In the last 25 years, traditional X-ray has been joined – and has also been challenged – by ultrasound, CT, MR and, some would say, nuclear medicine. They’re all vying for imaging dominance. Every modality, for instance, has a protocol to image the breast. How do you decide which is best and which should be used first? Where do these “turf wars” leave general radiology, particularly in light of CMS’ value-based compensation model? We asked a number of top doctors and leading radiology administrators what they think.

Jason Launders, director of operations for ECRI’s Health Devices Group, said, “While basic X-ray hasn’t changed significantly since Roentgen X-rayed his wife’s hand in 1896, it was a miracle to see inside the human body back then, and today, 120 years later, it’s just as relevant. “It’s like a piece of paper. Slips of paper will always be around because they do the job well. We should not be distracted by bells and whistles of the other technologies and overlook X-ray. It is the baseline.”

Brad Hellwig, administrative director of radiology at Crouse Hospital in Syracuse, N.Y., agrees.“X-ray, and specifically digital radiography, will remain the highest volume imaging modality because it is the most appropriate way to assess a wide range of diseases, conditions and injuries,” he said.

‘Sticks and stones…’
Dr. Ezequiel Silva, chair of the American College of Radiology Economics Commission, put it this way: “Plain radiology is not going anywhere. It is a good first-line study for many things,” especially in the sense of its negative predictive value. “X-ray is always going to be good at showing whether an ankle is broken or not.”Another mainstay is the simple chest X-ray. “This will always be one of the most common exams,” said Steven Jung, director, diagnostic imaging at Alexian Brothers Medical Center in Elk Grove Village, Ill. “Every morning doctors in ICUs and CCUs order chest X-rays to see what is going on.”

Improvements and innovations help expand X-ray use
Chris Tomlinson, senior director of radiology at The Children’s Hospital of Philadelphia, adds that “with new technologies including positioning devices, better mobility, flat screens, etc., the X-ray modality is actually becoming more flexible and keeping up with the more advanced things.” “We are seeing more advanced technology coming into radiography,” said Launders. “A lot of work and research has gone into getting the best possible images. Flat-panel detectors can produce [an] image in a few seconds almost anywhere at [the] point of care, in the emergency department, wherever you are.”

And these ease-of-positioning factors apply in particular to the use of X-ray, especially fluoroscopy, in the operating room. “Another capacity where X-ray is going to hold is in the OR,” observed Jung. “Fluoro via C-arm is the best way to do many exams. This will continue to provide it with a fair amount of volume.”

A shift in the use of fluoroscopy
Regarding fluoroscopy, there are some indications that use may be shifting. “Use of fluoroscopy at our hospital has been declining,” noted Hellwig. “We have a stroke center and we still conduct barium swallow exams to evaluate stroke patients before discharging them. It’s also used to evaluate urinary systems and for fertility testing. But in areas where fluoroscopy is used, it makes sense to purchase a radiography/fluoroscopy system so you have the best of both worlds.”

Tomlinson also observes that “MR has become able to pick up some of the fluoroscopy studies. That’s because it gives you more information to better diagnose. So there is some movement away there.” But while acknowledging “some shifting of modalities,” he “rejects the notion that X-ray and fluoro are no longer useful.”

Further application of tomosynthesis
According to Dr. Samir Mehta, associate professor of orthopaedic surgery and chief of the Orthopaedic Trauma Service at the Hospital of the University of Pennsylvania, “If there are any areas where basic X-ray may be superseded to a degree, one may be by tomosynthesis. Of course, it won’t replace X-ray, but tomosynthesis is increasingly useful for subspecialists in particular. If I have a patient with pelvic issues, I may be inclined to forego the plain X-ray and get the tomosynthesis examination because I get the X-ray ‘plus.’ ”

With tomosynthesis, Mehta adds, “I am not increasing my costs significantly and I am getting a tremendous amount of information. In some ways I’m doing away with the next exam, which in the pelvis situation might be a CT scan. The digital tomosynthesis gives me as much information as a CT would for certain conditions.” So while we will always have X-ray, said Mehta, “this is one example of a situation where we could go right to tomosynthesis.”

