Advanced Visualization: Adding CCTA Post-Processing to the Physician Skill Set
Coronary tree rendering using the Ziosoft coronary analysis thin client application. Source: Ziosoft
Even with experienced technologists and user-friendly software, cardiologists who want to read coronary CT angiography (CCTA) exams still need to understand the basics of CCTA post-processing.

Data Interaction
Workstation engineers are increasingly designing applications to mimic the way doctors think, consequently, these tools have become more user-friendly in terms of workflow, says Tony DeFrance, MD, medical director of CVCTA Education, San Francisco.

On many workstations, for example, CCTA studies present from the starting screen in 3D and 5 mm maximum intensity projections (MIPs). Workstations can automatically segment coronary arteries, subtract surrounding tissue and color-code plaque. Features such as echocardiographic views and user-friendly functional programs operate with a few button clicks.

“Automation has helped speed the review, but the physician will always need to manually manipulate the data,” DeFrance says. “The way I teach is the physician has to get on the workstation and manipulate the data in a very systematic way. This is not still-shot technology. You can’t just look at a picture that the tech took and make a diagnosis. It is dependent on the physician to manipulate the data.”

An essential feature of the workstation is multiplanar imaging, which means physicians have an almost unlimited number of angles to view the data. A lesion in the left anterior descending (LAD) artery, for example, can be viewed in a thousand different angles. “The ability to interact with the data on the workstation—to take a lesion and twist it and turn it and look at it from all these different angles—is one of the most powerful things about cardiac CT,” he says.

Physician champion
CCTA is still an emerging technology and reimbursement for the exam is not guaranteed. But as with any new technology, facilities need to have a physician champion who believes in it. “Once you’ve crossed that bridge, then CCTA is a service like any other offered at the hospital,” says John McB. Hodgson, MD, chairman of the cardiology department, Geisinger Health System in Wilkes-Barre, Pa.

The physician champion should be involved in the learning process, in setting up protocols—especially for technologists and nurses—and in helping with negotiations between radiology and cardiology. “Then you have to have a good workstation,” says Hodgson. “All of them are good and are improving each month.” He uses workstations from GE Healthcare and TeraRecon.

The next hurdle is deciding how to integrate this new test into the paradigm of referring physicians, including cardiologists. These physicians will most likely be more familiar with ordering a stress nuclear or echo exam. “They have to be educated as to why CCTA may be more appropriate in some situations,” says Hodgson.

At Geisinger, Hodgson has four high-end scanners (GE LightSpeed VCT) across the health system, which includes four hospitals and 15 community practice sites, sometimes hundreds of miles apart. Everyone is on a common EHR and Hodgson has developed an imaging consulting service to avoid precertification delays. “When someone orders a test to rule out ischemia, it comes through the EHR and prompts the physician to answer a few questions. The request gets routed to the noninvasive imaging pool where the right test is determined and the precertification is acquired.”

In-house or outsource
Hodgson is in the beginning stages of setting up a core 3D lab. A central server (GE Healthcare) will house the data from the four scanners. What’s the driving force? Economies of scale in terms of reading (one reader) and post-processing (one tech), he says.

Another business model is outsourcing the 3D post-processing, which is what Borgess Medical Center adopted when it experienced a dramatic increase in its CT volume. “It got to the point where our technologists were treating the 3D modeling as a deferrable task, something that could be done later,” says Tom Mushett, radiology director at the 424-bed facility in Kalamazoo, Mich.

Mushett and colleagues considered the option of creating an in-house 3D lab, but cost and space were an issue. While the department uses advanced visualization software from Vital Images, it would have needed a thin-client server for an in-house 3D lab for cardiologists to manipulate and rotate images at other workstations. In addition, Mushett’s analysis indicated he would need to hire at least one more technologist. “Like everyone, we’re under budget constraints. Rather than make the investment in technology and people, we decided to look for a different solution,” he says. The department opted to outsource the 3D post-processing to 3DR Laboratories in Louisville, Ky., about a year ago and has further increased its CCTA volume without additional employees.

Peripheral CTA
Robert S. Schwartz, MD, medical director, Minnesota Cardiovascular Research Institute at the Minneapolis Heart Institute, teaches 3D post-processing. He and others are currently developing course material for peripheral CTA. “Peripheral is important because it is well reimbursed and it is a very important technique for understanding atherosclerotic disease as well as planning strategies,” he says.

Schwarz is working with specialists to understand their needs. At some point, they will put a 3D workstation in the operating room so that diagnostic-quality 3D images can be readily available for any surgical questions or issues that arise.

No matter which vascular territory cardiologists are working in, they need to understand the fundamental advantages and limitations of 3D CCTA technology. Artifacts can look like lesions. Manipulating the data at a particular angle can seemingly increase calcium. Technologists—even experienced ones—will have questions. “Most important, cardiologists need to have skills to know how to get that “better view of the lesion in the LAD or get around that calcium or process the image in a way that allows them to answer a difficult question that has come up because of some artifact or some difficult physiologic or anatomic situation,” Schwarz says.

“We’ve moved beyond standard orthogonal, coronal, axial and sagittal imaging,” DeFrance says. “Now we have a thousand different angles and that changes everything. Good training, excellent techs and a sophisticated workstation are the backbone of this new era.”