A few days ago, someone told me that MRI was now a mature modality and would soon take a backseat to sexy newcomers like PET/CT. This prediction reminded me of the tale about the US Patent Commissioner who, in 1899, was reputed to have submitted his resignation to President McKinley, claiming that everything had been invented. Chalk this one up as another completely false urban legend. Charles Duell did not offer his resignation and even noted in his 1899 report to McKinley an increase of 3,000 patents over the previous year and 60 times the number of patents issued in 1837.

In a similar vein, reports of MRI’s transition to middle age appear to be, at best, wildly premature. On average, one in nine Americans will get an MRI examination this year. Since 1996, the modality has grown at an average of 15% to 18% per year, depending on your source of statistics, and recent trends may support higher percentages.

But what about the Deficit Reduction Act of 2005 (DRA)? Will that cause a freeze in growth just like that experienced by the market in the early 1990s because of concerns about the Clinton health plan? To the contrary, the effect of the DRA should be an increase in utilization. Diagnostic information is a commodity, and when the cost of an examination is lowered, but information content remains the same (or increases), usage increases. The DRA is projected to reduce MRI reimbursement for federally funded patients by 25% to 40% depending on type of examination. About one third of US health care costs are paid by Medicare and Medicaid, and many private contracts are tied to Medicare rates. Thus, a substantial increase in volume should result from this price reduction at nonhospital sites.

But is there anything new out there for MRI? For those who attended RSNA 2006, the answer is a most unambiguous "yes." Consider just a few trends.

Take Me Higher

The era of low-field MRI appears to be drawing to a close. The advent of 3T has demonstrated substantially better head and body imaging, which, in turn, has generated even more pressure on scan providers to improve examination quality per time unit. Price reductions at 1.5T now make it economically unattractive to purchase superconducting systems with less than this field strength. It should not come as a surprise that most vendors no longer offer tubular systems lower than 1.5T in the US market. Sales of open-sided units are a pale ghost of the halcyon days in the mid-1990s when these designs represented as much as 50% of the annual market. At the RSNA, it was hard to find any references to Profile (0.2T) or Ovation (0.35T) from GE Healthcare, Waukesha, Wis; Concerto (0.2T) or Magnetom C! (0.35T) from Siemens Medical Solutions, Malvern, Pa; or Panorama 0.23T from Philips Medical Systems, Andover, Mass.

Philips Medical did give substantial booth space and sales emphasis to the Panorama 1T, its current top-of-the-line open-sided MRI. Also supporting this trend, Hitachi Medical Systems America, Twins-burg, Ohio, unveiled a work-in-progress—a high-field, open-sided unit called Oasis. Although there was no disclosure of field strength, images in the booth indicated approximately 1.2T based on the fat-water chemical shift. Similar in design to Philips’ Panorama 1T, Oasis appears to be a superconducting system on its side, with a weight projected to be in the 30,000-pound range.

Toshiba America Medical Systems, Tustin, Calif, was the latest company to announce a commercial 3T whole-body system, joining GE Healthcare, Philips Medical, and Siemens Medical. Now, approximately 700 whole-body 3T systems are installed in the United State, the large majority of which are in standard clinical practices and not research sites. I am informed that in the United States, 3T has been outselling the entire open-sided market for some time.

Philips Medical made quite a splash announcing a rampable 1.5/3T system. Its Achieva XR offers a variable magnet that can be installed with 1.5T electronics but can be later ramped to 3T if the owner desires. Because the company’s 1.5T and 3T magnets are identical from the outside, such a transition would not require any visible changes to the magnet. Of course, new electronics and coils would have to accompany the change (as of yet undisclosed pricing), but this option could help relieve the decision pressure for some buyers who are undecided between 1.5T and 3T.

Throughput ... and Comfort

Many vendors recognize that speed without examination compromise is the key to success. As a method to improve patient throughput, GE Healthcare showed a set of breast imaging coils permanently mounted on its Signa detachable table. It also demonstrated the InSightec focused-ultrasound therapy unit mounted in a detachable table. Such dedicated systems should provide improved efficiency for breast scanning and MRI-guided therapy.

The Total Imaging Matrix (TIM) concept announced by Siemens Medical a few years ago appears to be gaining momentum. More vendors are offering combination coils to allow greater anatomical coverage without patient reposition. Virtually all high-field vendors also offer the ability to increase the number of independent data channels to at least 16. We are approaching the time when patient setup may not vary substantially for a range of examinations. With a blanket of selectable coils above and below the patient and with a large number of data channels, the operator can prepare the patient in a uniform way and then choose the coils that will best match the imaging needs. What a useful tool for increasing throughput.

