A Clearer Clinical Picture: Exploring Shifts in Use, Cost of Diagnostic Imaging
New technology entering the medical field often portends a number of positive developments: Improved diagnosis and management of a disease; faster or more efficient service at the point of care; and a more affordable option for patients. However, one drawback that clinicians can usually count on is a messy transition from the old technology to the new, a scenario currently taking place as rheumatology moves away from conventional X-ray to more advanced imaging techniques, including ultrasound and MRI.
Eugene Kissin, MD, associate professor of medicine and fellowship program director in the division of rheumatology at Boston University Medical Center, summed up the slow-moving nature of this transition. “I would not say that we are moving away from X-rays, but rather adding ultrasound and MRI information to complement what we learn from radiographs,” he said in an interview with Healio Rheumatology. “Both MRI and ultrasound use in clinical decision-making require additional knowledge of when these imaging modalities are most helpful in rheumatic diseases.”
While the newer diagnostic approaches have largely been embraced in the clinic and community for their improved ability to display important signs of disease progression, insurance carriers and regulatory bodies have been loath to throw their full weight behind them. A key reason for hesitancy is cost, according to Veena K. Ranganath, MD, MS, RhMSUS, of the department of rheumatology at Ronald Reagan UCLA Medical Center. “In terms of cost/benefit ratio, there is no doubt that the value of novel imaging techniques is there,” she said. “However, the financial barriers include the cost of the machines, the cost of training people to use them, and the fact that reimbursements have decreased significantly.”
Another issue to consider is that more frequent use of imaging increases the need for communication between radiology and rheumatology. Specifically, an uptick in imaging use fosters disagreements regarding who has what responsibilities, not only in diagnosing and monitoring the patients, but billing them as well.
Amanda E. Nelson, MD, MSCR, associate professor of medicine in the division of rheumatology, allergy, and immunology at the University of North Carolina at Chapel Hill and faculty member at the Thurston Arthritis Research Center, addressed the clinical reality that both rheumatologists and radiologists must consider. “No modality is perfect for all applications, and selection of an imaging study should be guided by an understanding of these strengths and limitations,” she said.
Underscoring all of this is that research surrounding novel imaging techniques in the rheumatology field is insufficient. Data are emerging, but those data range from surveys about musculoskeletal ultrasound (MSUS) training programs to disease-specific studies that only paint part of the clinical picture. The rheumatology and radiology communities await the type of comparative randomized clinical trials that can offer conclusive answers regarding the efficacy of each of these procedures across the myriad disease states under the rheumatology umbrella. In the meantime, clinicians must scuttle along as best they can with whatever guidance is available.
Setting the Stage
As is frequently the case with advances in medicine, the European Union has adopted novel imaging techniques more readily than the United States. In 2013, Colebatch and colleagues published the EULAR recommendations for the use of MSUS, MRI and other imaging tools in the diagnosis and management of rheumatoid arthritis. Inflammation observed with ultrasound or MRI can help predict progression, according to one of the 10 key recommendations from the authors. They also noted that ultrasound and MRI are better than clinical examination in detecting joint inflammation, and that MRI of bone edema may predict radiograph progression of RA.
“More frequent use of ultrasound and MRI allows for earlier detection of inflammatory arthritis given the increased sensitivity of these modalities to detect erosions compared to X-rays and to detect synovitis/enthesitis compared to clinical examination,” Sobia Hassan, MD, assistant professor in the division of rheumatology at Rush University Medical Center, told Healio Rheumatology in an interview.
With regard to ultrasound, Hassan noted the “bedside availability and portability” as valuable assets. “In addition, multiple joints can be imaged in one sitting — unlike MRI — and dynamic maneuvers can be utilized to provide more diagnostic information.”
Another bonus is that ultrasound-guided injections allow for more accurate delivery of medication and the ability to perform diagnostic aspiration in otherwise challenging scenarios such as in deep joints or in obese patients, according to Hassan. “Ultrasound-guided injections have been shown to be associated with less procedural pain,” she said.
It is important to note that although the authors of the EULAR recommendations suggest that ultrasound and MRI may ultimately be more effective than both clinical examination and other types of imaging, they do recommend conventional radiography in certain situations. For example, X-rays “of the hands and feet should be used as the initial imaging technique to detect damage,” they wrote. “However, ultrasound and/or MRI should be considered if conventional radiographs do not show damage and may be used to detect damage at an earlier time point (especially in early RA).”
A Look at Clinical Data
In a recent study from Bauer and colleagues published in BMC Musculoskeletal Disorders, researchers assessed joint-specific associations between power Doppler ultrasound (PDUS) and clinical swollen joints in patients with RA from various BMI groups, including obese individuals. Results showed that while demographics and clinically-determined disease activity were similar among BMI groups, significant differences were reported for PDUS scores (P = .02). Moreover, PDUS was associated with a significantly lower positive predictive value of swollen joints among patients across a range of BMIs (P = .02).