CT usage continues to encroach on basic X-ray
“The general category of abdominal pain is just one area where CT is better, because technology has evolved,” says Silva. “One example might be kidney stones. Years ago you might do an X-ray and an IVP or a pyelogram. Now, if a patient comes in with clinical suspicion, you are probably going to do a CT scan or a CT urogram, which results in a 3-D reconstructions/virtual view, which is a very good new exam. That’s an example of where, in 2016, there is a different diagnostic algorithm than we might have seen in 1986.”

Jung also sees this pattern, in reporting that “we still do some lower and upper GI work and IVPs, but a lot of that has migrated over to CT.” Of course, there are some things that straight X-rays will never do well. So, for evident reasons, as Launders puts it, the things that have already shifted away from X-ray are anything involving soft tissues. And, as Silva observes, “things like stroke care and any study of the brain will involve MRs that provided more appropriate, actionable information for diagnoses.”

But the rush to use ‘higher’ modalities first may be moderating
As Launders explained, plain film X-ray has previously lost some ground to the “competition,” partly because of the reimbursement structure in the U.S. and partly because people wanted to use the CT scanner as much as possible when it was new. “That may not be what was happening, but that’s the perception, anyway,” he said. As a result, CT use dramatically exploded.

“But now there are questions about [the] need for that and things are shifting somewhat back to radiography,” he said. As Jung sees the issue, the key is that there is now focus on radiation dosage and overutilization in CT scans, which may bring some of that work back. “With the radiation dose of CT high on the radar, especially in pediatrics, choosing ultrasound for an app scan study is becoming the first choice, rather than doing a CT on the child," he said. Another segment that Jung believes ultrasound may take away from CT is in cases of younger females with pelvic pain. Ultrasound should be the choice before CT, because of the reduced risk to the organs of a young female.

Launders summarized the issue: “In the past, physicians called for things to be done on the high-end stuff. This may have led to some inappropriate CT and MR use. Now things are being corrected because people are being more careful and spending more time looking at the referrals. Historically, the radiology department just scanned everyone they were given. Radiologists have really looked into it and stopped some inappropriate studies. That’s changing things. People were skipping the triage steps and going straight to more advanced studies like MR. Now X-ray can be seen as a triage step before going to [the] use of more expensive technologies. That could drive more studies to X-ray technology.”

The impact of costs and reimbursement rates
Then there is the issue of cost-benefit. It has been a “temptation for doctors to go straight to MR because they will get their answers quicker,” Launders said. “But the MR is a far more expensive study to do. To fit-out an MR room, you are looking at $2 million to cover a maximum of 20 patients per day, while a radiographic room can be fit-out for $300,000 and you can probably do 40-60 patients a day in it. The costs are very different. Even if you look at interventional radiology with state-of-the-art systems. They are very expensive and we are seeing that they are being used more for interventions and not for diagnostics.” These cost issues are becoming particularly relevant in the new health care financial climate.

Launders added: “The big thing driving health care today is obviously the shift to the value-in-care model from the fee-for-service model. This is going to drive some significant changes in how medical equipment is used. Until now, people have been paid to do more procedures. They have done more diagnostic testing in order to get paid in return. The change is that insurance companies and Medicare will be paying on [the] basis of the value of the care, not for the number of individual diagnostic tests and procedures, but one fee to treat that patient.

That is really a fundamental difference in health care finances, and I believe it is going to affect radiography quite considerably. It will keep X-ray around because there is much less cost in providing that service. “The pressures of the new reimbursement policies will ensure prominence of X-ray, in addition to the fact that outcomes are clearer, faster, involve less radiation and are cheaper.”

Tomlinson agrees, saying, “If hospitals are going to be paid for an indication — if someone comes in for an appendicitis episode — and you get paid once for that appendicitis treatment, not for how many X-rays, CTs or MRs you take, the key is not overusing expensive modalities. How will the cost be made appropriate for what the indication is?

If you are going to get paid $5,000 for a case of appendicitis regardless of how you treat it, and you can answer the question with an ultrasound, why would you do an MR? You are spending more for information you don’t need. Once you determine it is appendicitis, you are going to go to surgery. Did you answer it five different ways, with a more expensive test? That is not necessary. In a value-based world there is no reward in using higher-end tests and overusing resources. Even in ED and urgent care, if the patient is footing the bill, they want you to answer the question. They do not want you to overkill the modality.”

