Intraoperative position system guided with augmented reality improves the learning curve of endovascular navigation in endovascular-naïve operators.

Objective: This study aimed to compare the completion of gate cannulation task performed by participants of varying experience using fluoroscopy, the Intraoperative Positioning System (IOPS)-a United States Food and Drug Administration-cleared endovascular navigation system that has been developed to reduce dependence on fluoroscopy-or an investigational augmented reality electromagnetic navigation technology based on IOPS.

Methods: The task consisted in the cannulation of the gate of a GORE Excluder AAA endoprosthesis bifurcated aortic stent graft (W.L. GORE & Associates) deployed into a three-dimensional printed abdominal aortic aneurysm model connected to a 7000 MDX flow pump (Sarns Inc/3M) reproducing physiological conditions. The procedure was performed in a hybrid operating room (GE Allia IGS 7). Each participant performed the cannulation task with fluoroscopy, standard IOPS guidance with flat screen display (IOPS-FS), and the investigational IOPS with augmented reality headset (IOPS-AR), in a randomly assigned order. All participants used the same sensorized guidewire and steerable 6Fr catheter during their three cannulation tasks. A total of 26 participants were classified in three groups of experience: Group 1 (endovascular naïve; n = 13), Group 2 (surgeon in training; n = 12) and Group 3 (one expert surgeon). Primary endpoints included cannulation time and technical success, which was defined as the advancement of the catheter over the guidewire within the main body of the aortic stent graft within a maximum 15-minute cutoff time for each trial.

Results: In group 1, the mean cannulation time was shorter using IOPS-AR vs fluoroscopy (4.3 ± 4.4 vs 7.1 ± 4.9 minutes; P = .04), but not statistically different when comparing IOPS-FS and fluoroscopy (6.3 ± 4.5 vs 7.1 ± 4.9 minutes; P = .63). In group 1, technical success was 77% with fluoroscopy and 92% with both IOPS-FS and IOPS-AR (P = .59). In group 2, although there was no significant difference between cannulation time among the three different endovascular approaches, there was a trend towards shorter cannulation times with IOPS-FS or IOPS-AR as compared with fluoroscopy (mean time of 2.5 ± 0.9, 4.4 ± 4.0, and 5.2 ± 4.5 minutes, respectively). In group 2, technical success was 92% with fluoroscopy and 100% with both IOPS-FS and IOPS-AR (P > .99). The expert vascular surgeon repeated the cannulation task four times for each endovascular approach, with 100% technical success and no difference in mean cannulation time between the imaging modalities (P = .89).

Conclusions: Augmented reality allows for reducing the gate cannulation time as compared with fluoroscopy in participants with no previous exposure to any endovascular procedure. This suggests that augmented reality can be beneficial for individuals early in their career and can mitigate the learning curve. As individuals become experts, their ability to adapt to different endovascular modalities increases, eliminating the learning curve altogether.