“This study suggests that in an obese RA patient, a clinically assessed [swollen joint] is less likely to represent true synovitis (as measured by PDUS),” the researchers concluded.
“More frequent use of ultrasound can help increase accuracy of diagnosis in many rheumatic diseases, and accuracy of joint and soft tissue injections,” Kissin said. “However, greater use of ultrasound also increases the risk of misattribution for ultrasound findings to clinical symptoms, mistakes in ultrasound interpretation leading to faulty diagnostic conclusions, and risk of increasing the cost of service without increasing the quality of service if ultrasound use is not prudent.”
Yet another caveat about ultrasound is that it is operator-dependent and labor intensive, according to Ranganath. “You need to learn how to get the right images,” she said. “You have to learn where the bony landmarks are, and what the normal anatomy looks like.”
George R. Matcuk Jr., MD, associate professor of clinical radiology and fellowship director of musculoskeletal radiology at the University of Southern California Keck School of Medicine, believes that the rapid advancement of technology is improving both diagnosis and follow-up while making advanced imaging a more integral part of the care of a patient with rheumatological conditions.
“For example, with ultrasound, there is gray scale for evaluation of the anatomy and identifying areas of synovitis, tenosynovitis, and even erosion,” he said. “Color, power, and new Doppler imaging techniques allow for evaluation of hyperemia associated with active synovitis,” he said. “Newer technologies are emerging, such as contrast-enhanced ultrasound with microbubble agents that can be even more sensitive and quantitative, in which you can see curve enhancement patterns.”
Although Bauer and colleagues looked at nine different joints in their patient population, Ranganath stressed that ultrasound technology can offer several views of numerous joints. “Once an operator is able to get the right views of each joint, the value of this information is immeasurable,” she said. “You can pick up on diagnoses you would not have been able to with just a physical exam, and show them to the patient in the room.”
There is one key advantage to this type of real-time information, according to Ranganath. “Patient buy-in is huge,” she said. “When they can see what is happening right there in the exam room, they are more likely to go along with the diagnosis and treatment plan.”
In a literature review of MRI and MSUS in RA, Matthew and colleagues found that compositional MRI may have utility in understanding the association between synovitis and cartilage proteoglycan loss. MRI may also be a valid predictor of clinical and radiographic damage endpoints, according to findings published in Current Opinion in Rheumatology.
“The most important thing to consider is that by the time you are able to definitively see erosive changes on X-ray, the patient most likely has irreversible disease,” Matcuk said. “The advantage of ultrasound and MRI is that these 3-D modalities have higher resolution that allows you to see inflammatory arthritis before it progresses to erosive disease.”
These modalities also allow clinicians to see active synovitis, according to Matcuk. “When you see this, you can get them on a DMARD to prevent irreversible damage,” he said. “This is huge in terms of changing outcomes for patients.”
However, it is important to note that beyond cost, MRI is not without its drawbacks. Nelson cited gadolinium administration for MRI, or radiation with computed tomography and positron emission tomography. “For MRI in particular, implants, pacemakers, claustrophobia, discomfort and body size can be problematic,” she said. “While these are not limitations for ultrasound, it is limited in its use for deeper structures, such as subchondral bone, internal joint structures like the ACL and deep vascular tissues.”
The Importance of Training
The obvious way to neutralize so many unanswered questions is to increase and improve training with imaging technology. In their study in Arthritis Care & Research, Torralba and colleagues sent a needs assessment survey about MSUS to 113 rheumatology fellowship program directors, along with a curriculum survey that was sent to the lead faculty. Results showed that 94% of programs taught MSUS, with 41% having a curriculum in this area. Further data showed that 95.3% of program directors wanted formal curricular adoption of MSUS, while 65.7% preferred it to be optional.
“Rheumatology fellows are eager to learn about musculoskeletal ultrasound,” Ranganath said. “But right now, it is not a strict requirement for their rheumatology training. They have so many obligations just for their rheumatology training alone, it is often difficult to find time to learn about imaging in depth.”
Other findings from Torralba and colleagues showed that lack of divisional interest (P = .046) and fellow interest (P = .012) were reported in programs without a formal curriculum for MSUS.
Ranganath suggested that even in training programs where the fellows want to learn, cost becomes a factor. “These machines are expensive,” she said. “Then the institution has to generate resources to support the fellowship program, Additionally, the instructors should be qualified in musculoskeletal ultrasound to teach and to bill. This is a challenge even in large institutions.”
