Three Part Question
In an [adult presenting after wrist trauma with clinical signs of a scaphoid fracture but normal initial x-rays], can [ultrasound] detect the presence of an [occult scaphoid fracture]?
A 24-year-old man presents following a fall on an outstretched hand. He has clinical signs of a scaphoid fracture but his initial x-rays are normal. You wonder if ultrasound could be used to identify an occult scaphoid fracture.
Medline, CINAHL, and Embase databases using Ovid interface in June 2010.
Mesh terms (scaphoid bone) and (ultrasonography).
Keyword search: (scaphoid bone OR Carpal bones OR Fractures OR wrist injuries OR un-united) AND (Ultrasonography OR Ultrasonics OR Interventional Ultrasonography OR Ultrasound OR Sonography OR High-intensity Focused OR High-intensity focussed).
Twenty-one relevant articles were identified, five of which were found to address the clinical question
|Author, date and country
||Study type (level of evidence)
|Munk et al,|
|58 patients with unilateral wrist injury and clinical signs of scaphoid fracture attending orthopaedic clinic||Prospective observational study.
||All patients had x-rays taken initially and at 10-14 days||10/58 Patients had a fracture diagnosed on x-ray. US examination had a sensitivity of 50% and specificity of 91%||Poor gold standard. Little information provided about the criteria used for US diagnosis|
|Hauger et al|
|54 patients attending x-ray department with clinical signs of scaphoid fracture and normal initial x-rays||Observational unblinded diagnostic study.
||Positive x-ray or on the basis of clinical suspicion patients could have MRI, CT scan or bone scan||Six patients showed cortical disruption on US scan, five of these patients had a fracture on repeat x-ray. Sensitivity 100%, specificity 98% ||Unblinded. Inclusion criteria did not include anatomical snuffbox tenderness|
|Senall et al,|
|18 Patients attending an emergency department with traumatic wrist injury, a high clinical suspicion of fracture and normal x-ray ||Observational diagnostic study.
||Positive x-ray. X-rays performed on attendance, 2/52 after injury and then monthly as long there was clinical suspicion||Nine patients had positive x-rays. Sensitivity of US was 78%, specificity was 89%||Small study. Different radiologists performed the US, no attempt made at assessing interobserver reliability|
|Herneth et al,|
|15 patients with clinical signs of scaphoid fracture attending an orthopaedic clinic||Observational diagnostic study.
||Positive MRI scan at time of presentation was considered gold standard||Nine patients had positive MRI scans. Sensitivity of US was 78% with a specificity of 100%||Small study. Five of nine of fractures visible on initial x-ray|
|Fusetti et al,|
|24 Consecutive patients with clinical suspicion of scaphoid fracture and normal xrays attending a hand clinic||Observational diagnostic study
||CT scan gold standard||Five patients had fracture on CT scan. For moderate suspicion on US sensitivity was 100% and specificity 79%. For cortical disruption on US sensitivity was 100% and specificity 95% ||Small study. Patients assessed at review clinic rather than on presentation. US performed by radiologist|
As long as people keep falling over it is reasonable to assume that suspected scaphoid fractures will remain a very frequent cause of attendance at emergency departments. The risks of non-union and avascular necrosis in this bone are such that any clinical suspicion of fracture requires appropriate immobilisation even if the x-rays appear normal. Common practice is to treat the patient with a cast and then reassess and repeat the x-ray 10–14 days following the injury. At this time, it is hoped that decalcification at the fracture site may make any fracture more apparent; however, in the presence of clinical suspicion a negative x-ray at this point is still insufficient to rule out a fracture and treatment must continue while further investigation is arranged. An investigation that could reliably rule out an occult scaphoid fracture at the time of presentation could save the patient the inconvenience of having an immobilised wrist and save the health services the time and cost of reviewing the patient and arranging further investigations.
All of the included studies are small and subject to some limitations. The ultrasound scans have been performed by radiologists with no attempt to assess inter or intra-observer variability, and it is impossible to extrapolate how well it could be applied by an emergency physician with less training. The criteria for a ‘positive’ scan and the gold standards applied were variable and not always clearly defined. There is also a concern that it may be possible to visualise the scaphoid waist but not the poles of the scaphoid, especially in a patient who has a tender snuffbox and a limited range of movement at the wrist due to pain. Given these variables and the small number of patients in each study it is not surprising that the results were variable.
While a very specific test would confirm the fracture the patient would still have to undergo x-rays to determine where the fracture is, look for any associated injuries and check on the progression of healing. A very sensitive test would be more useful in allowing the fracture to be ruled out so that the patients could be treated for a soft-tissue injury. From the evidence available, ultrasound does not appear to be consistently specific or sensitive enough to change clinical practice.
Clinical Bottom Line
While there is some evidence that ultrasound can be used to detect occult scaphoid fractures it is not consistently sensitive enough to rule the condition out.
- Munk B, Bolvig L, Kroner K, et al. Ultrasound for the diagnosis of scaphoid fractures. Journal of Hand Surgery 2000;25B:4:369-371.
- Hauger O, Bonnefoy O, Moinard M, et al. Occult fractures of the waist of the scaphoid: Early Diagnosis by High-Spatial-Resolution Sonography. American journal of Roentgenology 2002;178:1239-1245.
- Senall JA, Failla JM, Bouffard A, Holsbeeck M. Ultrasound for the early diagnosis of clinically suspected scaphoid fracture. Journal of Hand Surgery 2004;29A:400-405.
- Herneth AM, Siegmeth A, Bader TR, et al. Scaphoid fractures: Evaluation with high-spatial resolution US- initial results. Radiology 2001;220:231-235.
- Fusetti C, Poletti PA, Pradel PH, et al. Diagnosis of occult scaphoid fracture with high-spatial resolution sonography: a prospective blind study. Journal of Trauma 2005;59:677-681.