Objective: The objective of the study was to determine the diagnostic accuracy of pediatric emergency physicians in diagnosing clavicle fractures by bedside ultrasound (US).
Methods: This was a prospective study of pediatric emergency department (ED) patients with suspected clavicle fractures conducted in a tertiary-care, freestanding pediatric hospital. A convenience sample of patients younger than 17 years underwent bedside US for detection of clavicle fracture by pediatric emergency physicians with limited US training. Ultrasound findings were compared with standard radiographs, which were considered the criterion standard. Pain scores using the validated color analog scale (0-10) were determined before and during US. Total length of stay in the ED, time to US, and time to radiograph were recorded.
Results: Fifty-eight patients were enrolled, of which 39 (67%) had fracture determined by radiograph. Ultrasound interpretation gave a sensitivity of 89.7% (95% confidence interval [CI], 75.8%-97.1%) and specificity of 89.5% (95% CI, 66.9%-98.7%). Positive and negative predictive values were 94.6% (95% CI, 81.8%-99.3%) and 81.0% (95% CI, 58.1%-94.5%), respectively. Positive and negative likelihood ratios were 8.33 and 0.11, respectively. Pain scores averaged 4.7 before US and 5.2 during US (P = 0.204). There was a statistically significant difference between mean time to US (76 minutes) and mean time to radiograph (107 minutes) (P < 0.001).
Conclusions: Pediatric emergency physicians with minimal formal training can accurately diagnose clavicle fractures by US. In addition, US itself is not associated with an increase in pain and may reduce length of stay in the ED.