Collaborative Effort: Where We Stand With Imaging Modalities in Rheumatologic Diseases

The rheumatology community has relied on radiographs as the primary imaging modality for several decades and for good reason. It is reliable, easy to use, familiar to patients and relatively cost-effective. However, advances in technology have brought a host of modalities into the fold, from previously existing approaches — such as MRI and PET — to those with just a decade or so of use, like musculoskeletal ultrasound.

“X-ray is still our workhorse in radiology,” John A. Skinner, MD, assistant professor of Radiology at the Mayo Clinic in Rochester, Minn., told Healio Rheumatology. “The vast majority of us like to start with a radiograph. It gives us a good overview of the lay of the land and often can answer the question.”

That said, despite the relatively short time it has been in use, many clinicians have found musculoskeletal (MSK) ultrasound to be a strong contender to become a mainstay of imaging in rheumatology. However, because of a number of hurdles, cost and difficulty of use among them, that road could be a long one.

“We see an equal spread of MRI and radiography at the Mayo Clinic, with a moderate amount of CT,” Skinner said.

It is a matter of how sensitive a test is for Hareth Madhoun, DO, assistant professor in the Division of Rheumatology and Immunology at The Ohio State University Wexner Medical Center.

“X-rays are generally the preferred method of evaluating patients with joint pain. However, they are by no means the most sensitive way of evaluating patients with inflammatory arthritis.” he said. “If I have a patient with whom I strongly suspect rheumatoid arthritis (RA) from history, labs and physical exam, my preferred test is an X-ray,” he said. “X-rays can give you a wealth of information; such as whether there is any soft tissue swelling, joint space narrowing, erosions, etc. but it is not always black and white. Occasionally, we have patients whose physical exam and X-rays are non-revealing. In those instances, I may look to another modality, such as MRI or ultrasound, which is more sensitive in picking up inflammation in a joint.”

Healio Rheumatology spoke to sources to gain perspective about which approaches should be used in specific patient populations and clinical situations. Factors ranging from specificity and operator skill to cost and toxicity tend to muddy the picture. It is for this reason that Skinner stressed that a good place to start is an open dialogue between rheumatology and radiology to achieve the best possible outcomes.

Dialogue Between Specialties

“We try to educate our rheumatology colleagues in terms of radiology,” Skinner said. “In most practices, there is an ongoing dialogue between specialties, what we need and what we can do for each other. We have many discussions with rheumatology. On a quarterly basis, we go over cases together.”

It is difficult to account for all possible interventions in all possible patient groups with the guidelines, according to Skinner.

“Ultrasound is often not presented as an option,” he said. “There are questions about dual-energy CT. Different centers have different resources and different capabilities in terms of imaging options. Guidelines are never perfect.”

With this in mind, Skinner discussed questions frequently raised by rheumatologists.

“We discuss case patients, and a lot of times the questions are about the diagnosis,” he said. “Rheumatology is a challenging field, and it can be confusing in the places where imaging and the patient history overlap between multiple diagnoses.”

Skinner said that as a radiologist, he tries to think about what else can be done from an imaging standpoint to help make the diagnosis.

“I have to rely on my experience and the literature to come up with a diagnosis, or at least steer the rheumatologist toward the most likely diagnosis, but we often see mixed information,” he said. “Collaborative discussions help us to explore the possibilities and figure out exactly what we are looking at.”

He said each centers should develop site-appropriate protocols for better imaging of rheumatology disorders based on common indications and available technology.

“We try to get highest resolution we can without missing important anatomy,” Skinner said.

Emergence of MSK Ultrasound

Musculoskeletal ultrasound has emerged as an important way of achieving the dual goal of high resolution while gaining perspective on the whole anatomy of a joint.

Gurjit S. Kaeley MD, professor of Medicine, director of Musculoskeletal Ultrasound in the Division of Rheumatology; Sunita Dodani, MD, PhD, adjunct associate professor of Epidemiology, College of Medicine and College of Public Health and Health Professions at the University of Florida, and senior consultant and research affiliate at the Mayo Clinic, Jacksonville, Fla., and colleagues raised the question of whether this modality was innovation or overutilization in research presented at the American College of Rheumatology (ACR) meeting last year.

