The American College of Radiology (ACR) and the Society of Nuclear Medicine and Molecular Imaging (SNMMI) have released the first in a series of requirements for physicians and clinicians interested in gaining proficiency in reading PET/MRI scans.
The first document, published online March 5 in the Journal of Nuclear Medicine, provides guidance on how much training and firsthand experience is needed to adequately read and interpret images from PET/MRI brain scans.
"The practice of PET/MRI requires a solid knowledge base of PET techniques and MRI techniques," the document states. "Although the ultimate goal is mastery of simultaneously acquired brain PET/MRI examinations, the physician's education, training, and experience should encompass dedicated brain PET [or] brain PET/CT as well as dedicated brain MRI."
Clinicians looking to achieve competency in PET/MRI will need to know about imaging protocols, contrast agents, and MRI sequences, as well as biologic and clinical parameters such as anatomy, physiology, normal variants, and disease states.
"We wanted to set some parameters -- whether it be a single individual or two individuals reading these studies -- to make sure there is competency in both modalities," said committee co-chair and ACR representative Dr. Rathan Subramaniam, PhD. "Since this marries both PET and MRI, unless someone has very good training in both modalities, it will be a tough task for one individual to report on both of them."
Credentialing guidelines
The initiative to develop competency criteria for PET/MRI began in June 2013, when ACR and SNMMI created a 12-member task force with six members from each organization.
The key recommendation in the document is that physicians and clinicians meet certain credentialing guidelines for the modality they are interpreting, be it PET, MRI, or both.
"We need to make sure that all the people reading these images, a single individual or two individuals together, are competent as a combination of both modalities. That is the most important part," explained Subramaniam, who is an associate professor of radiology, oncology, head and neck surgery, and health policy and management at Johns Hopkins University.
To view a table summarizing the initial and continuing competency criteria, click here. (Table provided with permission of the Journal of Nuclear Medicine.)
The competency criteria create three different categories of readers based on their expertise. There are both initial and continuing competency criteria for clinicians who are board-certified within two years of completing their residency or fellowship; individuals without board certification but with more than two years of a residency or fellowship; and those with board certification in another medical specialty.
"It sets different levels of expectations for those people, taking into consideration what they have done with their training," he said.
For example, the joint statement recommends that a board-certified clinician have experience with at least 30 FDG-PET and 300 MRI brain scans to reach initial competency levels. A noncertified clinician would need to interpret at least 50 FDG-PET and 500 MRI brain exams for initial competency. Finally, a person with expertise in another medical specialty would need to interpret at least 100 FDG-PET and 750 MRI brain exams.
There are similarly scaled requirements for continuing competency for each of those three reader categories.
The committee also noted that there is more to determining competency than just the number of scans a person has interpreted.
"Simple numerical criteria are not an optimal measure of competency," the authors wrote. "Documentation of competency by the use of objective, outcome-based tools related to clinical experience is preferable."
The statement also recommends that clinicians who are proficient in either PET or MRI should combine their expertise in PET/MRI interpretations. The two physicians should reach a consensus on the final interpretation and issue a joint report or two separate correlative PET and MRI reports.
Essentially, when interpreting PET/MR images, two heads are better than one.
"PET/MRI not only has multiple [MRI] sequences, but also the sequences change and are fine-tuned for each body part," Subramaniam added. "For example, the brain has different [MRI] sequences than the head and neck; there is different fine-tuning in the chest; and there would be a different approach with the liver. So our feeling is that if someone is not trained in both PET and MRI, we want to ensure that at least one individual will have knowledge about different sequences that could be applied to the different body parts with MRI."
From the top down
So why address PET/MRI of the brain first?
"One of the biggest benefits of PET/MRI has been with the brain, partly because MRI provides excellent soft-tissue resolution," Subramaniam said. "There is a huge need because the CT part of the PET/CT cannot provide those attributes."
The board will also craft competency guidelines for other parts of the body, moving from the brain down to the torso.
"The brain is the first one in this series," he said. "We are just getting started on the head and neck right now."
The PET/MRI head-and-neck competency document will take six to nine months to complete, he predicted. This is less time than it took to complete the PET/MRI brain guidelines because the committee now has a prototype framework from which to work. The final draft will then be reviewed and eventually approved by the boards at ACR and SNMMI.
The committee plans to proceed to the chest, liver, and other body parts and organs that would benefit from PET/MRI. Pediatric PET/MRI will most likely have its own competency credentialing document, Subramaniam noted.
