Side-to-side anastomoses have been described to connect the left internal mammary artery (LIMA) to target coronary vessels for the purpose of reducing surgical damage on LIMA . Song et al. also introduced a technique of side-to-side anastomosis with just ten proportional stitches by hand for artery grafts . In vein sequential grafting, side-to-side anastomoses are usually performed for the middle anastomoses . In this study, to address the issue that oversize graft-host diameter ratio at end-to-side anastomoses might induce adverse hemodynamic condition and reduce graft patency; we applied a side-to-side anastomosis technique to connect the distal end of the GSV to small target arteries in sequential CABG. We retrospectively compared the intraoperative graft flow parameters of patients who had distal end side-to-side anastomotic reconstruction and found that the reconstruction improved graft flow remarkably as compared to the end-to-side anastomoses. The comparison in this study was performed on data collected from the same patients. End-to-side anastomoses were initially performed on the 14 patients. Due to poor graft flow at the distal end, we revised the problematic distal end anastomoses into side-to-side anastomoses, resulting in significant improvement of intraoperative graft flow. We believe that the reason for the initial poor graft flow are not technical errors, instead are associated with unfavorable hemodynamic characteristics at the end-to-side anastomotic site.
Anastomotic geometry has been shown to have substantial effects on the hemodynamics near anastomotic areas. It is believed that small areas of low wall shear stress (WSS) and very few areas of high WSS gradient in the graft indicate a beneficial hemodynamic pattern for optimal graft flow [7–9]. In an end-to-side anastomosis, the vortex flow at the anastomotic heel produces large area of low WSS and high WSS gradient in the graft . This type of hemodynamic pattern impedes graft flow. Multiple reports have suggested that the hemodynamic pattern associated with side-to-side anastomoses might be more favorable for graft patency compared to that of end-to-side anastomoses. Frauenfelder et al. used computational fluid dynamics to simulate pulsatile blood flow in venous grafts based on patient angiography datasets and found that flow stagnation zone exists in end-to-side anastomoses but is absent in side-to-side anastomoses . Sankaranarayanan et al. analyzed the hemodynamic patterns of different anastomotic configuration by computational fluidic dynamics and found that side-to-side anastomoses have smoother blood flow and smaller spatial gradients of WSS than end-to-side anastomosis . Thus, the improved graft flow dynamics we observed likely results from a smoother blood flow pattern at the reconstructed side-to-side anastomoses as compared to a vortex flow pattern at the initial end-to-side anastomoses.
In our study, the diameter ratio between the GSV and the small target coronary arteries was approximately 4:1 in the 30 patients. Graft-host diameter ratio can influence graft hemodynamics at end-to-side anastomoses substantially [2–4]. In a computational simulation model of CABG, Qiao et al. reported that large graft-host diameter ratio (1.46) produces better hemodynamics than equal or small (0.8) ratio when the diameter of a target vessel is assumed to be 2.5 mm . We believe that the beneficial effects of large diameter ratio described in the report are based on the assumption that the target vessels have normal size, and the beneficial effects are limited within certain range of the diameter ratio. As the diameter ratio increases, other factors such as graft transfiguration might induce adverse effects on hemodynamics significantly, leading to an unfavorable flow patter for graft patency. In our study, the graft-host diameter ratio was approximately 4:1. Our data show that end-to-side anastomoses with such oversize diameter ratio resulted in unsatisfactory intraoperative graft flow parameters that required prompt revision, and the side-to-side anastomosis reconstruction markedly improved graft flow and PI.
The patient 3-month follow-up examination showed improved cardiac function, suggesting that patient outcomes of this technique are satisfactory. Kieser et al. conducted a study of long-term follow-up of sequential CABG and found that the patency rate of side-to-side anastomoses is much higher than that of end-to-side anastomoses . To assess the long-term patient outcomes of our distal end side-to-side anastomosis technique, we need to include a control group of patients receiving end-to-side anastomoses at the distal end of sequential grafts. However, in the present study, we had to revise the problematic distal end-to-side anastomoses, which produced unsatisfactory graft flow parameters, into the distal side-to-side anastomoses. Therefore, we were not able to include such control group in this study. The comparison based on the data collected from the same patients suggests that side-to-side anastomoses at the distal end of sequential vein grafts might produce a hemodynamic pattern favorable for long-term graft patency, indicating that distal end side-to-side anastomoses might be a superior anastomotic configuration than end-to-side anastomoses in general. We are now using animal models to compare the long-term graft patency of end-to-side versus side-to-side anastomosis at the distal end of grafts.
The CK-MB levels in our patients significantly dropped 48 hours after the operation, indicating the absence of serious complication associated with the surgical procedure. In this study, we were unable to directly compare CK-MB levels in patients with end-to-side anastomosis versus side-to-side anastomosis due to the absence of a control group with end-to-side anastomosis. However, according to our experience, the average CK-MB levels on arrival to the surgical ICU and 48 hours post-operation in patients undergoing routine CABG in our department are not significantly different from the CK-MB levels described in this report, suggesting that the distal end side-to-side anastomosis procedure do not appear to cause more postoperative complications compared to standard CABG procedure.