RESUMEN
OBJECTIVES: Since 2000, we anastomosed the saphenous vein graft to the right coronary artery system using the stump of the right internal thoracic artery as inflow. The long-term results of patients where the right coronary artery was grafted with the right internal thoracic artery or the ascending aorta as saphenous vein inflow has not been reported. METHODS: From 2000 to 2018, 699 consecutive patients had right internal thoracic artery elongated with saphenous vein (I-graft group, n = 358, 51.2%) or saphenous vein from the aorta (Ao-graft group, n = 341, 48.8%) on right coronary artery system. Inclusion criteria were age ≤75 years, bilateral internal thoracic arteries as a Y graft on the left system (three-vessel disease, n = 603, 86.3%) or as a left internal thoracic artery on left anterior descending and right internal thoracic artery elongated with saphenous vein on the right coronary artery system (two-vessel disease, n = 96, 13.7%), only 1 saphenous vein per patient. Propensity-matching identified 272 patients per group. One-hundred and twenty-two patients underwent coronary computed tomographic angiography to asses grafts patency after a median follow-up of 88 (65-93) months. RESULTS: In the paired samples, there was no difference in the early outcome. Ten-year survival and freedom from death, non-fatal acute myocardial infarction and repeat revascularization were higher in I-graft group: 90.6 [standard error (SE): 2.0] vs 78.2 (SE: 5.3), P = 0.0266, and 85.2 (SE: 2.4) vs 69.9 (SE: 5.3), P = 0.0179. Saphenous vein graft, at a long-time follow-up, showed a higher patency rate (81.6% (SE: 7.0) vs 50.7% (SE: 7.9), P < 0.0001) and a smaller internal lumen diameter (2.7, standard deviation: 0.4 vs 3.4, standard deviation: 0.6 mm, P < 0.0001) when right internal thoracic artery was the inflow. CONCLUSIONS: Grafting the right coronary artery with saphenous vein may entail higher patency rate and better outcome when the inflow is the right internal thoracic artery than when is the ascending aorta. Prospective randomized data are needed to test this hypothesis.
