173 / 225
DOS Kongressen 2017 ·
173
The diagnostic accuracy of ultrasonography com-
pared to conventional radiography for diagnosis of
extremity fractures in the emergency department:
a pilot project
Nissa Khan, Gerhard Tiwald
Ortopædkirurgisk Afdeling, Køge Sygehus ; Akutafdeling , Holbæk Sygehus
Background:
To investigate the chance to adopt ultrasonography (US), a non-
invasive technique entailing no exposure to ionizing radiation, for diagnosing
fractures performed by non-radiologist physicians.
Purpose / Aim of Study:
To compare diagnostic accuracy of radiography and
US for diagnosis of suspected extremity fractures. An emergency physician re-
ceived minimal standardised training in the use of US to evaluate fracture suspi-
cion in extremities. Results were compared with routine radiography. The patient
was treated according to the latter. The hypothesis was that US is as sensitive
and specific as conventional radiography in diagnosing fractures in extremities.
Materials and Methods:
The study was conducted over a 3-month period and
included 25 patients with a clinical suspicion of an extremity fracture. The inclu-
sion criteria were a clinical trauma history and a suspected extremity fracture
on objective examination. The exclusion criteria were fractures not localised in
an extremity, hemodynamic instability, open fractures, neurovascular lesions,
deformities indicative of fractures and bones containing orthopaedic hardware.
The X-rays were analysed by a radiologist while a blinded emergency physician
performed the US.
Findings / Results:
Prevalence of fractures was 44%. The sensitivity of US in
detecting fracture was 63.64% (95% CI: 30.79;89.0) and the specificity was
100% (95% CI: 76.84;100). The positive predictive value of US was 100% and
the negative predictive value was 77.78% (95% CI: 61.56;88.44).
Conclusions:
US of suspected extremity fractures is useful and accurate to
invalid a fracture suspicion but not reliable to diagnose a fracture. However, due
to the small study population, more studies are required before US can be rec-
ommended as a screening modality. For future studies, a higher level of training
of the ER doctors should be considered.
No conflicts of interest reported
125.