Media stories about patients being overexposed during CT scans, estimated risks of cancer incidence from CT scans and the increased use of CT imaging has generated considerable attention on the topic of medical radiation exposure. New techniques and technology, however, have proven successful in reducing the standard patient dose by 50 to 90 percent. Now, the only impediment to lower dose is for these largely software-based techniques to become a routine part of clinical practice.
Prospective, retrospective
Smith-Bindman et al in the December 2009 Archives of Internal Medicine concluded that the increasing number of CT scans in the U.S. (70 million in 2007) could potentially lead to “15,000 excess deaths” as a result of cancer. They found that the average radiation dose emitted to patients undergoing coronary CT angiography (CCTA) was 22 mSv. However, new dose-lowering techniques can reverse this trend, as they result in exposures of 1 to 5 mSv.
Prospective EKG gating, tube current modulation and scanners with faster temporal resolution deliver much lower radiation levels without suffering degradation in image quality. Prospective gating, also known as step-and-shoot, involves the x-ray tube turning off for a brief part of the cardiac cycle, resulting in no overlapping images. This technique can reduce dose by 70 to 80 percent, says James P. Earls, MD, director of cardiovascular CT and MRI at Fairfax Radiological Consultants in Virginia. Conversely, retrospective gating exposes the heart to four to five overlapping regions of x-rays, which considerably increases a patient’s radiation exposure.
“Prospective gating is a big deal,” says Tony DeFrance, MD, clinical associate professor at Stanford University Medical School in Stanford, Calif., and director of CVCTA Education, a cardiac CT training center, in San Francisco. His facility utilizes the Aquilion One (Toshiba America Medical Systems) 320-slice CT scanner, which acquires images in roughly 300 msec in a single rotation, offering 40 to 50 percent less dose than a 64-slice CT exam at about 3 to 5 mSv. At Stanford, prospective gating is used in 85 percent of CT exams.
At the Medical University of South Carolina (MUSC) in Charleston, S.C., imagers carefully adjust tube settings depending on a patient’s body type, says U. Joseph Schoepf, MD, director of cardiovascular imaging. Patients with a body mass index (BMI) of 25 or less with a moderate heart rate receive 20 percent less kilovolts (100 kV) than the normal tube output of 120 kV. For pediatric patients and smaller adults, tube currents can be reduced to 80 kV, resulting in a “substantial radiation dose savings.”
In an effort to reduce dose at the Lenox Hill Heart and Vascular Institute in New York City, technologists utilize prospective gating on its 256-slice scanners (Brilliance iCT, Philips Healthcare). Exams take three to five seconds, emit a radiation dose of 2.5 to 5 mSv and yield a 75 percent reduction in dose, says Harvey Hecht, MD, director of cardiovascular CT.
| FDA Steps Up Its Scrutiny of CT Radiation Dose |
In March, the FDA invited stakeholders to educate the agency regarding CT imaging radiation dose. While speakers expressed support for the FDA’s focus on radiation exposure, they raised a range of concerns, including the timing and coordination of implementing changes that the FDA might impose and who would be accountable to whom. Sean M. Boyd, from the FDA’s Center for Devices and Radiological Health, who moderated the meeting, contends that a national consolidation of information is appropriate, but it is “not clearly decided who should maintain and manage it.” Also, ensuring that facilities use such data to improve internal processes lies outside FDA’s regulatory scope, he says. As a result, figuring out accountability for new stipulations remains a work in progress. Cardiovascular Business asked Boyd to discuss the recent meeting and the FDA’s increased interest in medical radiation exposure. Cardiovascular Business: What is the FDA’s overall impression of the meeting?Sean M. Boyd: Our overall impression of the meeting was very positive. We heard a lot of great ideas and feedback on how FDA and industry can improve the safe use of medical imaging equipment through improved product features, better operator training and improved quality assurance. CVB: What has the agency learned from the meeting? SB: We learned that there are many groups with similar goals that are interested in working toward reducing medical imaging exposure. No single group has the answer or can accomplish the work on its own, and the FDA believes its role is to provide leadership and coordination of the varied efforts. Specifically, participants confirmed that improved alerts and access controls are needed in equipment. We have a direct role here in working with industry to improve equipment safety. We are encouraged to see that industry is proactively addressing some important issues through its Dose Check initiative (alerts) and efforts to standardize capture of equipment settings and dose information using DICOM structured reports. Another important federal partnership will be between the FDA and CMS, in order to influence training and quality assurance practices within the clinical community. CVB: What are some challenges for a workable solution? What is a reasonable timeline for this to be clinically routine? SB: Challenges are primarily coordinating the many activities that will promote patient protection. The timeline that FDA has control over will be to define requirements for medical imaging manufacturers to improve equipment safety. Over the coming months, we will develop guidance for medical imaging premarket submissions. Influencing clinical practice will be a more difficult task. With this, we need to promote efforts of the various professional organizations that will improve provider understanding of medical imaging dose to ensure exams are justified and dose is optimized. CVB: What is driving the momentum in this area? SB: Our understanding of medical imaging and its significant contribution to the U.S. population’s total exposure has changed as a result of the National Council on Radiation Protection and Measurements (NCRP) report*. This report, along with heightened awareness of the public—due to media coverage of adverse events—and a real interest among the medical imaging community to improve practice, will sustain efforts. * The NCRP Report No. 160, published March 2009, updates a 1987 publication. It estimates the total U.S. exposure to all sources of ionizing radiation has increased seven-fold since 1980. The increase was due mostly to the higher utilization of CT and nuclear medicine. These two imaging modalities alone contributed 36 percent of the total radiation exposure and 75 percent of the medical radiation exposure of the U.S. population. The report does not attempt to quantify the associated health risks nor specify the actions that should be taken in light of these data. |
Iterative reconstruction
Low-dose techniques carry an inherent risk for noisy images. Researchers have been testing various image reconstruction techniques that are better able to filter out noise in low-dose imaging. With conventional filtered back projection, dose must be increased to lower the image noise and achieve better image quality. In iterative reconstruction, the computer makes assumptions about image data, continually comparing its assumptions to the raw data until there is a match. With this process, there is no need to increase dose. In fact, dose can be lowered significantly, while maintaining image quality.