In their study published in Joint Bone Spine, Forien and colleagues explored ultrasound in RA with a targeted purpose in mind: They suggested that standardization of scoring and settings procedures is necessary before being universally accepted as a marker of disease activity.
For Matcuk, standardization of the routine should be the bedrock of training programs. “Ultrasound technicians are excellent with the imaging techniques, but they are still not as familiar with complex anatomy associated with musculoskeletal ultrasound,” he said. “To build a successful musculoskeletal ultrasound practice, it requires the musculoskeletal radiologist to be in there with the patient and help train the ultrasound technologists on the anatomy and pathological findings to look for to really make an accurate diagnosis. It is also very important to have excellent communication between physicians in radiology and rheumatology to combine the imaging and clinical information to provide the best care for each patient. This has to be factored in when talking about training and its relationship to the cost of examination.”
A further point on standardization is that the level of training between institutions varies greatly, according to Hassan. “For someone like me who has experience with using ultrasound, I would appreciate further efforts to standardize and simplify imaging protocols for the various rheumatic diseases,” she said.
For Ranganath, familiarity with an ultrasound machine is one thing; the advanced expertise of normal and abnormal anatomy to make an accurate diagnosis is another. “You need to spend a lot of time with an ultrasound machine to really understand how it works,” she said. “Both rheumatologists and radiologists are essentially learning nuances in a whole new area of research.”
Programs like the Ultrasound School of North American Rheumatologists (USSONAR) are beginning to proliferate but are far from common, even in big academic medical centers, according to Ranganath. “Even after completing a program like that, which is pretty intense, you still need a lot more experience to feel comfortable,” she said.
Communication Between Specialties
Teh and colleagues wrote a paper published in Radiologic Clinics of North America, which outlined what rheumatologists are looking for in the results of an imaging analysis. They suggested that more clearly defining rheumatology needs may help communication with radiology. It is their opinion that rheumatologists are looking for structural changes and damages within the joints, synovitis, tenosynovitis, bone marrow edema and bone erosions; therefore, understanding which regions should be imaged is critical, as is understanding diagnostic criteria for RA, osteoarthritis and other disease states. Rheumatologists are also using imaging to monitor for remission or determine signs of inflammatory activity or structural progression.
“One of the gaps I see, which may or may not be generalizable, is that of the question being asked,” Nelson said. “Sometimes the question was one of inflammation, and the report focuses only on damage, or vice versa, or a detailed comparison to prior imaging is not made or reported, or the wrong structure is focused on.”
She cited the example of reporting on the intervertebral discs rather than the sacroiliac joints. “Some of the communication issues between the ordering clinician and the one reading the study are related to the electronic health record,” she added. “There is also the disconnect between the order and the procedure performed, and/or between the ordering and interpreting provider.”
With an increasing number of nonradiologists performing MSUS, some radiologists may feel that rheumatologists are encroaching on their territory and taking away potential business, according to Hassan. “This may also have implications with regards to storage of ultrasound images, as the hospital’s central picture archiving and communication system used by radiology may not be made accessible to rheumatologists,” she said.
Matcuk tried to focus on the most obvious positive that may result from two specialties being involved. “They might see something we don’t see, or vice versa,” he said.
The hard reality of limited clinical visit time is an important consideration for Ranganath. “Rheumatologists hardly have enough time to do everything they need to do in a follow-up 20-minute visit,” she said. “Trying to add an ultrasound to that adds a whole new set of challenges. You have to pull the machine into the exam room, link it with the system, and document the entire report once you are done with it. Rheumatologists may not have the time to scan even though it is of value.”
“Conversely, radiology is busy, as well, with plenty of work to do,” she added. “They are often accepting of us performing ultrasounds and billing. However, this is institution dependent.”
An additional factor is that the level of communication between the specialties can vary greatly between institutions and practices. “In many academic settings, there are regular conferences between the two departments where cases involving advanced imaging can be discussed in greater detail,” Hassan said. “Although the clinical details of the patients are discussed in detail during these conferences, many radiologists complain that pertinent clinical information is often scant or left out when referring physicians place initial imaging orders.”
Hassan believes that access to greater information regarding diagnosis and treatment responses would allow radiologists to provide a more methodical approach to imaging analysis and reporting. “At the national level, there has been increased collaborative efforts between the radiology and rheumatology governing bodies,” she said. “The American Institute of Ultrasound in Medicine now recognizes the American College of Rheumatology’s RhMSUS certification and accreditation programs. This helps to set important standards and improve the overall quality of diagnostic ultrasound and ultrasound-guided invasive procedures amongst all specialties.”
