Computed tomography is enjoying a rebirth.

The staid technology of the last three decades has become rejuvenated with the advent of multislice CT. The ability to acquire more images — from four to 16 “slices” per rotation — with greater clarity and reduce scan times for patients has prompted healthcare providers to abandon single-slice CT equipment with no regrets.

Four-slice CT technology, which debuted four years ago, represents a vast improvement over single-slice CT equipment. In addition to providing more diagnostic information, multislice CT images patients faster and adds new clinical applications that could not have been performed with as much accuracy as previous CT technology.

“Now we can do physiological imaging in some areas, cover large volumes very, very quickly, get much higher resolutions and get greater image detail to find small abnormalities much more clearly,” says Jay Cinnamon, M.D., director of neuroradiology at Emory University (Atlanta).

The ability to scan patients faster and create sharper images is enhancing CT’s standing in radiology departments and converting more advocates in other areas of the hospital where medical imaging can be a beneficial tool to patient care.

U.S. CT market
The ascension of multislice CT prompted market research firm Frost & Sullivan (San Jose, Calif.) to herald the coming “extinction” of single-slice CT in a February report. The study predicted that multislice CT’s growth essentially would render single-slice technology obsolete by 2006.

Total revenues in the domestic CT market reached approximately $1.3 billion last year — up 24 percent from 2000 — thanks in large part to multislice CT sales. The report projects the CT market in the United States will exceed $2.3 billion in revenues by 2008.

“Market activity in this sector has become increasingly complex as a result of the rapid evolution of CT,” notes Frost & Sullivan medical imaging analyst Monali Patel. “Dynamic growth of the multislice CT market is countered by the steep descent of single-slice CT, which is headed toward extinction.”

The report calculated single-slice CT revenues of $134 million in 2001, which represents a sharp decline from single-slice CT’s peak of $600 million in 1998. The report blames the decline on market saturation, product maturity and the lack of product innovation “that prevents single-slice from competing with the overpowering presence of multislice scanners.”

Multislice CT has advanced quickly to account for 83 percent of total market sales since its 1998 introduction. Revenues topped $1 billion in 2001 and are expected to produce double-digit growth during the next four years. The Frost & Sullivan report projects multislice CT sales will peak at more than $2.4billion in 2007, before slipping back slightly in 2008, but remaining well above the $2 billion mark.

The reasons for multislice CT’s popularity include the technology’s ability to accommodate more clinical applications, moving beyond motionless organs to include a beating heart in cardiac imaging and CT angiography (CTA). Reduced scan times also mean greater patient throughput, which helps healthcare facilities reach break-even points sooner on the capital equipment and technology investments, and patients are happier, too.

“For general purpose radiology, everyone will take the two-slice [CT], if they are short on dollars,” says Jonathan Murray, global manager for premium CT at GE Medical Systems (GEMS of Waukesha, Wis.). “They also will come up to the four-slice or the eight-slice [CT] for general purpose radiology. It will allow for thinner slices and wider coverage.”

Speed and coverage
The two primary issues with multislice CT are speed and coverage. Image a patient as quickly as possible to collect a maximum amount of information for a specific region of the body, a particular organ or an entire body. Speed and coverage are balanced with how long a patient can hold his or her breath when imaging certain organs.

Rotation speed is critical in order to freeze and image the motion of fast-moving objects, such as the heart. Rotation speed of 0.5 seconds now is standard in cardiac imaging. When imaging a motionless liver or performing a brain perfusion study, coverage — imaging the entire organ — is the priority. Rotation speed of 0.75 seconds to 1 second in these applications is acceptable.

The proper balance of speed and coverage also extends to patient welfare.

“As coverage increases, the [radiation] dose decreases,” notes Kieran Murphy, M.D., director of interventional radiology and neuroradiology at Johns Hopkins Hospital (Baltimore). “So there are benefits to patient dose and physician dose with increased coverage.”

With the availability of faster rotation speeds, the new frontier for multislice CT — particularly 16-slice technology — is cardiac imaging. The goal eventually is to perform a cardiac study in one breath hold in the range of 17 to 20 seconds to maintain a steady heart rate.

“The heart is probably the most demanding application,” says William Kulp, manager of CT

marketing for Philips Medical Systems (Bothell, Wash.). “To a certain extent, coverage is important, because you want to cover the whole heart. If you can image the whole heart in one revolution, that is the Holy Grail for CT.”

Philips offers the Mx8000 multislice CT scanner, which the company inherited when it acquired Marconi Medical Systems Inc. (Highland Heights, Ohio) last year. The company began first shipments of its 16-slice Mx8000 IDT in the middle of June.

GEMS added to its CT portfolio last year, when the company acquired Imatron Inc. (So. San Francisco) and its electron beam tomography technology for coronary imaging.