“What we tried to clarify was that although the rapid rise of utilization of interventional MSK ultrasound could reflect innovation in care, it is not clear if the addition of MSK ultrasound decreases costs over the long term,” they told Healio Rheumatology. “This is the subject of our future studies.”

They retrospectively investigated diagnostic and interventional use of this modality within the Medicare population during 2011 to 2013. Musculoskeletal ultrasound was used more commonly for interventional than diagnostic purposes by all providers except podiatry, according to the results. Interventional use was highest in orthopedics. For non-radiology groups, a cumulative growth of 205% was reported for interventional use of the modality. The growth was 273% for diagnostic purposes.

“Marked dissociation of interventional MSK ultrasound services from diagnostic services was seen in most MSK specialties,” the researchers concluded. “Lack of corresponding diagnostic billing could be due to bundling of diagnostic services or inability to acquire and bill diagnostic ultrasound. Although the rapid rise of utilization of interventional MSK ultrasound could reflect innovation in care, it is not clear if addition of MSK ultrasound guidance on a population level leads to better value care. Medicare reimbursement of interventional MSK ultrasound procedures has been reduced, which may lead to a better value proposition and moderated utilization.”

“Many American rheumatologists have incorporated MSK ultrasound into their clinical practice,” Veena K. Ranganath, MD, MS, associate clinical professor in the Division of Rheumatology at University of California, Los Angeles, said. “There is still much work to be done in how best to incorporate MSK ultrasound into a doctor’s busy practice, but I do think this is an imaging modality that most likely will see widespread use.”

“MSK ultrasound has been around for a decade or 2, but it has only been over the past 5 years or 6 years that we have seen a rise in its use,” she said. “We are seeing it in rheumatology, sports medicine and orthopedics.”

The use of this modality in rheumatology is different compared to the way it is being used in orthopedics or sports medicine, according to Madhoun.

“Ultrasound in rheumatology has many utilities. One of the primary reasons that I use in my practice is to look for inflammation within joints and tendons, synovitis, erosions, etc.,” he said. “Ultrasound is a more sensitive method of evaluating patients for synovitis. It can help us capture people who have subclinical disease that we cannot pick up on our routine physical exams.”

Madhoun suggested there are more advantages than disadvantages for this approach, but that the disadvantages are not insubstantial.

“The time to scan and how to interpret the scan are current obstacles,” Ranganath said.

Patients often end up getting an MRI in situations for which ultrasound would be preferred, according to Skinner.

“This is largely the result of an education gap for providers who simply do not think of ultrasound,” he said. “It is worth noting that in other countries, ultrasound is used more commonly because they have less MRI access.”

Cost also may fall into the disadvantage category, according to Madhoun.

“Most rheumatologists are facing a reimbursement issue when it comes to using ultrasound,” he said. “It depends on the patient’s insurance coverage.”

From the provider’s perspective, it is the initial investment in an ultrasound machine — $30,000 or $40,000, depending on the quality — that frequently is a deterrent, according to Madhoun.

“A lot of companies have started lending or leasing the machines to offset the initial investment,” he said.

Diagnosis vs Intervention

A compelling argument surrounding MSK ultrasound pertains to how it is being used. Namely, for diagnostic or interventional purposes. Kaeley and Dodani laid out the particulars.

“Rheumatologists are often interested as to whether arthritis is active,” they said. “One feature of ultrasound that gives insight to this is to determine how vascular the aberrant pannus is and hence infer disease activity. When ultrasound is used to examine structures and answer clinical questions, it is being used for diagnostic purposes. Ultrasound can visualize needles with a high degree of fidelity and can be used to place the needle accurately in joint and soft tissue structures. This technique is often referred to as interventional ultrasound. Ultrasound-guided injections have been shown to be more accurate in a number of joint and tendon areas.”

In their study, they noted a discrepancy between diagnostic and interventional use in the United States.

“We saw rapid rates of increase in interventional utilization in subspecialties, such as orthopedics,” they said. “We are searching for reasons for over-utilization of interventional ultrasound among many of the musculoskeletal care-related specialties. Some of the reasons for overuse could be regional barriers to providers doing MSK ultrasound diagnostically, as well as adding MSK interventions to boost revenue.”

