This poster is published under an
open license. Please read the
disclaimer for further details.
Type:
Educational Exhibit
Keywords:
Experimental, Digital radiography, Trauma, Bones, Diagnostic procedure, Cost-effectiveness, Socio-economic issues
Authors:
B. Gibney, L. Murphy, D. Ryan, D. Hynes, P. J. MacMahon; Dublin/IE
DOI:
10.1594/ecr2016/C-1036
Findings and procedure details
143 patients presenting to the Emergency Department with acute wrist trauma and clinical suspicion of wrist fractures were immobilised and referred for follow up at the orthopaedic outpatient clinic for clinical and radiological review.
The average age was 41 (median 38,
range 16-82) and the male:female ratio was 1:1.55.
At clinic,
the patients had standard 4-view plain radiographs and lateral tomosynthesis of the injured wrist (Fig.
3).
Plain film imaging was performed using a GE Discovery XR656 and digital tomosynthesis was captured with the same system using VolumeRAD software.
The average number of tomosynthesis slices obtained was 46 (range 20-132).
The images were reviewed by an orthopaedic registrar and a consultant emergency radiologist.
39 fractures were identified in 36 patients (25% of patients).
The identified fractures consisted of 11 scaphoid (Fig.
4),
15 distal radius (Fig.
5) and 13 other carpal/metacarpal fractures (Fig.
6).
Eleven of the fractures identified with DTS were not identified on initial radiographs (Fig.
7).
Two fractures were not seen on DTS and were identified later,
a triquetrum fracture on plain radiographs 4 weeks after trauma and a scaphoid fracture seen on a second DTS 5 weeks after trauma.
During the trial period a number of adjustments were made to the imaging protocol which subjectively improved ease of review and diagnostic accuracy.
These included switching from lateral tomograms to DP tomograms and angling the forearm 20 degrees above the horizontal plane of imaging and placing the hand in ulnar deviation (Fig.
8).
Discussion of the experience and impressions of those using the new imaging technique are as follows:
Radiographers’ Experience
This new technique requires some training to utilise but once training has been received it is simple to perform.
Difficulty in taking satisfactory images is low and similar to that of plain radiography of the scaphoid once one is familiar with performing a tomogram.
Similarly,
the time taken to perform imaging is approximately the same as plain radiography of the scaphoid.
The only issue with tomography is the time taken for the series of images to be formatted.
Computer processing can take up to 5 minutes for each tomogram and this could potentially introduce delays for radiographers.
Orthopaedic Experience
Adjustment to the new technology can be slow with many preferring initially to use plain radiography images over tomograms.
Some issues arise from simple technical problems which can be easily overcome with adjustments to PACs station settings.
Others require practice such as discerning fractures from the artefacts of overlying structures.
DP tomosynthesis is preferred as it is the plane usually used to view plain films of scaphoid.
Tomosynthesis has been particularly sensitive for non-scaphoid occult wrist fractures.
It is useful to have an additional imaging modality which can be accessed during clinic hours rather than rescheduling and potentially incurring prolonged patient waiting times for typical second line imaging modalities.
Radiologists’ Experience
Experience in wrist tomosynthesis is required to be able to confidently diagnose abnormalities due to relatively pronounced DTS specific artefacts which can simulate fracture lines.
The lateral position is good for distal radius and triquetal fractures but DP best for scaphoid,
capitate and probably hamate.
A combination of both is probably optimal.
On our PACs the DTS imaging presents as a series of single images rather than a series such as in other cross sectional imaging modalities but this could likely be fixed in the future.
It means zooming/windowing only affects the single image this is performed on rather than the whole series.
There is definite room for improvement in the technology,
namely reducing the effective slice thickness and overlying artefacts.