Atlanta, GA - Interventional cardiologists learning to perform carotid angiography through serial training on a virtual-reality simulator (VRS) show clear improvement in important indicators of performance over the course of testing, researchers say [1]. The study, published in the May 2, 2006 issue of the Journal of the American College of Cardiology, should help reaffirm the role of VRS systems in what the study authors refer to as a "high-stakes" field of carotid interventional training.

"We talk a lot about doctors having learning curves with new procedures," senior author Dr Christopher Cates (Emory University, Atlanta, GA) explained to heartwire. "This is finally the first study to actually measure doctors' learning curves while learning a new procedure. This opens a whole new world of being able to quantify and measure procedural performance on a simulator, prior to working on patients."

This opens a whole new world of being able to quantify and measure procedural performance on a simulator, prior to working on patients.

Previous studies have shown that residents trained on VRS made significantly fewer objectively assessed intraoperative errors compared with those who underwent standard training and that much of the steep learning curve of endovascular procedures can be achieved on simulators, rather than vulnerable patients. In the current study, Dr Amar D Patel (Emory University) and colleagues sought to establish whether VRS training using the VIST simulator (Mentice AB, Gothenburg, Sweden) could be used as a measuring tool to assess improvement in performance and reduction in errors over successive VRS sessions.

In the study, 20 experienced interventional cardiologists who had not previously performed carotid stenting completed an instructional course on carotid angiography and then performed five serial simulated carotid angiograms. During the VRS sessions, procedure time, fluoroscopy time, contrast volume, number of cine-loop recordings, and composite catheter handling errors were recorded by the simulator.

For all study participants, significant improvements were seen for procedure time, fluoroscopy time, contrast volume, and composite catheter handling errors between the first and last simulations. No significant differences were seen in the number of cine-loop recordings across the five sessions. Internal consistency of the simulator was high, except for fluoroscopy time, while test-retest reliability was high for composite catheter-handling errors.

Cates believes the study should help boost the role of simulator training for delicate interventional procedures and especially endovascular interventions. "This should, hopefully, help to gain acceptance [of VRS] within the medical community, although there are still those who resist objective measurement of procedural skill, saying it's not important," Cates observed.

Dr David Dawson (University of California, Davis), an early pioneer of virtual-reality medical simulators, who was not involved in the study, agrees. "I think that there is widespread enthusiasm for the use of simulators for the early phases of procedural training, but most doctors remain wary of using simulators for evaluation of skills or assessment of procedural competence," he commented to heartwire. "More validation work is required in these areas."

According to Dawson, the study by Patel et al helps clarify the role of simulators in training for endovascular procedures. "Limited information is available to validate many widely held assumptions, including the very fundamental question of whether or not people can acquire new skills by using the current versions of commercially available simulators. This study . . . is one step forward, showing that interventional cardiologists improve on their simulator performance with practice."

As more such studies are published, Dawson added, "there will be an increased awareness and acceptance of the role of simulation in training cardiovascular specialists." Subsequent research, Dawson says, will be to determine whether simulation-based training also improves subsequent performance in actual clinical practice.

For Patel, Cates, and colleagues, the next step will be to put other VR-simulation systems through the same kinds of rigorous validation methods "to make sure they can measure true technical skill and be good enough for 'high-stakes' assessment," Cates told heartwire. "Otherwise it's just playing on a video game: if the simulator doesn't have metrics, [it can't] measure the true technical skill of the operator."