A training program in portable fluoroscopy for the detection of glass in soft tissues.

Abstract

The role of emergency department (ED) bedside fluoroscopy is incompletely defined. One potential application is the evaluation of glass foreign bodies (GFBs). We studied the effect of a training intervention on the ability of emergency physicians to detect GFBs in an established chicken leg model. We also studied performance one week after training. Assessment of this teaching intervention involved nine emergency medicine residents with minimal prior hands-on fluoroscopy experience. Zero, one, or two clear glass shards no larger than 1.0 mm in any dimension were inserted into fresh chicken legs. Each subject scanned 12 to 15 legs without being given any feedback to determine baseline accuracy. Next, 27 to 30 more legs were immediately scanned with feedback given after each response. Subjects scanned 15 more legs one week later without feedback as follow-up. Data were analyzed using mixed-model analysis of variance. At baseline, overall sensitivity and specificity were 0.40 and 0.61, respectively. At follow-up, sensitivity and specificity were 0.93 and 0.91, respectively. Subjects were significantly more accurate during training and follow-up than at baseline. This level of proficiency reached during training was maintained and continued to improve one week later. Emergency medicine residents previously inexperienced in fluoroscopy may, through a brief training intervention, acquire the skill to accurately detect small pieces of glass in a chicken leg model, and maintain proficiency one week later. Further studies of emergency department fluoroscopy in GFB management in humans are warranted.