Pediatric Emergency Departments Find Ways to Lower CT Use

Daniel M. Keller, PhD

February 13, 2012

February 13, 2012 — Contrary to reports of an overall increase in the use of computed tomography (CT) in emergency departments, researchers found no trend for increased CT use in 2 tertiary care pediatric emergency departments of a large pediatric healthcare system during a recent, 8-year period. In fact, lead investigator Margaret Menoch, MD, from the Department of Pediatrics and the Department of Emergency Medicine at Emory University School of Medicine in Atlanta, Georgia, and colleagues found that between 2003 and 2010, in areas where alternative non-radiation-based imaging modalities were available, CT use decreased and the use of alternative modalities increased. Dr. Menoch and colleagues released their results in an article published online February 13 in Pediatrics.

CT imaging exposes patients to relatively large doses of radiation, and the potential risk for radiation-induced malignancy is of particular concern in children, who may be most vulnerable because of rapid cell proliferation in their developing tissues. Various reports have shown increased use of CT for both adult and pediatric patients during the past decade or so. Contributing to the trend of increased CT use is that it is widely available in emergency departments, it is fast, it gives intricate details of anatomy, and physicians often do not recognize the harm that may go along with it.

Dr. Menoch explained to Medscape Medical News that previous studies, including ones involving adults, have seen increasing trends in CT use based on somewhat older data, but in light of her study's more recent time frame, she "found that there is no increase at our institution. And specifically, in certain chief complaint categories, we actually found a decrease."

For this study, the investigators performed a retrospective electronic chart review of pediatric emergency department patient records within their healthcare system and categorized annual CT, magnetic resonance imaging (MRI), and ultrasound imaging use rates according to anatomic location and the chief complaints of head injury, seizure, and abdominal pain, as well as the acuity of the visit. Visits were categorized by chief complaints according to the various International Classification of Diseases, Ninth Revision, codes constituting each complaint. Each form of imaging was counted separately if more than 1 method was used during a visit.

During the study period, orders for CT scans were at the sole discretion of the ordering physician. All patient visits were included except in the cases of transfers or unknown transfer status because of the potential for having a CT scan before arrival.

No Increase in CT Use Either Overall or by Anatomic Location

A CT scan was performed in 5.5% of the 987,932 eligible pediatric emergency visits, of which 63% were head CTs, 20% were of the abdomen/pelvis, 3% were of the cervical spine, and the remaining 14% were categorized as "other." The median age (range, 8 - 9 years) and sex distribution of patients (range, 55.3% - 57.7% boys) were similar year to year.

Despite an increase in high-acuity visits overall during the study period (β, 0.9; 95% confidence interval [CI], 0.76 - 2.2), as well as admission disposition (β, 0.87; 95% CI, 0.63 - 2.3), the rate of CT use overall or by anatomic location did not change significantly. CT use declined significantly for the chief complaints of seizure (β, −0.974; 95% CI, −1.437 to −0.897) and head injury (β, −0.927; 95% CI, −1.707 to −0.726). There was no significant change in the rate of head MRI.

The proportion of high-acuity patients in the head injury and seizure groups increased significantly during the study period. There was a significant decrease in the proportion admitted for head injury (β, −0.8; 95% CI, −1.52 to −0.12), but not for seizure.

Although there was no significant decline in abdominal CTs for the chief complaint of abdominal pain (β, −0.64; 95% CI, −0.83 to 0.08), the rate of abdominal ultrasound increased (β, 0.886; 95% CI, 0.247 - 0.786), especially after 2007.

There were no changes in the number of head CTs or MRIs to evaluate ventriculoperitoneal shunts.

Limitations of the study are its retrospective nature, with inherent concerns about data retrieval, and the fact that it was performed in the pediatric emergency departments of 1 academic healthcare system. In addition, the study evaluates only imaging ordered in the emergency departments and cannot account for imaging after hospital admission or for imaging of stable patients deferred to the outpatient setting.