“What Imatron can get done in a single heartbeat, takes multiple heartbeats for traditional CT scanners to do,” says Murray. GEMS’ top-of-the-line CT is the 16-slice LightSpeed Ultra.

At Johns Hopkins Hospital, Murphy has been utilizing multislice CT as an interventional tool in a fluoroscopic mode to guide needles under near real-time imaging into the brain and spine and to navigate around and within the spinal cord. The CT scanner generates axial images at 13 or 14 times per second — and as fast as 26 times per second — creating the sense of real-time motion.

“I have an image that is of greater detail than I cannot obtain in any other way,” Murphy adds. “I have imaging that is so rapid that it allows me to pass it under great control and sensitivity to nerves, arteries, veins and whatever and into lesions as small as one or two millimeters.”

Murphy also has been navigating the ventricles of the brain with this CT technique, using the Aquilion eight-slice CT scanner from Toshiba America Medical Systems Inc. (TAMS of Tustin, Calif.). Johns Hopkins is scheduled to have a 16-slice Aquilion installed during the third quarter.

This capability “will revolutionize neurosurgery,” Murphy predicts, “because these [applications] have always been things that required open surgery in the past.”

Expanded applications
CT fluoroscopy is just one application that flourishes with multislice CT technology.

Emory’s Cinnamon adds the example of CT pulmonary angiography for people who are suspected of having blood clots in arteries or in the lungs. Single-slice CT technology generally took 30 to 50 seconds to scan the lungs. One obstacle is that there are patients who simply cannot hold their breath for that length of time.

Multislice CT can scan the lungs in 12 seconds and cover the lungs with thinner sections to produce greater detail than before, Cinnamon says.

In the area of acute stroke and perfusion imaging, a physician can use multislice CT to gather data from the brain more rapidly with the help of a contrast agent. With the thinner images, a physician can gauge the level of perfusion, even before the brain demonstrates any changes that would be detected with a routine CT scan.

“We are able to identify — literally within minutes after a stroke has started to develop or a blood vessel has become occluded — how much territory is at-risk and where the occlusion is,” adds Cinnamon. “Since time is brain, the faster the diagnosis can be made of the amount of territory at-risk and the site in the arterial tree that is diseased, the faster appropriate intervention can be guided to relieve the occlusion and salvage some brain.”

Images from an entire cervical spine or the neck can be obtained in a matter of 15 seconds. With the aid of computer post-processing, detailed images are available within five minutes.

“Now it is taking some things away from plain film and x-ray diagnosis and moving them more into the CT realm,” Cinnamon says. “CT can do a better job in identifying disease or ruling out disease and can do it much faster than plain film.”

Into the ER
With its expanded applications, multislice CT also may become an option in an emergency room for use in CT angiography (CTA) or noninvasive coronary imaging, if a person arrives with chest pains. The procedure is analogous to the head scan performed on the potential stroke patient.

Multislice CT and CTA also save valuable time in patient treatment. With higher temporal resolution, examination time has been reduced.

“With the reliability of the systems to do CT angiography, we are finding neurosurgeons taking patients straight from CT into the operating room,” says Bryan Westerman, TAMS’ clinical science

manager for CT. “Normally, they would have gone from CT to a conventional digital subtraction angiogram and then to the OR.”

Westerman adds that healthcare providers also are opting for multislice CT, as the thinner slices show more diagnostic information when imaging joints and extremities.

“People are starting to use CT for musculoskeletal applications that normally they would have gone to MR for,” Westerman says. “The reason nobody looked for this sort of application earlier was that they didn’t have 0.5 millimeter slices available. When you are looking for small structures — tendons, ligaments, bone and cartilage — you need that.”

In the realm of pediatric imaging, scanning times of eight to 10 seconds with multislice CT have helped Johns Hopkins Hospital reduce by 60 percent the number of pediatric patients who require sedation.

“Nearly every kid can hold still for that long,” Murphy adds with a hint of humor.

With multislice CT expanding into so many medical imaging applications, Murphy says the technology has given healthcare providers “the equivalent of a Swiss Army knife.”

Voluminous data
Frost & Sullivan’s report on multislice CT debated how the market would react to more slices and faster scan times and some potential obstacles.

“This generation of scanners,” the document notes, “may have additional consequences of steep learning curves and severe workflow obstacles that stem from trying to manage enormous volumes of data, not to mention an expected higher price tag. While these barriers may not halt sales altogether, it is unlikely that the newer technologies will be embraced to the same degree as the most recent generation of multislice scanners.”

The amount of data generated from a multislice CT exam is overwhelming. With a single-slice CT, a radiologist reviewed eight to 10 sheets of film with eight or 12 images per sheet. A four-slice CT scanner will produce 300 to 800 images from one exam.

