Preliminary experimental study of a mechanical connector allowing vascular anastomosis.

BACKGROUND

Aortic clamping time needed to carry out anastomosis during aortic surgery is one of the major predictive factors of morbidity/mortality. Because of the difficulties inherent in this technique, the duration of aortic clamping can be increased significantly with laparoscopy, which may explain why this kind of surgery is not often undertaken. We designed an original connector that will help us perform automated end-to-end vascular anastomoses. This anastomotic device was tested on a cadaver, test benches, and finally on pigs.

METHODS

The connector consists of a bare-metal stent with spikes covering its outer surface. Once the prosthesis has been introduced into the arterial lumen over a 15-mm length, the connector is placed into the prosthesis; then, upon inflating a balloon and employing stent spikes, the prosthesis can be anchored in the artery. A feasibility study was carried out, first on cadaveric arteries and then on bench tests. In vivo, using this connector, the infrarenal aorta of 7 pigs was replaced with a 6-8-mm-diameter prosthesis. One to 42 days after implantation, angiography was performed before explanting the prosthesis. Evaluation criteria included anastomosis performance time, leakproofing capacity, tensile strength, patency, and histologic changes induced by the device.

RESULTS

On the cadavers, manual traction tests showed that the anastomoses were satisfactory. Median force needed to rupture an automated anastomosis was 15.85 N. No leakage was seen under a systolic/diastolic pressure of up to 250/180 mm Hg. All pigs benefited from automated anastomosis. Once the designed device had been assessed, the median connection time was 2 minutes. In 4 cases, aortic thrombosis was found, 1 of which was septic. The implanting ancillary was responsible for 2 posterior aortic wall perforations, which extended the clamping time; therefore, this device should be modified. Anastomosis patency and tensile strength tests were satisfactory. Histologic results showed the connector incorporation, integration of the spikes in the arterial wall, and the absence of inflammation in the aortic wall.

CONCLUSIONS

These preliminary studies confirm that the connector is quick and reliable for performing in vivo arterioprosthetic anastomoses. Further studies are needed to improve the ancillary device, which will enable its use it in laparoscopic and conventional surgery.