This poster was originally presented at the RANZCR Annual Scientific Meeting 2011, October 6-9, in Melbourne/AU.
Congress:
RANZCR ASM 2011
Keywords:
Musculoskeletal bone, Musculoskeletal joint, Emergency, Digital radiography, Conventional radiography, Decision analysis, Athletic injuries, Trauma
Authors:
N. Kutaiba1, J. Cruickshank1, D. Arhanghelschi1, A. Al Joboory2; 1Ballarat/AU, 2Melbourne/AU
DOI:
10.1594/ranzcr2011/R-0104
Conclusion
Discussion
We have reviewed the literature for current guidelines for radiography of suspected shoulder dislocations.
The studies demonstrated that routine ordering of radiographs before and after reducing dislocated shoulders is no longer necessary.
Emergency physicians are often certain of both diagnosis and reduction of shoulder dislocations3,4.
In addition,
certain patient characteristics such as age,
mechanism of injury and recurrence can help in determining which patients need x-rays to rule out possible associated fractures5,6,9,10.
Harvey et al questioned the need for post-reduction x-rays first after concluding that they did not add much to the management of shoulder dislocations7.
He was followed by three research groups who conducted studies to examine this idea and to question the need for pre-reduction x-rays as well3,5,9.
All these groups conducted further prospective studies to validate their findings and produce clinical guidelines for selective radiography of suspected shoulder dislocations4,6,10.
In addition,
Kahn et al investigated whether post-reduction x-rays would have any additional clinical value and concluded that ED management did not change even when new fractures were seen on post-reduction x-rays11.
Shuster et al proposed the Banff Shoulder Dislocation Guideline (Fig.
1) for selective elimination of pre-reduction x-rays which relies on the Emergency Physician’s certainty of diagnosis and reduction4.
Hendey et al proposed a guideline for selective elimination of pre- and post-reduction x-rays based on mechanism of injury and recurrence (Fig.
2)10.
Emond et al proposed the Quebec shoulder dislocation rule for selective elimination of pre- and post-reduction x-rays utilizing age in addition to mechanism of injury and recurrence (Fig.
3)6.
All three algorithms provide a general guideline as to when radiography can be used confirm the diagnosis and reduction of shoulder dislocations.
Validation of these findings in other populations is still necessary and larger numbers are needed to ensure that significant fractures are not missed.
All of these studies address the issue from an emergency management perspective and the effect of selective elimination of imaging on further management and follow-up is not studied yet.
Whether further x-rays were done for these patients in an outpatient setting was not reported.
Current practice in our hospital is to refer non-complicated patients with shoulder dislocations,
which are reduced in the emergency department,
for outpatient orthopaedic follow-up.
Further x-rays might be ordered by the orthopaedic team if no x-rays were taken in the emergency setting.
The retrospective evaluation that we conducted in our emergency department showed that ordering x-rays for patients with suspected shoulder dislocations was left to the emergency physician’s discretion.
No clear guideline was used and more post-reduction than pre-reduction x-rays were ordered.
However,
significant fractures were not missed and only 10.3% (4/39) of patients were treated without any x-rays taken.
3 of these patients had recurrent dislocations and the fourth one had an atraumatic first dislocation.
There was a similar prevalence of clinically significant fracture-dislocations of 8% in our study compared to the reported data in these studies except for the study conducted by Emond et al where they had a prevalence of 18%6.
This could be related to the smaller sample size of our study population which was similar to the sample sizes of the other studies.
3 patients in our study group had Hill-Sachs fractures reported on post-reduction x-rays but not on the pre-reduction x-rays.
These findings did not change acute management and were similar to findings by Kahn et al11.
Limitations
Our retrospective evaluation was limited by many factors.
First,
a small sample size was studied and the prevalence of significant fractures was low.
Whether this will change with a bigger sample size is still not clear and further evaluation should be done to confirm it.
Second,
the retrospective nature of our evaluation does not allow for complete assessment of information.
Some data were missing and poor documentation was seen in some of the records.
Third,
x-rays were taken based on the emergency physician’s discretion and the level of certainty was not documented.
One can argue that experienced emergency physicians are certain of their diagnosis and reduction but order x-rays routinely for medico-legal confirmation.
Fourth,
follow-up for patients who did not have post-reduction x-rays was not studied.
This follow-up might have been already organized since most of such patients had recurrent dislocations.
Fifth,
length of stay for these patients was not assessed.
It has been proved already that there is a significant reduction in length of stay when pre- and/or post-reduction x-rays are not done10.
Finally,
we have not addressed other types of imaging,
such as ultrasound,
in the emergency management of suspected shoulder dislocations.
The role of ultrasound is still reported in case series and has not been evaluated formally in a primary diagnostic accuracy study12,13.
Conclusion
Current recommendations suggest selective radiography for shoulder dislocations.
ED physicians are selective in ordering x-rays for shoulder dislocations with more post-reduction x-rays ordered than probably needed.
Prospective evaluation of these recommendations is needed in an Australian population to validate clinical guidelines for ordering x-rays without affecting clinical management.