Mathiessen and colleagues investigated ultrasound for the detection of small osteophytes in hand osteoarthritis (OA). They suggested ultrasound can be more beneficial in the detection of OA in the hand at an earlier stage than with conventional radiographs. They explored associations between ultrasound-detected osteophytes at baseline and incident radiographic OA features in joints without radiographic OA at baseline. The study included 78 patients in Norway. They used two sonographers in collaboration to determine baseline ultrasound-detected osteophytes. Baseline radiograph data indicated 301 joints were normal. However, 86 of those joints had concurrent osteophytes when assessed by ultrasound. The researchers suggested ultrasound-detected osteophytes at baseline strongly predicted incident radiographic OA during follow-up. Data at follow-up indicated 47% of the joints with preliminary sonographic osteophytes developed OA by radiography. Conversely, 17% of the joints without baseline osteophytes developed radiographic OA. The strongest association was reported in the development of joint space narrowing.

“In the current analysis, we demonstrated that ultrasound-detected osteophytes in joints assessed as normal on radiographs were a strong predictor for development of radiographic OA in the same joint 5 years later,” the researchers wrote. “These results support the use of ultrasound as a promising clinical tool for early detection of hand OA.”

“Ultrasound-guided procedures have many advantages. It allows visualization of the target,” Madhoun said. “You can see the adjacent nerves and arteries, joint effusions and needle trajectory. It helps with the accuracy in placing the needle into the joint. Historically, ultrasound-guided procedures have a higher rate of success when compared to ‘blind injections.’”

Diagnostic Tools

An important area of question is whether MSK ultrasound can be used solely as a diagnostic tool or as a tool to assist in various interventions. For Skinner, the diagnostic use is preferable when a particular diagnosis is suspected.

“If you have non-specific knee pain, ultrasound is not typically the best choice,” he said. “But, if you have a specific tendon rupture or synovitis, ultrasound can be useful in zeroing in on the target.”

Diagnostic uses of ultrasound include biopsies, numbing joints and injecting steroids.

“There are times when imaging modalities may not provide us with adequate information to make a definitive diagnosis,” Madhoun said. “In those instances, obtaining a sample of joint fluid is essential. Synovial fluid analysis can help us distinguish non-inflammatory fluid, such as from OA, trauma, etc., from inflammatory fluid, such as gout, rheumatoid or infectious arthritis. Using ultrasound-guided procedures can help to visualize the fluid and decrease the odds of a “dry tap.” Overall, it offers a better diagnostic yield.”

MSK ultrasound as an interventional tool offers a number of advantages, according to Madhoun.

“The use of MSK ultrasound has given us access to joints that were traditionally not available to rheumatologists, such as injections of the jaw, hip and sacroiliac joints,” he said. “Fluoroscopy used to be one of the only means of reaching those joints.”

One concern with MSK ultrasound is it can be operator dependent, according to Skinner.

“It requires a lot of skill, but there is sometimes great variability from operator to operator,” he said.

For Madhoun, it is a matter of practice making perfect.

“Ultrasound is indeed technician-dependent,” he said. “Some people are just better at it than others when it comes to looking for inflammation within the joint. Are there occasionally false positives or negatives? Absolutely. But ultrasound is becoming more standardized now, so that is not as common.”

He said that these interventions can take years to perfect.

“No matter how much you train, you have to keep using it,” he said. “If you do not have this technology at your clinic or institution, [then] you will be limited in your ability and comfort.”

Another advantage is real-time results are available, as opposed to waiting days or weeks for results of an MRI.

“In many ways, MSK ultrasound is still being under-used,” Skinner said. “At the Mayo Clinic, we have great access to MRI, but ultrasound provides a cost savings.”

This does not mean that cost is not a concern, according to Madhoun.

“Some ultrasound procedures do not get reimbursed,” he said. “We provide it anyway because it is safer than doing procedures blindly, and it is still cheaper than a CT scan or MRI. You can use it any time and as often as you would like.”

Use of MRI

Collins and colleagues investigated whether changes in semi-quantitative biomarkers from knee MRI during the course of 24 months could predict OA progression through 48 months. Bone marrow lesions, cartilage thickness, cartilage surface area, effusion-synovitis, meniscus morphology, meniscus extrusion, entophytes size, and Hoffa-synovitis were the factors that underwent analysis. The researchers reviewed data from 194 patients and 406 controls.

