUND This study aimed to evaluate the accuracy of emergency nurse practitioner (NP)-performed point-of-care ultrasound (POCUS) for the detection of soft tissue foreign bodies (FBs). METHODS Following a 2-h training session, ten NPs were assessed on their ability to detect various FBs in an experimental model. FBs (wood, metal and plastic) were inserted randomly into eight experimental models (uncooked chicken thighs) by an independent observer. Control experimental models had no FB inserted, but all had a 1-cm incision made on their surface. NPs, blinded to the type of model, were then assessed on their ability to detect the FBs by ultrasound examination using high-frequency linear transducers (Toshiba Nemio). Models were also scanned by two experienced emergency physicians (EPs) as a further control. RESULTS Overall, NP-performed POCUS detected 47 of the 60 foreign bodies with a sensitivity, specificity, positive predictive value and negative predictive value of 78.3%, 50%, 82% and 43%, respectively, compared with 83.3%, 75%, 90.9% and 60% for EPs. Sensitivity for detecting specific types of FB was 95%, 85% and 50% for wood, metal and plastic, respectively, for NP-performed POCUS, compared with 100%, 100% and 50% in the EP group. CONCLUSIONS NPs with no previous ultrasound experience can detect soft tissue FBs with accuracy comparable to that of EPs in an experimental model. Test sensitivity was high for wood and metal foreign bodies. Specificity was generally low.