Geometry assessment of coronary artery anastomoses with construction errors by epicardial ultrasound.

OBJECTIVE There is concern about the quality of the distal anastomosis in off-pump coronary artery bypass grafting. We investigated the impact of specific construction errors on anastomotic geometry using epicardial ultrasound. METHODS Twelve ex vivo pressure perfused porcine and five isolated post-mortem human hearts were used to construct 35 internal mammary artery to coronary artery anastomoses, either without (n = 7) or with a standardized construction error (oversutured toe, oversutured heel, cross-over or purse string; each error, n = 7). The anastomotic geometry was visualized and measured by a 13 MHz ultrasound mini-transducer. Impression cast material was used to validate anastomotic geometry. RESULTS All 28 errors were visualized properly. Two unintended construction abnormalities were observed. In the porcine heart, the ratio of anastomotic orifice area and outflow corner area was 1.3+/-0.2 (mean+/-standard deviation) in the control group and reduced in the error groups: oversutured toe, 0.6+/-0.2 (P = 0.001 oversutured heel, 0.9+/-0.2 (P = 0.037); cross-over, 0.4+/-0.2 (P < 0.001); purse string, 0.3+/-0.2 (P < 0.001). None of the errors reduced the area of the inflow or outflow corner itself compared to the recipient coronary artery. In the human heart, all construction errors as well as wall plaque were visualized properly. In all anastomoses, ultrasound geometry corresponded to cast geometry. CONCLUSIONS Ex vivo, epicardial 13 MHz ultrasound enabled accurate visualization and assessment of four different construction errors in the coronary anastomosis. All errors reduced the area of the anastomotic orifice, but not the inflow or outflow corner.