Robert McHugh, MS4
Bedside Ultrasound vs. X-Ray for the Diagnosis of Forearm Fractures in Children. Ultrasound in Emergency Medicine. Rowlands, R., Rippey, J., Tie, S., Flyn, J.
Rationale: Suspected forearm fracture is a common presenting problem to ED, represents 2.2% ED visits at Princess Margaret Hospital. XR current gold standard imaging modality for diagnosis. Benefit of US over XR is ability to view region in multiple planes. US also has ability to visualize soft tissue structures and capture dynamic images. At superficial regions, like the forearm, bone cortex is clearly visualized. US does not have exposure to ionizing radiation. Major disadvantages of US are related to user dependence including education, experience, technology.
Given increasing usage of ultrasound (US) by emergency physicians, could bedside ultrasound imaging be an alternative to traditional XR in evaluation of paediatric forearm fractures.
Study:
Primary Aim – Can US safely exclude forearm fractures in children
Investigators – Physicians with minimal pervious US experience after training program
Secondary Aim – Compare pain and discomfort associated with XR and US
Secondary Aim – Determine acceptability of US as diagnostic tool for parents and patients
Design – Prospective, nonrandomized diagnostic study
Demographics – 419 children age 0-16 years old with suspected fracture forearm, excluding those with open fracture or outside imaging prior to arrival. Average age 9.3 (sd 3.5), 57% male.
Investigators – 25/33 eligible doctors (48% pediatric background, 52% ED and pediatric background) completed commercially available 80-minute online module and quiz followed by 2-hour hands-on session. Reported that vast majority with no previous US experience (28% had taken previous eFAST course).
Protocol – Informed consent obtained. Scanned with Sonosite M-Turbo with HFL50xl15-6 MHz linear transducer on MSK preset. Obtained images of radius and ulna in four planes (dorsal, ventral, lateral, medial). Defined fracture as presence of cortical disruption. XR performed after completion of US and interpretation report. Definitive diagnosis based on consultant radiologist’s report of XR. Pain scores obtained 5-minutes after clinical exam and again after US and XR. Modality preference was asked at completion of all imaging.
Results:
US diagnosed forearm fracture with sensitivity of 0.915 and specificity of 0.876
· 234 patients had fracture diagnosed on XR. US correctly diagnosed 214/234, sensitivity of 0.915. Missed US fractures included: 4 displaced fractures of distal radius, 6 subtle buckle fractures of distal radius, 1 undisplaced ulna styloid fracture, 9 midshaft radius/ulna, bowing fractures or radial neck fracture.
· 185 without fracture on XR. US negative for fracture in 162/185 cases, specificity of 0.876. 60.9% (14/23) false positive results occurred in operators with less than 10 scans at the time.
· Post hoc analysis including only studies after operator completed minimum of 10 scans increased overall sensitivity to 94.9% and specificity to 91.4%
No difference in pain associated with study between the two different modalities
· 369 complete pain scores recorded. Pain score during clinical exam (mean 5.5) significantly greater than imaging (US mean 3.7, XR mean 3.8). No significant difference between imaging modalities.
Family preference for US over XR
· 362 families provided feedback on modality preference. 52.8% with a strong preference for US, 18.5% for XR, 28.7% no preference. Reasons cited included: radiation exposure to children and difficult with XR study requirements.
Discussion: Good agreement between US and XR, particularly given the large proportion of US naïve participants. Missed mid-shaft fractures that were retrospectively clinically obvious may be in part due to the focus on distal forearm throughout the training program. Would benefit from further stratification of fracture subtype and US findings. Concerning that there were a large number of different types of fractures missed. sStudy did not include any way to follow up with patients to determine if false positives on US could actually have been missed on XR. There were similar pain scores, signifying that pain is not barrier to US. Acknowledge that learning curve exists and accounting for experience improves sensitivity and specificity. Ultimately, US was preferred by patients and with increased experience may become viable substitute for conventional XR.