The Balancing Act PET/CT: Ensuring Image Quality, Controlling Dose

Creating excellent images while keeping the radiation dose at a safe level for the patient always has been a challenge of medical imaging. And since PET/CT essentially marries two traditionally distinct fields, without established radiation guidelines they may not always be used in the safest and most effective manner. Therefore, new methods for imaging cancer patients without overexposing them to radiation are being developed.

A study from the University of Southern California recently found that PET/CT imaging of lymphoma patients is an effective method to evaluate response to radiation therapy, and may help patients avoid unnecessary follow-up treatments. The combined imaging modality “is more accurate in defining a complete response to radio-immunotherapy treatment,” say the researchers. “Specifically, PET/CT is able to correctly label inactive scar masses, which are sometimes mistaken for active cancer when CT is used alone.”

For this reason and with more proven clinical results in the oncology arena expected soon, Paul Shreve, MD, of the PET Medical Imaging Center in Grand Rapids, Mich., organized the recent Best Practices in PET/CT conference in Sonoma, Calif. The meeting was intended to help create consensus on PET/CT protocols for body oncology imaging.

Vast variability

“There is still a great deal of variability in how PET/CT is understood as a modality and how the scans are performed in different practice settings,” Shreve says. Interpreting physicians may have nuclear medicine, general radiology, or organ-system subspecialty radiology backgrounds, and may practice in academic or large or small private practice settings. “By bringing together interpreting physicians of all these different backgrounds, we were able to learn from each other and find many common concerns.”

Conference attendees, Shreve says, agreed that reducing patient radiation dose per scan involves both reducing the PET radiopharmaceutical dose and reducing the redundancy of CT scans performed. “We shouldn’t do two CT scans when we do PET/CT, unless it generates useful information.” Unfortunately, patients are still undergoing non-contrast CT followed by contrast CT. “There is no reason to do that,” he says. “Along the same lines, patients should not have a CT followed by a PET/CT as separate examinations. When at all possible, they should get one optimized diagnostic CT scan as part of their PET/CT procedure. That is part of an overall strategy to reduce unnecessary radiation exposure.”

Another important question is how much of the body needs to be scanned. The CPT code-driven “skull base to mid thighs” axial coverage doesn’t make sense for some cancers, depending upon the nature and location of the primary cancer.  Scanning so much of the body routinely and uniformly “adds unnecessary radiation exposure, problems with false positive findings on both the PET and CT images, and additional interpretive workload for a very small yield of true positive findings,” Shreve says.

Debating dose

David Townsend, PhD, professor of Medicine and Radiology, and director of the Cancer Imaging and Tracer Development Program at the University of Tennessee in Knoxville, was a speaker at the Sonoma conference. He points out that the combination of PET/CT highlighted the dose issue by requiring all patients for a PET scan to also undergo a CT scan.

Overall, the PET whole-body dose is about 7 mSv for a 10 mCi (370 MBq) injection of FDG, whereas a clinical CT scan can be up to three times that—depending on the system used. “So, the dose from CT is the big concern,” he says. “Many facilities reduce the CT dose by running the x-ray tube at low current.” As a result, some clinicians are tempted to call these scans non-diagnostic, since they’re not done at the very top beam currents, but that’s not correct, Townsend says. “These scans are still diagnostic.”

This is a tricky issue, he admits, because of the potential implications. “There’s no question that the majority of radiation dose an individual receives over a lifetime will come during radiation procedures in hospitals. The issue is, do some patients get scans they don’t need?  This isn’t easy to answer.”

The litigious environment in medicine in the United States today is partly responsible, Townsend says. A physician might not believe a CT scan will be helpful, but feels safer ordering the scan to avoid litigation in the case that the scan reveals something unexpected.

Another issue is reimbursement. If a patient undergoes a CT scan which then indicates a need for a PET scan, the patient might be referred for what is now a PET/CT scan. The second CT scan won’t be reimbursed since the patient just had one, so instead a low-dose (non-reimbursable) CT scan is acquired for the PET/CT. That helps to reduce the radiation exposure, but Townsend would rather see referring physicians send patients for clinical PET /CT in the first place and avoid the double CT.

Education and standardization

Wider use of PET/CT requires greater education, Townsend says. While some physicians with some good experience under their belts can start to feel that PET/CT can do anything, other physicians actively resist it. Townsend has heard of some cases in which a physician refused to refer patients for PET/CT even though it would have helped with their cancer staging and workup. A total of 21 million SPECT scans were performed in 2006 in the United States, compared with 1.5 million PET and PET/CT studies. “Clearly, there are still many cases where patients don’t get a PET scan that would be beneficial.”

Many cancer patients receive radiation therapy that involves large doses, so the additional CT dose exposure is not really a factor. “It shouldn’t be forgotten that these are patients with serious disease,” Townsend says. “The risk-benefit ratio is then favorable for having a PET/CT scan. For those [patients] receiving very high radiation for treatment, additional imaging scans won’t contribute all that much to their overall exposure.”

A major surprise from the Sonoma conference, says Townsend, was the lack of consensus from the different centers regarding the PET/CT protocols: for example, how much radiotracer to inject and how long to wait between injecting the tracer and scanning the patient. Protocols are hard to standardize, he says, because imaging technology has been changing so rapidly. Newer protocols don’t necessarily apply to those imaging centers with older equipment.

For example, Townsend says his facility has a recent system that, for a whole-body scan, requires an acquisition at only four bed positions for 1 to 2 minutes of scan time, compared with seven bed positions and 3 to 4 minutes with some of the older PET/CT scanners. Some standardization, however, is critical, he says. Clinical trials that pool data from different centers will give meaningless results if the data are from different generations of scanners and involve dissimilar protocols.

In the future…

Shreve says many conclusions reached during the Sonoma conference will be published after August in Seminars in Ultrasound, CT and MR.

Meanwhile, “we are not at the stage to mandate to the community,” Townsend says. “We are still trying to establish guidelines and best practices. With the huge explosion relative to what PET used to be, a lot of places, to some extent, don’t know what they’re doing. You can find appalling practices. Meanwhile, we see patients daily who don’t have access to PET/CT technology who could benefit from it.”

In the long-term, Shreve expects to see more and more applications of radiation therapy planning and better understanding of how to use PET in therapy monitoring in the course of chemotherapy. Although “FDG will still be our workhorse,” he says, in the future a small amount of PET/CT imaging may be done with different tracers. He also anticipates the possibility of growth in cardiac PET/CT.