The authors noted that "it is estimated that >85% of CT imaging performed in children in the [emergency department] is done at primarily adult facilities," so the study does not address trends that may exist in those settings.

"I think the point is to show how we're practicing [pediatric emergency medicine] help make the adult [physician] population more aware of alternative ways to practice and to expand the understanding of the risks of CT into their world," Dr. Menoch said. "Because over 80% of pediatric patients present to a nonpediatric facility, and more CTs happen at adult emergency rooms, I think this is where the behavior needs to be centered on changing and the awareness needs to be focused."

Although the study could not show causality for the findings, Dr. Menoch speculated about some of the reasons for the decrease in the use of CT in her pediatric emergency department, and possibly elsewhere. "First off, it could be that families and physicians are more cognizant of the increased radiation risks that are associated with CT usage," she said. Furthermore, recent studies have demonstrated more efficient ways to use head CTs for seizure in closed head injury in combination with increased awareness of alternative imaging or observation. She said observation is a "tool that we're using frequently now...which is really just opening people's eyes to [the fact] that there are...alternatives to imaging everyone who comes in with a seizure and a head injury. You don't have to practice that way."

David J. Brenner, PhD, professor of radiation biophysics and director of the Center for Radiological Research at the College of Physicians and Surgeons of Columbia University in New York City, who was not involved in the study, commented to Medscape Medical News that overall CT usage has been rising steadily by about 5% to 6% a year in the United States during about the same period as in the study.

He cited the figure that CT usage as a whole in the United States is about 81 million scans each year. From 1996 to 2007, CT usage overall in emergency departments "went up pretty steadily, from about 3.2% of all visits to about 14% of all visits," he said. "If indeed the case that pediatric use was not going up, that was bucking the overall trend.... So it's surprising and gratifying, I think."

Unfortunately, he noted, almost all the published studies tend to be done in "pretty good academic institutions, and where people are very aware of the issues associated with pediatric CT." However, most emergency department children's CT scans are done in a less academic setting, where people may not be so aware. "And it's a big weakness that we don't actually have a real way of quantifying that," Dr. Brenner said.

He said that one way of getting the word out to the medical profession is through the efforts of Image Gently, an initiative of the Alliance for Radiation Safety in Pediatric Imaging, a coalition of 68 healthcare organizations formed to encourage safe, high-quality pediatric imaging.

Dr. Brenner mentioned 2 ways to minimize radiation exposure: lower the dose, or do fewer scans. To accomplish the former, he said, most modern CT scanners incorporate new techniques, 1 being automated exposure control, in which the scanner optimizes the amount of radiation delivered based on a patient's body habitus. The operator can set how much noise he or she is prepared to accept in the image, and the machine adjusts the dose to produce a scan with that signal-to-noise ratio. As it scans down the length of body, it can adjust for bony areas, and it varies the current as the machine rotates around the body to optimize the X-ray dose.

In addition, good decision rules can reduce the number of scans, "and you also need people to actually take note of the decision rules," Dr. Brenner advised. As an example, he said that at Massachusetts General Hospital in Boston, when a physician orders a scan, the computer asks the reason for the scan, lists a level of appropriateness of a CT scan in that context, and gives alternatives for imaging. Ultimately, the physician makes the decision about whether to do the scan, but Dr. Brenner said the system has worked "quite well," and CT use in the hospital's emergency department has leveled off.

The number of CT scans is going up in general because "it's a darn good tool," he said. "It's become the gold standard in many fields" because it is fast, has good specificity, and gives information about alternate diagnoses if present.

However, "a third of CT scans don't need to be done. That's really where we need to focus our efforts to make sure that we don't do CT scans when we don't need to, because inherently they are a high-dose examination," Dr. Brenner warned.

The study was not commercially funded. The authors and Dr. Brenner have disclosed no relevant financial relationships.

Pediatrics. Published online February 13, 2012. Abstract


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.