New Feature – Trauma Radiology

We’re proud to announce a new feature on the ETM Blog – trauma radiology posts with our new CT scroller, which will allow us to bring you real CT images of trauma cases.

Below is the first series of images from a case which highlights the need to examine your stable trauma patients carefully, to not dismiss seemingly innocuous clinical findings, and to maintain a healthy degree of skepticism for the “best available evidence” and clinical decision rules.


A 62 yo woman is brought to a rural ED after falling asleep at the wheel and crashing her car, apparently at around 60km/hr.  There was no actual impact, she veered across the road, and ploughed diagonally across a 1 metre ditch, mounting the far side of the ditch causing significant damage to the undercarriage of the car. Seatbelt worn, airbags deployed.  Her only complaint was of pain in “both hips”.

Her primary survey was essentially normal, with normal vital signs, a a soft nontender abdomen and negative FAST. She had very mild left 3rd-4th rib tenderness. On log roll she had no midline spinal tenderness. C-spine was cleared clinically and she had no neurology. Compressing the pelvis caused no pain, and her hips examined normally, however on actively flexing her hips (raising her feet off the bed) she complained of mild diffuse lower abdominal discomfort.  Chest x-ray and pelvic x-ray were normal.

Thoraco-lumbar x-rays were not ordered due to the absence of high risk features on mechanism, the absence of back pain and the “normal” spinal examination.

However due to the abdominal discomfort on hip flexing, a CT abdomen was ordered to rule out an occult intra-abdominal injury (eg occult pelvic fracture, or retroperitoneal bleed).

Here is the CT:

[radiologystack cats=13,12,14]

[DDET What are the findings on the CT?]

There is no abnormality of the abdominal viscera or pelvis, but there is a burst fracture of L3, with involvement of the body, lamina and pedicles. This involves all 3 columns of the spine and is considered an “unstable injury”.


This case raises several interesting questions:

[DDET Can the thoraco-lumbar spine be “cleared” clinically in trauma?]

Unlike the NEXUS and Canadian C-Spine clinical decision rules, no such guidelines exist for clearance or imaging of the thoraco-lumbar spine.  A recent paper by Gill et al proposes that trauma patients who are sober, have not received opiate analgesia pre-hospital, have no pain or tenderness on log-roll, no neurology, and no concurrent cervical fracture, can be cleared clinically of lumbar (but not thoracic) fracture. This patient fit these criteria, yet had an unstable lumbar fracture. Clearly these criteria are not 100% sensitive, and the authors do acknowledge that prospective trials are needed in this area.  So in short, the answer is “no”.  [/DDET]

[DDET So when should thoraco-lumbar x-rays be done on trauma patients?]

Some practitioners argue that AP and lateral thoraco-lumbar views are part of a standard “trauma series” of Xrays and should be done routinely in addition to the chest and pelvic x-rays.  Some argue that the clinical findings above are enough to “clear” the lower spine and imaging is not required. The jury is out on this topic, and it is the subject of debate in the current literature.  In practice you need to weigh up the mechanism with the signs, and always err on the side of caution. If the patient qualifies for any imaging, (eg chest or pelvic x-rays), then do spinal imaging as well. Whilst the prevalence of occult or asymptomatic cervical, thoracic and lumbar injuries is low, the consequences are so devastating if missed, and the radiation dose so low, that relying on clinical clearance in the absence of a “proven” (I use that word cautiously) decision rule is risky.  As to which imaging modality to use (ie CT vs plain films), well that’s a discussion for another post!


[DDET I’m confused! Should we just “pan-scan” everyone and be done with it?]

A “pan-scan” is a non-contrast CT brain and C-spine, followed by a contrast scan of the chest, abdomen and pelvis, with reconstructions of the thoraco-lumbar spine. In most Australian major trauma centers, the use of the “pan-scan” for major trauma patients is standard practice. The rationale for doing this has to do with the selection bias of the patient population (more likely to have multiple and occult injuries), the need for expedient diagnosis, and the real rate of missed injuries in major trauma, which can be as high as 10-38%.    The majority of missed injuries in these patients are benign, but picking them up early means saving time, possibly reducing length of stay, (and therefore costs), and saves the patient having piecemeal investigations.

Anecdotally having been attending the weekly trauma meeting at a local trauma center for the last couple of months, I’ve seen enough serious internal injuries that would have been missed without a pan-scan that I’m becoming convinced of the benefit. On the flipside, in smaller and rural ED’s (like the one this patient was taken to), to get a pan-scan can be a bit of a big deal. The radiographers may not have a protocol for it, they may not be used to setting their machine up to get optimal images and reconstructions, and the radiologists are often appalled at the apparent lack of “necessity”. You may find yourself being told “we don’t scan everyone for everything like they do in the city”.  So have a plan ready in case of resistance when you order a pan-scan in a smaller ED. Have some stats on your side, know who meets the criteria for “major trauma”, and be considerate to local practice.  If you only need to CT one body region to decide if the patient needs transfer to a major trauma centre (where they will get pan-scanned again), just do that part.


