Tag Archives: ct scan

Trauma Mortality vs Cancer Mortality from CT Scans for Trauma

Trauma professionals worry about radiation exposure in our patients. A lot. There are a growing number of papers dealing with this topic in the journals every month. The risk of dying from cancer due to CT scanning is negligible compared to the risk from acute injuries in severely injured patients. However, it gets a bit fuzzier when you are looking at risk vs benefit in patients with less severe injuries. Is it possible to quantify this risk to help guide our use of CT scanning in trauma?

A nice paper from the Mayo clinic looked at their scan practices in 642 adult patients (age > 14) over a one year period. They developed dose estimates using a detailed algorithm, and combined them with data from the Biological Effects of Ionizing Radiation VII data. The risk level for injury was estimated using their trauma team activation criteria. High risk patients met their highest level activation criteria, and intermediate risk patients met their intermediate level activation criteria.

Key points in this article were:

  • Average radiation dose was fairly consistent across all age groups (~25mSv)
  • High ISS patients had a significantly higher dose
  • Cumulative risk of cancer death from CT radiation averaged 0.1%
  • This risk decreased with age. It was highest in young patients (< 20 yrs) at 0.2%, and decreased to 0.05% in the elderly (> 60 yrs)

Bottom line: Appropriate CT scan use in trauma evaluation is challenging. It’s use is widespread, and although it changes management it has not decreased trauma mortality. This paper shows that the risk of death from trauma in the elderly outweighs the risk of death from CT scan radiation. However, this gap narrows in younger patients with less serious injuries because of their very low mortality rates. Therefore, we need to focus our efforts to reduce radiation exposure on our young patients with minor injuries.

Related posts:


  • Comparison of trauma mortality and estimated cancer mortality from computed tomography during initial evaluation of intermediate-risk trauma patients. J Trauma 70(6):1362-1365, 2011.
  • Health risks from low levels of ionizing Radiation: BEIR VII, Phase 2. Washington DC: The National Academies Press, 2006.

Gunshots And CT Scan Of The Abdomen

Abdominal gunshots and CT scanning are usually thought to be mutually exclusive. The usual algorithm generally means a prompt trip to the operating room. But as with many things in the management of trauma, there are always exceptions. The key is to understand when exactly one of those exceptions is warranted.

Exception 1: Did it really enter the abdomen? Gunshots have enough energy that they usually do get inside. However, freaky combinations of trajectory and body habitus do occur. There are three tests that must be passed in order to entertain the possibility that the bullet may not have made it inside your patient: physiology, anatomy, and physical exam. For physiology, the patient must be completely hemodynamically stable. Anatomically, the trajectory must make sense. If the known wounds and angles allow a tangential course make sense, then fine. But if there is a hole in the epigastrium and another next to the spine, you have to assume the bullet went straight through. Finally, the physical exam must be normal. No peritonitis. No generalized guarding. Focal tenderness only in the immediate area of any wounds. If all three of these criteria are passed, then a CT can be obtained to demonstrate the trajectory.

Exception 2: Did it enter an unimportant area of the abdomen? Well, there’s really only one of these, and that’s the area involving the right lobe of the liver and extending posteriorly and lateral to it. If the bullet hole(s) involve only this area, and the three tests above are passed, CT may confirm an injury that can be observed. However, there should only be a minimal amount of free fluid, and no soft tissue changes of any kind adjacent to bowel.

Exception 3: A prompt trauma lap was performed, but you think you need more information afterwards. This is rare. The usual belief is that the eyes of the surgeon provide the gold standard evaluation during a trauma lap. For most low velocity injuries with an easily understood trajectory, this is probably true. However, high velocity injuries, those involving multiple projectiles, or complicated trajectories (side to side) can be challenging for even the most experienced surgeon. Some areas (think retroperitoneum or deep in the pelvis) are tough to visualize completely, especially when there’s blood everywhere. These are also the cases most likely to require damage control surgery, so once the patient has been temporarily closed, warmed and resuscitated, a quick trip to CT may be helful in revealing unexpected shrapnel, unsuspected injuries, or other issues that may change your management. Even a completely unsurprising scan can provide a higher sense of security.

Bottom line: CT of the abdomen and gunshots to that area may actually coexist in some special cases. Make sure the physiology, anatomy and physical exam criteria are passed first. I also make a point of announcing to all trainees that taking these patients to CT is not the norm, and carefully explain the rationale. Finally, apply the concept of the null hypothesis to this situation. Your null hypothesis should state that your patient does not need a CT after gunshot to the abdomen, and you have to work to prove otherwise!

Hypotensive Patients And CT Scan – Part 2

Yesterday, I went on a rant about taking hypotensive patients to CT. The bottom line is that this is a generally bad idea. However, we all know that there are no absolutes, especially in trauma.

So yes, there are two cases where one could justify taking a hypotensive patient to CT scan. Here they are:

  1. You believe that your patient has a catastrophic brain injury which is responsible for the hypotension. You would like CT confirmation so you can begin to withdraw support and terminate any other interventions.
  2. Your patient has sustained a cervical spinal cord injury and has neurogenic shock. You have started fluid resuscitation and are considering a pressor to normalize blood pressure, but would like to continue your diagnostic routine.