Both ultrasound and MRI are considerably more expensive than conventional X-ray, and it is uncertain whether that will change any time soon. An associated factor is that insurance carriers are not yet 100% on board with covering ultrasound or MRI in the rheumatology space. Further complicating the cost issue is a change in the CMS 2019 proposed Medicare Physician Fee Schedule that could reduce reimbursements for MSUS by as much as 90% by 2029.
“For me, the biggest obstacle is probably the cost of some of the newer medications,” Matcuk said, suggesting that it is important to look at diagnostic imaging not in a vacuum, but as part of a larger treatment continuum. “As treatments get better and better, the cost of drugs has gone up. It makes a lot of sense to try some of the less expensive medicines first, but if the patient is not doing well on those, an MRI or ultrasound can help make a decision to potentially intervene with newer but more expensive medication.”
Kissin outlined another layer of financial complication. “The cost/benefit ratio for diagnostic imaging depends on many factors and is difficult to calculate as the same test could cost vastly different amounts depending on the patient’s insurance status,” he said. “For example, I have seen the same musculoskeletal ultrasound evaluation cost as little as $30 or as much as $2,800.”
The benefit also varies widely depending on how the imaging technology is applied, according to Kissin. “Based on data from CMS for the year of 2016, the number of comprehensive ultrasound examinations of the musculoskeletal system billed to Medicare by rheumatologists in the United States ranged from 11 to 8,463,” he said. “I suspect that these numbers would affect the cost/benefit ratio.”
Nelson zeroed in on specific clinical applications as a function of cost. “For things like large vessel vasculitis, where we previously had little more than a sedimentation rate to go on, we are now able to get much more detailed information from an imaging study that can guide therapy,” she said. “On the other hand, there is essentially no indication to get an MRI for clinically diagnosed knee OA, although this is frequently done. Ultrasound can be helpful for evaluation of subclinical synovitis and for monitoring treatment response, but the indications and intervals are not well-established.”
The most problematic issue with insurance coverage of imaging is that it is in “constant flux,” according to Nelson. “While X-rays are essentially universally covered, there is a lot of variability in what prior authorization, peer-to-peer conversations, and other approval issues need to be dealt with for more advanced imaging studies,” she said. “In some cases, it is impossible to get an indicated study paid for, such as a PET scan for nonmalignancy indications, while other less essential studies — like the MRI for knee OA — may be covered based on seemingly arbitrary rules at the insurer level.”
Regardless of how any of these issues evolve, one thing is certain, according to Ranganath: “Ultrasound and MRI are not going away,” she said. “Rheumatologists will continue to use them, particularly ultrasound, and bill for them, despite not being appropriately compensated.”
With that in mind, she suggested that more research into all facets of diagnostic imaging in the rheumatology space — efficacy, cost/benefit, workflow issues, and communication concerns, among others — will go a long way in answering the questions from both specialties. “This is the reason I do the research I do,” she said. “I see an ultrasound machine as an extension of a physical exam. Now, we need to generate enough research to prove the value of it beyond a shadow of a doubt.” – by Rob Volansky
- Bauer EM, et al. BMC Musculoskeletal Disorders. 2017;doi:10.1186/s12891-017-1406-7.
- Colebatch AN, et al. Ann Rheum Dis. 2013; doi10.1136/annrheumdis-2012-203158.
- Caporali R, et al. Ann Rheum Dis. 2018;doi: 10.1136/annrheumdis-2017-211458.
- Forien M, et al. Joint Bone Spine. 2017;doi:10.1016/j.jbspin.2016.08.003.
- Matthew AJ, et al. Curr Opin Rheumatol. 2016;doi: 10.1097/BOR.0000000000000282.
- Teh J, Ostergaard M. Radiol Clin North Am. 2017;doi:10.1016/j.rcl.2017.04.001.
- Torralba KD, et al. Arthritis Care Res (Hoboken). 2017;doi:10.1002/acr.23336.
- For more information:
- Sobia Hassan, MD, can be reached at 1611 West Harrison St., Suite 510, Chicago, IL 60612; email: firstname.lastname@example.org.
- Eugene Kissin, MD, can be reached at 72 East Concord St., Evans-506, Boston, MA 02118; email: email@example.com.
- George R. Matcuk, MD, can be reached at 1500 San Pablo St., 2nd Floor Imaging, Los Angeles, CA 90033; email: firstname.lastname@example.org.
- Amanda E. Nelson, MD, MSCR, can be reached at 3300 Thurston Bldg., CB #7280, Chapel Hill, NC 27599-7280; email: email@example.com.
- Veena K. Ranganath, MD, MS, RhMSUS, can be reached at 200 Medical Plaza Driveway, Los Angeles, CA 90095; email: firstname.lastname@example.org.
Disclosure: Hassan, Kissin, Matcuk, Nelson and Ranganath report no relevant financial disclosures.