Does this scenario resonate?
Father: Doctor, I am so worried. Junior has a horrible injury to his head.
Doctor: Oh, goodness, I am very sorry to hear about that.
Mother: We need a CT scan, doctor.
Doctor: OK, we can certainly discuss that. First of all, though, tell me what happened.
Patient: Happy, coo, happy, coo, goo.
Father: This afternoon, sister and big brother were having a pillow fight and — sob, sob, sob — they caught Junior straight in the face, and he crashed into the couch. Then he turned white and passed out.
Mother: Which is why we need a CT scan, Doctor.
Patient: Happy, coo, happy, coo, goo.
Doctor: (who is examining child and see no signs of trauma): I can understand your concerns. Before we decide on that CT, let me ask a few questions. First of all, how long was your child unconscious?
Father: It must have been five minutes at least.
Big brother: Dad, it was like two seconds.
Mother: Did I mention that I am a lawyer?
Doctor: Very well then, I'll see you all after the CT scan.
Painful, right? And, of course, this sort of mentality does not exist just for pediatric head trauma; we encounter it with all sorts of other clinical conditions. We EPs now have powerful ammunition to answer back to unreasonable family requests for patients with minor pediatric head trauma. Not only can we describe the current state of overutilization, in which the imaging rate (25-50%) is much, much higher than the rate of clinically significant brain injury (<1%), but we also have triple-barreled evidence to back us up. This knowledge, as we know, involves radiation risk, good, solid evidence-based prediction rules for risks of significant brain injuries after head trauma, and a viable alternate management strategy for a substantial number of patients.
We all know that medicine, like life itself, almost always has tradeoffs. Most treatments, even unassuming ones like oxygen, have side effects. And most medical tests hold the potential of unintended consequences. It's undeniable that there's been an explosion in CT use, and that this is a concerning trend. Mounting evidence shows that CT exposure in childhood results in a small but real increased risk of cancer later in life. Miglioretti and colleagues recently reported in JAMA Pediatrics, for example, that CT use doubled in children under 5 and tripled in children ages 5-14 between 1996 and 2005. The authors extrapolated the number of future cancers that might be attributed to CT radiation from their data. They estimate that the approximately four million pediatric CTs done annually in the United States will ultimately lead to nearly 5,000 lifetime cancers.
Another recent study from Australia by Matthews et al helps give us a bit more perspective. (BMJ 2013;346:f2360; http://bit.ly/1dqnypW.) The investigators retrospectively reviewed the national health database records of more than 10.9 million Australians, and identified nearly 700,000 people (6%) who had at least one CT scan performed during childhood, and those were linked with cancer diagnoses one year or more after CT exposure. They found, after some statistical adjustment, that the group of children exposed to CT scans were 24 percent more likely to develop any cancer during the study time period and that their risk increased (about 16% or so) with each additional CT scan. The peril was greater for children exposed at younger ages, and was linked to many different types of malignancies, including cancers of the brain, skin, blood, and gut.
Other recent evidence demonstrates that while the word is out about radiation risk from CT, not everyone is in the know. A recent research letter by Caverly and colleagues published in JAMA Internal Medicine reported that only 31 percent of surveyed Veterans Affairs patients who were undergoing a CT understood that radiation and cancer were potential risks. (2013;173:588.) Another survey study by Boutis et al found that approximately half (357/752; 47%) of parents whose children were seen in a pediatric ED for head trauma knew something about the ionizing radiation risk with CT. (Pediatrics 2013;132:305.) Furthermore, 90.3 percent of patients expressed a desire to be informed of the radiation risks associated with CT, and the disclosure of risk had a significant effect on reported willingness to undergo CT (90.4% before, 69.6% after).
Hopefully everyone is aware of the landmark 2009 Lancet article from Kuppermann et al (2009;374:1160) done on a cohort of more than 40,000 U.S. children with head trauma. The study identified six clinically important data points for kids under 2 and six more for kids 2-18. What you might not realize is the amount of granularity in the PECARN rules; the large cohort allows for risk estimates even when multiple risk factors are present, and many of these are extremely low (less than 1%).