[DDET But what about the radiation dose?!]

In the past, most of the data that was used to calculate risk of cancer from CT scans came from monitoring people exposed to nuclear radiation (eg Hiroshima survivors). Modern data however has shown that risk of cancer while not zero is much lower than previously thought. Even in pregnancy, the risk of fetal loss, malformation or childhood cancer risk does not seem to increase until the radiation dose goes above 40-100mGy.  Depending on the scanner, the average dose of a pan-scan is around 25mGy.  The risk of malforation is highest in the period of organogenesis (between 8-15 weeks). Given the risk of missed injury to the mother, and the cumulative doses conferred by multiple scans, it seems reasonable that a pregnant major trauma patient with a good clinical indication can undergo a single pan-scan with the risk-benefit ratio clearly in favour of this approach.  For those who don’t meet major trauma criteria, have low risk mechanisms, or are not critically unwell, a consensus should be reached between the treating clinicians and the mother to decide on the choice of imaging modality, and this discussion documented in the notes.

The American College of Radiology has an excellent paper on this if you need the stats to discuss with your team.

In elderly patients (over 60) the risk of radiation from a pan-scan compared to the risk of missed injury weighs heavily in favour of pan-scanning.


[DDET Why did this patient have abdominal pain?]

Abdominal pain is an uncommon symptom of lumbar spine fracture in the absence of associated abdominal visceral injury.  Burst fractures of L1 (Chance fractures) are associated with pancreatic and duodenal injuries, which may cause pain, however the lower lumbar vertebrae less so.  The psoas muscles originate from the transverse processes of T12-L5, so it’s likely in this case that when the patient flexed her hips, the pull on the lumbar vertebrae moved the fracture which was perceived as pelvic discomfort.


So there you go, the first Trauma Radiology case on the ETM blog, with many learning points.

What are the take home messages?

  • There is no clinical decision rule or criteria that can be used to reliably clear the thoraco-lumbar spine.
  • If a trauma patient has unexplained symptoms or signs, don’t stop until you find the cause!
  • The radiation dose of thoraco-lumbar x-rays is so low as to not be a reason for not ordering them. In most cases, including in pregnancy, if the patient meets criteria for being a “major trauma”, is unwell, elderly or has unexplained clinical findings and your radiology department can handle it, a pan-scan is likely to be the most appropriate option.




2 Comments on “New Feature – Trauma Radiology”

  1. Brandon O

    Great post. I just want to make one point of order, which is what exactly we mean when we note the presence of an “unstable” spinal injury like this.

    In almost all cases, this term is refers to certain radiological findings which we’ve agreed (by general consensus) indicate a spine that, without intervention, has the potential to induce neurological deterioration from to normal movement. How many of the injuries within this a priori category ACTUALLY lead to neurological harm (ranging from minor and transient paresthesias to total transection) is not really clear; I haven’t seen any literature that examines this directly, but it’s not terribly high. And when you drill down to the sort of unimpressive presentations where we usually end up “missing” those injuries (the classic “occult” fractures), the numbers are even smaller; sequelae are very rare and serious, permanent sequelae are exceptionally so.

    So I suppose my point is that, although we should recognize the sort of “misses” you describe here, the natural reaction is to assume this was almost a catastrophe had we sent that patient along. And while it could have been, it probably wouldn’t; a miss with no consequence is the rule, not the exception, because the “unstable” fractures are mostly not very unstable.

    Obviously, you may still elect to manage these patients with high suspicion, for reasons ranging from the legal to the merely cautious, but it should be done with open eyes as to the real risks.

    1. trawmadoc

      Thanks Brandon
      Fair points regarding “stability”. Having been in the ED when a patient who had a barely noticeable C3 fracture missed on CT, had no pain, so had their collar removed, turned their head and became instantly quadriplegic several years ago, I remain ultra-cautious about the stability issue. on the flipside, many of us have seen the multi-column spinal fracture walk in to ED several days post injury with no neurology. Like you point out, the actual risk is very low, but the consequences are so grave should they be the 1 in _____ (insert large number here) who sustains neurological damage post a “miss”, that I always err on the side of caution with spines. All spinal fractures should be referred to the nearest spine service, due to the many variables (and variations in management between surgeons) that exist, and I think using the terms “stable” vs “unstable” while waiting for surgical opinion is really just academic, and of limited clinical use for ED folk. The risk of movement is very hard to quantify, and in the not too distant past, we used to do flexion-extension views on people with neck injuries, may of whom would have had fractures not visible on plain x-rays!

      Thanks for commenting.

Leave a Reply