Study maps CT radiation doses in 100K exams

By Eric Barnes, staff writer

July 27, 2015 -- Have you ever wondered how your facility's CT radiation dose levels stack up against those of other institutions? A study in Radiology describes the experience of researchers from five University of California (UC) campuses who measured dose in nearly 86,000 adults and children undergoing common CT scans.

The results showed that reference dose levels at the high end of the scale -- 75th percentile -- are generally higher than 75th-percentile doses reported in Europe, where reference doses are well-documented. In fact, some European centers are achieving lower doses than even this study's 25th percentile, the authors wrote. However, for the U.S., the numbers represent a kind of data that has been hard to find.

In contrast to Europe, few U.S. studies have assessed CT doses by anatomic area for a large number of consecutive scans. As a result, the U.S. lacks large-scale reference dose information on par with Europe's, explained Dr. Rebecca Smith-Bindman from UC San Francisco, along with colleagues from four other UC campuses.

"We provide some kind of ballpark of what dose levels you should try to stay around," Smith-Bindman told in an interview. "I don't say our numbers are optimal, but they are achievable at facilities that want to achieve them. If I'm going to scan a patient, I have a responsibility to get below that level, and if not, I need to think about the reasons."

Building benchmarks

The University of California Dose Optimization and Standardization Endeavor (UC DOSE) is a collaboration to evaluate and optimize dose at UC medical centers in Davis, Irvine, Los Angeles, San Diego, and San Francisco, the authors wrote (Radiology, May 19, 2015).

The centers pooled data, agreed on measurement, and created benchmarks and reference levels to track performance and changes over time, according to Smith-Bindman.

The researchers summarized CT data from 12 facilities at the five universities, reporting doses as CT dose index volume (CTDIvol), dose-length product (DLP), and effective dose (mSv).

They prospectively collected radiation dose metrics from 199,656 consecutive CT exams acquired during 2013 in 83,181 adults and 3,871 exams in 2,609 children. In all, there were 34 scanners from five manufacturers; this included one eight-detector-row machine and five 16-detector-row scanners, while the rest had 64 detector rows or more.

Software (Radimetrics, Bayer HealthCare Pharmaceuticals) was used to measure and track dose data. The group reported results for exams of the head, chest, or abdomen; combined chest and abdomen; neck; and sinus. These regions account for 79% of all CT scans in adults and 71% of CT scans in children, the authors wrote.

Automatic tube current modulation was used in 92% of scans in adults and 88% of scans in children. The most common area imaged in adults was the abdomen (32%), followed by the head (16%) and combined chest and abdomen (14%). For children, the head led with 33% of all exams, followed by the abdomen (18%) and chest (9%). Multiphase scans accounted for 42% of exams in adults and 14% in children.

Median effective radiation doses
Exam type Effective dose for 25th percentile (adults) Effective dose for 75th percentile (adults) Effective dose for 25th percentile (children) Effective dose for 75th percentile (children)
Single phase 1 mSv 3 mSv 1 mSv 3 mSv
Multiphase 3 mSv 8 mSv 3 mSv 6 mSv
Single phase 5 mSv 13 mSv 2 mSv 4 mSv
Multiphase 12 mSv 29 mSv 2 mSv 7 mSv
Single phase 6 mSv 16 mSv 3 mSv 6 mSv
Multiphase 15 mSv 32 mSv 4 mSv 10 mSv

The dose ratings offer benchmark data that CT facilities can use as a starting point for evaluating their own dose levels when other detailed measures are not available, Smith-Bindman and colleagues wrote.

The American College of Radiology (ACR) has created summary diagnostic reference dose levels as part of its CT certification program, "but these levels are based on submissions of a small number of self-identified 'best' cases, rather than on typical cases or random cases from a facility," Smith-Bindman said.

What's more, the UC dose levels are lower than those from ACR, she noted.

"If you compare our numbers to ACR's, our summary data are 20% to 65% lower -- and I think they are very achievable," Smith-Bindman said. "For the ACR accreditation doses, people are supposed to be submitting their best work, but in my mind [ACR] isn't setting targets as low as they could be."

The study's strength lies in its large sample size, according to the authors. In terms of limitations, the study did not address some scans covering multiple regions, and future technical improvements may permit even lower doses. The doses provided "do not necessarily reflect best practices," they cautioned, noting that many centers get by with even lower levels.

Beyond this analysis, the data show enormous dose variability between centers. There is also variability within the same facility for the same exam and sometimes even the same patient, Smith-Bindman said. An upcoming study will look into it.

"This paper is admittedly very boring, because telling people what dose reference levels are isn't exciting," she said. "But I get asked these questions all the time: 'What should the radiation dose be?' Here we at least begin to provide some answers."

Copyright © 2015

To read this and get access to all of the exclusive content on create a free account or sign-in now.

Member Sign In:
MemberID or Email Address:  
Do you have a password?
No, I want a free membership.
Yes, I have a password:  
Forgot your password?
Sign in using your social networking account:
Sign in using your social networking