Compare that to the birth of CT in the early 1970s, when the technology took eight minutes to produce an image.

“Now we are producing eight slices per second,” says Philips’ Kulp. “If you plot where we think it is going, it will double again in the next three to four years.”

Kulp and other CT enthusiasts see computer-aided detection (CAD) technology as playing a pivotal role in helping radiologists pour through the vast amounts of data generated by multislice CT.

“By no means will [CAD] replace radiologists, but it will help the radiologist have more confidence in what he or she is seeing and sorting out the wheat from the chaff.”

Emory’s Cinnamon shares that view on CAD. “Mortal humans simply cannot handle analysis of so much data,” he adds. “We have gone from 100-slice studies to 800-slice studies and we are doing them faster. There is no way people can handle this data without computers assisting us and helping to identify pathologies and helping to screen images.”

Siemens Medical Solutions (Iselin, N.J.) is working with its clients to develop a data model which essentially would recommend the most appropriate medical images to store and those to discard. Markus Lusser, Siemens’ segment manager for CT, says the collaboration is making progress.

“Their feedback is that they produce less data to store in a PACS than they produced on a four-slice machine,” he adds. “If you do it smartly, you create less archiving data than if you do it traditionally by storing everything.”

Siemens introduced its Magnetom Sensation16 multislice CT scanner late last year and currently is marketing the system commercially.

Human factors
With vendors conquering 16-slice CT technology with 0.4-second rotation, R&D teams are working on the next advance to 32 slices, 64 slices, 256 slices and beyond. At what point, is enough enough?

“There may come a point where there is a limit, but I don’t think we are at it,” opines Cinnamon. “I think what will happen is that there will continue to be an increase in the number of slices, but there will be an increase in the field-of-view.”

Instead of having two or three centimeters per coverage per rotation, for example, Cinnamon believes image coverage eventually may increase to 10 centimeters. With that field-of-view, he sees imaging vascular phases and physiological aspects of the liver with contrast as one beneficiary.

With the potential for more slices and faster rotation also come physical concerns of the patient.

For example, using 16-slice CT technology, it would take approximately 10 seconds to image a patient’s thorax with a table feed of 72 millimeters per second. If 32-slice CT technology were used with a rotation of 0.4 seconds, the table feed would have to be 144 millimeters per second.

“There is a physical limitation for the patient,” Siemens’ Lusser explains. “The patient would run so fast through the gantry that you run into motion artifacts caused by the high table feed. Also, blood is not fast enough to run through the patient,” if contrast is used.

In addition, if a scanner were to be designed with more than 16 slices using conventional CT technology, the radiation dose for this type of scanner would have to be dramatically higher to match diagnostic image quality, creating an unhealthy environment for patient and physician.

“There are races going on for thinner and thinner slices, but to really create an image with 0.4 or 0.5 millimeter collimation — which is diagnosable — you need such a dose that it is not worthwhile,” Lusser says.

The future
So what is next for multislice CT?

From a market standpoint, Peter Arduini, GEMS’ general manager of global CT, sees cardiac and oncology imaging as two catalysts for multislice CT technology.

“Those are two major disease areas that will push CT technology — and the evolution of the components of the system — based on what will make the biggest difference,” he says.

On the technology side, Lusser sees the probability for a flat-panel CT design whereby the patient remains stationary and a wide array acquires one organ image in one rotation.

“Then you can enter environments of dynamic volume CT where you do dynamic studies and all kinds of new applications for the heart, the brain and the kidney,” Lusser adds.

Philips’ Kulp also sees the advent of large-area detectors — as large as the heart and other organs — allowing healthcare providers the ability to perform whole-organ perfusion.

At GEMS, Murray says the company is exploring many paths, as it “challenges” its engineers to increase and maximize rotation speed, coverage and resolution. “We have phantom studies where we can fit an entire heart inside the field-of-view of a large x-ray detector,” he notes.

Johns Hopkins’ Murphy has been dabbling with 256-slice technology with 25 centimeters of coverage. The technology — which generates 256 slices per rotation — currently is in the prototype stage at TAMS.

Murphy says one potential application would be the installation of a chemotherapeutic jell into a head and neck tumor or bone cement into a vertebral body.

“Until you get that extensive coverage,” he adds, “you are limited to where you can put a liquid, because you don’t know where it is going or if passes out of the slice you are in.”

TAMS’ Westerman speculates that it will be four to five years before 256-slice technology is installed in radiology departments. The X factor is whether a facility can handle the amount of data from such a CT scanner.

“We need to get reasonably inexpensive data processing, which is fast enough to handle that data stream,” Westerman says.

For a medical imaging technology that has had little change through most of its tenure, CT’s developing technology has the potential to become an even greater contributor to patient care.