A close look at the PECARN data could help justify ED observation rather than CT. A perusal of the article — Figure 3 in particular — or a visit to www.MDCalc.com to see the figure could allow you to give parents nuanced and detailed numbers and advice. Did you know that the risk of clinically important brain injury is only 0.2 percent for younger children whose only positive PECARN predictor is severe mechanism of injury (such as fall from height greater than three feet)?
Nathan Kuppermann, MD, MPH, explains: “The goal of the PECARN study was to identify a group of low-risk children that you could safely evaluate without CT scans. We found that approximately 20-25 percent of CT scans obtained in children are in those with very low risk of brain injury, with NO PECARN risk factors, and therefore the CT scans can be avoided in these children. Even in those with just one of the PECARN risk factors, however, such as a history of loss of consciousness (as in the example above), or a history of significant mechanism of injury, the risk of brain injury remains very low.
“This suggests that many children with just one of the PECARN risk factors (excluding altered mental status or signs of skull fracture, both of which carry a higher risk) could be observed for a period of time in the ED before deciding whether or not to obtain a CT scan. If they did not develop signs of brain injury during a period of observation and remained well-appearing, many of these children with “isolated” PECARN risk factors (that is, no other signs or symptoms of brain injury) could be discharged home after a period of observation without CT scans. Discussing risks and benefits of obtaining CT scans with parents (“shared decision-making”) is an important concept in current-day clinical decision-making when tradeoffs exist and a clinical decision is not clear.”
Finally, and this may be news to some of you, we now have increasing evidence that ED observation is a safe and effective alternative to immediate CT in some children after minor head trauma. A recent article built on prior work by PECARN investigators on this topic with the results of their study of the time-dependent characteristics of ED observation for minor pediatric head injury in three EDs. (Ann Emerg Med 2013 July 24. [Epub ahead of print].) This prospective observational study examined the outcomes of 1,381 children (86% capture rate) with minor blunt head trauma, and used the same inclusion/exclusion criteria and outcome definitions as the Kuppermann study. They compared 676 children (49%) whose physicians noted a plan of ED observation in 676 patients (49%) with those whose doctors either ordered an immediate CT or did an evaluation and immediate discharge.
Observed patients received fewer CTs, and longer observation times were associated with a decrease in CT rates. In fact, after adjustment, every hour of ED observation reduced CT use by about 70 percent. And, importantly, there were no misses; all eight children with significant brain injuries were picked up immediately. Given that there were only eight such outcomes, the study was underpowered to comment on brain injury as a study outcome. But, nonetheless, Dr. Kuppermann said, “this study offers further evidence that the strategy of observation before CT decision-making is effective in reducing unnecessary CT scans in selected children after blunt head trauma. The challenge remains incorporating shared decision-making, then disseminating this strategy widely with the help of electronic health records.”
Some parents, of course, will not have any interest in such advice, and EPs may need to rely on the threat of ionizing radiation in those cases to help them make an appropriate risk/benefit calculation. So knowing all this, perhaps the hypothetical scenario we painted above could have gone a bit differently.
Doctor: Before we get the CT scan, however, I'd like to review the data with you. Fortunately, we have some really good medical evidence to guide us here. Because it sounds like your child likely had such a brief period of being unconscious, which was probably a breath-holding spell, and is otherwise quite well, his risk of having a significant head injury is close to zero.
Mother: I don't care about your medical evidence.
Doctor: Well, then, can I tell you about the risk of ionizing radiation from the CT scan? The ionizing radiation that Junior will get in CT could increase his risk of a cancer later in life.
Father: Oh, my.
Doctor: Perhaps I could offer a different option. What if we observed Junior here in the ER for an hour or two? We can keep him for a while and if anything seems amiss, well, we'll re-consider that CT scan.
Patient: Happy, coo, happy, coo, goo
Mother and father in unison: Oh, yes, thank you, Doctor!
Doctor: Phew. One day soon, I'll need to talk patients into getting a CT.