The pelvic X-ray in paediatric blunt trauma- Another example of "What NOT to do"?

Annals of Emergency Medicine has another great article on one of my favourite topics in medicine - What Not to Do.

The article, titled Sensitivity of Plain Pelvis Radiography in Children with Blunt Torso Trauma by Kwok, et al and from the Pediatric Emergency Care Applied Research Network (PECARN), is a prospective observational study from 20 sites of 12,044 children with blunt torso trauma.

451 (3.7%) had a pelvic fracture and/or a dislocation. Sixty-five underwent operative intervention and 21 were hypotensive on initial presentation.

Using this dataset the authors examined the utility of the pelvic x-ray. In these blunt torso trauma patients, the sensitivity of a pelvic x-ray was 78%. Another way to say it: pelvic x-rays missed about 25% of fracture dislocations identified on CT.

For the 21 (4.7%) of the hypotensive patients (7 of whom died), only 17 had pelvic x-rays, and only 14 (82%) identified the pelvic fracture or dislocation. In the entire dataset, there were 5 patients with fractures that were missed on pelvic x-ray that required operative intervention. 

Low-level falls, falls from bicycles, and falls down stairs very rarely resulted in fractures or dislocations. Patients who could ambulate never required an intervention.

The authors reported that children are at low risk for pelvic injury if they: are able to give a reliable exam (not altered, not distracted, not intoxicated), lack a lower extremity injury, lack an abnormal pelvic physical exam, and had NO other indication for a CT Abdo-Pelvis.

So we should only go by physical exam? Well it turns out that the physical exam for “pelvic instability” also performed very poorly, with 89 (0.8%) of the patients without fractures being documented as “unstable” and only 4 of 14 (36%) patients with open-book fractures being documented as “unstable.” That seems like an unacceptably low specificity or sensitivity.

I like this study for several reasons.

First, it is a great example of the value of reading the literature. You show me a person who claims they’ve treated 65 significant paediatric pelvic injuries, and I’ll show you a person whose math skills or memory sucks. There are just not that many out there.  But you don’t need to treat this many to be knowledgeable. A quick read of this paper gives anyone some experience on an important condition.

Second, it supports my previous biases:

1) Patients who are sick, physiologically unstable, and difficult to evaluate patients require evaluations with high sensitivity and specificity (CT scans).

2) Patients who are well, physiologically stable, and who can be clinically evaluated can safely be evaluated with serial and tertiary exams.

Reflexively ordering insensitive and nonspecific radiology (pelvic x-rays), produces unhelpful results that don’t improve patient management and require providers to order the correct tests down the road.

My one-sentence take-homes from this data are:

1) Don’t trust the pelvic exam or the pelvic x-ray on patients who are "unstable" or difficult to evaluate. 

2) Don't waste time, cost, and radiation by getting pelvic-x-rays on “stable” patients who are getting a CT abdo-pelvis anyway.

3) Don't get pelvic x-rays on "stable" patients who are NOT getting a CT abdo-pelvis who you can then monitor and test for ambulation.


Kwok MY, Yen K, Atabaki S, Adelgais K, Garcia M, Quayle K, Kooistra J, Bonsu BK, Page K, Borgialli D, Kuppermann N, Holmes JF. Sensitivity of Plain Pelvis Radiography in Children With Blunt Torso Trauma. Ann Emerg Med. 2014 Jul 29.