Three Part Question
In [patients with blunt thoracic injury and suspected sternal fracture], does [bedside ultrasound perform better than standard chest and sternum x-rays] to [accurately diagnose sternal fracture].
In the middle of the night, a 48 years old male is brought to your emergency department after a car accident. His seatbelt was fastened and he complains of moderate to severe sternal pain. He is alert, has no shortness of breath, his vital signs are normal and your secondary survey is otherwise negative for any other injury. You suspect your patient has an isolated sternal fracture. After ordering an electrocardiogram, cardiac enzymes and administering medication to relieve your patient, you wonder if bedside ultrasound would perform better at diagnosing sternal fracture than standard chest and sternal x-rays.
The search was performed on October 27th 2014.
A. No Best BETs was found on this topic.
B. Cochrane library
'(Sternum fracture OR sternal fracture) AND (echography OR ultrasound)' No Cochrane review was found on this subject.
C. www.clinicaltrials.org was searched for an ongoing trial on this topic. No relevant trial was found on this subject.
D. MEDLINE (trough Pubmed)
1) (sternal OR sternum AND fracture) Filters: Humans: 794
2) (((ultrasonography OR echography))) Filters: Humans: 335631
3) ((sternal OR sternum AND fracture) AND (ultrasonography OR echography)) Filters: Humans: 27 (with no time restriction)
1) sternal OR 'sternum'/exp OR sternum AND ('fracture'/exp OR fracture) : 1163
2) ultrasonography OR echography: 334752
3) 1 + 2 AND 'human'/de (with no time restriction): 34
Of the 27 articles found on Medline, 10 were potentially relevant. References of these 10 articles were screened for relevant papers. One new study was found and added to the papers to be screened. Of these 11 papers, 8 were available in full text version for analysis. Of these 8, 2 were excluded because they were written in German. Of the 6, 4 papers were considered more pertinent to answer the clinical question and were kept for final analysis. They are presented below. Of the 34 papers found on Embase, 10 articles were found to be relevant to answer the clinical question. There were cross-references for all these articles with those found on MEDLINE via Pubmed.
|Author, date and country
||Study type (level of evidence)
|Engin G et al.|
|23 patients with suspected SF after blunt thoracic trauma.||Retrospective observational study.||Diagnosis of SF by US compared to conventional CR.||Both US and X-ray detected SF in 16/23 (69,6%). US detected SF in 2/23 (8,7%) where CR were negatives and 2/23 (8,7%) where CR were suspicious. No fracture were seen in US and CR in 3/23 (13%).||Small single center retrospective cohort.
Not clearly specified who performed US (radiologist vs emergency physician vs other).
13/23 patients had associated injuries.
No gold standard comparison.|
|Hendrich C et al|
|45 patients with suspected SF after MVA||Prospective observational study||Diagnosis value of US compared to conventional CR for SF detection.||15 cases of SF were demonstrated by CR and US. 1 SF was detected only by US.||Very few details on recruitment are available.
Examiner were not blinded as US was performed after CR interpretation.
Small single center cohort.
No gold standard comparison.|
|Jin W et al.|
|50 consecutive patients with suspected SF after chest trauma.||Prospective observational study||Diagnosis of SF by US compared to CR and bone scintigraphy.||When comparing CR to US: 12/50 SF were detected on CR compared to 31/50 for US. No SF were found with either CR or scintigraphy when US was normal.||US was performed by 2 musculoskeletal radiologists (which decreased external validity and application in an emergency setting).
No gold standard comparison.|
|You JS et al|
|36 consecutive ED patients with suspected SF after blunt chest trauma.||Prospective observational study.||Diagnosis of SF by bedside US compared to CR.||20 SF were diagnosed on CR. 24 SF were diagnosed with US (17 of these had positive CR, 7 had negative CR). 3 false positives on CR were diagnosed with US.||Small single center cohort.
No gold standard comparison (CT scan or follow up US to demonstrate callus formation).
Calculation of sensibility and specificity are not valid because there is no gold standard comparison.
We found four small cohort studies. All were favouring the use of US to diagnose SF when compared to CR although there was no gold standard comparison like CT scan or follow-up US to demonstrate callus formation. Only one study (4) compared bedside US to CR, but radiologists interpreted images. In the remaining studies, radiologists performed US or it was not specified. This situation decreased external validity and application in an emergency setting and possibly favoured US sensibility. CR, Chest radiographies; ED, Emergency department; MVA, Motor vehicle accident; SF, Sternal fracture; US, Ultrasonography.
Clinical Bottom Line
In patients with blunt thoracic injury and suspected isolated sternal fracture, US seems superior than CR to diagnose sternal fracture. Although no study has clearly assessed the diagnosis accuracy of US performed and interpreted at bedside and no study had a gold standard comparison. More studies will be needed and these results should be validated prospectively in a multicenter cohort with gold standard comparison.
- Engin G, Yekeler E, Guloglu R et al US versus conventional radiography in the diagnosis of sternal fractures. Acta radiologica 2000 May;41:296-9.
- Hendrich C, Finkewitz U, Berner W. Diagnostic value of ultrasonography and conventional radiography for the assessment of sternal fractures. Injury 1995;26:601–4.
- Jin W, Yang DM, Kim HC et al. Diagnostic values of sonography for assessment of sternal fractures compared with conventional radiography and bone scans. J Ultrasound Med. 2006 Oct;25:1263-8.
- You JS, Chung YE, Kim D et al. Role of sonography in the emergency room to diagnose sternal fractures. J Clin Ultrasound 2010 Mar-Apr;38:135-7.