“Both loss of cartilage thickness and cartilage surface area worsening over 24 months were independently associated with knee OA progression over 48 months,” the researchers concluded. “In addition, changes over 24 months in semi-quantitative measures on MRI related to meniscus morphology, effusion-synovitis and Hoffa-synovitis were associated with progression of knee OA in multivariable models.”

“MRI is the second most commonly used imaging modality in most practices,” Madhoun said. “It is extremely helpful. It is great at picking up inflammation in joints that you cannot pick up on exam.”

Another advantage of MRI is the comprehensive nature of the procedure, according to Madhoun.

“With ultrasound you have to do every joint individually. A joint has to be evaluated from different aspects and at different probe orientations to get a comprehensive image, which can take a long time,” he said. “With MRI, you easily get a complete image of the joint.”

However, MRI can also be cost-prohibitive.

“Some insurance companies do not readily and easily approve MRIs,” Madhoun said. “Especially if you need to repeat it more than once.”

MRIs do have other limitations. “MRIs with contrast should be avoided in patients with advanced renal disease. Additionally, patients with embedded metallic objects, such as a pacemaker, should avoid MRIs all together.”

Role of PET/CT

Shah and colleagues presented results of a study at last year’s ACR Annual Meeting in which they investigated the role of PET/CT in general rheumatology practice with the intent of gaining insight into indications and diagnostic yield. They also aimed to determine the additional benefit, if any, of whole body CT.

The study included 80 PET/CT requests from eight rheumatologists between January 2008 and April 2015, according to the study. PET/CT was positive in more than half of the 27 cases reported (56%). New diagnoses were found in 10 of these patients, while scans found increase in disease activity in two patients and non-specific abnormalities in three. There were 12 true normal scans. PET/CT scanning failed to add additional information to whole body CT in 69% of patients. However, PET/CT yielded a direct diagnosis in one patient with Takayasu’s arteritis, three patients with aortitis and two patients with inflammatory arthritis that had not been detected in whole body scans.

“PET/CT has moderate diagnostic sensitivity and reasonable specificity in general rheumatology practice and often appears to provide no further information to [whole body/CT] WB-CT,” the researchers concluded.

Old and New Modalities

Most centers do not use CT for rheumatology with great frequency, according to Skinner.

“We will occasionally use dual-energy CT,” he said. “With that modality, we can see deposits of green pixels which represent uric acid in gout.”

He added that another area of question in which CT is useful in imaging of spondylitis.

“Contrast is helpful to see synovitis in both CT and MRI. CT can nicely depict erosive cortical changes,” he said.

Madhoun agreed that dual-energy CT has become something of a new modality in rheumatology.

“It can be helpful, but this is not something we rely on heavily,” he said, and added that although this method is not as expensive as MRI, there are still cost concerns.

“There is also a significant amount of radiation with CT compared to ultrasound and X-ray,” he said.

Researchers presented findings for a number of novel imaging modalities at last year’s ACR Annual Meeting. Tabechian and colleagues investigated B-flow, a novel vascular ultrasound “technique in which the high resolution aspect of gray scale ultrasound is optimized to demonstrate movement of one tissue, particularly of erythrocytes, against the background of stationary structures,” according to the researchers.

They suggested that B-flow may be comparable to power Doppler in terms of accurately depicting background non-vascular anatomy. The aim of the current study was to determine whether this approach could assess inflammatory arthritis in a cohort of 10 patients with RA. Fifty-one joints underwent analysis using power Doppler and B flow. Results indicated agreement between the two approaches. The median B flow score was 1, while the median PD score was 2. Perfect agreement between the two modalities occurred in 28 joints, for a percentage exact agreement of 54.9%. Higher power Doppler scores were reported in 39.2% of the cohort, while 5.88% of joints had higher B flow scores. Both modalities also yielded high rates of inter-reader reliability, with the intra-class coefficient at 0.978 for B flow and 0.964 for power Doppler.

“In this pilot study, B flow correlated well with [power Doppler] in the assessment of rheumatoid synovitis in small- and medium-sized joints with similarly excellent inter-rater reliability,” the researchers concluded. “Both ultrasonographers felt there were fewer sonographic artifacts with B flow imaging compared to [power Doppler] and this may explain the lower scores seen on B flow. The ability to clearly demonstrate the background gray scale anatomy while displaying the vascular structures of interest offers advantages over Doppler techniques as artifactual findings can be more clearly recognized and excluded from interpretation.”