The ERASIR I trial tested GE Healthcare’s Adaptive Statistical Iterative Reconstruction (ASIR) technique in three arms: filtered back projection alone, filtered back projection with ASIR and no other dose-reduction techniques (pre-protocol) and ASIR with other dose-reduction techniques such as a lower kVp (100), reduced tube current modulation based on BMI and minimal use of padding (post-protocol). The multicenter trial involved more than 1,100 patients being imaged on a LightSpeed VCT 64-slice CT scanner.
The mean effective radiation dose was 3.8 mSv for all filtered back projection studies, 2.6 mSv for ASIR pre-protocol and 1.3 mSv for ASIR post-protocol. The use of prospective gating was similar across the three arms, yet the use of ASIR was associated with a significant reduction in tube current (mA). “Despite the reduced radiation dose, image interpretability and signal-to-noise ratio of post-protocol ASIR and filtered back projection were similar,” says Earls, co-principal investigator. He adds that the low dose was applicable across all patients.
At Fairfax Radiology Consultants, Earls and colleagues use the ASIR technique in combination with other dose-reduction methods during 92 percent of CT exams on six outpatient scanners. The facility tracks dose reductions of 25 to 40 percent during each patient CCTA exam, and for patients with a BMI of 30 or less, tube current modulation is utilized, further reducing dose.
At last year’s Radiological Society of North America (RSNA) meeting, Siemens Healthcare unveiled its iterative reconstruction in image space (IRIS) technique. Schoepf and colleagues at MUSC use the software and say it decreases artifacts and lessens dose by 60 percent. “With the use of IRIS, you can reduce scatter so that coronary artery stents and dense calcifications actually look much more their real size,” Schoepf says. IRIS also has specific dose reduction benefits for pediatrics and the obese.
Philips intends to have its iterative reconstruction product (iDose) ready to ship in the second half of 2010, while Toshiba is expected to release its version—adaptive iterative dose reconstruction (AIDR)—in the near future. In addition, GE announced at the 2009 RSNA meeting its foray into the next-generation of iterative reconstruction—model-based iterative reconstruction (MBIR). GE also is sponsoring a multicenter research trial to assess the clinical efficacy of the technique as a method to improve image quality with lower radiation dose.
Tracking dose
Dose-reduction techniques, however, need to be employed by technologists or they are of no use. Researchers in the PROTECTION-1 trial found that dose-reduction techniques work, but the techniques are used infrequently (JAMA 2009;301:500-507). They also found discrepancies of radiation dose between facilities, and they defined the reduction of kV levels and minimizing scan field as the two most important dose-reduction techniques. The authors recommended improved education of physicians and technologists.
To better track the effects of dose, the National Institutes of Health (NIH) Clinical Center is incorporating radiation dose exposure reports into the EMR, which it hopes will lead to an accurate assessment of whether any cancer risk is associated with low-dose radiation exposure from imaging tests. Radiology and nuclear medicine at the NIH Clinical Center have developed a radiation reporting policy that will be instituted in cooperation with major equipment vendors, beginning with exposures from CT and PET/CT.
Vendors that sell imaging equipment to radiology and imaging sciences at the NIH Clinical Center will be required to provide a means for radiation dose exposure to be recorded in the EMR. In addition, the radiology department at NIH will require that vendors ensure that radiation exposure can be tracked by the patient in his or her own personal health record.
It seems that the cardiac imaging community is at a crossroads. The technology and techniques to reduce patient radiation exposure exist, but the routine use of these techniques is lacking. Early adopters and academic centers are prime users of low-dose techniques, but these have to filter out into the community at large. As medical radiation continues to be scrutinized by private payors and the U.S. government, imagers will no doubt step up their understanding and use of radiation reducing protocols.