Jung shares the view that the new value-based model is “definitely an issue.” CMS is “looking at this and cutting, and imaging [is] always a target for reductions. That very possibly could press things back toward X-ray because CT and MRI can require preauthorization for insurance to approve it," he said. And Silva also acknowledges that “as we see pricing or payments going down, that will put some stress on practice and use of the newest technologies.”

Nevertheless, while these could be factors, Mehta doesn’t think that the new pricing environment will necessarily impact modality usage rates one way or the other: “While X-ray is here to stay, I don’t necessarily agree that it is the automatic test that you are going to give to every patient. For instance, there are so-called Ottawa Rules for ankle injuries that establish that if you don’t meet three criteria, you don’t need to get X-rays. That is widely accepted. As we move forward, we need to be mindful of these binary style algorithms of care where, if it is [a] young patient, positive for X, Y, Z, where the answer may be MR, CT or digital tomosynthesis first. I think that the new guidelines could actually work both ways: they will make sure that we don’t reflexively order either an X-ray or an MRI.”

Appropriateness criteria: The right test at the right time
While all perceive the potential significance of these pressures, these experts agree with Mehta’s assertion that they should not push things in any single direction. And in every case, they make reference to the importance of the ACR appropriateness criteria guidelines as the way to achieve the appropriate balance of modality use. “The place to go to for specific uses is the ACR referral guidelines. These show how conditions are being channeled to the different modalities,” says Launders.

“Standardizing care pathways, using the same tests for indications, not letting people just choose X-ray, CT, MR, whatever, is the way to go,” adds Tomlinson. In Silva’s view, there are a number of benefits to this developing system: “You want patients to get the best study possible to diagnose their condition. In addition to getting the right exam for that case, we have the benefit of physicians learning what the best exam is for future purposes. They can go to links to look at literature to see how those determinations were made, making them better doctors.

Physicians appreciate this level of support. They want to do the right thing. Some might say doctors should do what they want to do. They could override the recommendation and go with something with a lower score. But they can also consult with the radiologists as to whether the appropriateness score is correct in that particular case. This can open up the basis for cooperation aimed at making the best decision. This can result in a broader collaborative process. “I think that Medicare and the health care system want the best for their beneficiaries, but they want to do it responsibly,” Silva continues.

“This is why I think clinical decision support is a nice way to implement that by bringing all the parties together. It’s good for patients because they get the right exam at the right time. It’s good for the physicians and radiologists because they can be confident that they are ordering and doing the right exam, too. The importance of appropriate use criteria is growing. There is some pushback from Medicare and some specific physician groups that want to delay it, or do it in a less than comprehensive manner. But it is the best combination we can put together for now. Putting together information to insure appropriate ordering is the goal. Patients are all-important and we want to use the best technologies to give them the most appropriate care.”

Flexibility is the key: No single modality, but the right blend
What does this reference to the appropriateness criteria guidelines mean for radiology modalities? A drive for flexibility and blended approaches, recognizing that all these are valuable tools for the range of indications experienced.

As Tomlinson put it: “It’s not as though MR is always better than an X-ray. Sometimes the best choice is an X-ray, sometimes it’s an MR, sometimes it’s a CT, from an appropriateness perspective. Appropriateness criteria stipulate things. I would challenge the assumption that MR and CT are always better. We should first lead with the question: What is the appropriate study?
What is [the] right study on the basis of what the patient needs? That’s the first thing that challenges [the] notion that X-ray, or any other modality, will be eliminated, or take over.”

Mehta is in full agreement: “These are evidence-based decisions that are consistent with the ongoing development of the appropriate use criteria determining a path of modality which needs to be based on clinical assessments of the individual case. They may point in the direction of X-ray, but in some cases not, and therefore it must be flexible. X-rays are a very simple, low-cost, high-yield, safe method of getting an idea of what is happening from a bone perspective. You can also see some soft tissue shadows.

You can see things like alignment. You can see things like the start of arthritis, or significant arthritis. Plus, based on what you see on the X-ray, you can determine what the next best step might be, whether it is [a] digital tomosynthesis examination, or a CT scan, or an MRI, or an ultrasound. These are all very reasonable options, but you need to know what you are doing first. I agree that X-ray is here to stay. It isn’t going anywhere. But if you start there, the question you then have to ask is, ‘What is the next test after X-ray?’ What is the next go-to, and where does that fit in?”