The effect of total arterial grafting on medium-term outcomes following coronary artery bypass grafting.

Abstract

BACKGROUND: While it is believed that total arterial grafting (TAG) for coronary artery bypass grafting (CABG) confers improved long-term outcomes when compared to conventional grafting with left internal mammary artery and saphenous vein grafts (LIMA+SVG), to date, this has not become the standard of care. In this study, we assessed the impact of TAG on medium-term outcomes after CABG. METHODS: Peri-operative data was prospectively collected on consecutive first-time, isolated CABG patients between 1995 and 2005. Patients were divided into two groups based on grafting strategy: TAG (all arterial grafts no saphenous veins) or LIMA+SVG. Patients who had an emergent status or underwent fewer than two distal bypasses were excluded. Medium term univariate and risk-adjusted comparisons between TAG and LIMA+SVG cases were performed. RESULTS: A total of 4696 CABG patients were included with 1019 patients undergoing TAG (22%). Unadjusted in-hospital mortality was 1.5% for TAG patients compared to 2.0% for LIMA+SVG (p = 0.31). The mean follow-up was 4.8 +/- 2.0 years for TAG patients compared to 6.1 +/- 3.0 years for LIMA+SVG patients (p < 0.0001). At follow-up total mortality (8% vs 19%; p < 0.0001), and the incidence of readmission to hospital for cardiac reasons (29% vs 38%; p < 0.0001) were significantly lower in TAG compared to LIMA+SVG patients. However, after adjusting for clinical covariates, TAG did not emerge as a significant independent predictor of long-term mortality (HR 0.92; CI 0.71-1.18), readmission to hospital (HR 1.02; CI 0.89-1.18) or the composite outcome of mortality and readmission (HR 1.00; CI 0.88-1.15). Risk adjusted survival was better than 88% in both TAG and LIMA-SVG patients at 5 years follow-up. CONCLUSION: Patients undergoing TAG appear to experience lower rates of medium-term all-cause mortality and readmission to hospital for any cardiac cause when compared to patients undergoing LIMA+SVG. However, after adjusting for clinical variables, this difference no longer persists suggesting that at median follow-up there are no mortality or morbidity benefit based on the choice of conduit.