ACR Bulletin

Covering topics relevant to the practice of radiology

New Pediatric CT Dose Benchmarks

The ACR DIR benchmarks are a step forward for pediatric dose optimization.
Jump to Article

We are constantly trying to balance this equation of what is acceptable image quality and what is the minimum detriment of radiation.

—Donald P. Frush, MD, FACR
December 20, 2021

Comprehensive benchmarks for pediatric CT protocols have been lacking in the radiology space for a long time. That changed in October 2021 with the publication of the new ACR Dose Index Registry® (DIR) benchmarks in Radiology — a working document to guide radiology facilities in adjusting pediatric CT protocols and resultant doses for their patients.

“This is the first paper of its kind,” says Kalpana M. Kanal, PhD, past chair of the ACR DIR and professor and section chief of diagnostic physics in the department of radiology at the University of Washington. “We now have benchmarks for the top 10 pediatric CT exams in the U.S., and hopefully this will help sites to optimize radiation dosage as a function of patient size and age.”

The process of compiling these benchmarks was a years-long endeavor, Kanal says. The DIR supplied investigators with approximately 1.5 million CT examinations as a function of patient age and size from a wide range of facilities with different characteristics, such as community clinics, urban hospitals, and academic sites. The publication’s authors analyzed DIR data from 2016 to 2020 to develop the diagnostic reference levels (DRLs) that radiology personnel can reference in their day-to-day work.

In the report, both DRLs and achievable doses (ADs) are provided so facilities are encouraged to compare their doses to this national benchmark and optimize their radiation dose delivery as needed. DRLs should be used to determine if a local facility’s CT dose indices are unusually high or low, but not be used as target doses. DRLs and ADs are not intended to be used for comparisons with dose indices for individual patients but rather for the whole patient population. Facilities now can analyze and compare their size and age-grouped dose indices with the respective size or age-grouped ADs and DRLs, as appropriate.

We now have benchmarks for the top 10 pediatric CT exams in the U.S., and hopefully this will help sites to optimize radiation dose as a function of patient size and age.

—Kalpana M. Kanal, PhD

“This information is current and comprehensive,” says Donald P. Frush, MD, FACR, professor of radiology and chair of the Image Gently® Alliance. “Our analysis indicated that much more than 90% of CT scans are going to be included in that set of 10 most common pediatric CT exams.”

As described in the publication, prior to these long-awaited benchmarks, little data existed on pediatric dose indices. The data set that was available was often “outdated, lacked data from diverse imaging settings and scanner manufacturers, or had limited statistical power.”1

“We are constantly trying to balance this equation of what is acceptable image quality and what is the minimum detriment of radiation,” says Frush. “Radiation exposure in any patient is something we should be mindful of, and it has been recognized that growing children are more susceptible to radiation effects than someone who is fully grown.”

According to Frush, these benchmarks help take the guesswork out of radiation doses for populations of pediatric patients and allow radiologists, RTs, and medical physicists to work together under well-researched guidelines that will keep their patients safe and ensure image quality. It is challenging, he says, for small radiology practices where pediatric scanning is very infrequent to analyze DRLs and proper dosing, so these benchmarks are especially helpful in that setting.

Frush notes, however, that it is important to remember that nothing in radiology is one-size-fits-all. Each case is different and each patient’s unique needs should always come first. DRLs are meant to be used to determine if a local facility’s dose indices are unusually high or low, rather than being used as target doses in all cases. DRLs and achievable doses are intended to compare a patient population as a whole, he says.

“People ought not to look at this paper as a measure of enforcement at all,” says Frush. “These benchmarks are provided as a tool for people to review and say, ‘We now know how we compare with a very large pool of data of how others do this.’ I think many practices today will find their imaging protocols are entirely harmonious with these DRLs, and there is a level of satisfaction and comfort to be able to reassure the pediatric patients and families they take care of.”

“Radiologists, RTs, and medical physicists all play a role in taking care of the patient and have to be committed to open communication and constant collaboration,” says Kanal. “The image quality has to be adequate to give radiologists confidence in their diagnosis and that they are not missing something on the CT scan. RTs are on the front lines, scanning the patient in real-time and communicating any limitations to the radiologist and the medical physicist. Lastly, the medical physicist provides the technical knowledge that will guide the radiologist and RTs in adjusting CT protocols. Regular protocol assessments ensure that nothing is missed, and all the different parts are moving together as one.”


1. Kanal KM et al. U.S. Diagnostic Reference Levels and Achievable Doses for 10 Pediatric CT Examinations. Radiology. 2022 302:1, 164-174

Author Ivana Rihter,  Freelance Writer, ACR Press