TORONTO - Reducing unnecessary variation in CT radiation dose can be done fairly easily by getting radiology professionals together to share best practices, even across a healthcare system as large as the University of California's (UC), according to a presentation at the annual American Roentgen Ray Society (ARRS) meeting.
Healthcare providers from across five UC facilities were able to smooth out wide-ranging variation in CT radiation dose simply by meeting to communicate about scanning protocols and share dose data, according to Nicole Wilson of the University of California, San Francisco. The project was badly needed: At least one UC site had a median CT dose that was nearly twice as high as the site with the lowest dose.
"One of our goals has been to standardize CT dose across all of the UC medical centers," Wilson said. "And we found that just by providing dose audit feedback and sharing best practices, doses were reduced."
Brainstorming session
Wilson and colleagues in 2014 gathered a group of radiologists, physicists, and technologists from five UC medical centers: Davis, Irvine, Los Angeles, San Diego, and San Francisco. Together, the group reviewed detailed CT radiation dose summary reports from the facilities and brainstormed techniques to lower and standardize doses.
The researchers collected radiation dose metrics for more than 58,000 consecutive CT scans of the head, chest, and abdomen across all five medical centers before and after the meeting. They measured radiation for CT head exams using CT dose index volume (CTDIvol, expressed in mGy) because the scans were single-phase; they measured radiation for chest and abdominal exams using effective dose (mSv) because most body scans are multiple-phase.
The medical centers were numbered 1 through 5, and Wilson presented study results using these numbers rather than the names of the facilities.
Before the meeting to discuss dose, median head CT dose was 44 mGy, 62 mGy, 33 mGy, 59 mGy, and 53 mGy, for centers 1 through 5, respectively.
"The highest site had a 90% higher median dose than the lowest," Wilson said.
Chest and abdominal median effective doses are shown in the tables below.
| Median CT radiation dose across UC sites, preintervention | |||||
| UC 1 | UC 2 | UC 3 | UC 4 | UC 5 | |
| Chest | 11 mSv | 9 mSv | 17 mSv | 8 mSv | 6 mSv |
| Abdomen | 11 mSv | 14 mSv | 21 mSv | 20 mSv | 11 mSv |
For chest CT, the highest-dose site had a 180% higher median dose than the lowest; for the abdomen, the highest-dose site had a 90% higher median dose than the lowest, Wilson said.
After the meeting -- also known as "the intervention" -- the median effective dose decreased across all five sites by 25% overall for both chest and abdominal exams, the researchers found.
| Median CT radiation dose across UC sites, postintervention | |||||
| Chest | Abdomen | ||||
| Dose | Reduction from preintervention | Dose | Reduction from preintervention | ||
| UC 1 | 9 mSv | 18% | 8 mSv | 27% | |
| UC 2 | 7 mSv | 22% | 10 mSv | 28% | |
| UC 3 | 13 mSv | 23% | 17 mSv | 19% | |
| UC 4 | 6 mSv | 25% | 13 mSv | 35% | |
| UC 5 | 6 mSv | 0% | 10 mSv | 9% | |
As for head CT exams, only UC 2 demonstrated a meaningful dose reduction after the meeting (from 62 mGy to 55 mGy, an 11% reduction).
Sharing is caring
Radiologists were more than willing to come together and share dose data, Wilson said. And something this simple had dramatic results.
"The group was enthusiastic," she said. "Radiologists found the dose feedback and sharing of best practices helpful, and they enjoyed developing concrete dose reduction strategies as a team."
Wilson's team plans to continue this work through a project called the Partnership for DOSE Collaboration, which includes 27 adult and pediatric hospitals and is funded by the Patient-Centered Outcomes Research Institute (PCORI) and the U.S. National Institutes of Health (NIH).
"We look forward to continuing to explore the barriers and facilitators to dose optimization," she said.
