07 April 2011

How things change -- CT scans and radiation

There was an interesting study published this week in the journal Radiology: 

Rising Use of CT in Child Visits to the Emergency Department in the United States, 1995–2008 (Abstract)

The results are not surprising to anyone who has been working in medicine in the US over the last fifteen years. Basically, in 1995, a kid visiting the ER had a 1.2% likelihood of getting a CT scan, and by 2008, that number was 5.9%.

I had written about this general phenomon not too long ago, in defense of the general increase of CT utilization in the ER, largely on the basis that CT is a better tool: it provides diagnoses in a rapid and timely manner, and excludes many potential life threats, saving lives and mitigating malpractice risk. That was largely relevant to the adult population, though, and kids are not, as they say, just little adults. The increase in scanning children is more dramatic, especially given the generally lower incidence of disease in kids compared to adults and the chonrically ill.

The experience I have had in the ER was comparable to the study findings. When I started training in the mid '90s, CT scans were time-consuming, offered less than excellent images and were relatively uncommon to get in general, let alone in pediatric populations. By the time I started practice in 2000, things had changed dramatically. The image quality and diagnostic information a scan provided was way better. The development of high-speed helical scans made it much easier to get them on kids without sedation, which greatly lowered the operational barrier to getting them. The overall accessibility and use of scans were rising and the pediatric use rose in lockstep. There was little concern about the risks of radiation or the downside of getting a scan on a kid if there seemed to be a clinical indication. This was largely (but not entirely) driven by percieved malpractice risk, especially in the hot-button liability scenarios of head injuries and potential appendicitis. The feeling was that the consequences, both human and financial, in missing an epidural hematoma in a child were so gargantuan, and the risk inferent in the radiation so trivial, that why not do it? We ordered scans willy-nilly, it seems in retrospect.

Appendicitis is a little different. There is the fact that you had other options, including ultrasound, observation, and surgical consultation. So CT was not used quite as profligately for potential appendicitis. But for some cases (and some docs) it did become the test of choice. Ultrasound is frustratingly insensitive, time-consuming and operator dependent. Surgeons are difficult to get to come to the ER for a consult. And, especially in the malpractice insurance crisis in the early part of the decade, many docs were uncomfortable with the risk inherent in watchful waiting. This is not to excuse, but to explain why CT rates shot up, at least in my experience.

But a very welcome countervailing trend has definitely arisen over the past four years or so, especially in but not limited to the pediatric population, as the risks of radiation with regard to lifetime cancer incidence have become better understood and better publicized. This has spurred a change in practice in our ER, driven by the clinical leadership in concert with our radiologists and pediatricians. I don't have numbers, but it seems like the use of scans for kids has slowly dropped and dropped over this time frame. 

We have been assisted in this by the development and validation of some good decision-making rules for selecting the kids with head trauma who are likely to benefit from a CT scan. There's also increased awareness and acceptance amoung patients and their families. It used to be that parents had an expectation of and even advocated for CT imaging. I've always been more of a diagnostic minimalist, and back in the day I would have to really go out of my way to convince some parents that a CT scan was not necessary. Now it seems all I have to say is that "We have learned that this is a lot of radiation to focus on a growing brain/set of genitals," and the parents are nodding and very accepting of my explanation that their child is not likely to benefit from a scan. We have also developed algorithms for the evaluation of abdominal pain which make it pretty clear that CT scans are discouraged unless there is a darn good reason. So things are changing, and for the better.

I recall a couple of years ago I blogged about a difficult diagnosis of appendicitis and the need to scan a young boy. It was complicated by the fact that the parents, from the Ukraine, has a remarkable fear of radidation, due to the Chernobyl incident. I had reassured the parents that "I had a five-year-old son, and if it were my son with the same pains, I would not hesitate to put my son in the scanner." At the time, that was true. I'm not so sure that I would make that assurance today. Now, I am much more likely to treat kids with observation (either in house or at home).

The pendulum swings, but ever so slowly.


  1. There's a good article in the Feb 25 issue of Science on this very question. I can't post the full article, but the summary is below.

    Science 25 February 2011:
    Vol. 331 no. 6020 pp. 1002-1004
    DOI: 10.1126/science.331.6020.1002

    News Focus

    Second Thoughts About CT Imaging

    Lauren Schenkman

    David Brenner directs Columbia University's Center for Radiological Research, where he focuses on exactly how radiation damage leads to cancer. He's become one of the most insistent voices in an imbroglio that is roiling radiologists, medical physicists, and the general public over the rising and largely unregulated use of computed tomography (CT) scans, and whether the technology can, in some cases, cause more harm than good. The risks are surprisingly unclear, given how old and how commonly used the technology is. Brenner found that each CT scan gives a patient a very small chance of developing cancer, although many radiologists and medical physicists say that for a single CT scan, there's no hard evidence of any raised cancer risk. But even the skeptics favor managing potential CT risks, if for no other reason than to reassure patients. As the debate rages, the number of CT scans administered continues to soar and shows no sign of slowing down.

  2. My four year old just got worked up for a possible appendicitis last week, her exam was quite convincing to the PA at the clinic but appendicitis was ruled out by ultrasound fortunately.

    I was glad the ultrasound was able to do the job, but if it hadn't I'd still be much happier with the CT scan that would have been done than with the appendectomy that would have been done twenty odd years ago when I started in health care.

    I forget the exact statistics but a ridiculous number of appendectomies used to be done on perfectly normal appendixes. I little extra radiation is far better than the little extra hole in the belly and the attendant risks that people used to have to go through.

    Certainly as technologies become available, like ultrasound, that decrease the risk further they should be used, but people complaining about CTs so much gets silly some times.

  3. It is not silly when people continue to be overdosed in little and big hospitals across the country - like ME.