• 
• 

Three Part Question

In [children presenting to the ED with suspected clavicle fracture], is [Ultrasound scanning as good as radiography] at [correctly diagnosing fractures]?

Clinical Scenario

A 12-year-old boy presents to the Emergency Department (ED), with pain and swelling around his left clavicle which occurred during a game of rugby. You explain his clavicle might be broken and an x-ray is required to confirm the diagnosis. His father points out that he has multiple x-rays in the past for other sporting related injuries and asks if the x-rays are necessary? You wonder whether Ultrasound (US) would be an alternative means for diagnosing a fracture of this bone.

Search Strategy

Medline from 1950 and Embase from 1980 to April 16th 2012 (NHS Evidence):
[{(Ultrasound) ti,ab} OR {(Ultrasonography) ti,ab} OR {exp Ultrasonography} or {(sonography) ti.ab} OR {(osteosonography) ti.ab}] AND [{(clavicle) ti,ab} OR {exp clavicle} OR {(fracture*) ti,ab} OR {exp Fractures, Bone} OR {(“Clavicle Fractur*”) ti,ab}]. Limits: Humans and (Age groups all children 0–18 years) and English Language.

The Cochrane Library April 2012: MeSH descriptor Clavicle explode all trees.

Search Outcome

The search produced 580 results of which 40 abstracts were read. From these abstracts four papers were found to be relevant and one further paper was found through searching references. These papers are summarised in the table

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Cross et al, 2010, USA	100 children aged 1–17 presenting to an urban, tertiary care ED with shoulder or clavicle pain due to recent trauma with no neurovascular complication and requiring x-ray. 43 fractured clavicles Diagnostic test: Clavicle US performed by Paediatric EP Gold standard: x-ray reported by a radiologist	Prospective study.	Sensitivity	93% (95% CI 80% to 98%),	Convenience sample with enrolment when researchers were in the department. 97% scans performed by one EP. Investigators did not follow-up patients so occult fractures may have been missed on x-ray which were reported as present on US, which might have influenced results. No standardised x-rays, some patients had dedicated clavicle x-rays while others had shoulder x-rays
			Specificity	86% (95% CI 74% to 93%)
Weinberg et al, 2010, USA	212 patients aged <25 presenting to one of two urban ED's with musculoskeletal injury requiring x-ray or CT investigation. 15 had clavicular injury with 60% fracture rate Diagnostic test: point-of-care US performed by Paediatric EP's Gold standard: x-ray or CT reported by radiologist	Prospective observational study.	Sensitivity	89% (95% CI 51% to 99%),	Convenience sample studied. Low prevalence of fracture in study group (24% fracture rate). Limited sample size for individual bone fractures, only 15 patients presented with suspected clavicle fractures. Two of the patients aged 18 years and over, 13 aged <18 years (see comment)
			Specificity	83% (95% CI 36% to 99%)
Moritz et al, 2008, Germany	653 patients aged newborn to 17 years with a history of fall who had both US and x-ray performed of the injured bone or bones. 56 had clavicular injury with a 66% fracture rate. Diagnostic test: US analyzed during the time of examination by a Paediatric Radiologist. Gold Standard: x-rays reported by Paediatric radiologists	Prospective study.	Sensitivity	97.3%	Small number (8.6%) of patients had suspected clavicle fractures. Radiologists not EP performing US. Hospital setting unclear—fracture attendances may be high. Depending on their clinical examination patients either had US or x-ray first, with the other modality as a second study, therefore patients were not investigated in the same way. Unclear whether US paediatric radiologists went on to report x-rays which could introduce bias
			Specificity	73.7%
Blab et al, 1999, Austria	49 children with suspected clavicle fractures presenting to the authors institution Comparative—no gold standard specified	Prospective study	Practicability	In three cases US not possible (2 refused and in 1 neonate the probe was too big)	Little detail in the paper about the setting of this study and who the investigators are (except for the correspondence address for department of paediatric surgery). Unclear who performed the US, how experienced they were in the use of US and whether they were blinded to the results of the x-ray. Unclear who reported the x-rays and whether they were blinded to the US report. No comment on the sample size (convenience or calculated?).
			Diagnostic findings	No significant difference
Chien et al, 2011, USA	58 patients aged 3 months to 16 years presenting to a tertiary care children's hospital with suspected clavicle fractures with 67% fracture rate Diagnostic test: Bedside US performed Paediatric EP Gold Standard: Standard view x-rays reported by radiologist	Prospective study.	Sensitivity	89.7% (95% CI 75.8% to 97.1%)	Paediatric EP's not blinded to clinical information of the patient making this a realistic application of US but introducing operator bias. Convenience sample used with more patients adding more power to the study. More training could have reduced the false positive and negative rates. US interest bias as potentially only those interested in US volunteered to recruit. Tertiary paediatric hospital may see higher fracture rates than other ED's
			Specificity	89.5% (95% CI 66.9% to 98.7%)

Comment(s)

A further paper by Hubner et al was excluded as the data on clavicle US was combined with skull US in their table of results, although they do describe ‘relatively good results’ for US of the clavicle. The authors of the paper by Weinberg et al were contacted for clarification on the ages of patients included for clavicle USA. Blab et al state that at the Department of Paediatric Surgery in Vienna, ‘osteosonography’ is the method of choice for diagnosing a suspected clavicle fracture. They feel that if Emergency Physicians can be trained in the use of US to assess the clavicle, x-ray could become a second line investigation in cases of uncertainty.

Clinical Bottom Line

Ultrasonography is sensitive at detecting clavicular fractures in children.

References

1. Cross KP, Warkentine, FH, Kim IK, et al. Bedside ultrasound diagnosis of clavicle fractures in the pediatric emergency department Academic Emergency Medicine 2010;17:687–93.
2. Weinberg ER, TuniK MG, Tsung JW. Accuracy of clinician-performed point-of-care ultrasound for the diagnosis of fractures in children and young adults Injury 2010;41:862–8.
3. Moritz JD, Berthold LD, Soenksen SF, et al. Ultrasound in Diagnosis of Fractures in Children: Unnecessary Harassment or Useful Addition to X-ray? Ultraschall in Med 2008;29:267–74.
4. Blab E, Geißler W, Rokitansky A. Sonographic management of infantile clavicular fractures Pediatric Surgery International 1999;15:251–4.
5. Chien M, Bulloch B, Garcia-Filion P, et al. Bedside Ultrasound in the Diagnosis of Pediatric Clavicle Fractures Pediatric Emergency Care 2011;27:1038–41.
6. Hubner U, Schlicht W, Outzen S, et al. Ultrasound in the diagnosis of fractures in children. J Bone Joint Surg Br 2000;82:1170–3.