Report from NASCI: Automated CTA-based tool could streamline coronary cath interventions

Dutch researchers have come up with a new way to plan percutaneous coronary interventions that relies on existing coronary CT angiography exams. It has the potential to replace what cardiac imagers deem an outdated, subjective standard.

Optimal diagnosis and endovascular treatment of coronary artery disease require defining proper viewing angles for the vessels involved before a catheterization or stenting procedure takes place. To a large degree, interventionalists still rely on their surgical experience to estimate subjectively the optimal angle on the fly during procedures using 2D fluoroscopy.

This subjective approach is time-consuming and can yield unreliable information while exposing patients to unnecessary radiation, said coauthor Johan H.C. (Hans) Reiber, Ph.D., director of the department of radiology's image processing division at the Leiden University Medical Center.

To overcome these limitations, Reiber's team developed XaViewCT, a user-friendly tool that determines optimal viewing angles for catheterization using available diagnostic coronary CTA data. Reiber released preliminary findings of the team's experience with the technique at the 2007 North American Society for Cardiac Imaging meeting in Washington, DC.

Reiber and colleagues automatically mapped out 83 coronary bifurcation areas from 15 coronary CTA data sets using XaViewCT. The method provided the viewing angles required for angiographic procedures by reducing vessel overlap and foreshortening, a visual effect that makes vessels appear shorter than they are. The investigators compared XaViewCT results with views provided by two experts and scored images on a scale of zero to 10, with six and above indicating clinically usable views.

XaViewCT produced clinically usable views for coronary bifurcations in 95% of the cases (mean score of 8.4) as compared with 98% for the experts (mean score of 8.7). But the automated technique produced clinically relevant views with less foreshortening than did the experts (XaViewCT = 16%; expert A = 47%; expert B = 38%).

Researchers found no significant differences between XaViewCT and the manually performed approach for selection of viewing angles. However, they found evidence of up to 30% grade difference between the automated CTA-based and the subjective 2D angiography approaches. XaViewCT could, for instance, determine more accurately than 2D angiography the length of a stent needed for a procedure, Reiber said.

"With the increase of available CTA data, (XaViewCT) is a valuable method for the planning of angiographic interventions," he said.

For more information from the Diagnostic Imaging archives:

Delayed-enhancement coronary artery imaging tracks vascular inflammation changes in stable and unstable CAD

Off-label use of drug-eluting stents leads to higher rate of complications

SCAI announces percutaneous coronary intervention guidelines for hospitals not equipped for cardiac surgery

Research determines timing, tools for optimal coronary interventions