Radiologists Urged to Think Beyond Reduction When it Comes to Pediatric Dose

Optimization, not simply reduction, should be the goal when it comes to reducing radiation exposure from CT scans—particularly in pediatric cases.

Optimization, not simply reduction, should be the goal when it comes to reducing radiation exposure from CT scans, particularly in pediatric cases, according to new research presented at RSNA 2012.

“We as radiologists in conjunction with the technologists and physicists are responsible in determining the appropriate factors for every CT examination,” said Charles Glasier, M.D. “From my view, radiologists need to be in the lead and set the parameters so the technologists have a relatively narrow range of techniques they can use in most patients.”

Overexposure remains an issue across the U.S. and the world as the number of CT examinations in children continues to increase annually. Presenters noted, however, that the number of exams is being reduced at pediatric hospitals. Balancing the risks for a particular patient can be complicated, they added.

Optimization over reduction is the goal of a new protocol presented by David B. Larson, M.D., M.B.A., of Cincinnati Children’s Hospital. He said the CT dose index (CTDI) should be converted to a size-specific dose estimate and used a real-world example to show that the actual dose a patient receives can vary by more than a factor of two if CTDIvol is used without taking into account the patient’s size.

“It’s not about reducing dose; it’s about finding that optimal target,” Dr. Larson said. “In order to do that in practice with the tools we currently have, you have got to measure a size-specific dose. You can’t just take the CTDI.”

CTDI is a metric that doesn’t take into account individual patients and is not meant to be a surrogate for patient dose, said Dianna Cody, Ph.D., a physicist in the Department of Imaging Physics at the University of Texas MD Anderson Cancer Center in Houston. “CTDI is something that can be very standardized and measured in an exact way in many different settings,” she said. “It’s about scanner output. It’s not really about patient absorption.”

Dr. Cody pointed to several factors that influence dose, including kilovoltage peak (kVp), milliamps and scan time, and pitch or table speed, and offered three dose reduction options: tube current modulation, dynamic collimation and iterative reconstruction.

To implement these options, she said, radiologists should make sure CTDIvol is displayed on all CT scanners—sometimes the display is turned off—and make sure dose report pages are sent to PACS. Dr. Cody also urged radiologists to consider joining the American College of Radiology Dose Index Registry.

“If physicians can develop a more intuitive feel for CT radiation dose issues, patients should benefit from that intuition,” Dr. Cody said. “Some sites may have CT scanners with capabilities for reducing radiation dose that are not fully utilized. I hope that after attending courses like these, practitioners will be proactive and will seek out and implement these options on the scanners in their care.”

Evolving Reconstruction Techniques Preserve Quality, Reduce Dose

Tang

Zbijewski

An iterative reconstructive method known as prior imaging constrained compressed sensing (PICCS) maintains the diagnostic quality of CT images while reducing dose when imaging for urolithiasis, according to research presented at RSNA 2012.

Jie Tang, Ph.D., found PICCS provided diagnostic image quality with a 19 percent mean radiation dose in sub-mSv CT scans for urolithiasis, while filtered back projection (FBP) images were non-diagnostic.

“Many people are concerned about radiation dose and overexposure from CT examinations,” said Dr. Tang, an associate scientist in the Department of Medical Physics at the University of Wisconsin-Madison. “We have demonstrated that with PICCS we are able to significantly reduce noise in the images and thus we are able to reduce radiation dose.”

An ultra-low dose (ULD) CT scan was performed immediately after standard-dose clinical scans were performed on 13 patients. The ULD images were reconstructed with FBP and PICCS, while the standard scans used only FBP. All 63 of the high-contrast stones smaller than 2 millimeters were detected in standard dose with FBP and ULD using PICCS. The ULD scans reconstructed only with FBP detected 59.

“We have to do real clinical evaluations to establish the true benefit of this new technique,” Dr. Tang said. “We spent a lot of time performing clinical evaluations, but now we ask all radiologists to evaluate many images and make their judgment based on their own clinical observations.”

Dual-energy, Cone-Beam Imaging Maintain Image Quality

In other research presented at RSNA 2012, Wojciech Zbijewski, Ph.D., determined that image quality can also be maintained at reduced dose in dual-energy (DE) and cone-beam (CB) CT exams using advanced iterative decomposition methods. These methods are suitable for applications in musculoskeletal, vascular and interventional imaging, Dr. Zbijewski said.

Researchers were concerned with situations where contrast agents such as iodine must be discriminated from bone, Dr. Zbijewski said. “Dual-energy imaging has an easier way of discriminating them than through single-energy, but it is a very specific imaging task that you want to discriminate,” said Dr. Zbijewski, senior research scientist in the Department of Biomedical Engineering, Johns Hopkins University.

“That means potentially you can actually help yourself in terms of reducing the dose by tailoring your reconstruction methods toward just discriminating tissues and not necessarily being concerned about some other aspect of image quality,” he added.

The group found that they could use dual-energy imaging at doses around 3 to 6 mGy. By using the optimized reconstruction methods, they maintained 95 percent discrimination for contrast for 5 milligrams per meter of iodine. Dr. Zbijewski noted the potential for low-dose DE-CBCT will directly benefit iodine-enhanced arthrography.