Propensity-score analysis of early outcomes after bilateral versus single internal thoracic artery grafting

OBJECTIVE: The use of bilateral internal thoracic arteries (BITAs) during coronary artery bypass grafting (CABG) improves long-term and event-free survival compared with single internal thoracic artery (SITA) grafting. It is controversial whether BITA grafting alters in-hospital adverse events after CABG. METHODS: Isolated CABG cases using BITA or SITA at a single US academic center between January 1, 1997 and June 30, 2006 were retrospectively reviewed. A propensity score was used as a covariate to balance the treatment groups (BITA and SITA) with respect to 44 preoperative risk factors. A multivariable logistic regression model tested whether treatment type was significantly associated with in-hospital death, deep sternal wound infection (DSWI), or hospital length of stay (LOS). RESULTS: There were 599 BITA and 10,212 SITA cases performed. Overall for all BITA versus SITA cases, adjusted mortality (0.8% vs. 1.7%, P = 0.85) was not different between the groups. However, adjusted incidence of DSWI (2.0% vs. 1.2%, P = 0.036) and LOS (6.7 vs. 6.1, P = 0.025) were significantly higher in BITA patients. Subsets analyses of obese patients and diabetic patients revealed no statistical differences for any of the outcomes between BITA and SITA. CONCLUSIONS: The long-term benefits of BITA grafting do not come at the cost of increased adjusted risk of in-hospital death. BITA grafting was associated with an increased risk of DSWI and a longer adjusted LOS. Neither obesity, nor diabetes significantly increased the risk of poor outcomes after BITA.