Reducing patient anxiety through equipment design is proceeding apace. The above-mentioned Philips Medical Panorama 1T and the Hitachi Medical Oasis appear to herald a new stage in higher-performance open systems. These units continue to offer the psychological advantages of open sides without as large a compromise in reduced image quality or longer scan times that are often the consequence of low fields. Equipment price and service costs are, as always, fundamentally linked to financial success.

Some tubular units also have expanded diameters and shorter bore length. Since 2004, the Siemens Medical 1.5T Espree has offered a 70-cm bore size, the same internal width as that found on most CT systems. The magnet is only 1.25 meters long, or less than 50 inches. However, there are trade-offs. This shortened magnet design does limit the field-of-view (FOV) down the bore direction to about 30 cm. However, at RSNA, the company announced a software upgrade, TIM CT, that provides scanning and simultaneous table movement—sort of a "spiral MRI." This allows larger FOVs without longer magnets. The advance is now in clinical testing and will be available as an upgrade on all Siemens Medical TIM units this summer.

Consider for a moment the changes in technology required to provide such an innovation. First, the system must have multiple array coils so signals are detected from that portion of the patient in the magnet at any given time. Second, the system must have multiple detection channels to interface with these coils. Third, the unit has to automatically switch from one set of coils to the next as movement takes place. Finally, the scan sequence must be fast enough to allow RF excitation and signal reception before the tissue is shifted outside the "sweet spot" of the magnet. This does not sound like a mature modality.

Competing Ideas

Is there anything on the horizon that might sap portions of the MRI market? After all, advances in ultrasound have been remarkable, and CT angiography (CTA) now produces some of the most spectacular images of flow that have ever been made. Will these impact MRI growth?

MRI’s low examination cost, noninvasive character, ready availability, and broad acceptance have secured its place in the diagnostic imaging world for at least the next 10 years. For proof, try to imagine a world without MRI. Joint studies would have to be done by arthroscopy at far higher cost and greater pain and risk to the patient. Neurological examinations would be far less effective by CT or far more expensive by PET/CT. Can you imagine trying to convince patients to have contrast CTs of the spine as the only method for studies of the cord and nerve roots? Although ultrasound and CTA have contributed mightily, they largely do so in areas outside the purview of traditional MRI.

Indeed, it appears that MRI may be in a position to garner significant new patient volume from CT. In 2007, it is projected that about 75 million CT examinations will be done in the United States, and about two thirds of these will be body examinations. For MRI, body examinations constitute only about 5% of total volume (estimated to be less than 2 million examinations in 2007). Studies have shown that MRI is diagnostically as useful as CT in the body and patient motion is no longer a technical problem. The remaining constraint to broader use of MRI in the body is cost. As the DRA brings MRI costs down by a larger percentage than CT costs, an increasing fraction of body CT is likely to transition to MRI.

New Territory

Where else can MRI secure additional patient volume? The following five areas may prove fruitful for MRI studies in the next 5 years.

1. Spectroscopy. 3T systems have shown that their additional signal-to-noise ratio (SNR) and greater dispersion (ability to resolve and quantitate more metabolites) can markedly enhance the quality and application of spectral analysis. It appears absurd that the federal government does not reimburse for this highly effective and very low-cost addition to MRI. The combination of morphology by MRI and chemical characterization by magnetic resonance spectroscopy is a natural marriage.
2. BOLD and related techniques. Blood oxygen level dependent (BOLD) studies allow MRI to detect real-time metabolic activity without the use of radionuclides or other invasive agents. I believe such methods can contribute information now available via PET but do so at one tenth the cost and at virtually no risk to the patient.
3. Breast. MRI has a unique ability to detect active cancer at an early stage when treatment is most effective. Although mammography and ultrasound will continue as more prevalent examinations, MRI can provide additional information that can extend life.
4. Radiation therapy planning and evaluation. MRI now has virtually the same spatial accuracy as CT and gives far better contrast for soft tissue. Further, it can detect earlier changes in tissue due to radiation since it can sense interstitial water and does not rely on tissue necrosis. For some patients, MRI may provide a better method for planning their radiation treatment and earlier evaluation of its effectiveness.
5. Prostate. According to the American Cancer Society, one in six men will be diagnosed with prostate cancer at some time in his life. Ten-year survival rates exceed 90% if the cancer is confined to the capsule, and a significant fraction of patients may benefit more from "watchful waiting" than from surgical intervention or other active therapy. MRI, combined with localized multichannel phased array coils, may offer a convenient and low-cost method to track the progress of disease in such patients.

As a clinical technique, MRI is now about 25 years old. I think it is more accurate to describe it as a Generation-X member than as a rapidly graying Baby Boomer. Like most young adults in their 20s, MRI is full of energy and looking for new heights to scale. Stand back, world. I predict more changes to come.