Supplementary Material: (08.09.2017)
Aortic graft anastomosis in less-invasive left ventricular assist device implantation
Seraina Anne Dual MSc Eng *1, Alissa Muller BSc Eng *1, Stefan Boës MSc Eng1, Oliver Brinkmann BSc1Eng, Séline Steffanoni MSc Eng1, Volkmar Falk MD Prof2, Mirko Meboldt Prof Eng1, Marianne Schmid Daners PhD Eng1, Simon Sündermann MD2
1 Product Development Group Zurich, Department of Mechanical and Process Engineering, ETH Zurich, Zurich, Switzerland;
2 Department of Cardiothoracic and Vascular Surgery, German Heart Center Berlin, Berlin, Germany
* Both authors contributed equally to this study and are considered co-first authors.
IN VITRO TEST BENCH
A series of anastomoses were constructed ex vivo on porcine aortas to determine the optimal design parameters for the anastomotic stent and the delivery device as well as to estimate the anchoring force of the anastomosis. Figure S1shows the test setup. Freshly excised aortas from the slaughter house were used for all tests. These were secured by cable connectors on two aluminum pipes of the same diameter. On each aorta, the anastomosis was constructed according to the anastomotic procedure. During the testing of different membrane materials and the sealing performance of the anastomotic stent, the tubes were filled with water and pressurized with a maximum aorticpressure of 150 mmHg.
Various membrane materials for covering the stent were tested on the test bench. As possible materials, we tried a biocompatible polytetrafluorethylene (PTFE) membrane, a multi-layer polyester membrane that is used in mountaineering, and a biocompatible Sefar filter membrane. With the purpose of sealing the anastomosis and securely connecting the cuff with the stent, we sewed the membrane to the stent. The Sefar membrane did not sufficiently seal the anastomosis. If using thePTFE membrane, the sewing tore the membrane at the needle holes, which led to a leakage of the anastomosis. The polyester membrane was used because it was more resistant to tearing.
The stent mesh design was verified by measuring the diameter of the stent, after implantation. The anastomotic stent was able to resist the elastic recoil of the aortic wall.The sealing of the delivery device was sufficient on the ex-vivo test bench for all steps up to the retrieval of the device. The final anastomotic stent was sufficiently leak-tight with water under maximum aortic pressure.
The trade-off between the overlap zone of the aorta and the cuff combined with a low blood-exposed non-intimal surface (BENIS) was assessed on the test bench. The drop-shaped cuff consists of a Nitinol wire of a defined length and the connection angle of the two wire endings. The choice of connection angle is restricted by the mechanical stability of the connection, while the length can change the overlap area. The length was chosen such that the stent was anchored sufficiently, which was verified in a pulling test.
In the dry test setup, pulling force was applied in the direction of the graft until the anastomoses broke. Subsequently, a spring balance was hooked to the open end of a graft. Force was applied to the spring balance at a 45° angle relative to the aorta until the anastomosis detached. The maximum pulling force was recorded. The experiment was repeated four times with the same anastomotic stent loaded on a new delivery system and deployed at different aortic sections. The maximum pulling force applied to break the anastomosis was 10 ± 4 N (mean ± standard deviation).
Figure S1: Experimental setup for pulling force measurement. A porcine aorta was clamped between two tubes. After constructing the anastomosis, sutures connected to the stent allowed to connect a spring balance to be connected for measuring the pulling force.