Southern Association For Vascular Surgery

Back to 2019 Abstracts


Predicting Aortic Endograft Shape and Spatial Position Pre-operatively Using 3D and Virtual Reality Technologies
Davis Moon, Caleb Stewart, John McGee, WC Sternbergh, III, Richard Milani, Korak Sarkar, Hernan A Bazan
Ochsner Clinic Foundation, New Orleans, LA

Background
We aimed to determine whether three-dimensional (3D) and virtual reality (VR) technologies could be applied to predict aortic endograft positioning utilizing an ex-vivo model.
Methods
A patient with a 5.6cm infrarenal abdominal aortic aneurysm (AAA) with suitable anatomy for endovascular aortic aneurysm repair (EVAR) underwent a successful AAA exclusion using the Medtronic Endurant II graft system (Medtronic, Minneapolis, MN). The aortic neck measured 21mm diameter and was nearly 50mm long. A 25 x 16 x 166mm main body and 16 x 16 x 124mm left iliac limb were utilized for successful exclusion of the aortic aneurysm. A digital 3D model of the luminal surface of the patient’s vessels was created using a pre-intervention computed tomography angiogram (CTA). From this digital model a semi-transparent cast was designed and created using stereolithography additive manufacturing techniques (Formlabs 2, Cambridge, MA). The cast was printed in two halves allowing for access to the lumen along its whole length. An identical aortic main body endograft and iliac limb were then deployed within the cast below the lowest renal artery. This was then fixed in its shape using a synthetic injectable polymer. The fixed endograft was removed from the cast and scanned as a virtual model using a 3D scanner (Einscan SE 3D scanner, Shinning 3D, Hangzhu, China, Figure 1). Once scanned, the model was exported to a 3D design software (Blender, Blender Foundation, Amsterdam, Netherlands, Figure 2).
Results
The virtual 3D model of the post-intervention CTA was created for comparison with the ex-vivo model. Both the ex-vivo and postoperative virtual models were co-registered using positional data and compared based on orientation (Figure 2). The ex-vivo fixed endograft model was superimposed with the post-operative CTA model. The ex-vivo pre-operative model closely mimicked that of the post-operative CTA model. Only minimal malalignment in the left iliac limb was noted.
Conclusion
We demonstrate feasibility in predicting how an aortic endograft will lay and be positioned pre-operatively utilizing a pre-intervention CTA and 3D printing and VR technologies. There was close overlap between the pre-operative model generated using 3D and VR and the post-EVAR CTA model. Ongoing studies are evaluating quantitative analysis of this model to verify congruency between the pre-operative model and actual stent graft position. Work such as this one will elucidate how 3D and VR technologies may play a role in vascular and endovascular surgery planning, particularly precision medicine based on patient-specific anatomy.


Back to 2019 Abstracts