Virtual Reality for Preprocedure Planning of Covered Stent Correction of Superior Sinus Venosus Atrial Septal Defects.

Abstract

Background: Covered stent correction (CSC) of a superior sinus venosus atrial septal defect is an alternative to surgery in selected patients, but anatomic variation means that assessment for CSC requires a 3-dimensional anatomic understanding. Heart VR is a virtual reality (VR) system that rapidly displays and renders multimodality imaging without prior image segmentation. The aim of this study was to evaluate the performance of the Heart VR system to assess patient suitability for CSC.

Methods: In a blinded fashion, 2 interventionalists reviewed preprocedural computed tomography scans using Heart VR to assess the feasibility of CSC, including the potential need for pulmonary vein protection. The total review time using VR was recorded.

Results: Using conventional imaging, 15 patients were deemed suitable for CSC, but at catheterization, 3 cases were unsuitable. Using VR, when both interventionalists agreed that a case was suitable for CSC (n=12), all proved technically feasible. In the 3 cases that were unsuitable for CSC, the interventionalists using VR were either uncertain (n=1) or did not agree on suitability (n=2). The strategy for pulmonary vein protection was correctly identified by interventionalist 1 and 2 in 9/12 and 8/12 cases, respectively. In cases where pulmonary vein protection was required intraprocedurally (n=5), this was correctly identified using Heart VR. Using VR, in 3 cases it was determined that pulmonary vein protection would be required, but this was not the case on balloon interrogation. VR data loading and review times were 82 seconds and 7 minutes, respectively. Verbal feedback indicated that Heart VR assisted in the assessment of case suitability.

Conclusions: Heart VR is a rapid and effective tool for predicting suitability for CSC in patients with a superior sinus venosus atrial septal defect and could be a feasible alternative to segmented virtual or physical 3-dimensional models.