“This is a small study, and I did not see any papers published in PubMed about this approach, but it is an interesting concept,” Ranganath said. “B-flow seems to correlate with [power Doppler ultrasound] PDUS, but not perfectly. If this approach can identify artifactual findings that can be more readily excluded, then this could be of importance. But it is too early to say much based on one small paper.”

“This is blood flow imaging without using Doppler,” Skinner added.

Fluorescence Optical Imaging

Mendonca and colleagues conducted a cross-sectional study that included 44 patients with psoriatic arthritis, 10 healthy controls and six patients with OA of the hands. The researchers aimed to compare power Doppler and spectral Doppler indices, and explore associations between spectral Doppler measurements and clinical parameters. They examined nail bed measurements and semi-quantitative gray scale and power Doppler scores for all nails. Results indicated that gray scale and power Doppler semi-quantitative scores were significantly higher among patients with psoriatic arthritis than controls. Resistance index measurements using spectral Doppler had high sensibility and specificity for psoriatic arthritis, according to area under the curve analysis.

“Nail [spectral Doppler] indices are significantly different in [psoriatic arthritis] patients independently on the presence of clinically evident nail involvement,” the researchers concluded. “These [spectral Doppler] parameters might find a place in early diagnosis, monitoring of disease activity and response to therapy in [psoriatic arthritis] patients.”

“Blood flow correlates with the amount of inflammation,” Skinner said. “We use Doppler to assess this. Right now, power Doppler does a good job of this, as well.”

Kawashiri and colleagues investigated indocyanine green-enhanced fluorescence optical imaging (FOI) in 25 patients with active RA. They aimed to compare this modality with ultrasound and serum biomarkers. Clinicians performed fluorescence optical imaging and ultrasound on the same day. They assessed 18 joints at an early, intermediate and late phase evaluation. Gray scale and power Doppler were used to score joints by ultrasound, as was bone erosion. At the time of imaging, 45 serum biomarkers were measured using a multi-suspension cytokine array. Results indicated significantly high fluorescence optical imaging scores in joints where bone erosion was detected by ultrasound compared to joints that showed no bone erosion on ultrasound. Results from individual patients indicated that the fluorescence optical imaging scores demonstrated a clear correlation with both gray scale and power Doppler scores. The imaging scores also correlated with DAS28-ESR. Correlations also were reported between fluorescence imaging and serum interluekin-6, VEGF and tumor necrosis factor-alpha. They concluded this approach correlates with ultrasound scores and serum biomarkers. “However, the significance of each phase of FOI may be different and need to be further clarified.”

“A lot of these approaches are fairly new, and we are still learning things about them,” Madhoun said. – by Rob Volansky

• References:
• Collins JE. Paper #1007.
• Kaeley GS. Paper #925.
• Kawashiri S. Paper #189.
• Mathiessen A. Paper #928.
• Mendonca JA. Paper #191.
• Shah K. Paper #136.
• Tabechian D. Paper #186.
• All presented at: American College of Rheumatology Annual Meeting; Nov. 7-11, 2015; San Francisco.

• For more information:
• Sunita Dodani, MD, MSc, PhD, can be reached at 850 8th Street, Tower II - Suite 6009, Jacksonville, FL 32209; email: sunita.dodani@jax.ufl.edu.
• Gurjit S. Kaeley, MD, can be reached at 655 W. 8th St., Jacksonville, FL 32209; email: gurjit.kaeley@jax.ufl.edu.
• Hareth Madhoun, DO, can be reached at the Department of Internal Medicine, Division of Rheumatology and Immunology, Davis Medical Research Center, 480 Medical Center Dr., Columbus, OH 43210; email: hareth.madhoun@osumc.edu.
• Veena K. Ranganath, MD, MS, can be reached at UCLA David Geffen School of Medicine, Rehabilitation Building, 1000 Veteran Ave., Room 32-59, Los Angeles, CA 90095; email: vranganath@mednet.ucla.edu.
• John A. Skinner, MD, can be reached at 200 1st St. SW., Rochester, MN 55905; email: grove.ethan@mayo.edu.

Disclosures: Dodani, Kaeley, Madhoun and Skinner report no relevant financial disclosures. Ranganath reports she has done investigator-initiated studies with G-tech and Pfizer and ad board meetings for Bristol-Myers Squibb.