What is the ideal number of CT scans for headaches in children?

Many, myself included, have written about the overuse of head computed tomography (CT) scanning in children. This concern has become more focused now that we have some data on the radiation risk of those scans. The bottom line is not that we should stop doing head CTs in children, but that we should always balance the risk against the benefit, just as we should do with all medical testing. In the case of CT, the risk is tiny, but it is not zero. That risk is worth taking if the benefit is substantially more than the risk; that is, if getting the information that the CT yields is a good exchange for the risk. Put another way, is the risk of not getting the information the scan gives us greater than the risk of doing the scan itself? As a PICU doctor I order a lot of head CT scans, and I think about this trade-off with each one.

A very common reason doctors order a head CT is to evaluate a headache, even when they know the chances of finding a serious cause for the headache, such as a brain tumor, are very small. In many situations those chances of finding something bad are near zero. So how should we analyze the risk/benefit ratio between doing the CT or not? Various experts have formulated recommendations for when a head CTis indicated to help guide us in our decision-making. These include any items in the following list. The key thing to keep in mind is that serious brain problems, especially a tumor, only very rarely appear without some other abnormality accompanying the headache.

  • Age less than 3 years
  • Sudden, explosive and severe headaches
  • Associated changes in mental status or performance
  • Headaches that awaken a child from sleep
  • Associated nausea and vomiting
  • A steadily worsening pattern of headache

recent study in Pediatrics, the journal of the American Academy of Pediatrics, assessed how well doctors (and families, too — parents often urge doctors to do CTs) are doing. The answer is not so well.

The authors of the study examined the claims records of over 15,000 children who had had at least two visits to a doctor with the complaint of headache — 25% of them received at least one head CT scan as part of the evaluation of their symptoms. Interestingly, children seen by a neurologist, a brain specialist, were only half as likely to get a CT scan. Children seen in an emergency department were four times more likely to get a CT scan than were children evaluated in a doctor’s office.

Some of these results are easily explained. It makes sense that a neurologist is more skilled in evaluating headaches and is therefore more comfortable not doing a CT than is a doctor who only occasionally treats headache. Children being seen in an emergency department are usually there because of some acute problem, so if they are there for a headache it is more likely to be new and sudden. Emergency department doctors rarely know the child, so they are probably more swayed to rule out a serious problem while they have the child there. Also, emergency room doctors order a lot of head CT scans anyway, probably too many, for minor head injuries. So they are already primed to order scans for headaches.

What is the ideal number of CT scans for headaches in children? Certainly we don’t want every scan to show an abnormality; if that were the case, it would mean we are not ordering enough of them. We want most of them to be negative to make sure we are catching all the serious problems. But the finding that a quarter of all children going to the doctor for a headache are getting a CT scan is disturbing — it’s too many.

Parents have a key role in this, especially since at least one study indicates that a major reason for doctors ordering all those scans is that parents are anxious enough not to trust either the doctor’s judgement or the standard list of indications for a scan, whether that’s for a headache or a bonk on the head. I have encountered that myself. Remember that the recommendations for head CT have been validated by research; if your child with a headache doesn’t fit any of them, it is best to wait and see how things go.

Christopher Johnson is a pediatric intensive care physician and author of Your Critically Ill Child: Life and Death Choices Parents Must FaceHow to Talk to Your Child’s Doctor: A Handbook for Parents, and How Your Child Heals: An Inside Look At Common Childhood Ailments.  He blogs at his self-titled site, Christopher Johnson, MD.

Comments are moderated before they are published. Please read the comment policy.

  • Frank Lehman

    Johnson said: “Parents have a key role in this, especially since at least one study indicates that a major reason for doctors ordering all those scans is that parents are anxious enough not to trust either the doctor’s judgement or the standard list of indications for a scan, whether that’s for a headache or a bonk on the head.”

    Here we go again with “blame the patient” (or the parent of the patient) for the physician’s inadequacies. Just like prescribing antibiotics when not medically appropriate.

    • Guest

      I don’t know what physician wronged you before, but you seem like a wretched soul because of it.

      Physicians are not machines. Medical decisions do not occur inside a vacuum. It’s critical to include the patient (or their parents) as part of the decision making. Ultimately it will be the physician pulling the trigger on ordering a test, but to suggest that patients have no say is disingenuous at best.

    • disqus_mmpOofx73c

      Actually, Johnson has a point here. I’ll speak from experience: my 2 year old son is hearing impaired, with many issues involving congestion and ear infections. A highly regarded implant specialist at NYU Medical Center recommended a CT scan, but my pediatrician and ENT strongly think we should hold off unless it becomes vital. Meanwhile we’ve done an MRI.

      I am the type who usually puts pressure to run every test and be proactive, but my ENT has taught me that not always is testing the answer, some things have to resolve themselves without medical intervention. If it were up to me the CT scan would have been done a long time ago, but I am happy that my pediatrician and ENT are holding me in check.

    • http://www.chrisjohnsonmd.com/ Chris Johnson

      Of course I’m not blaming the patient. And yes, in fact, families do sometimes demand antibiotics when they are not indicated.

      Doctors and families are a team. Both have rights and responsibilities.

  • PollyPocket

    A physical exam by a pediatric neurologist for those kids prior to scanning would probably slash those numbers dramatically. It’s a shame they are so few and far between.

    What are your thought in a fast T2 MR scan? Yes, cost is much higher, but it is fast enough, sedation is generally not required and there is no radiation risk.

    • Allison Falin

      Where I lived for the last 20 years, we had no pediatric specialists at all and it was a month or more wait to get a referral into see one (similar to what you are saying).

      While that is a great resource to have, utilizing more specialty care when a pediatrician or pediatric healthcare provider can do leg work. If more people would have the relationship with their healthcare provider to be able to trust their judgement (and they practice evidence based medicine) then I think the CT overuse would stop.

      I can recall my son being hit hard in the face and he had a nosebleed that just would not stop. I was perfectly fine with a facial x-ray, but the provider did a CT instead all the while admonishing me that it won’t show a concussion (yes, I knew that already). It was a HUGE waste of money to see if the child had a facial fracture.

Most Popular