Unpredictability of CT Dose Addressed
Knowledge of the intrinsic variability in radiation dose delivered to patients undergoing diagnostic CT imaging would aid efforts to evaluate and lower the radiation patients receive, said a presenter of a study Wednesday at RSNA 2013.
Radiation dose can be almost two-fold higher for a patient receiving an identical repeated CT in identical conditions, said Douglas G. Larson, M.D., during his presentation, "CT Dose Variability for Patients Undergoing Repeat Identical CT Scans: A Retrospective Analysis of 2606 Patients Undergoing 12,632 CT Scans."
"All we know is the average dose number is being brought lower," said Dr. Larson, an abdominal imaging fellow in the radiation department at Duke University School of Medicine. "Almost nobody is talking about the standard deviation. It turns out there is quite a bit of unpredictability in the CT system."
Dr. Larson discovered the minimum to maximum variation ratio of radiation dose ranged from 1.56 times to 2.02 times in patients. "We pushed all the same buttons, we did everything the same and one of the scans had double the dose of one of the others without necessarily imparting increased clinical value," Dr. Larson said.
Patient size, table height, scanner manufacturer and scanner model can all affect dose variability. No variation was seen with patient age or patient gender. "An important step in understanding CT variability is identifying the controllable – the things we can manipulate about the scan and the patient – as well as the intrinsic components of dose variability," Dr. Larson said.
Dr. Larson retrospectively examined the records of more than 2,000 patients who had undergone more than 12,000 combined repeat CT scans at his institution. He identified colon cancer, rectal cancer lung cancer and renal stone patients who underwent the same CT protocol at least twice between January 2007 and February 2013.
Data collected included study protocol; CT dose index (CTDI) and dose-length product (DLP); table height; pitch; noise index and effective mA; peak kilovolt (kVp); date; patient age and gender; and patient size.
The researchers also wanted to know the role played by the scanners themselves -- the manufacturer, the model and the unique ID of each. They found a significant statistical difference does exist.
"Even if the scanners look identical they don't actually perform the same," Dr. Larson said. "Each piece of equipment has its own dose distribution characteristics. If you look at different manufacturers or generations of scanners, everything behaves a little differently."
Other contributing factors likely exist, including particular CT technicians, which were not taken into account during his research, he said.
As the push for lowering radiation dose moves forward, Dr. Larson said the question involves more than simply reducing the average dose received by patients as a whole.
"I feel it's a real blind spot in all of the work that's happened to date," he said. "I fully support everything we've done to lower CT dose. But as we try to stretch the limits, it's time to start looking at some other factors as well."