But before you can even consider leaving your resuscitation room, you must ensure that there is no other source of hypotension. This means getting chest and pelvic xrays to look for hemothorax or fractures. It means getting a good FAST exam to make sure there is no significant hemoperitoneum. It also means making sure that any fractures are properly splinted and there is no uncontrolled external bleeding.

You can only go to CT scan once all of these other potential bleeding sources have been ruled out. If in doubt, you must proceed to OR to either stop the bleeding or prove that it does not exist.

Are there any other reasons to take one of these patients to CT that you can think of? If so, leave comments or tweet!

Related post:

Can I Take A Hypotensive Patient to CT?

Hypotension and CT scanners don’t play together well. For years I’ve cautioned against this, having seen a number of patients crash and burn in this area early in my career. But it’s a common error, and may jeopardize your patient’s safety. A paper that is now in press looked at this practice in a trauma hospital in Taiwan.

Patients who had blunt abdominal trauma were retrospectively reviewed. Those who remained hypotensive (SBP<90) after 2L of crystalloid were scruitnized. The CT scanner was described as being located in the same area as the ED resuscitation rooms. Furthermore, several physicians and nurses were present during scans, and a full selection of resuscitation equipment was available in the scan area.

Here are the factoids:

  • 909 patients were entered into the study
  • Only 91 patients remained hypotensive after initial resuscitation, and only 58 of these were scanned before definitive management
  • As expected, patients who were hypotensive after initial resuscitation had more serious injuries (ISS 22 vs 12), required more blood transfusions (938 vs 202 cc), and had a higher mortality (10% vs 1%).
  • There were no significant differences in comparing hypotensive patients who went to CT scan vs those who did not if they underwent some sort of hemostatic procedure (laparotomy, angioembolization)
  • In the hypotensive patients, time to OR in the CT scan group was 58 minutes vs 62 minutes for those who skipped the scan.
  • In the same patients, time to angio in the CT scan group was 147 minutes vs 140 minutes without a scan first.

The authors conclude that “hypotension does not always make performing a CT scan unfeasible.” (weak!)

Read this paper closely and don’t get fooled! It is very retrospective and very small. And if you look at the times carefully, you will see some funny business. How can time to OR or angio be virtually identical regardless of whether CT is used? Is it the world’s closest, fastest scanner? Probably not.

The authors showed that hypotensive patients have a ten-fold increase in mortality. They also recognized that definitive control of hemorrhage is the key to saving the patients. Unfortunately, there are factors in this retrospective study, such as various biases and some undocumented factors that make their two patient groups look artificially alike. This gives the appearance that the CT scan makes no difference.

In reality, the fact that there is no difference in times ensures that there is no clinical difference in outcome. To really answer this question, this kind of study must be done prospectively, and must have an adequate population size. 

Bottom line: Don’t even consider going to CT with hypotensive patients. Even if you have the fastest, closest scanner in the world. Shock time still kills, and most CT scan rooms are very poor resuscitation rooms. If your patient is unstable in the ED, do your ABCs, get a quick exam, then transport to the area where you can get control of the bleeding. This will nearly always be your OR.

Reference: Hypotension does not always make computed tomography scans unfeasible in the management of blunt trauma patients. Injury, in press, 2014.

The Pan-Scan For Trauma

Diagnostic imaging is a mainstay in diagnosing injuries in major trauma patients. But the big questions are, how much is enough and how much is too much? X-radiation is invisible but not inocuous. Trauma professionals tend to pay little attention to radiation that they can’t see in order to diagnose things they can’t otherwise see. And which may not even be there.

There are two major camps working in emergency departments: scan selectively and scan everything. It all boils down to a balance between irradiating enough to be satisfied that nothing has been missed, and irradiating too much and causing harm later.

A very enlightening study was published last year from the group at the University of New South Wales. They prospectively looked at their experience while moving from selective scanning to pan-scanning.They studied over 600 patients in each cohort, looking at radiation exposure, missed injuries, and patient injury and discharge disposition variables.

Here are the interesting findings:

  • Absolute risk of receiving a higher radiation dose increased from 12% to 20%. This translates to 1 extra person of every 13 evaluated receiving a higher dose.
  • The incidence of receiving >20 mSv radiation dose nearly doubled after pan-scanning. This is the threshold at which we believe that cancer risk changes from low (<1:1000) to moderate (>1:1000).
  • The risk of receiving >20 mSv was lower in less severely injured patients (sigh of relief)
  • There were 6 missed injuries with selective scanning and 4 with pan-scanning (not significant). All were relatively minor.

Bottom line: Granted, the study groups are relatively small, and the science behind radiation risk is not very exact. But this study is very provocative because it shows that radiation dose increases significantly when pan-scan is used, but there was no benefit in terms of decreased missed injury. If we look at the likelihood of being helped vs harmed, patients are 26 times more likely to be harmed in the long term as they are to be helped in the short term. The defensive medicine naysayers will always argue about “that one catastrophic case” that will be missed, but I’m concerned that we’re creating some problems for our patients in the distant future that we are not worrying enough about right now.

Related posts:

Reference: Comparison of radiation exposure of trauma patients from diagnostic radiology procedures before and after the introduction of a panscan protocol. Emerg Med Australasia 24(1):43-51, 2012.